4K EFI Test Rig - IAC Valve

[Banjo]

Most of You are well aware of my interest in ECUs & trigger systems, to produce the ultimate power & reliability, of our K series engines, designed & built over 50 years ago.  I've been building & testing, various trigger wheels: & have recently moved onto optical trigger discs, built inside existing "gutted" K series distributors.  My final experiment, will be with a flywheel located trigger wheel; where I initially started years ago, counting flywheel ring gear teeth; until I realised there was no ECU capable of counting odd numbers of teeth.  (not the case these days)

Simple Flywheel 36:1 trigger system

In doing these experiments, I've not wanted to reinvent the wheel, so to speak; or make mistakes or, "go down rabbit burrows" ; only to have to retreat.  However, this research, takes a lot of time on the net, reading articles & posts on other forums, about this subject. Recently, I've been getting AI to do this research for me; & to date, I've not found anything it has written, that was not accurate. 

So today, I asked AI to write a short article about the practicality of the number of teeth on a trigger wheel, & why the industry has standardized, on teeth numbers between say 24 to 72 teeth per crankshaft revolution. 

just look at the sheer numbers of different mechanical arrangements people have come up with.

Trigger Wheels Pictorially

Below, is what AI produced . . . . .

____________________________________________________________________________________________________

The Automotive ECU Crankshaft Trigger Wheel: Design, Functionality, and Optimal Tooth Count

An In-Depth Guide to Crankshaft Trigger Wheels and Tooth Count Selection for Engine Control Units

Introduction

The heart of an internal combustion engine is a symphony of precisely timed events, each dependent on accurate position and speed measurements of key components. At the center of this orchestration lies the crankshaft trigger wheel, a seemingly simple but crucial element that enables the engine control unit (ECU) to synchronize ignition timing, fuel injection, and variable valve control. In this article, we will explore the design and function of automotive crankshaft trigger wheels, examine the importance of tooth count, and discuss what range of teeth per revolution proves most practical and effective in modern vehicle applications.

Understanding the Crankshaft Trigger Wheel

The crankshaft trigger wheel, sometimes referred to as a reluctor wheel or timing wheel, is a toothed disc mounted to the engine's crankshaft. As the crankshaft rotates, the trigger wheel passes by a sensor—typically a magnetic or Hall effect sensor—which reads the passing teeth to generate electronic signals. These signals are interpreted by the ECU to determine the crankshaft’s position and rotational speed in real-time.

Function in the Engine Management System

The main roles of the trigger wheel are:

·         Crankshaft Position Sensing: The ECU uses the signals to determine the crankshaft's angular position. Precise knowledge of this position is vital for optimizing ignition and fuel injection events.

·         Engine Speed Measurement: By counting the number of teeth passing the sensor per unit time, the ECU can calculate the engine's RPM with high accuracy.

·         Synchronization: The trigger wheel often includes a unique pattern—commonly a missing tooth or unique gap—that allows the ECU to determine the crankshaft’s absolute position within its cycle, differentiating between top dead center (TDC) for cylinder 1 and other rotation points.

Design Variations of Trigger Wheels

Trigger wheels can be found in several configurations and with different numbers of teeth, depending on the engine's complexity and the precision required by the ECU. The most common designs include:

·         Single-tooth Wheels: Rare today, these provide only a basic signal per revolution and lack precision for modern high-performance or emission-controlled engines.

·         Multi-tooth Wheels with Missing Teeth: Widely used, such as the 36-1 wheel, which has 36 equally spaced teeth with one missing. The missing tooth provides a unique positional reference.

·         Dual-track Wheels: Some advanced systems use two concentric tracks of teeth for greater positional accuracy or for distinguishing between crankshaft and camshaft positions.

Sensor Types

Most trigger wheels work with either variable reluctance (VR) sensors or Hall effect sensors:

·         VR Sensors: Generate an alternating voltage as teeth pass by a magnetic pickup.

·         Hall Effect Sensors: Use an electronic effect to produce a digital signal, and are less susceptible to electrical noise.

