Les Banki's project

When I was looking for solutions to get rid of the waste spark for my Electropolished WFC Replication Project, I found these two videos mentioning circuitry developed by Les Banki:

http://www.youtube.com/watch?v=UfGRPJWOGvo

http://www.youtube.com/watch?v=JomHUdpQbvw

It turns out that Les developed a lot of circuits for running engines on hydroxy gas, most of which are posted on the overunity.com forum. I collected all Les' attachments and Les was also kind enough to send me the .PCB files, which are made with Protel PCB for Windows, which can be found on the internet with a little bit of searching.

All Les' project files can be found here. There is also a zip file, which contains all the files, as well as a gallery with all the images, which can also be accessed as individual files.

What follows are the contents of Les' word files, slightly edited and enhanced with links and images.

-- Arend --

Update March 2nd, 2014

I received new files from Les, along with the following e-mail:

-- begin email --

This is a MAJOR update on my WFGP. (Water Fueled Generator Project)

As you may know, I have designed a LARGE number of electronic circuits (as well as mechanical designs) over the years! (spanning over 3 computers!) MANY of those designs relate to "water fuel" which were published on several forums. They were meant to be 'building blocks', either used independently or in certain combinations, for experiments made by those who were/are willing to DO things, instead of just talking!

Note that only a few of those designs are used for the current project.

It took several days to 'clean up' (edit) all electronic circuits and documentations which are being used in this current WFGP.

Some had only minor changes while others had major ones.

Further, there are also additional circuits I designed recently which have NOT been published before. Such as:

1. Control panel circuit diagram, pcb layout and control box description

2. Infrared transmitter & receiver circuits used by the two stage electrolyzer water refilling system

3. Hall switch circuit with a buffer stage which eliminates RF interference pick-up!

Note that ALL the electronics sections for this project are now COMPLETE. (tested and working)

In order to avoid future confusion, I have created 4 Folders which only contain files relating to this current WFGP:

'WFGPdoc' contains all text files (circuit descriptions, etc.) -- is complete

'WFGPsch' has all circuit diagrams -- is complete

'WFGPpcb' has all pcb files in the original PFW format -- is complete

'WFGPphotos' there are some photos of the ECU now but will have many more of the entire set-up, as the work progresses.

All related text files have been edited/updated. The edited versions are all dated 2014 and are the only ones which are to be used for this current project.

So, if anyone intends to duplicate this project (or part of it), should only use files from these 4 Folders!

[...]

Note also that I have designed/made a semi-automatic "groove cutting machine" (similar to a CNC router) which is now operational. (It will be pulled apart later in order to paint it since MDF is sensitive to moisture/water!) Further, I also designed/made an automatic vacuum work holding system for this machine. There are NO clamps, NO bolts, NO holes in the work table! Work piece is fitted & released in seconds!

There will be photos of this machine as well.....

-- end email --

All these files can be found here:

http://www.tuks.nl/WFCProject/LB_WFGP/

Overview & explanation of my complete Engine Control Unit design.

In order to AVOID future misunderstandings, I decided to write this overview/explanation but first, I wish to make some VERY IMPORTANT statements and I ask ALL readers: Please make sure you UNDERSTAND them!

The Engine Control Unit (ECU) design I am presenting to the public, IN ITSELF, is NO GUARANTEE to produce enough Hydroxy to run any engine!

The VOLUME and QUALITY of Hydroxy will depend almost ENTIRELY on what kind of electrolysis system is used. (and perhaps some additional factors not covered here) [some examples: ‘brute force’ low voltage, series cells, high voltage series cells, (with or without ‘resonance’ drive), Hasebe replica, Stan Meyer type ‘resonance’ cells, etc., etc.]

My INTENTION (with the numerous building blocks of the ECU) is to provide anyone who is willing to ‘get their hands dirty’ with the necessary CONTROL ELECTRONICS to achieve their goal.

In essence, what I am saying here is:

IF we are to use the old, rather crude and VERY inefficient (around 26%) Internal Combustion Engine at all, we need to provide it with ignition sparks at the correct times, supply fuel (in this case, HydrOxy) at the correct times and in correct volumes.

Further, the fuel pressure needs to be held steady (pressure regulation) and the power required to create the HydrOxy also needs to be supplied AND controlled (limited).

The need for all this control is INDEPENDENT of the method used for generating the required volume of Hydroxy!

In other words, REGARDLESS of which method of HydrOxy generation is employed, the supply & controls described above are ESSENTIAL. However, you have probably noticed that I offer additional circuits as well, not absolutely necessary but desirable for a smooth working control system and power back up (for example: automatic battery charger circuit).

There is also a convenient control panel where all adjustment are made and pressure, current and voltage levels are SET and DISPLAYED.

There is no denying that this design aims at PRODUCTION! (not just for experiments)

Anyone who is not too familiar with general physics, electrolysis, ICE (Internal Combustion Engine) and electronics technology, could be forgiven for perhaps questioning the need for the number of circuits presented!

It is also likely to give an impression of unnecessary complexity and create confusion.

Note that I choose the name Engine Control Unit (ECU) deliberately as its functions are similar to that of the existing systems used by car manufacturers.

However, all unnecessary functions of the ‘standard’ ECU have been left out!

On the other hand, its functions are expanded to include the power supply AND control to create the FUEL itself, HydrOxy.

All circuit sections are mainly ANALOG, using common, cheap and readily available components. (NO ‘microprocessors’, NO complex software programming!)

In simple terms, (detailed info in the respective circuit descriptions) here is a list of the circuits I have developed/designed and their intended use:

Hall switch – tiny pcb, mounted on the engine. With a small permanent magnet attached to the exhaust valve’s ‘rocker arm’, it supplies pulses to the Ignition/Injection control module. These pulses indicate the piston’s position in the engine’s work cycle.

Capacitor Discharge Ignition (CDI) module – when connected to an ignition coil, it creates the required high voltage (20,000V+) to fire the spark plug.

Ignition & Injection control module – supplies the control pulses to the CDI module (WHEN to deliver the sparks) and the drive pulses to the injection solenoid.

Automatic RPM control – automatically brings engine speed from start-up to the correct RPM where the generator supplies approx. 240V with a frequency of 50Hz.

Feed-back control loop – to maintain a STEADY frequency (50Hz) and voltage (240V) output with varying loads.

Auto start – simple circuit which activates the remote control for 3 seconds to start the generator when the set gas pressure is reached.

Pressure regulator module – decides the desired pressure ‘scale’ (PSI, kPA or whatever), Sets and displays the pressure limit and continuously monitors and displays (on the control panel) the actual pressure.

Battery charger – automatic charger, used to maintain FULL charge AT ALL TIMES on a stand-by battery which will be necessary once mains power is no longer connected. (for re-start after maintenance stops)

Power supply (regulator) module – supplies +12V-1, +12V-2, -12V, +5V and -5V to the various modules and sensors.

Water level sensor & pump driver 1 – used to automatically detect the minimum water level in the electrolyzer unit and refill to the set maximum level when necessary.

Water level sensor (& pump driver) 2 – used to detect the minimum (Danger!) water level in the flash-back arrestor and SHUTS DOWN the electrolyzer power supply! Can also be used (with a second pump) for automatic re-fill of the flash-back arrestor.

Relay board – a universal AC/DC 30A relay with a 12V DC coil, transistor driver and indicator LED. Can be configured for either start-up/run or for general HIGH power switching and is used mainly with the timer & timer interphase circuit.

Timer & timer interphase module – while NOT essential, it is VERY ‘handy’, particularly for REPEATED experiments. It eliminates time measuring errors and a lot of ‘guess work’. Also eliminates large mechanical power switches! It can also be used to stop the engine/generator after a pre-set time (up to 24 hours!) ''' Test oscillator''' – it is powered up ONLY during set-up (when the engine is not turning there are NO pulses from the Hall switch) it provides the pulses needed for testing.

However, since this oscillator is NOT used during normal operation, if desired, it could be used to flash the LED which indicates power SHUT DOWN to the electrolyzer in the event the flash-back arrestor’s water level drops too LOW.

Control panel – See circuit description for the functions which can be SET and DISPLAYED.

Closing notes:

Once again, as indicated in this overview, not all circuits are being used at the same time. I have tried to cover everything I could think of which I thought would the necessary and/or desirable to control. This would give a choice of options, if you like.

I wanted ALL electronic controls in place and available when (or if) they were needed, just so I would NOT need to run “back to the drawing board” (computer) to do more designs in the middle of the physical work with engines!

Coming up with the concepts, developing/designing the circuits, drawing the circuit diagrams and pcb layouts, writing the circuit descriptions (and other writings), chasing part samples needed for the designs, etc., has taken over 5 years of hard work.

If I have missed or left out something, let me know! I will do my best to fill the gap!

Finally, judging by the number of posts on several Forums, the ignition and ”waste spark” issues have received a fair bit of attention lately. Quoted here is part of just one such post to help me illustrate my point:

“I just converted a 5 HP Briggs with a 4 to 1 gear reducer hooked to the crankshaft (3 gears, all 2 to 1 reduction) and added an old set of points from a 69 gmc that have the condenser built in (had a great time at the auto parts store buying these), welded a bolt to the last gear to trigger the points, I mounted the gears on a plate that ties into the 4 bolt pattern on the rear of the engine, slotted the holes in a circular pattern so that I could adjust the timing on the fly, added a coil and a coil resistor, hooked it up to a battery and fired up my CHEAP, SIMPLE!!! ignition system (the old timers had this all figured out already, why reinvent the wheel, keep it simple),….”

So what IS my point?

Well, just about every engine BRAND and MODEL is different. Some may be able to be modified like that above, some won’t. And so, here is the BIG question:

Which of the two options (below) do you prefer?

1. The above mechanical method: gears, rods, welding, etc., which (as described) is a power hungry Kettering system, drawing between 5 and 10A! (60 – 120W)

OR,

2. Fit a small magnet to the exhaust valve’s rocker arm, attach the tiny Hall switch pcb to the engine block and then turn a potentiometer on the control panel to set your desired ignition point, continuously variable +/- 45° from TDC, while the engine is running!

Oh, by the way, my CDI system draws only 0.5A. MAXIMUM power draw is 6W!!)

Les Banki

(Electronic Design Engineer)

Water Fuel & LBE Technologies

The Hall Switch

high res

PCB

Ignition system for small engines running on Hydroxy ONLY

It should be obvious that with Hydroxy as the ONLY fuel, the use of 2 stroke engines are ruled out since they require oil to be mixed with their fuel for lubrication.

Therefore, only 4 stroke engines will be considered in this brief.

First, some engine data.

The crank shaft on a 4 stroke engine turns twice (720º) for every ‘work’ cycle. Since most (if not all) small engine designs use a magnet on the fly wheel (which is mounted on the crankshaft) to generate the ignition sparks, 2 sparks are delivered for every work cycle. The second spark (which is delivered during the exhaust stroke) is NOT needed and so it is called “waste spark”. With hydrocarbon fuels it is harmless.

However, with Hydroxy ONLY, this “waste spark” MUST be eliminated.With hydrocarbon fuels, ignition usually takes place around 8º before TDC to allow some atomization of the fuel before the actual ‘explosion’, which occurs approximately 10º after TDC.

If Hydroxy is ignited at ANY point before the piston has reached TDC, the explosion takes place at that INSTANT. (There is NO delay or atomization here since it ‘burns’ about 1000 times faster than hydrocarbon fuels and it could be said that it is not ‘burning’ but exploding!) The force of the explosion instantly tries to push the piston DOWN when it is still trying to come to the top to complete its compression stroke!

That is most undesirable!

When the ignition is delayed (retarded) to the point where the explosion usually occurs with hydrocarbon fuels (around 10º after TDC) then the piston’s downward movement is reinforced and useful work is gained.