The Importance of Tooth Count

The number of teeth on a trigger wheel fundamentally affects the ECU’s ability to interpret engine speed and position:

·         Resolution: More teeth per revolution increase the position data’s granularity, allowing finer control of ignition and fuel events.

·         Signal Frequency: Higher tooth count means more signals per revolution, which can improve accuracy, but also increases the processing demand on the ECU and the system’s susceptibility to electrical noise or missed pulses.

The choice of tooth count, therefore, is a critical balance between achieving sufficient resolution and maintaining reliable signal integrity and manageable ECU processing overhead.

Practical & Effective Tooth Count Ranges

Modern automotive practice has settled on specific ranges of teeth per revolution as being optimal for most applications. Factors influencing this choice include engine speed range, ECU processing power, required precision, and physical constraints.

Common Tooth Count Configurations

·         12-1 and 24-1 Wheels: Used in many early electronically controlled engines, these provide a moderate level of resolution suitable for simple sequential or batch-fire ignition and fuel control. With 12 or 24 signals per revolution, they offer a balance between precision and simplicity.

·         36-1 and 60-2 Wheels: These are now considered standard in many OEM and performance applications. A 36-1 wheel has 36 teeth with one missing, while the 60-2 has 60 teeth and two missing. The missing teeth create distinct reference points, enabling the ECU to quickly synchronize with the crank’s position.

·         More than 60 Teeth: Rarely used in automotive crankshaft applications due to excessive signal frequency at high RPM, which can overwhelm sensor and ECU capabilities.

Ideal Range for Most Applications

For the majority of modern engines, the most practical and effective range for a crankshaft trigger wheel is typically 24 to 60 teeth per revolution, often with one or two missing teeth for synchronization. Here’s why:

·         Resolution vs. Processing: This range provides sufficient resolution for precise ignition and fuel injection control, even at high engine speeds, while not overburdening the ECU with an excessive number of signals.

·         Signal Integrity: With too few teeth (less than 12), the ECU may struggle to interpolate crank position accurately at low speeds or during rapid acceleration. With too many teeth (more than 60), the risk of missed pulses and signal noise increases, especially at high RPMs.

·         Industry Practice: The 36-1 configuration is especially popular among OEMs and aftermarket ECUs for its reliability, ease of synchronization, and proven performance in diverse environments.

Special Considerations and Advanced Applications

Some high-performance engines, racing ECUs, or specialized applications may deviate from the standard range. For example:

·         Motorsport: Racing engines sometimes use higher-tooth-count wheels in conjunction with advanced ECUs capable of processing very high-frequency input to enable extremely precise control at extremely high RPMs.

·         Low Cylinder Count Engines: Engines with fewer cylinders or unusual firing orders might use custom wheels to ensure the ECU can distinguish cylinder phases effectively.

·         Redundancy and Diagnostics: Some trigger wheels incorporate multiple tracks or additional reference marks to enhance diagnostic capabilities or provide backup signals in case of sensor failure.

Installation and Machining Constraints

Physical factors also play a role in tooth count selection:

·         Wheel Diameter: A small diameter restricts the maximum number of teeth that can be reliably fabricated and sensed. Too many small teeth could lead to manufacturing defects or sensor misreads.

·         Mounting Location: Space in the timing cover or crank pulley area may limit wheel size and, thus, tooth count.

·         Sensor Technology: The sensor's resolution and sensitivity will also dictate the practical limits for tooth count and spacing.

Conclusion

The crankshaft trigger wheel is a vital component in modern engine management systems, translating the mechanical motion of the crankshaft into precise digital signals for the ECU. Selecting the optimal tooth count is a balancing act: too few teeth and the system loses precision; too many and it risks complications from noise, missed pulses, or processing overload. For most automotive applications, the most practical and effective range is 24 to 60 teeth per revolution, with configurations like 36-1 and 60-2 leading the way due to their proven reliability and performance across a broad spectrum of engines.

Advancements in ECU and sensor technology may gradually shift these boundaries, but for now, adhering to this range ensures robust, accurate, and efficient engine control in both OEM and aftermarket contexts. Whether building a custom engine or tuning a contemporary performance car, understanding the role and design of the crankshaft trigger wheel—and the best practices for tooth count—remains foundational to achieving optimal results.

P.S.  This was produced by AI in about 5-8 seconds.   It used the AI dialogue line just above the start of a new Microsoft Word document.

P.S.S.  I'm surprised; that with small micros, becoming faster & cheaper; that someone simply has not just built a tooth counter program & hardware, so that the main ECU processor does not have to be interrupted, every time a tooth rising or falling edge appears.  The line above in the AI's conclusion; "Advancements in ECU and sensor technology may gradually shift these boundaries" ; may be alluding to just that possibility.

Cheers Banjo

 

 

Edited September 8, 2025 by Banjo

[Banjo]

After a lot of time spent, investigating different triggering systems for my 5K engine, I've come up with a simple arrangement, by fitting an optical CAS distributor module, inside a K Series Bosch dissy, which is a lot wider than the more common Denso model dissy.

I'm really happy with the Speeduino's performance on the bench, so it is now time to fit it to my  4K-U in my KE30, 2 door sedan, daily drive.

I've just been checking all the external temp sensors on the car today, to hook into the ECU. However, I've just realised, my KE-30, doesn't have a TPS on it.  It will need one, for the Speeduino to function properly.

I have a TPS of  "unknown origin" hooked up on my 5K engine on a test bed in the garage, but don't want to pilfer that. 

It's a bit crude, but it has worked well on the 5K on the test stand, for a few years now.  This olde TPS rotates from the center of the carby butterfly shaft.

 

 

Just looked on ebay & there are plenty of TPS units there for many makes of engine. They don't however indicate which direction they rotate in.

Currently I just want to test this ECU, using ignition only. Once that's sorted, I'll add the 7K EFI inlet system I have, complete with injectors. 

Has anyone made up a TPS arrangement for a K Series carby engine, that works OK, & could pop a photo up, of how they have mounted it, & what TPS unit they used.

The stainless steel timing disc I currently have, has 24 camshaft trigger pulses per revolution, & one CAS pulse. I would like a S.S. disk with 36, 60 or 72 holes around the disk. Chased everyone here in Australia, & nothing available; so might have to bring one in from the USA. 

There is one, that was used on Nissan vehicles which has 360 slots punched around the perimeter. Heaven knows how they do it; as the disks are only 50mm in diameter.

It anyone has one of these disks they want to part with, for $$$, then give me a yell.   The 360 slot disc, on the camshaft, would provide a crankshaft RPM resolution of 2 degrees, whereas my current 24 tooth disk, provides, just 30 degrees. If I do get hold of a 360 hole disk, I'll have to divide down the frequency a bit, as Speeduino ECU setup, will not accept a number greater than 255, for the number of trigger teeth.

Cheers Banjo

Edited October 13, 2025 by Banjo

[Banjo]

Altezzaclub put me onto a Camry TPS that He has used; & that are reasonably priced on ebay.

The only disadvantage is, that they are not rotated from the center.

Center driven ones are available. This Honeywell TPS I found online, has a 6mm dia. shaft in the middle of it; however, at $ 220.00 ea., its a bit rich.

Link to Above Honeywell specs.

Big issue mounting a TPS on a carby K series engine; as it has to sit out there in the middle of nowhere.

I made a mounting bracket out of 3mm thick aluminium, as below.  It is important, that the center of that big round hole, where the Camry TPS is mounted, is perfectly in line with the carby butterfly shaft center.

 

The two top holes arrowed, attached to the top of the carby.  The bottom single mounting point, was the extension on the inlet manifold, which normally, will support the standard air filter housing leg.

 

Add an EFI Solutions 4W plug & harness wire back to the ECU under the dash; & it looks like this.

Works really well !  Only three of the four wires are used in a standard TPS, ECU installation.