Now, consider what would happen if the waste spark was NOT eliminated. As stated above, the crankshaft revolves twice for every ‘work’ cycle. (The first revolution covers the intake and compression stroke and the second one the power and exhaust stroke.) Thus, the second spark (‘waste spark’) occurs just before (the same degree of advance as the wanted spark, about 8º) before TDC at the end of the exhaust stroke.

But when the ignition pulse is delayed to be after TDC, the waste spark will occur at the beginning of a new ‘cycle’, where the intake valve has just started to open.

So now, with a slightly open valve there is an open path to the fuel line (Hydroxy), and there comes a spark! Guess what happens… Guaranteed back fire!

And I can assure you that even the most minute opening will allow the ‘flame front’ to propagate back to the supply line. How do I know? Experience. Lots of it.

Further, let me tell you that I have personally not found ANY method of stopping back fires to propagate back to the electrolyzer, EXCEPT water. While most people call them bubblers, I prefer the name “flash-back arrestor”, since that is their true role. If you doubt the above statements about stopping flash backs traveling back to your electrolyser and DESTROY it, by all means, ignore the advice. Not only will you DESTROY your electrolyser but very likely injure or even KILL yourself and/or others!

An example of engine calculations:

I bought a new, 118cc, one cylinder, 4 stroke petrol engine for Hydroxy experiments. Its rated max. output is 4 horsepower (2960W) at 3600RPM. For the ease of calculations, lets round up the capacity to 120cc (0.12L) This is the maximum volume of air/fuel mixture it can suck in during its intake cycle.

As stated before, the engine’s ‘work’ cycle number is half the crankshaft revolution.

Thus, at 3600RPM, the number of fuel intakes is 1800/minute. 1800 x 0.12L = 216L/minute

However, as only 1% of QUALITY Hydroxy (mixed with 99% of air) is needed to obtain the same power as petrol, this 120cc engine should require only 2.16L/minute of Hydroxy to run at 3600RPM!! (Naturally, it would require less at lower speeds. It remains to be seen if it will require more under full load than this calculated volume.)

Now a few notes about the necessary ignition delay and how to achieve it.

In one article it was suggested that one could use a 555 IC to delay the ignition pulse. Yes, that could be done but it would only be correct at ONE speed.

The reason is obvious:

Ignition advance/delay is related to piston position, NOT time. It is expressed in ‘degrees’ but for hydrocarbon fuels it is varied slightly with engine speed. (due to its relatively slow burning) With Hydroxy, ignition will take place at the same ‘degree’, (same position of the piston) regardless of engine speed.

At this stage, a couple of things are clear already:

One: on my test engine (and I dare say on most, it not all, small engines) it is not possible (meaning: NOT practical) to eliminate the ‘waste sparks’.

Two, there is NO provision for ignition timing adjustments, neither mechanical, nor electronic.

In other words, the existing ignition systems used on small engines are USELESS for Hydroxy. We need a NEW electronic ignition system, complete with ADJUSTABLE delay.

So how can that be done?

Again, two revolutions of the crankshaft is 720º (two circles but one ‘work cycle’).

The camshaft, (controlling the valves) however, turns only ONCE, which is 360º.

In electronic terms, that is 100%.

We want to delay the ignition timing from where it is now, say, from 8º before TDC to 10º after TDC. That is a delay of 18º.

The equation is: 360 : 100 = 18 : X Re-arranging it: 360X = 1800, X = 5

In other words, 18º is 5% of 360º.

We need to delay our original ignition pulse by 5%, irrespective of frequency. (the ‘frequency’ here is the engine’s revolution)

The above example serves to illustrate the difference between the ‘old’ and the ‘new’ settings, assuming that the degree settings relate to the camshaft revolution, 360º.

However, as I understand it, the ignition advance/retard degrees are usually expressed in terms of crankshaft degrees (720° - two revolutions of the crankshaft)

In that case, the above percentage of 5% is halved. Then, 18º is 2.5% of 720º

Since we need a NEW ignition system, this ‘delay’ will no longer relate to the ‘old’ setting. A new signal is taken from a sensor (Hall switch) mounted on the engine, detecting the intake (or exhaust) valve’s position. Using the signal from this sensor, the ignition spark could be made to occur anywhere but we want it approx. 10º (or more) after TDC (adjustable within a few degrees)

Of course, our reference is still TDC.

When we express all that in electronic signal terms, the intake stroke (piston travels from TDC to BDC) is ¼ of the engine’s work cycle, which is 25% of our wave form. (90º of the work cycle and 180º of crankshaft rotation)

If we transform the delays from degrees to percentage, we get the following figures:

10º ATDC is a delay of ~1.39%

25º “ ~3.47%

So, if we want the adjustment range of 10º - 25º, the percentage difference is 2.08%.

[We can also calculate the elapsed time this translates to, for any given speed. For example: at 3600RPM, the ‘frequency’ is 30Hz. One period is 1/30 = 0.0333sec. Thus, a 1.39% delay means that the piston has traveled (from TDC) for 463.3µs to reach the position of 10º ATDC (relating to crankshaft revolution)]

One simple way to implement these delays is to use a PWM (Pulse Width Modulator) circuit, which is my preferred choice. (How this is done will be described in detail in a technical “circuit description”.)

It needs to be pointed out that the ignition system for Hydroxy ONLY (not just a booster) will be very different from ignition systems for hydrocarbon fuels.

It will be significantly simpler.

There will be NO need for high energy sparks, multiple sparks, etc.

Further, there will be NO such thing as UNBURNED fuel remaining in the cylinders!!

In short; when we get to the larger engines (cars), the first thing we have to do is to rip out the “computer” and install our own system, incorporating electronic injection as well. (Perhaps another option could be to completely re-program the ‘computer’, provided that one could obtain the original programming software from the manufacturer, which, I would say, is HIGHLY unlikely!)

I am in favor of electronic injection (but ONLY for Hydroxy) for two reasons:

1. I reason that if we allow Hydroxy to flow continuously, some of it may disappear during the other ¾ of the engine’s work cycle. (the intake stroke is only ¼ cycle)

2. If Hydroxy is ALWAYS present in the intake manifold, we may risk a damaging back fire.

I am aiming at a mainly analog design, using parts available everywhere and are dirt cheap!

Should a fault occur, it will be quick, easy and cheap to repair.

Les Banki

(Electronic Design Engineer)

Water Fuel & LBE Technologies

Important notes about the Ignition & Injection control circuit and Test Oscillator

While this control circuit is relatively simple, testing and adjustments DO require some test equipment AND a certain degree of knowledge in electronics. Unless you have both, (or know someone who does and is willing to help you) you should NOT attempt to duplicate this circuit.

If you decide to go ahead, you should be aware of the following:

Unlike the CDI module (which creates the ignition sparks), this Ignition/Injection control circuit cannot be tested/adjusted without a dual trace oscilloscope !

Even if you buy a ready made board, it may still need to be re-adjusted (perhaps only slightly) to suit your particular engine type!

The reason is that the position of the pulses delivered by the Hall switch depends on how and where the activating magnet is attached to the engine.

This design is based on Hall switch US2881UA, made by Melexis.

It has very high sensitivity and is Bipolar. That means both polarity (N & S) of the magnet can be used for switching but that too will affect the pulse position slightly. (However, other types of Hall switches can also be used, with or without modifications.)

For the above reasons, the EXACT pulse position in the engine’s working cycle can only be determined by electronic measurements when everything is in place and the engine is turning! (by hand or by starter motor)

If all this appears to be somewhat complicated, well, it is!

But there is more. The pulse input circuit has a relatively high input impedance (determined by R10, 100k). If the Hall sensor end of a long wire is left open (not terminated) it is prone to pick up interference which upsets operation. (For example, should it pick up mains hum (50Hz), it may deliver sparks at mains frequency rate, without the engine turning!) Once the Hall sensor is connected, everything is fine since it has a LOW impedance output. Still, I suggest you keep the connecting cable (3 wires) to the Hall sensor as short as practical. Shielded cable is recommended.

For set-up & testing purposes, the pulses normally coming from the Hall switch must be substituted with some other signal source.

To eliminate the need for a dedicated pulse generator, I am offering a very simple design of a 4046 (PLL) based square wave oscillator. (VCO) (Note: It does not have to be low duty cycle pulse since the input of the control circuit ONLY responds to the RISING EDGE of the waveform.)

I choose to put this simple circuit on a separate (small) circuit board and it is intended to be PERMANENTLY attached to the system but only connected (powered up) during set-up and testing. There is not much to be said about this very basic circuit but perhaps I should mention that its frequency range is restricted to approx. 1Hz - 40Hz.

The restricted range also eliminates the possibility of setting incorrect frequencies.

Les Banki

(Electronic Design Engineer)

Water Fuel & LBE Technologies

Capacitor Discharge Ignition (CDI) module

high res

Introduction

This brief explanation may assist those who are not familiar with Capacitor Discharge Ignition.

A capacitor (usually about 0.5 – 2uF) is charged to about 300 – 350V. The formula for the stored energy in each charge is: E=1/2·C·V²

In words: the energy (E, in Joules) = capacitance (C, in Farad) multiplied by the voltage (V) across the capacitor, squared, multiplied by 0.5 (or divided by two, same thing)

For example: in the CDI design I am presenting here, the capacitor is 1uF. It is charged to 330V before it is allowed to discharge. The stored energy (E) in each discharge is: 1/2·1·330² = 0.05445 joules or 54.45 mjoules. [Incidentally, the required minimum spark energy for Internal Combustion Engines (ICE) is said to be about 25 mjoules. (millijoule = 1/1000 joule)] As you can see, a 1uF capacitor delivers more than twice that minimum.

Of coarse, lowering the voltage decreases the energy. (AND also the average charging power required!) In case the capacitor is charged to only 300V (which will still deliver at least 20,000V to the spark plug), the energy in each charge is 45mjoules. (Still almost twice the claimed minimum.)

Again, note that the voltage (V) in the formula is squared. This means that even a relatively small change in the voltage (increase or decrease) results in a significant increase or decrease of stored energy. However, keep in mind that HydrOxy requires considerably LESS spark energy.

(Even a very low energy electrostatic discharge spark is sufficient to ignite HydrOxy!)

Capacitor Discharge Ignition differs significantly from the well known (and old!) ‘Kettering’ system. Instead of feeding the primary winding of the ignition coil with LOW voltage (12 – 14V) and HIGH current (5 -10A), HIGH voltage (300 – 350V) and LOW current is dumped into the primary winding from a charged capacitor. Thus, the POWER requirement of the CDI system is only a fraction of the Kettering system! The design I am presenting here uses about 6W while the Kettering type ignition use 50 – 120W, depending on the design. The above mentioned 6W power consumption is for a system requiring a maximum of 50 discharges per second which corresponds to 6000RPM for a ONE cylinder engine.

Naturally, for multi cylinder engines the power consumption will increase somewhat as the number of charge/discharge cycles increase.

The high voltage (300 – 350V) needed for a CDI system is obtained by using a DC – DC converter. There are several types of DC – DC converters but all of them use an inductor or transformer of some sort. Such a transformer (or inductor) usually has to be custom designed. All designers face this problem. Most (if not all) manufacturers are not willing to design/make just a few pieces. Unless one is prepared to order large quantities, they are NOT interested. Thus, the cost of any new design is very much an issue!

There is a very good reason for telling you all this. Everyone who intends to duplicate this circuit needs the following information:

While investigating several options, I discovered that several types of commercially made DC – DC converters are available to power CFL’s (Compact Fluorescent Light). One of these little beauties are sold here by Oatley Electronics for a grand sum of \$4.00!! www.oatleyelectronics.com (one could hardly get a transformer for that price!)