Twas time to eventually fit my COPs mounting frame, to ensure I can fit the Bosch dizzy in there, underneath it, as I'm now not using the trigger disk, on the crankshaft pulley.

So all good, but a couple of return coolant lines to reroute first.

Getting close to the first fire up, as I now have an acquired a Nissan CAS dissy, to fit into a Bosch K Series dizzy housing, which Altezzaclub sourced at His local wreckers, ThankYou !

This particular one has a 54mm dia. S.S. disc, with 360 slots around the edge.  That will provide excellent resolution,; for the ECU, for the position of the crankshaft, at any time.

Cheers Banjo 

 

 

 

Edited October 3, 2025 by Banjo

[Banjo]

Quote

This particular one has a 54mm dia. S.S. disc, with 360 slots around the edge.  That will provide excellent resolution,; for the ECU, for the position of the crankshaft, at any time.

Well sounded like a really good idea, at the time; but the realisation of it, was like going down a "rabbit burrow"; with so many twists & turns. The Nissan vehicles were one of the first to introduce the 360 slot disc, although it had to be 54mm in dia.; whereas all the others; including my 24-1 S.S. disc, were 50mm in dia. The optical reader module was also slightly larger, to accommodate the 54mm disc.  This posed a problem; as the module for the 50mm disc, just fitted inside a "gutted" the larger K Series 'Bosch dizzy housing, so how was I going to squeeze in the 4mm larger Nissan module, therein.

However, I was buoyed & encoraged by Taz-RX, who had previously done this.

https://www.rollaclub.com/board/topic/10940-sis-5kte-ke55/page/67/

It required a lot of cutting & filing; drilling & tapping; to generally get the module, concentric around the center shaft.  It also required a completely new base plate to be fashioned, to attach the optical module to.

 There are two critical measurements, when attempting this project. One is that the center shaft of the dizzy, is completely concentric with the big hole it passes through, in the module. This results in all the slots in the punched S.S. disc, passing directly under the infra-red source & reader head, on the top right corner of the above pic.  

The second critical dimension is that the disc pass clearly through the middle of the slot in the reader head. The gap in the reader head, is only 1.5 mm wide. The disc is 0.5mm thick. That only leaves a gap on either side of the S.S. disc of 0.5mm.

So just when I got it all right, & was fitting the dizzy cap I had made; I found the reader head on the module , fouled with the inner edge of the dizzy cap; requiring some delicate work with a rotary cutter & drill.  

Anyway; looked good when finished & now I had to make it work.

I then made up a new stand for the dizzy, so I could test it, on the bench, as with 360 slots around the edge of the S.S. disk, You don't need any vibration to upset the "cleanness" of the delivered pulses to the ECU.

I was dissappointed to find that the Speeduino firmware does not cater for the Nissan 360 slot disc I had, which had 4 off CAS slots per revolution. 

One slot which is the true CAS pulse; is wider that the remaining three (3); which are all the same width. 

Looked like I might have to revert to using a /2 frequency divider, to reduce the effective trigger pulses to 180 per camshaft revolution; & also block off the three (3) CAS slots I did not require.  (tiny little slivers of black insulation tape worked well) 

I then came up with an idea of cheating, by setting the firmware trigger settings, to advise the ECU that this dizzy's outputs where generated by the crankshaft; & not the camshaft.  That also allowed the teeth per revolution to be set to 180, as Speeduino firmware will not accept a trigger number greater than 255.

Well it worked, & I now have some serious testing to do on the bench, to ensure that it works under all conditions, & scenarios.  

P.S.  Not having to divide the trigger output of the module by two (2); resulted in a trigger resolution of 2 degrees, which is excellent.