But there is a catch. Its output is over 500V! (unloaded) That is WAY too high for a CDI unit! Loading alone does NOT bring the voltage down to the desired value AND its output changes with load changes! So, its output needs to be REGULATED. There are basically two ways to do this. One is to regulate the bias to the two driver transistors, the other is to regulate the input voltage to the unit. I have tried both. When regulating the bias, both transistors need to be heath sinked since they are no longer turned on fully and so they run hot. I found regulating the input voltage to be a lot better option.

Since this design is based on this particular CFL inverter (or rather, its transformer), everyone who intends to duplicate this design will face the same practical ‘problem’. My circuit and pcb layout for this CDI system is built around this transformer. I have actually bought a large number of these inverters. (there is no point designing pcb’s for just a few units). I strip these units, discard the original (round) circuit board and transfer the components to my pcb. I found this to be by FAR the easiest and cheapest way to obtain the desired DC – DC converter! In any case, even if I choose a custom designed transformer, duplicators would still have no choice but obtaining THAT particular transformer.

In case this is not acceptable to some of you, you are on your own and you have to “roll your own” design!

Needless to say, I will sell completed units and perhaps kits, too.

Regulating the output was relatively easy. However, during extensive testing I discovered that in case of certain possible fault conditions (more on this later) the DC – DC converter draws excessive currents which over heaths the inverter transformer and destroys the driver transistors.

Therefore, I have added a fairly complex protection circuit which I developed/designed. It gives full protection!

Detailed circuit description

Let’s start with the CFL inverter described above. While the manufacturer/supplier does not offer any kind of description (they hardly ever do!), it is easy enough to figure out how it works. (it is not important) All I want to say is that it is a clever, simple design which seems to be very efficient and works well. It runs at about 100kHz.

Looking at my circuit diagram (pcb pdf,PCB), the components used from this inverter are: TF1, L1, Q7, Q8, C9, R26 and R27.

The output is full wave rectified by HV ultra-fast diodes (UF4007) D3, D4, D5 and D6. C10 (10n, 630V) is filtering the HV output. This 330V (or 300V) output is connected to one side of capacitor C12 (1uF, 400V). The other side of C12 is connected to the “hot” side of the ignition coil primary. The other side of the coil is grounded. (as usual) In other words, the other side of capacitor C12 is grounded through the ignition coil.

The capacitor’s stored energy is discharged into the coil as follows: SCR1 (TYN816) is connected between the high voltage output and ground. Its Gate is triggered by transistor Q9 (BC547), wired as an emitter follower. When ignition pulses (from the ignition module) are fed to its base (through R28, 1k), it turns on and its emitter supplies the trigger current from the 12V supply rail, through collector resistor R29 (390 ohms). When Q9 is turned on, some current also flows through R30 (470 ohms) in addition to the SCR’ gate trigger current. The low value of R30 and C11 (0.1uF) shunt spurious transients which could cause false triggering.

When SCR1 is triggered, it becomes (for all practical purposes) a short circuit. Through this ‘short circuit’ the capacitors energy is discharged to ground.

The discharge current also flows through the ignition coil’s primary which is transformed (1:100) and creates a secondary voltage well in excess of 20.000V! (depending on the type of coil used).

Regulating the inverter’s output voltage

As I have stated above, I choose to regulate the inverter’s input voltage.

It is a standard ‘series pass’, OP amp based regulator. (IC1B, LM324) It drives Q5 (BC547) and Q6 (TIP31B) in a Darlington configuration. The HV (300 – 330V) output is attenuated by R23 (680k) and R24 (15k) and connected to pin 6, IC1B’s inverting (-) input. It is also connected to the emitter of Q6, which is the output of the regulator. The non-inverting input is connected to the slider of P1 (10k), which, with R25 (3k3) forms a voltage divider to restrict the adjustment range of P1. This, in turn, limits the high voltage at the output of the inverter. Since OP amp IC1B is a “virtual earth” amplifier, its inverting (-) and non- inverting (+) inputs are practically at the same voltage. Therefore, the voltage appearing at pin 6 (regulator’s output) will be the same as the voltage on pin 5, the slider of P1.

The voltage divider R21 (990k) and R22 (10k)/C6 (10n) provide a convenient low voltage, low impedance test point TP2 for adjustment/test purposes of the HV output.

Protection circuits

First, I will try to explain why the somewhat complex protection circuit is necessary.

Please look at the circuit diagram. You will see that C12 (the 1uF capacitor which supplies the spark energy) is connected between the HV output and, through the ignition coil’s primary winding, to ground.

Now, consider what happens if C12 goes short circuit. (In other words, there is a short placed on the DC – DC converter’s output!) The poor thing will try to supply power into a short circuit! (with plenty of current but almost NO voltage!) As a result, current draw from the power supply will increase dramatically. This causes the driver transistors AND the transformer to over heath, until something gives!

Consider now an open circuited C12. There is NO stored energy to discharge. Then there is NO charge time to consider. Remember that SCR1 (TYN816) is also directly across the HV output.

Normally, when SCR1 fires to discharge the energy in C12, the current flowing through SCR1 is eventually reduced below its ‘holding current’ so it ‘drops out’. (stops conducting) When there is NO capacitor, (same as an open circuit capacitor) there is NO periodic discharge, the DC – DC converter is continuously supplying current so SCR1 will NOT drop out. This means an INDEFINITE ‘short circuit’ (in form of a continuously conducting SCR) across the HV output.

Further, the EXACT same condition will also occur if the wire to the ignition coil’s primary is broken or disconnected. (or if the coil goes open circuit)

IC1A is used to detect the presence/absence of the HV. R1 (12k) and R2 (680k) form a voltage divider between the HV output and ground. The voltage developed across R1 is a fraction of the HV and it is fed to the non-inverting (+) input pin 3 of IC1A, used here as a comparator. A fixed voltage (approx. 2.9V) is applied to the inverting (-) input (pin 2) from the voltage divider R4 (68k) and R5 (22k) which is filtered by C2 (10uF). Under normal operating conditions the output of this comparator is HIGH.

Should the voltage on the non-inverting input (pin 3), which represents the HV output, decrease significantly (below the voltage on pin 2, the inverting input) or disappear completely due to a fault condition, the output of the comparator IC1A (pin 1) will go LOW. This output is connected to the Gates of DMOS transistors Q2 and Q4 (2N7000), through R6 and R15, respectively (both 100 ohms). (Note: for this application bipolar transistors are un-satisfactory. Their off-state collector-emitter leakage is too high.)

IC1C and IC1D are wired as square wave oscillators. Since the normally conducting Q2 and Q4 are connected between the inverting (-) inputs and ground, both oscillators are DISABLED. (C3 – 3.3uF and C4 – 1uF are the timing capacitors) In the (sampling) oscillator IC1C, the charge/discharge times are separated. This gives (with the component values shown) approx. 2 seconds HIGH and about 25 seconds LOW signal at IC1C’s output (pin 8). Through D2 (4148) and R13 (10k) this signal is fed to the base of Q3 (BC547) which is used as an inverter. Q1 (2N7000) is connected between ground and the non-inverting (+) input (pin 5) of voltage regulator IC1B. Its Gate is connected to the collector of Q3. When Q3 is conducting, Q1 is NOT. (NO Gate voltage – it is shorted by Q3) When Q3 is NOT conducting, Q1 gets its Gate drive from Q3’s collector through R14 (10k). Q1 is now conducting, bringing the voltage on the non-inverting input (pin 5) of the regulator (IC1B) to 0V. As a result, the regulator’s output is also zero.

NO INPUT VOLTAGE to the inverter means NO current draw. In other words, this is NOT a current limiter. The inverter is completely OFF, drawing NO current.

As long as the fault condition exists, oscillator IC1C continues its 2/25 seconds ON/OFF routine. Its output is inverted by Q3 which then turns Q1 OFF/ON. So, when Q1 is OFF, the regulator (and the inverter) is working normally. When Q1 is ON (conducting), the regulator (and thus the inverter) is cut off. In this condition, there is NO current draw.

In layman’s terms, this is what happens: Due to a fault condition, (capacitor C12 open or short circuit, ignition coil primary open circuit, wire to the coil broken or disconnected…) oscillator IC1C is ENABLED and is producing a 2 seconds ON and 25 seconds OFF signal. This signal ENABLES/DISABLES the regulator supplying the inverter. The 2 seconds ENABLE signal is for SAMPLING. Is the fault still there? Yes. OK, cut power OFF for the next 25 seconds. Then, SAMPLE again (for 2 seconds) to check if the fault has been cleared or not. If not, this oscillator will continue its 2/25 second routine INDEFINITELY.

Since power is applied for only 2 seconds (SAMPLING) and there is NO power for 25 seconds, no harm is done! If the fault has been cleared, the oscillator is disabled and the regulator/inverter once again works normally.

Since indicator LED1 for the sampling oscillator is only turned ON for 2 seconds (and OFF for 25 seconds) there is a need for continuous indication of a fault condition.

That is the role of oscillator IC1D. Under normal working conditions it is disabled by Q4 (2N7000) which is shorting its timing capacitor C4 (1uF). It is wired as a square wave oscillator which, under fault conditions, flashes LED2 ON/OFF about 3 times per second (~3Hz).

Under normal operating conditions, the inverter’s regulated supply voltage output is around 6.4V. Current draw is about 0.5A. With a short circuit placed on the output, the current rises to around 1 - 1.2A. Should the regulator transistor Q6 go open circuit, the inverter simply stops operating. However, should it decide to go short circuit, (unlikely, due to the moderate current draw of only 0.5A) the full rail voltage of 12V would be applied to the inverter and its output would rise to over 500V! This, in itself, should not be a problem, except for two things: 1. Capacitor C12 (rated at 400V) might go short circuit (which would activate the protection circuit described above). 2. The ignition coil would produce excessive secondary voltage which could cause internal insulation break down.

Testing and adjustment

A number of TP’s (test points) are provided for testing and adjustment(s) purposes.

There is only ONE adjustment to be made on this pcb, to set the inverter’s output voltage to the desired value (usually somewhere between 300 – 330V depending on the ignition coil used). Connect a voltmeter (set to 600V or 1000V range, depending on the meter) between TP3 (ground) and TP1 (HV) and disable discharge triggering TP4 by shorting it to TP3. Now adjust P1 to the desired voltage (300 – 330V) You could also use TP2 (and TP3) to adjust to 3 – 3.3V (100:1 attenuator, provided mainly for oscilloscope connection to eliminate the risk of damaging its input)

The regulator’s output voltage (supplying the inverter) can be measured at TP5. For a 330V output it should be around 6.4V.

The operating temperature of the inverter transformer and its driver transistors are a very comfortable 47°C and 45°C, respectively, measured in ambient temperature of 30°C!

Les Banki

(Electronic Design Engineer)

Water Fuel & LBE Technologies

Ignition & Injection control module

Important notice:

The control electronics described below is NOT suitable for engines with manual (pull cord) starters, UNLESS the pull cord is EXTENDED to provide MORE revolutions!

Reason: Recent physical tests of several manual start generators reveal that their pull cord starters only produce two (2) or 3 revolutions of their crankshaft.

That is only ONE (or 1.5) engine work cycle. In order to establish & stabilize the necessary waveform for proper operation, AT LEAST ONE (or more) cycle is required. During this process NO injection or ignition pulse(s) are allowed.

In other words: At least ONE engine cycle (2 crankshaft revolutions) is needed just to establish the correct waveform, WITHOUT injection or ignition pulses! Then, additional revolution(s) are necessary for starting.

Electric starters can deliver as many revolutions as needed and thus solve the ‘problem’. Needless to say, this involves a battery which will be needed anyway as a start-up supply for the electronics and generation of Hydroxy.

Background

In my thesis, “Ignition system for small engines”, I briefly outlined why a new ignition system is needed with Hydroxy as the ONLY fuel and what are the technical requirements for such an ignition system.

To start with, here are a couple of quotes from an EXCELLENT web site which briefly explains (with moving animation!) ignition technology:

“It is interesting to note that one complete engine cycle takes two revolutions but that individual valves and spark plugs only operate once in this time. Hence their timing needs to be taken from a half engine speed signal, which is the camshafts speed.”

“If the timing disk is attached to the crankshaft, there is a need in some engine configurations to have a sensor on the camshaft so that the igniter knows which ½ of the four-stroke cycle the engine is in.”

(http://www.gill.co.uk/products/digital_ignition/Introduction/6_4stroke.asp)

Most (if not all) existing small engines use a magnet mounted on the fly wheel which gives two pulses for every engine cycle. (thus generating “waste” sparks)

Electronically dividing by two would NOT solve the problem since another signal would be needed to determine which one of the two pulses we want and which one we don’t.

Only ONE sensor is needed IF it gives only ONE pulse for every ENGINE cycle. It does not matter WHERE in the engine’s cycle this pulse originates because it can be electronically ‘moved’ anywhere in the engine’s 360º (100%) cycle. The obvious choice of a sensor is a ‘Hall effect’ switch.

Since modern engine blocks are non-ferrous (aluminium) alloys, the Hall switch can be placed on the outside of the engine block. For example, it can detect the position of a magnet which is fitted to the valve’s ‘rocker’ arm.

As the ‘rocker’ arm/magnet moves in and out of certain positions, the Hall switch turns on/off.

This sensor must detect the position of the cam shaft (which makes one revolution per engine cycle), NOT the crank shaft (which rotates twice for every work cycle). Looking at most small engine designs, it seems that accessing the cam shaft is easiest at the exhaust/intake valves under the valve cover.

Thus, the mechanical modification consist of fitting a small magnet to the rocker arm of the exhaust (or intake) valve but because the magnetic field is blocked by ferrous metal, if the valve cover is made of steel, it must be either replaced with a non-magnetic one (like aluminium), or, cut a hole in it which is then covered by some non-magnetic material.

[Note, lamare : here are some images of my hall sensor construction]

I used a very small magnet, around 5 mm in diameter, and fixed that to the exaust rocker:

For a hall sensor, I used the TLE 4905 L, which I mounted on the engine using a piece of copper tube:

I soldered the tube on the rocker cover:

I have adjusted the exact position of the sensor / tube has been a few times, before I got the optimal position, such that the sensor is activated / deactivated when the valve is about half way pressed.

The finished sensor construction:

[end note]

Surely, the above modification should be easier than making a 2:1 gearing!

My choice is to place the magnet on the EXHAUST valve’s (rocker) arm for the following reason:

The EXHAUST stroke is the LAST of the engine’s working cycle. From this point, everything is in the correct order. First the INTAKE stroke, during which the gas will be injected at the correct time (and duration, determining the speed) Now the pulses from the Hall switch can be correctly delayed by the required amounts for the Injection and Ignition functions. Following the INTAKE stroke is COMPRESSION, at the end of which IGNITION takes place.

As the pulses from this single Hall switch usually do not occur exactly where we want them, they need to be ‘moved’ (delayed) to deliver the desired injection/ignition functions at the correct times.

First, some engine basics:

Engine speed is expressed as RPM (Revolutions Per Minute). In electronics, however, the unit of time is SECOND. Since there is 60 seconds in a minute, (it was the last time I checked! (() the engine’s RPM is divided by 60 to get the engine’s ‘frequency’, in Hz.

Example; an engine running at 3600RPM (crankshaft speed), divided by two is 1800 engine cycles per minute. Divide that by 60 gives 30Hz. It means that the spark plug is going to fire 30 times per second.

One thing is CERTAIN.

IF the fuel is Hydroxy ONLY and when there is a sufficient volume of it mixed with air in the cylinder and a spark occurs, it WILL explode, regardless of correct or incorrect timing!!!

Due to this fact, irrespective of all other design changes, when starting the engine, injection/ignition pulses MUST be inhibited for at least ONE cycle in order to establish and stabilize the saw tooth waveform.

In practical terms this means that the engine would start on the 2nd (or 3rd ) revolution of cam shaft (4th or 6th revolution of crank shaft) which would NOT be within the firs pull of the cord!! (for manual start) For this reason electric start is needed. (or pull cord extended!)

Once again: Since 1 Hz is 2 crank shaft revolutions, the RPM is 120. 2 Hz is 4 revolutions and thus 240 RPM.

Principle of operation:

1. The first task is to convert the pulse train from the Hall switch to ‘frequency’. [(so that ONE ‘period’ is ONE engine cycle. (4 stroke)] This can be done by either digital or analog means. (or a combination of the two) Both have advantages and disadvantages. While the modern “buzz” word is “microprocessor”, there is no need for it here with its complex software programming. Basically, the two main issues with any design are: simplicity and cost. My choice for this design is analog. It is a relatively simple and low cost design.

2. The engine’s frequency is transformed into a LINEAR saw tooth waveform. This saw tooth is fed to a comparator. The output is a variable duty cycle square wave. [This is the basic principle of the analog Pulse Width Modulator (PWM)]

3. The rising or falling edge of this variable duty cycle square wave is used to trigger the desired ignition/injection action.

New ignition & injection control design

Close examination of my previous design revealed two problems:

1. When starting the engine, an ignition pulse was allowed to occur following the very first pulse from the Hall switch! [At least two (2) pulses are necessary to create & stabilize the desired waveform!]

2. Frequency-to-Voltage converters are inherently slow. When rapidly changing the frequency, large amplitude variations are unavoidable. This plays havoc with the injection & ignition timing to the point of being useless!

Note: all this occurs ONLY at very low STARTING frequencies. (1Hz – 5Hz)

First of all, I have drawn a chart I named “4 stroke engine timing cycle”:

high res

Not only does it aid the understanding of a 4 stroke engine’s work cycle and the injection & ignition process but it also serves as an essential reference for the initial set up of the injection & ignition timing circuits for the engine used.

Like the previous design, this new design is also based on a Hall switch, activated by a magnet attached to the exhaust (or intake) valve’s rocker arm.(Thus, the pulses always occur at the same “degree” of the engine’s work cycle.)

The control circuit’s basic principle of operation also remains the same: creating a linear saw tooth waveform from the pulses supplied by the Hall switch.

However, the new design DOES NOT USE a Frequency to Voltage converter (F/V) which has an inherently slow response. Further, the method of generating the linear saw tooth voltage has also changed. Instead of fast charging the timing capacitor and slowly discharging it with a constant current source, (producing a falling slope) it is now continuously charged by a simple constant current source, producing a rising slope.

The main reason for this change was/is to obtain a Ground (0V) referenced waveform which is much easier to manipulate. (for example in a feed back loop)

The new circuit (pcb.pdf, PCB) operates as follows:

The pulse train from the Hall switch, IC1* (on its own pcb), is fed to monostable IC1A’s rising edge input A (pin 4, 4538 dual monostable) The output pulse width is set to 100µs by R3 (10k) and C3 (10n). These pulses (from pin 6) are fed to the falling edge B input of IC1B (pin 11), the ‘clock’ input (pin 14) of IC6 (4017) and also to the ‘logic’ input (pin 8) of track & hold IC5 (LF398) through attenuator resistors R20 (3k3) and R21 (1k).

Monostable IC1B’ (4538-2) output pulse width (100µs, at pin 10) is set by R4 (10k) and C4 (10n). These pulses operate switch Q2 (2N7000) which discharges timing capacitor C5 (0.1µ). Continuously charging C5 is a simple constant current source, comprising of Q1 (BC327), D1 (4148), R6 (3k3) and R7 (270k). (Charging time constant is determined by C5 and R7.)

The resulting linear saw tooth is buffered by IC2 (TL071), wired as a unity gain voltage follower. The buffered saw tooth (output of IC2) is fed to both inputs of IC3 (TL071), a Voltage Controlled Amplifier (VCA) through R8 and R9 (both 10k). The ‘gain’ control elements are P-channel JFETs Q3 and Q4 (both J174), operating in their ‘ohmic’ (linear) region, their D-S voltage being restricted to a few tens of milli volts by the inputs of the OP amp. (IC3)

Since the VCA (IC3) is inverting, the next stage, IC4A (TL072-1) is another inverting amplifier with unity gain (-1). Its output signal (from pin 1) is now of correct polarity and is fed to comparator/error amplifier IC4B’ (TL072-2) inverting (-) input (pin 6). Its non-inverting (+) input (pin 5) is fed with an adjustable voltage from the voltage divider R16 (8k2), P1 (10k) and R17 (22k). These component values give a range of approx. 6.56 – 9.55V (for 12V supply) and 2.73 – 3.98V (for 5V supply) for setting the saw tooth amplitude to the desired level.

The error signal (from the output of IC4B, pin 7) is fed, through R19 (10K) to the input (pin 3) of track/hold IC5 (LF398). At the peak value of the saw tooth, IC5 is put in the Hold mode for the duration of the following cycle. This DC voltage is fed from its output (pin 5) to the gate of Q3 (J174) JFET.

It is this voltage which sets the gain of the VCA.

Thus, the VCA (IC3), the unity gain inverter (IC4A), the comparator/error amplifier (IC4B) and the track/hold stage (IC5) form a feed-back loop to correct the amplitude of the waveform as the frequency changes. [Suppose the starting frequency is 1Hz and the saw tooth P-P amplitude is set to 8V. (3.33V for 5V supply) When the frequency has increased to 2Hz, the amplitude has DECREASED to 4V (1.66V for 5V supply) and needs to be amplified (x2) to bring it to 8V. (3.33V for 5V supply) At 4Hz, the amplitude is 2V (0.83V for 5V supply) and it is amplified by 4 to bring it to 8V (3.33V for 5V supply) and so on….)

Unlike the inherently slow Frequency to Voltage converter, this method stabilizes the waveform amplitude after only ONE completed cycle. This is due to the fact that amplifier action is FAST! (device propagation delays are measured in micro or nano seconds!)

The amplitude AND polarity corrected saw tooth waveform from IC4A’ output (pin 1 & TP9) is fed, through R36 (10k), to the inverting (-) input (pin 2) of another comparator, IC9A (LM393-1). A voltage divider is formed by R34 (22k), P4 (10K) and R35 (33k) and the non-inverting (+) input (pin 3) of IC9A is connected to the wiper of P4 (10k), thus:

P4 is the IGNITION PULSE POSITION adjustment.

The above component values of the voltage divider allow a wide range of adjustment. However, to suit different engine designs, the values of R34 & R35 may be altered.

Since we now have a rising ramp, the voltage on the inverting (-) input (pin 2) of the comparator IC9A (LM393-1) remains below the set point on the inverting (+) input (pin 3). As a result, the output of IC9A stays HIGH until the set point is reached. When that happens, the output snaps LOW. This falling edge pulse triggers monostable IC8A (pin 5). The output pulse width is set to 100µs by R38 (10k) and C17 (10n) and is available from pin 6 (Q output of IC8A).

These 100µs pulses are fed to the base of ‘emitter follower’ ignition trigger transistor Q9 (BC 547) through R39 (2k2) and R28 (1k). Q9’ collector resistor R29 (390 ohms) supplies the necessary trigger current for SCR1 (TYN816). R30 (470 ohms) & C11 (0.1µF) help to reduce/eliminate spurious noise pulses from the SCR’s gate. (Note: Q9, R28, R29, R30, C11, SCR1 and capacitors C12A,B,C are on the CDI module)

Capacitors C12A,B,C (1µF 400V) are continuously charged to 330V by a DC-DC converter. The stored energy is then dumped into the primary winding of the ignition coil. Thus, it is a CDI (Capacitor Discharge Ignition) system.