Cheers Banjo

 

 

 

Edited October 13, 2025 by Banjo

[Banjo]

A number of people have sent me PMs, asking about comments I have made in the past, about errors & stability of pulses generated in my experiments, from various trigger wheels & CAS arrangements.  These comments are only related to those, who want to control their ignition and/or injection, using a "sequential system". (injecting & firing one cylinder at a time;  (1-3-4-2 )

This results from the crankshaft having to rotate through 720 degrees, (2 revolutions); to fire all cylinders in sequence.  In "wasted spark" mode, two cylinders are fired together every revolution. (1 & 4, on the first crankshaft revolution, & 2 & 3, on the second crankshaft revolution, although there were some olde British designed & manufactured 4 cylinder engines which had a firing order of (1-2-4-3).

https://www.youtube.com/watch?v=ZCUftAm3Qag

As any engines get older & older, they inevitably wear; & timing chain/sprockets & timing belts stretch & loose.  This results in timing errors, in anything that is "camshaft driven; such as valves, & distributors, which require precise timing.  It is impossible to correct valve timing, except by replacing camshaft sprockets, chains, & tensioners.  Other items that are camshaft driven; items like the fuel pump, & oil pressure pump, don't need precise timing considerations at all. 

Injection & ignition timing is however critical, to get the best performance from your engine. There are a couple of places where this slop, wear, & inaccuracy of timing can occur.  One is obviously the sprockets on the crank & camshaft; & the chain that joins them. The distributor, is driven at right angles to the camshaft from a set of helical gears & a dog joint to the oil pump, in our K Series engines. The dog joint & slot is a wear point. Next time you have you distributor out, take a look at it, & you'll see the tell wear mark, on the bottom of the dizzy shaft.  However, this dog wear does not effect the timing; as the dizzy is driven by the meshed gears on the bottom of the dizzy shaft, & the camshaft.

I recently did a test on my 5K engine, which had the sprockets & chain replaced about 4-5 years ago. I jammed the "locked" distributor internally, so it could not turn at all.  I moved the crank in a clockwise direction & took a reading, from the large degree wheel attached to the flywheel; & noted the reading. I then turned the crank in the opposite direction; & again took a note of the degree wheel reading. I then subtracted one from the other, & the result was nearly 4 degrees. Bear in mind, that this was a static test, in that the engine was not running, & there was no oil  pressure to drive the timing chain tensioner, to take up the "slop" in the timing chain. Dynamically, I'd expect the error would be about say 2 degrees.  It could be much greater, in a tired olde K Series engine.

A better way to see this "slop" in the camshaft chain & sprocket setup, is to run the engine at idle, & watch the timing light, illuminate the TDC mark on the camshaft sprocket cover. If it is rock solid, then your engine is in a good state; (timing wise) If it is wandering about, back & forth; then there is wear.

I have come up, with a system, that overcomes  that timing error, in a "sequentially fired & injected engine".

It does this by synchronizing the varying C.A.S. (camshaft angle sensor/pulse) which is moving around; with a crankshaft generated pulse, which cannot move around, as it is rock solid, because it comes from either the crankshaft pulley, or flywheel; both of which, are one with the crankshaft.

The following rough sketch, describes this graphically.

 

The top line (A) of the C.A.S  (Crank Angle Sensor) pulses are created by, & generated in the distributor. 

Timing wise they will move around relative to the crankshaft, as depicted on line "A". The camshaft angle sensor (CAS), will produce this once every two revolutions of the crankshaft.  The second line "B" is the crank angle sensor, which is rock solid, as it cannot move around horizontally, but does turn up every revolution.

However, if we logically "AND", or "NAND" these two pulses together in a logic IC (integrated circuit), we produce an extremely stable single pulse, every two (2) revolutions of the crankshaft, which becomes our new stable C.A.S signal, generated by the crankshaft, in co-ordination with the camshaft.

The only proviso of this arrangement; is that the distributor produced pulse must be wider than the crank pulse, so as it moves around laterally/timewise; that the two pulses will always overlap, & allow the logically generated C.A.S. pulse on line "C" above; to be produced.

So where do we need to place this crank pulse, operationally, described above as line "B" ?