In practice, ALL generators are run at a CONSTANT speed somewhere in the range of 2000 to 4000RPM (depending on design). (Remember, this ignition system is designed specifically for ONE cylinder generators.)

Injection

In the general practice of the “art”, the injection solenoid is sometimes turned on just before the intake valve opens to gain a little extra time for injection. (But it must be turned off before the valve closes.) The volume of gas injected is determined by gas pressure AND the solenoid’s ON-time.

Just as with the ignition, injection timing is related to the engine’s work cycle.

The injection circuit is almost identical to the ignition circuit with a couple of differences. Just like the ignition circuit, the injection comparator (IC9B-LM393-2) gets its linear saw tooth signal from the output of IC4A (pin 1 & TP9), connected through R30 (10k), to its inverting (-) input (pin 6). The non-inverting (+) input (pin 5) is connected to the wiper of P2 (10K) which is part of a voltage divider. [R28, (68k) – P2 (10k) – R29 (15k)]

P2 is the injection STARTING POINT adjustment.

Again, because of the rising ramp, the output voltage from IC9B (pin 7) is HIGH until it reaches the set point. Then it snaps LOW and this falling pulse triggers monostable IC8B (pin 11). Its output pulse length is determined by P3 (100k), R32 (15k) and C15 (68n). With these component values it is adjustable from about 1ms – 8.2ms.

P3 is thus the SPEED (RPM) control of the engine.

Note: If using the automatic RPM control module, monostable IC8B is now used to create short, fixed time constant trigger pulses to the VCT on the auto RPM board. Now R32 is 22k, C15 is 1n and P3 is replaced with a shorting link. (time constant approx. 22us) The negative going pulses are taken from the Q-bar output (pin 9) and fed to control pin 5 of the VCT. All solenoid driver components are omitted and are fitted to the auto RPM board instead.

The MTP3055E MOSFET has an in-built protection diode. (If MOSFETs without in-built diodes are used, an external diode MUST be fitted!)

In addition, in order to eliminate the very large (close to 100V!) BEMF pulses generated by the solenoid coil, diode D3 (4007) must be fitted across the coil.

ANY power MOSFET can be used which can comfortably switch 2A or so. (It is most unlikely that you need to use a higher current solenoid.)

I have tested 2 gas injector types: one draws just under 1A and the other just under 2A. Since the ON resistance of most power MOSFETs are in the ‘milli ohm’ range, there is no need to put the MOSFET on a heath sink. Further, the injector solenoid is only ON for ¼ of the total time (during INTAKE) and thus its “duty cycle” in only 25%. (max.)

Blocking injection/ignition pulses at start up:

An ACCURATE method is a digital pulse counter feeding a retriggerable monostable, which in turn enables/disables the monostables delivering the injection/ignition pulses.

The pulses from the Hall switch (or from the monostable) are clocking decade counter/decoder IC6 (4017), which counts the pulses. At the desired number, say, 3 pulses, the counter is reset and at the same time it delivers pulse number 3 to IC7B (4538-2), configured as a retriggerable monostable.

The time constant of this monostable output pulse must be at least 3 pulse periods from the Hall switch, in order to prevent it from terminating its timing cycle. This means that at the very low cranking speed of the engine, say 1Hz (120RPM), it needs to be just over 3 seconds.

In practice, the time constant can be set to 4.7 seconds (approx.), using a 1µF capacitor (C13) and a 4.7M resistor (R26). (T = R*C, 4.7*1 = 4.7s)

This is the way it works:

When power is applied, all stages are RESET. (Hall switch, counter, monostables, etc.) In particular, the retriggerable monostable IC7B’s Q output (pin 10) is LOW when reset. It disables injection pulse monostable IC8B by pulling its CD (clear) input (pin 13) LOW by forward biasing D3. (diode ‘clamp’) Likewise, ignition pulse monostable IC8A is also disabled by pulling its CD (clear) input (pin 3) LOW. Thus, since IC8A & B are disabled, NO injection or ignition pulses could occur, even if trigger pulses are supplied.

Details:

By cranking the engine, the Hall switch starts delivering pulses. The counter, IC6 (4017), counts these pulses. When the pre-set number is reached (say 3), it resets itself and at the same time it triggers the retriggerable monostable IC7B, with a set timing period of 4.7 seconds. In the reset stage, its Q output (pin 10) is LOW.

For normal operation, R27 (10k) keeps the CD inputs of the injection & ignition monostables IC8A & B HIGH. During the pulse count after a reset, the CD inputs of IC8A & B (pins 3 & 13) are held low by IC7B’ LOW output and the clamp action of D2, disabling them. When the set count is reached, IC7B’ output goes HIGH and that enables operation of IC8A & B by removing the clamp.

Repeat: from reset, during the pulse count, IC8A & B are disabled. When the set count is reached, IC8A & B are enabled to operate normally, allowing injection & ignition pulses to be delivered to their respective circuits.

As the engine picks up speed, the pulse period from the Hall switch gets shorter. However, due to the long time constant of 4.7 seconds, even if the engine speed remained at a very low RPM indefinitely, the monostable timing period could still not terminate, so ignition pulses would be delivered as normal. During normal operation, monostable IC7B receives only every third (3rd ) (or higher) pulse as trigger pulse from the counter. (This stage can be considered as a “missing pulse” detector.)

When the engine has stopped, it takes 4.7 seconds (after the last trigger pulse) to terminate the monostable’s timing cycle and once again clamp the ignition pulse delivery. In practical terms it means that the engine can be re-started after about 5 seconds with a guaranteed blocking of the first 2 (or 3, as required) ignition & injection pulses!

Note: 3 pulses from the Hall switch constitute 2 full engine working cycles. (Between 1 and 2 is ONE and between 2 and 3 is TWO full engine work cycles.) Keep in mind that the third (3) pulse is the start of the THIRD cycle and subsequent ignition pulse.

Testing and adjustments:

First, a variable frequency (1Hz to 40Hz) pulse generator is connected to TP4. (Hall switch input) To test & adjust this circuit, a DUAL TRACE oscilloscope is essential. When using BOTH probes, I suggest you connect only ONE of the ground probes as the circuit is sensitive to ‘ground loops’. (they may upset not only the measurements but also the operation of the circuit itself)

Set up adjustments of this new design are considerably simpler than the previous one!

Apart from the CONTROL PANEL adjustments of the IGNITION point (P4) and SPEED (P3), there are only two other (pre-set) adjustments: 1. Saw tooth amplitude, using P1 (set DC voltage at the wiper of P1 to 8V) 2. Injection STARTING point with P2

This Ignition/Injection control circuit, together with the CDI module and the Hall switch, forms a COMPLETE ignition system.

Before you decide to go ahead building/using it, you should also read “Important notes about the Ignition & Injection control circuit and Test Oscillator”.

Notes: VCA (IC3) MUST be a single OP amp package. For some strange reasons [which I had no time (or desire!) to investigate], it refuses to work properly with an OP amp from a dual or quad package!

The saw tooth amplitude DOES vary somewhat. It is highest at the low frequency end (1Hz), decreasing to a steady value in the desired frequency range 15 - 30Hz (1800 – 3600RPM) where most engines will be running. However, this small amplitude change can be ignored since the engine spends very little time in the cranking mode (very low RPM) and the ignition point there is not critical.

What IS important is that it is STABLE in the required running RPM range.

Les Banki

(Electronic Design Engineer)

Water Fuel & LBE Technologies

AC power cotrol circuit description.doc

AC power control circuit description

This circuit was developed/designed for powering high voltage, multi cell electrolyzers. (However, it can also be used as a high power light dimmer, motor speed controller, etc.)

It employs AC phase control, using a Triac. (which is a very robust and reliable device!) Phase control necessitates the use of a random phase optoisolator Triac driver (like the MOC3021) and the design of a synchronized zero crossing detector and a current limiter. Using the components indicated on the circuit diagram, it will run comfortably at 240VAC at 10A – which is 2400W! It can be scaled up by using some components with higher ratings.

Through a bridge rectifier, the phase controlled mains AC power is applied DIRECTLY to the electrolyzer! (NO transformer!)

Some may not like the idea but the advantages of a transformerless circuit for this purpose FAR outweighs the disadvantage of a LIVE electrolyzer. (I am talking about a 120 cell, 240V unit.) LIVE or NOT, you keep your hands OFF a 240V device, AC or DC! Period!

The only technical ‘problem’ I have encountered is the necessary isolation of the current limiter control signal from the ‘live’ mains. (More about this later.)

Zero crossing detector:

From the mains transformer’s secondary (15V), diodes D1 and D2 (4007) feed pulses to the inverting (-) input (pin 6) of the zero crossing comparator IC1B (LM324) through a voltage divider R1 (1k) and R2 (1k), clamped to +12V by Zener diode ZD1. [This is done to avoid the pulse amplitude exceeding the power supply (12V)].

The non-inverting (+) input (pin 5) is biased by R3 (100k) and R4 (470ohm) to about 56mV (0.056V). Narrow, positive pulses appear at IC1B’s output (pin7). These are fed through R5 (10k) to discharge switch transistor Q1’s (BC547) base. A constant current source [comprising of Q2 (BC 327), LED1 and R7 (3k3)] is continuously charging capacitor C1 (0.1u). Q1 periodically discharges C1, thus a LINEAR saw tooth waveform is formed. (the fast discharge part of the waveform occurs during zero crossing of the sine wave)

IC1C is a unity gain ‘voltage follower’ (low output impedance buffer). Its output (pin 8) is connected to the non-inverting (+) input (pin 12) of the PWM comparator (IC1D). The inverting (-) input (pin 13) is biased by R8 (1M2) and R9 (10k) to around 12mV (0.012V). This corresponds to the maximum output power for this type of circuit. (which is about 95%) In other words, this is the output control pin. The control voltage is applied to this point through diode D4 (4148).

In this configuration, (as required for the purpose of current limiting) when the voltage increases, output power decreases.

The output of the PWM comparator IC1D (pin 14), through R10 (10k) and R11 (1k), drives the Gate of Q3 (2N7000). D3’s Anode is connected to the junction of R10 & R11.

Power OFF/ON control is achieved by applying a LOW signal to the Cathode. This removes the drive signal from Q3’s Gate. (This is used for timer operation as well as pressure control.)

The LED in the optocoupler MOC3021 (IC2) is connected between the positive supply rail and the Drain of Q3, with resistor R12 (470R) limiting the current.

The remaining components around the optocoupler and the Triac are for limiting dv/dt, suppression of transients and to reduce RF interference. (C3, C4, C5, R13, R14, R15, VDR1 and L1)

Now to the current limiting:

There are a number of options. Some are ‘sophisticated’ (read: expensive), others are not. Since high accuracy or high speed is hardly required for this application, I have adapted an unusual approach to current limiting.

Using the principle that current flowing in a conductor is heating it, I have a ‘dedicated’, short pcb track in the current path to the load (electrolyzer, etc.). Over this track I place a diode (like the 4148 which has a thin glass envelope) with heath conducting compound (same as used on heath sinks) for good thermal conduction. It is well known (or should be!) that a semiconductor junction forward voltage drop exhibits a temperature coefficient of about -2mV/C°. This is amplified, say, 10 – 1000 times, depending on the required ‘sensitivity’ (Hysteresis). [In this circuit the amplifier is IC1A. R20 (100k to 3M3) and P2 (2M) sets the gain.]

Referring to the circuit diagram:

You find diodes D6 and D7 (both 4148), R16, R17, R18 and R19 (all 4k7) in a ‘Wheatstone Bridge’ configuration. D7 is the temperature/current sensor. P1 (500R – 25 turn) sets the current limit by balancing the bridge at the required temperature of the pcb track/diode (D7) and thus the current. [P1’s wiper is connected to the positive supply rail (12V)]

This adjustment works as follows:

As an example, suppose the track & diode temperature (at the required current) is, say, 10°C ABOVE the ambient temperature inside the equipment. (Keep in mind that the other diode of the bridge, D6, IS at that ambient temperature.) So, -2mV/C° x 10 = -20mV. P1 needs to be set so that the bridge ‘leg’ D7 is in, is 20mV HIGHER* than the point where the bridge would balance if both diodes were at ambient temperature. (* Yes, HIGHER, since the semiconductor junction forward voltage drop is NEGATIVE. When the junction temperature increases, forward voltage DROPS.)

Now, consider the effect of the inverting amplifier IC1A.

The ‘leg’ of the bridge where D7 is, is connected to the inverting (-) input (pin2). If the voltage at this inverting (-) input is LOWER than the non-inverting (+) input, there is output from IC1A (pin 1). [If much lower, the output SATURATES! – (close to the supply rail voltage.)] If it is HIGHER, the output is ZERO! This is EXACTLY the condition we want! [As stated above, the sensor diode (D7) is connected to the inverting (-) input of IC1A but since its power supply is only single rail, its output simply cannot go negative.]

Adjust the current limit as follows:

First, connect a 60 - 100W light bulb as a load. (This minimum load is required to get the circuit operating correctly)

Now apply power. The ‘current limit’ LED might be on or off. If it is ON, turn the pot. clockwise until the LED is just OFF. [This is the point where the diode bridge is balanced. (Max. sensitivity.)]

Now connect the desired load (up to 2400W) and keep turning the pot clockwise and monitor the current on the panel meter. (up to 10A) Done.

To sum it up: The set up I described above could hardly be simpler!

It can be used for AC or DC (or ANY complex waveform) current as the thermal inertia in the copper track/diode body is MUCH larger than the waveform(s) time constant.

The diode body material gives electrical isolation between the mains voltage and the low voltage control circuits.

Bulk price of 1N4148 diodes in Australia is around 1 cent each! A 2 cents current limiter! (2 in the bridge)

Note: The above set-up is on 2 pcbs. High voltage mains components on one, control circuit is on the other. This gives flexibility as I have also developed/designed another current limiter, (on its own pcb) based on Allegro’s ACS712ELCTR-20A-T Hall effect current sensor. Naturally, it gives the required isolation, however, its output is AC and it needs to be processed (and filtered) differently.

Les Banki (Electronic Design Engineer) Water Fuel & LBE Technologies

Battery charger circuit description

Background

One might ask: what has a battery charger got to do with an Engine Control Unit? Well, EVERYTHING and NOTHING!

Remember our goal: generators using water as the ONLY fuel, producing electricity.

Now, imagine that you have such a generator running and you have already disconnected from the ‘mains’ power (‘grid’ for US readers!). This generator will be running 24/7 but as all engines need maintenance stops sometimes, how are you going to re-start it?? Or, rather, the questions are: What power source is going to power up the electronic control circuits? How are you going to generate enough Hydroxy gas to start the engine?

The answer is a LARGE BATTERY, of course, which will need to be kept FULLY CHARGED AT ALL TIMES!

Unfortunately, my personal experience over the years (to this day!) with ready made battery chargers is that batteries CANNOT be left connected to them indefinitely! That is why I designed (about 30 years ago!) my own AUTOMATIC battery charger which CUTS OFF completely when the battery is fully charged. To such a charger, a battery can be left connected INDEFINITELY. This simple design is very versatile and has been modified (slightly) a few times over the years to suit various applications.

AC power control circuit description

This circuit was developed/designed for powering high voltage, multi cell electrolyzers. (However, it can also be used as a high power light dimmer, motor speed controller, etc.)

Through a bridge rectifier, the phase controlled mains AC power is applied DIRECTLY to the electrolyzer! (NO transformer!)

The only technical ‘problem’ I have encountered is the necessary isolation of the current limiter control signal from the ‘live’ mains.

All word files as text

Automatic engine RPM control circuit description 2.doc

Automatic engine RPM control circuit description

Since the speed (RPM) of ALL Internal Combustion Engines (ICE) are governed ONLY by the volume of fuel they receive (in this case, HydrOxy), their RPM can be accurately controlled with electronic fuel injection.

Note that when an engine is powering a generator, its RPM determines the generator’s output voltage AND frequency. It should be obvious that in order to keep the output voltage and frequency constant, the engine needs to be turning at a CONSTANT speed. (actual RPM depending on design)

I have conceived the following concept of how to bring the engine speed from idle to the required RPM automatically. I have developed the theory of the set-up, sequence of events, choice of components and their practical implementation. I have then set up the circuit on a ‘bread board’, de-bugged and fine tuned it.

Note that the circuit section which brings the engine speed from idle to the correct RPM automatically is an OPEN loop system! It CAN NOT correct changes in output voltage & frequency due to load changes! That is done by a dedicated feed-back loop circuit which is also incorporated.

Principle of operation:

Unlike car engines, generators are NOT required to respond to quick throttle changes.

Suppose the engine is required to run at 3600 RPM (30Hz) to produce the correct voltage & frequency, considering the minimum response time 250ms/Hz, it would take 7.5s to reach 3600 RPM from the first firing. So, we can choose to reach the required RPM in say, 10 seconds. (or more)

After powering up all the electronics, the VCT (Voltage Controlled Timer) produces a minimum timing period which represents engine idle. Engine speed is increased by gradually increasing the injection solenoid’s ON time. To do this automatically, a VCT (mentioned above) is required.

A linear voltage slope is created by a constant current source charging a capacitor. This slowly increasing voltage is used to control the VCT and thus the RPM.

To start this process, an unmistakable “engine has started” signal is needed.

The most “unmistakable” signal is the VERY loud noise of the ‘explosions’ created by an internal combustion engine (ICE). To eliminate ‘false’ starts, two explosions are detected within a certain time frame (say 2 seconds).

The sound amplifier & 2 timers trigger an R-S flip-flop (F/F) which SET the latch, activating the creation of the VCT control voltage slope.

A sample of the output voltage from the engine/generator is fed to a frequency switch. When the generator’s output frequency reaches 50 Hz (at approx. 240V), this frequency switch operates a SET-RESET (S-R) flip-flop (F/F), activates the Hold function and ‘saves’ the control voltage. After a short delay it erases the voltage slope which has created it. The circuit is thus ready for another “start engine” cycle.

The ‘saved’ voltage is switched OUT and a feed back loop switched IN to maintain constant output frequency despite load changes on the generator’s output.

Sequence of events in more detail:

After the circuit is powered up, the VCT (Voltage Controlled Timer, IC7- C555), which determines the injection solenoid’s ON time, runs on its minimum time constant. The minimum value of the control voltage range (at pin 5) is chosen to be less than the required level for idling. (about 0.5 – 0.7 ms)

To enable idle adjustment for different types of engines, an adjustable voltage is ADDED to this minimum by IC10D (LM324)

A properly set up C555 makes a perfect VCT! It is also very simple and cheap to boot!

It has, however, specific trigger requirements for this purpose. In order to achieve the widest possible control voltage range, the negative going trigger pulses (to pin 2) must go close to 0V. [That is because whatever the voltage on the control pin is, the 1/3 and 2/3 ratio of the trigger (pin 2) and threshold (pin 6) is ALWAYS maintained!]

Further, the width of the trigger pulses should be much shorter than the basic R-C time constant of the C555’s output pulse.

In other words, for this application, the trigger pulses need to be: short, negative going (from the +12V rail) to very close to 0V. (Ground) The Q-bar outputs of the 4538 dual monostable (pins 7 & 9) supply just such pulses!

The monostable IC8B (4538) on the ign/inj. board which was used for manual injection drive can now be used with a MUCH shorter time constant to create narrow , negative going trigger pulses for the VCT. (IC7- C555) These pulses are taken from the normally HIGH Q-bar output (pin 9).

Repeat: The timing range needed for the injection solenoid (to go from start to the required RPM) is supplied by this VCT. It is calculated to be about 1 – 8.33ms for 3600 RPM, or 1 – 10ms for 3000 RPM.

A sound operated switch detects if the engine has started, by detecting the VERY loud sounds of the first two (2) ‘explosions’. Its second output pulse SETs the RS F/F (IC4A-4013). Its Q output (pin 1) goes HIGH which turns ON transistor switch Q4 (BC547) and turns transistor Q3 (2N7000) OFF. The role of Q3 is to allow charge or discharge of capacitor C25 (1uF).

The F/F’s output thus switches HIGH which is inverted to LOW by Q4. This LOW turns the clamp transistor Q3 OFF. Now the capacitor is allowed to charge and a linear voltage ramp is generated.

Once again: at power-up, F/F IC4A (4013) is RESET. Its output is LOW. This is inverted by transistor Q4 to HIGH. This turns ON the clamp transistor Q3 which acts as a short across the capacitor.

In this condition, (when its control voltage on pin 5 is between 0 – 0.25V)

the VCT produces minimum pulse width. To obtain the correct idle timing for a particular engine type, an adjustable ‘offset’ voltage is added.

To minimize/avoid false triggering by spurious noise (other than the engine ‘firing’ sound), the sensitivity of the microphone amplifier is heavily reduced. Further, the sound operated switch uses two (2) timers. (555) TIMER 1 triggers on the first ‘firing’ of the engine and supplies TIMER 2 with power. If a second ‘firing’ occurs within the set time constant of TIMER 1, it triggers TIMER 2. The short output pulse from TIMER 2 is the valid trigger which SET the F/F and starts the voltage sweep. (linear ramp)

While the trigger pulse from the microphone amplifier is connected to both timers, only TIMER 1 will trigger on the first ‘firing’ sound since TIMER 2 has no power yet. When the second ‘firing’ signal arrives within the set time constant of TIMER 1, TIMER 2 will trigger because TIMER 1’s output is now supplying power to it. To prevent TIMER 2 from triggering as power is applied to it, the RESET (pin 4) has a short delay. This delay can be timed to prevent TIMER 2 from triggering for a desired period.

In other words: Should a second ‘firing’ of the engine (or some other unexpected loud noise) occur earlier than it should, the circuit will not produce an output pulse. Thus, voltage sweep can NOT start and the engine remains at idle.

With Q4 ON and Q3 OFF, C25 is allowed to receive charge from the constant current source comprising of Q2 (BC327), D3 (4148), R39 (3k3), R40 (100k) and P5 (1M). (Required time constant is adjusted by P5) IC10B buffers the voltage on C25 and IC10C inverts it (gain of -1) before it is fed to voltage ADDER IC10D.

The charging time constant of the linear ramp is set to the desired maximum time for the engine RPM to reach its operating point where it delivers 50Hz at approx. 240V. The ramp voltage at that instant is saved by the T/H, just before the frequency operated switch erases the ramp by RESETing the F/F. (which in turn discharges the capacitor)

The increasing voltage slope output of IC10D is fed to the input (pin 3) of S/H IC6 (LF398) and the output of IC6 (pin 5) is connected to the control pin (5) of the VCT (IC7-C555), through electronic switch IC12A (“a” – “ax’) and buffer IC10A (LM324).

It is this voltage which determines the time constant of the VCT, the injection solenoid’s ON time and thus engine RPM.

In this configuration, Sample & Hold (IC6-LM398) is used as Track & Hold. (Using the LM398 S/H as a T/H is only a matter of control logic timing.)

As the voltage slope is gradually increasing towards the correct value for the required RPM, IC6 is in the Tracking (Sample) mode. That requires a HIGH on ‘Logic in’ (pin 8). This HIGH is supplied by the Q output (pin 13) of IC4B (4013) which was SET

 (pin 8) by C6 (0.1uF) and R12 (1M) at power-up.

Since the logic control must be HIGH for Sample and LOW for Hold, the T/H needs to “sample” continuously (i.e. track), the logic control voltage must be held HIGH while “tracking” and LOW for Hold.