Most ECU manufacturers suggest somewhere well before the most advance one could expect to require.  Very few engines, or ECUs, would require an advance, under any condition; greater than say 40 deg BTDC No: 1 cylinder. I've always set mine somewhere between 90 - 60 degrees BTDC No: 1 cylinder.  Be warned that many ECU manufacturers use & stipulate a number that is after TDC; so the setting in the ECU, may actually be 270 - 300 deg ATDC (which is 360 - 90 or 360 - 60).

Unless you are using a distributor that distributes the spark via a rotor button & individual spark plug HT wiring, where the positioning of the distributor, is critical, the setup is pretty easy.  Set the crankshaft in the desired position, where the pulse for line "B" above is produced.  Then simply insert the distributor, & rotate until, the CAS pulse (line "A") above is also present, at the same point in time.  Ensure duration of  "A pulse"  overlaps pulse "B", & pulse "C" will be produced cleanly & reliably.  

Hope that is all clear.

Cheers Banjo 

Edited November 14, 2025 by Banjo

[Banjo]

My last post above, was the theory of how You can stabilise the CAS pulse, that is produced by the camshaft driven dizzy, but now it was time to turn the theory into practice. I  built an interface "circuit board", that slotted between the dizzy Nissan optical distributor module & the Speeduino ECU.  The interface board accepted 360 pulses per dizzy rotation, as well as the single CAS pulse from the dizzy.  In addition, there was a single "crankshaft generated" pulse to line up with the CAS pulse from the dizzy, every second revolution of the crankshaft.

There are twelve teeth on the cog on the bottom of K Series distributor; so you choose one, where the two pulses basically will align; & the dizzy is located, so the 4 way electrical plug & socket faces away from the engine, so it is accessible.

I positioned the crankshaft at an angle where it's pulse occurs, at a point before top dead centre, greater than the "very maximum advance", You are ever going to use.  Anywhere between 60-110 deg BTDC No: 1 will suffice. N.B.  A lot of ECUs actually note this point as being ATDC.  So  60-110 deg BTDC becomes 250-300 deg ATDC. 

Then all you have to do is rotate the dizzy by hand, until both pulses align.  I fitted a couple of LEDs on the interface board I built, so this was easily accomplished.  As per the diagram below, in my previous post, it is important, that the pulse created by the CAS in the dizzy, is wider than the pulse generated by the crankshaft.  The crankshaft pulse cannot move around. The dizzy CAS pulse can move around, dependent on wear & slop in the camshaft sprocket & chain; as well as the helical gear between the bottom of the dizzy & camshaft.  This exercise was to measure the angles of overlap, to ensure the CAS pulse was extremely stable.

This needs to be checked with a degree wheel on the crankshaft pulley; or the flywheel, (which is exposed on my engine test bed arrangement).

I was a bit concerned as the 54mm dia. slotted stainless steel disk in the Nissan module is extremely thin (0.5mm) & I did not think I would be capable of filing out a tiny hole to make it wider, (if it was necessary) without destroying a disk; whose replacement cost can be around $ 100 !

As it turned out, that was not necessary. The single pulse width from the crankshaft was 4-5 deg of crankshaft duration.

The single pulse from the dizzy CAS pulse was 15-16 degrees of crankshaft duration.  I then carefully rotated the dizzy, until the crankshaft 5 deg pulse occurred right in the middle of the 15-16 degree duration of the of the dizzy CAS pulse.

That means that the CAS pulse generated by the dizzy, can move around + & - about 5 degrees,  without having any control over the stability of the resultant pulse, when these two pulses are "logically", AND'd together.

So next step will be to fire the engine up, & put the dual channel oscilloscope on both pulses & the resultant "AND'd" pulse, & see whether the theory all plays out in practice.

Each of the 360 & single CAS pulse from the dizzy, are clean & square, as demonstrated in the picture above, on the oscilloscope.  Those vertical axis squares on the screen are 2 volts each; so the pulses are exactly 5V in amplitude.

The development of something like this, is not something that can easily be accomplished on a running engine. I built a little test stand to mount the K Series Bosch dizzy, with the Nissan 360-1 trigger S.S. disk, so I could work on it's design, in the comfort of my office/workshop.