This condition (Tracking or Sample mode) is maintained until the generator’s output frequency has reached 50 Hz and the voltage is approx. 240V.

This is the correct RPM for the engine.

It is the control voltage at this instant which is saved and then held by the feed back loop.

From this point on, the injection solenoid will stay open always for the same length of time, (provided that there are NO load changes) holding the engine speed constant.

The command for this action is delivered by frequency switch IC2 (LM2907- N8).

The linearly increasing voltage ramp is stopped when the output frequency of the generator has reached 50Hz. Transformer TR2 (240V/6V) reduces the generator’s output amplitude to a suitable level which is fed to the input (pin 1) of ‘frequency switch’ IC2 (LM2907N-8). A further 11:1 reduction of the signal level (amplitude) is done by R4 (100k) and R5 (10k). C1 (0.1u) is an AC coupling capacitor (blocking DC) and R1 (10k) is holding the input side of C1 at 0V (DC) potential. (These components are essential for correct operation.)

Switching frequency is adjusted by P1 (1M).

Hysteresis is provided by resistor R8 (10k). Switching point is indicated by LED1 (R). R3 (3k3) is a pull-up resistor for the comparator’s output in IC2 (LM2907N- 8). (Without it, the LED glows faintly even without a frequency input!)

P7 (10k) sets the switching voltage level for the comparator in the frequency switch IC2. P7 is adjusted FIRST to match the control voltage of the VCT for the desired RPM of the generator used. Then, the “switching frequency” is adjusted by P1 to match. (to ‘trip’)

The “stop pulse” from the ‘frequency switch’ also operates the T/H.

Since the output of IC2 (pin 5) is latching at this point, a one-shot (IC3D- 4093) is required to supply only a single pulse to RESET R-S F/F IC4B (4013) AND trigger monostable IC5A. The Q output (pin 13) of F/F IC4B is now LOW. This LOW, through attenuating resistors R14 (3k3) and R15 (1k) to the Logic input (pin 8), puts S-H IC6 (LF398) in the HOLD mode.

At the same time (as mentioned above), monostable IC5A is also triggered on the rising edge input A (pin 4). Its pulse output time constant is set by R15 (10k) and C8 (4n7) to 47µs. This output pulse is connected to the falling edge input B (pin 11) of IC5B. So, at the end of the pulse from IC5A, IC5B is triggered and its pulse time constant is set by R16 (10k) and C9 (10n) to 100µs.

This is the pulse (from pin 10) which RESET R-S latch (F/F) IC4A (through D2).

After RESET, IC4A’s output is LOW which cuts OFF Q4 and turns ON Q3. Q3 (when turned ON) represents a short across capacitor C25 and thus erases the voltage slope previously created. (Note, however, that before that occurs, the needed voltage has already been saved by putting S-H IC6 in the HOLD mode) So, once again, the role of IC5A & B is to create the necessary delay between saving the wanted voltage and then erasing the slope.

The circuit is now ready for an other “start engine” process.

VCO for testing the servo loop (generator output/automatic setting of RPM):

The VCO section of a 4046 PLL (IC13) is used for this function.

First, the generator/engine RPM (where it produces the correct voltage and frequency-for example 240V/50Hz) must be determined. (This will depend on design/type/size/model/manufacturer, etc.) From this RPM, the fuel intake time is calculated as follows:

RPM (crank shaft) divided by two (2) to get engine work cycles in minutes. Divide this by 60 to get ‘work cycles’ per second. Now take the reciprocal (1/xx) of this number which is TIME (in seconds) for one (1) work cycle. Divide this number by 4 since the engine’s fuel intake lasts for only ¼ of the work cycle. This is the max. time the fuel intake valve is open.

Example: 3000 RPM : 2 = 1500 ‘work cycles’ (per minute). 1500 : 60 = 25 ‘work cycles’ (injections/ignition pulses) per second. Reciprocal of 25, (1 : 25 = 0.04) seconds for each ‘work cycle’ (40 ms) 0.04 : 4 = 0.01 seconds (10 ms) maximum fuel intake valve opening time.

The next step is to set the VCT control voltage to produce the required injection pulse duration. For the example above, the control voltage needs to be 6.5V (approx.) to deliver 10 ms injector ON time. Note that the VCT in this design, with the timing component values shown on the circuit diagram (C27 – 3n3, R50 – 3M9) produces the following pulse widths/control voltages:

1 ms/~0.8V 8.33 ms/~5.7V 10 ms/~6.5V 15 ms/~8.3V 20 ms/~9.5V

Thus, this circuit can also be used for a low RPM (1500) engine/generator by supplying the VCT with a 9.5V control voltage to deliver 20 ms wide injection pulses.

OK. So how do we set the control voltage range for the required RPM? By adjusting the ‘trip point’ of the comparator in IC2, using P7 (10k).

Then, adjust P2 so the VCO’s output frequency is 50Hz with the desired control voltage applied to pin 9 (TP8).

The VCO’s output (pins 3+4) is connected to the ‘AC input’ of frequency switch IC2 (LM2907-N8) and its Voltage Control input (pin 9) is connected to the VCT’s control voltage (pin 5 of IC7-C555).

With C13-1µF, R19-2M2 and P2-50k (approx. 36k), we get 8V control voltage for 50Hz.

The role of test (service) LEDs are as follows:

LED 1 is normally OFF. Frequency switch IC2 (LM2907-N8) turns it ON when the generator output frequency has reached 50Hz. It remains ON as long as the frequency is about 50Hz. It turns OFF if the frequency falls a certain amount (Hysteresis) below 50Hz.

LED 2 indicates the state of S/R-F/F IC4B. Since this F/F is SET at power-up, LED 2 will be ON. Frequency switch IC2 turns it OFF when the 50Hz limit is reached.

LED 3 indicates the LOUD sounds of the engine firings. It comes ON momentarily at system power up, then it is OFF. Once the engine has started, it will stay ON as long as the engine is running.

So, LED 1 is OFF and LED 2 is ON after power-up. As the 50Hz limit is reached, LED 1 turns ON and LED 2 turns OFF but note that during normal operation they are never on at the same time!

LED 4 indicates the state of R/S-F/F IC4A. It is OFF after power-up since IC4A is RESET. Since IC4A is repeatedly SET by the sound timer IC9, LED 4 stays ON. It turns OFF when power is removed and then re-applied (power-up).

Note that when IC4A is repeatedly SET, it removes the short across capacitor C25. This allows it to be charged by the constant current source and the voltage slope is created once again. However, this no longer matters as the wanted control voltage has been previously saved and a short has been placed across the HOLD capacitor.

Method of holding the RPM constant with varying load:

Suppose the correct RPM (for a 50Hz output frequency) was reached with only a light load (or no load) on the generator’s output. Logically, it can be assumed that when load is applied (or increased), the engine should require more fuel to maintain the same RPM.

Since the VCT controls the injection solenoid’s timing, its control voltage must be increased by the feed-back loop to adjust the solenoid timing to maintain the 50Hz frequency output. (If the load is reduced, solenoid opening time must be reduced.)

However, the initial fixed timing output generated (and saved) by the auto- RPM circuit for the correct speed is to be used as a ‘reference’ and a correction voltage added to or subtracted from it.

A Frequency-to-Voltage (F/V) converter, as the name implies, generates a frequency dependent voltage.

Adding a comparator to the F/V converter creates a frequency switch as well. In the frequency switch (IC2 – LM2907-N8) a frequency dependent voltage is also available at pin 3. To avoid loading this voltage (which represents the frequency) it needs to be buffered. This is the control voltage to the VCO (IC13-4046).

If, due to increased load the engine slows down, the generator’s output FREQUENCY goes down. So does the output voltage of the F/V converter. To bring the output frequency up to the correct point again, the engine needs more fuel which means longer opening time for the solenoid. If the load is reduced, the engine will speed up, the generator’s output frequency goes up. To bring it down to the correct level, the engine needs less fuel.

In other words: the feed-back needs to act in opposite direction. Increased load – decrease in frequency = need to increase solenoid ON-time.

Feedback loop:

In order to develop/design a feedback loop, the engine’s function needed to be temporarily substituted with a circuit who’s frequency can be altered. Since the test VCO (4046) in this circuit is already voltage controlled, to change its frequency, (without altering its present control voltage) the R- C time constant of the oscillator must be altered. When altering its frequency this way, the change in its control voltage can be observed.

The voltage from the F/V converter (IC2- pin 3) is filtered by capacitor C4. By reducing its previous value from 1uF to 0.1uf, a LARGE ripple is produced.

However, using a capacitance multiplier (IC1B), a 10x decrease in response time is gained as this stage has a gain of 1000 but a capacitance value of only 4n7 (C30). It eliminates the ripple, without affecting response time!

This (now ripple free) voltage is fed to the inverting (-) input (pin 6) of error amplifier IC11B. The non-inverting (+) input (pin 5) has an adjustable voltage divider which sets the reference point for the loop.

It is adjusted to be the same as the VCT’s control voltage for 50Hz.

During the time the auto RPM circuit is slowly bringing the RPM to the correct point, the error amplifier is also operating but its output is not used yet.

As soon as the correct RPM is reached, the ‘saved’ voltage is switched OUT and the output of the feedback amplifier is switched IN and is now used as the control voltage for the VCT!

Change-over switch ‘A’ is used to first CONNECT the voltage control line [control pin (5) of the VCT & VCO (pin 9)] to the output of the S/H (pin 5- IC6), then DISCONNECT and CONNECT the voltage control line to the output of the ‘error amplifier’ IC11B instead.

The ‘common’ (“a”) of switch ‘A’ (pin 14 of IC12A – 4053) is connected, through buffer IC10A, to the control pin (5) of the VCT (IC7-C555) AND the voltage control input (pin 9) of the “50Hz test VCO”. (IC13-4046).

 “ax” (pin 12) is connected to the output (pin 5) of S/H-IC6-LF398.

 “ay” (pin 13) is connected to the output (pin 7) of  ‘error amplifier’

IC11B (LM358).

When the logic control for ‘A’ is LOW, “a” is connected to “ax”. [Output (pin 5) of S/H (IC6) is connected to the control (pin 5) of the VCT (IC7) AND voltage control pin 9 of the VCO (IC13)]

When the logic control for ‘A’ is HIGH, “a” is connected to “ay”. [The voltage control line (VCT & VCO) is connected to the output of ‘error amplifier’ IC11B.]

Further, IC12C is used to put a short circuit between the capacitor terminal (pin 6) and 0V (Ground), preventing the charge on the Hold capacitor to obtain a negative voltage value! [Note that this circuit requires a dual (±12V) power supply.]

After a short delay of at least 1ms, [created by R26 (1M) and C16 (0.1u)], logic HIGH is also applied to switch ‘C’. This delay is necessary to allow time for the transfer to take place, before placing the discharging switch ‘C’ across the Hold capacitor. [“c” (pin 4) is connected to Ground (0V) and “cy” (pin 3) is connected to the Hold capacitor terminal (pin 6) of S/H (IC6)]

Note that the output impedance of the S/H amplifier is very low (less than 4 ohms) but the ON resistance of the switch is typically about 100 ohms (at Vdd = 12V). The overall impedance of this point is still very low so it is extremely unlikely to pick up interference.

In other words: With logic HIGH applied to both switches (IC12A & C), the output of S/H-IC6 is disconnected from the control pin of the VCT and connected to the output of ‘error amplifier’ IC11B instead.

Now, the control voltage to the VCT is supplied by the feed back loop.

The output of voltage follower (buffer) IC10A is LOW impedance which drives the Control Voltage pin of the VCT (IC7 – C555).

Note also that the output of the S/H will gradually diminish due to “droop” and eventually becomes 0V. (‘Hold’ capacitor completely discharged)

However, the LF398 Sample & Hold IC, in the HOLD mode, has a nasty habit of discharging the Hold capacitor not only to 0V but continue down to about -10.76V!!! In this application, this action plays HAVOC with the circuits that follow!

The remedy is put a short (switch IC12C) across the Hold capacitor just after the ‘saved’ value has been transferred to the feed back loop.

After system power-up, the electronic switches (IC12A & C) controlling the transfer of the saved control voltage, need logic LOW.

To hold switch ‘A’ LOW at power up, R25 (100k) is connected to Ground. [C (pin 9) is also connected to Ground through R26 (1M) and R25 (100k)] However, the turn-ON (by F/F – IC3A & B latching bistable, through D1) of IC12C needs to be delayed relative to ‘A’ to allow sufficient time for the ‘transfer’ of the control voltage to take place. This is achieved by using a simple RC network of C16 (0.1u) and resistor R26 (1M).

When the engine is running and has reached the correct RPM, a momentary (100µs) pulse is supplied by IC5B to reset R/S latch IC4A (which erases the voltage ramp). After this action, control switches IC12A & C need continuous logic HIGH as long as the engine is running!

To create a continuous logic HIGH from a single pulse, a latching bistable (F/F) is needed. IC3 (4093) is a quad, 2 input NAND Schmitt trigger. IC3D is used as a ‘one shot’ to trigger IC4B and IC5A. (IC3C is not used)

IC3A & B form a latching bistable (F/F).

Power-on RESET (falling edge) is done by R20 (1M) and C14 (0.1µF) and is SET by also a falling edge pulse from IC5B’ [pic] output (pin 9) which is normally HIGH. Output of the latch is taken from pins 3 & 6 and through D1 (4148) is connected to the logic control pins of electronic switches IC12A,B,C.

Finally, perhaps I should mention that the “autorpm” is necessary for yet another reason:

Since the electrolyzer’s power supply is set for MAXIMUM power, the generator’s engine may not require all the gas continuously. Therefore, the pressure regulator will cut power OFF when the set pressure is reached, turning it ON again when pressure drops. (Hysteresis) In other words: the power to the electrolyzer is either fully ON or fully OFF. For an electrolyzer current set to 10A (2400W), this means that the LOAD on the generator’s output (in addition to other loads) will abruptly vary by 2400W!!

Thanks to the ‘autorpm’ control circuit, RPM remains steady.

Les Banki (Electronic Design Engineer) Water Fuel & LBE Technologies

Circuit description for phase control power supply.doc

Circuit description for power supply and control of HIGH VOLTAGE, multi-cell electrolyzers.

This circuit was developed/designed for powering high voltage, multi cell electrolyzers. (However, it can also be used as a high power light dimmer, motor speed controller, etc.)

Through a bridge rectifier, the phase controlled mains AC power is applied DIRECTLY to the electrolyzer! (NO transformer!)

Some may not like the idea but the advantages of a transformer less circuit for this purpose FAR outweighs the disadvantage of a LIVE electrolyzer. (I am talking about a 120-150 cell, 240V unit.) LIVE or NOT, you keep your hands OFF a 240V device, AC or DC! Period!

The only technical ‘problem’ I have encountered is the necessary isolation of the current limiter control signal from the ‘live’ mains. (More about this later.)

Zero crossing detector:

The non-inverting (+) input (pin 5) is biased by R3 (100k) and R4 (470ohm) to about 56mV (0.056V). Narrow, positive pulses appear at IC1B’s output (pin7). These are fed through R5 (10k) to discharge switch transistor Q1’s (BC547) base. A constant current source [comprising of Q2 (BC 327), LED1, R6 (15k) and R7 (3k3)] is continuously charging capacitor C1 (0.1u). Q1 periodically discharges C1, thus a rising LINEAR saw tooth waveform is formed. (the fast discharge part of the waveform occurs during zero crossing of the sine wave)

IC1C is a unity gain ‘voltage follower’ (low output impedance buffer). The rising saw tooth needs to be inverted so that minimum control voltage to the comparator gives minimum power output and maximum control voltage results in maximum power. (This is required by the feedback loop.)

The output of the unity gain buffer IC1C (pin 8) is connected to the inverting (-) input (pin 9) of the unity gain (-1) inverter IC5C (LM324).

[Note that IC1 (LM324) is running on a single supply while IC5 (also LM324) needs a dual supply. (±12V)]

Its non-inverting (+) input (pin 10) is connected to ground through R12 (5k1). This stage (IC5C) also performs level shift. (R9-68k, P2-10k, R11-10k) (P2 is adjusted so the bottom tip of the now falling saw tooth is at 0V.)

The output of IC5C (pin 8) is connected directly to the PWM comparator’s (IC1D) inverting (-) input (pin 13).

The non-inverting (+) input (pin 12) is the output power control pin. (TP16)

It is connected to the output (pin 7) of error amplifier IC5B, through R8 (10k).

The output of the PWM comparator IC1D (pin 14), through R14 (10k) and R15 (1k), drives the Gate of Q3 (2N7000). D3’s Anode is connected to the junction of R14 & R15.

Power OFF/ON control is achieved by applying a LOW signal to the Cathode. This removes the drive signal from Q3’s Gate. (This is used for timer operation and pressure control.)

The LED in the opto coupler MOC3021 (IC2) is connected between the positive supply rail and the Drain of Q3, with resistor R16 (470R) limiting the current.

The remaining components around the opto coupler and the Triac are for limiting dv/dt, suppression of transients and to reduce RF interference. (C3, C4, C5, R17, R18, R19, VDR1 and L1) The Triac used here (see circuit diagram) is BTA26-600B, rated at 26A, 600V.

Now to the current limiting:

Current limiting a HV AC phase control circuit is a bit “tricky”! This is due to two main factors: 1. The need for isolation from the 240V mains voltage. 2. Since ‘SCR waveforms’ are distorted sine waves, measuring current (voltage) with reasonable accuracy can only be done by using a TRUE RMS – DC converter.

Further, unlike ordinary negative feedback, it is needed ONLY for limiting and thus acting only in one direction: DOWN. In other words: once the desired current limit is set, the feedback is to act only to REDUCE the current to the pre-set level. This means that if a connected load draws LESS than the set limit, the feedback is NOT operating and full power is applied to that load. Should the load try (for whatever reason) to draw more current than the set limit, the feedback (error amplifier) starts operating in order to maintain the set current.

As mentioned above, the error amplifier’s output (IC5B-pin7) is connected to the non-inverting (+) input (pin 12) of the PWM comparator (IC1D) through R8 (10k).

This is the way it works:

The error amplifier (IC5B) needs a reference voltage on its non-inverting (+) input. The reference voltage corresponds to the SET current limit.

It is created by the low output impedance unity gain buffer IC1A, R28 (4.7k) and P1 (10k). Its output (pin 1) is connected [through R27 (10k)] to the error amplifier’s (IC5B) non-inverting (+) input (pin 5). Since the P-P amplitude of the saw tooth waveform is about 8V, R28 (4.7k) limits P1’s adjustment range to just over 8V. (0 – 8V)

Note: 0V on the PWM (IC1D) comparator’s control pin (12) corresponds to NO power and 8V (or higher) means FULL power which this phase control unit is able to deliver.

Load current is measured by a Hall effect current sensor (ACS712T-ELC-20A) which is a 20A device, gives 2.1kV RMS voltage isolation and has a sensitivity of 0.1V/A. Thus, for 10A, it’s output is 1V.

It is placed in the AC loop, between the Triac and the bridge rectifier for the cell.

Its output is AC coupled (C8-0.47uF) to the input of the RMS to DC converter AD737, (IC4) through a 5:1 attenuator R20 (300k) – R21 (75k). This is due to the 200mV input limit for the AD737 in this configuration.

Note that the output of the AD737 is negative.

Inverting amplifier IC5A (gain of -50) converts this to positive polarity and a correct amplitude needed by the error amplifier IC5B. (Gain of -5 is to compensate for the 5:1 attenuation, plus a further gain of -10 to bring the signal to the approximately same level as the control signal for the PWM comparator.)

OK. As long as the error amplifier’s inverting (-) input is LOWER than the reference voltage on its non-inverting (+) input), it goes into saturation. Then its output is about +10.75V. Since this output is connected (through R8-10k) to the PWM comparator’s control pin (12), power output is at MAXIMUM.

Now, should the load try to draw more current, the error amplifier’s inverting (-) input will exceed the reference voltage on its non-inverting (+) input and the amplifier comes out of saturation and adjusts its output voltage to maintain the set current limit.

An ‘old fashioned’ 10A moving coil meter is used to monitor the current because of its almost perfect integration! (which is due to mechanical inertia of its coil movement)

Adjust the desired current limit as follows:

Turn the current limiter control P1 to MINIMUM - fully anti-clockwise. Apply power to the system. Connect the desired load. While watching the Meter, start turning P1 clockwise, SLOWLY. Observe the rising current.

At the point where the current no longer increases, you have reached the natural current draw for that particular load. Now back down the setting to the point where the current stopped rising.

If you connect a very HEAVY load (which would naturally draw a LOT more current than what you want), as you turn the control, the meter reading will increase all the way to the limit of the meter (10A) AND what the system is designed for. (also 10A) Naturally, you can stop the current at ANY level, up to the maximum.

System fuses are rated at 15A (50% overload) just in case there is a mishap!

Note that this power supply requires a MINIMUM of 60-100W load to operate.

While testing a ‘stand alone’ unit (powered by its own low voltage power supply), I discovered that if the load is already connected when power is applied, there is NO regulation, which means FULL power is applied!

(Without regulation, the power drawn by the electrolyzer might be FAR higher than desired and would at least result in blown fuses!)

The reason for the lack of regulation at power-up is that the power supply lines are not yet stable, thus the circuit cannot be expected to work properly!

ONE way to fix this is to delay the Triac control pulses for 1–2 seconds at power-up. This will allow the power supply rails to stabilize.

There are many ways to introduce such a delay. I choose a small MOSFET (2N7000) for the clamping of the pulses, controlled by an RC network with the correct time constant for the delay. It also functions as a fast ON/OFF switch, controlled by the pressure regulator circuit and/or the timer.

This stage works as follows:

MOSFET Q4 (2N7000) is connected between the input of optocoupler IC2 (MOC3021) and Ground. When its Gate voltage is close to zero (0.12V), Q4 is OPEN circuit. If a voltage greater than its threshold (about 2V) is applied to its Gate, Q4 turns ON and shorts the control pulses to Ground.

Resistors R29 (100k) and R30 (10M) form a voltage divider which keeps Q4’s Gate voltage at around 0.12V. Capacitor C18 (3.3uF) is connected across R30. Before power-up, C18 is held discharged by R30. At power-up, for a moment, a discharged capacitor acts like a short circuit. Since one side of C18 is connected to the +12V supply rail, the other side will also be at almost the same voltage but only momentarily.

Thus, for a brief period of time, Q4 is turned ON and shorts the control pulses to Ground. When C18 has charged up, Q4 turns OFF and circuit operation returns to normal.

Q4 is also used as an electronic switch. Through R31 (1k), D4 & D5 (both 4148) and SW1 (optional, manual override switch), positive voltage is applied to Q4’s gate which turns it OFF.

D3’s function is to isolate the R30-C18 network from the applied control voltage pulses. It allows rapid ON/OFF control, independent of the power-up time constant of R30-C18.

Note that since the capacitance of C18 is fairly high (3.3uF) and the discharge resistor R30’s value is large (10M), after power is turned off, it takes 15-30 seconds to sufficiently discharge C18.

Les Banki (Electronic Design Engineer) Water Fuel & LBE Technologies

Circuit description for phase control power supply for HV multi-cell electrolyzers.doc