A little 3500 RPM electric motor drives the dizzy, as depicted below.

The electric motor is fully controlled by a knob from a "dual gang" potentiometer.

One gang, controls the electric motor speed, & the other sends a TPS (Throttle Position Sensor) signal to the Speeduino ECU.

As the electric motor will run right up to 3500 rpm, then I can effectively test the ECU at engine speeds up to 7K rpm, which is a lot more than my little 5K engine will ever be asked to do !

The Speeduino ECU is connected to TunerStudio software; which has a dashboard with all the guages You need for fully testing the ECU.   TunerStudio, also incorporates a trigger pulse data logger, where you can view the stability & duration of all of the trigger pulses.

Both diagnostic functions are really appreciated, when developing something dynamic like this exercise. 

P.S.  When I get this running in my KE30, I will still need a Speeduino ECU for further bench testing.

I don't need the hassle of pulling my ECU in & out of the KE30 constantly; so I managed to purchase another Speeduino PCB very cheap ($10). I've just about finished populating it with components, so I will be soon have two (2) to work with.  The later version (V0.4.3d) printed circuit board, is about 2/3rds the size of my other board, (V0.3.7) & has a single 40 way plug in terminal block, so removing it & refitting it, to the vehicle wiring loom is much easier.

P.S.  When You look at the pricing of the big brand commercial ECUs; the cost really prevents the likes myself & others, from "dabbling" in this way.   The Speeduino ECU, is the most low cost effective ECU system, I have come across, & is "open sourced", so it is always getting better & better, & has a very active forum, where You can always get assistance & information.  

Cheers Banjo 

 

 

 

 

  

Edited November 14, 2025 by Banjo

[Banjo]

I've built up another interface board, so I can still investigate this optical dizzy system on the bench, with the new Speeduino & a second optical dizzy I've built, with a 24-1 punched stainless steel disk.  I ran into a few issues, as the pulses coming out of the Mitsubishi optical module, I fitted inside a K Series dizzy body, had me scratching my head.

Mitsubishi Optical Distributor Module: 

There are a number of issues that I have come across, with the above modules.  This is largely due to the fact; that little specific & detailed information is available; as these modules are sold as replacement parts, & expected to be simply replaced, & the engine wiring harness, just plugged back in.  For those of us, that are experimenting with them; or using them in other ways, to process their trigger & CAS output signals, it is not entirely clear, even with A.I. querying.

1.       Are the modules powered by a +12 volt or + 5 volt DC supply ?

2.      What is the arrangement of the four (4) connections to the module ?

3.      Are output pulses, in a switched +12 volt or +5 volt format; producing positive going pulses ?

4.      Are the output pulses simply a “open collector” output where the output goes to ground, for the duration of the negative pulse, & requires external fitting of a suitable load resistor, to either, a +12V or a, +5V DC supply ?

5.       Is the pulse produced, when I.R. (Infra Red) light passes through the punched disk, or when it is blocked ?

6.      Is there any internal logic, that prevents the leading or trailing edges of the CAS & trigger pulses, to coincide ?

ECU microprocessors that are fed these pulses; usually are only interested in the leading & trailing edge of these pulses. This is because they are usually connected to “interrupter inputs”, on the microprocessor.  This technique is designed to remove the reading of these pulses from the constant running code.  The code to read the interrupter input, is not in the program, running continually.  When an input  changes state (either hi to lo, or lo to high), on one of these “interrupter inputs”; then the running code stops & jumps to the programming code that  handles this input. The program then resumes to the constantly running cyclic code, after it has finished reading & acknowledging the pulse has arrived.

1.      My research has indicated that most, if not all optical modules; are  +12 volt DC powered.

2.      The second consideration is to find out what the four (4) connections are, as to get it wrong, when wiring it up; could possibly result in the module becoming inoperative, (defective, U/S).

I still do not know, if all optical distributor module manufacturers, all use the same arrangement ?

However, the one I am using, (model 93740928) suitable for a Mitsubishi engine is connected as follows.

Holding the module, with the  IR reader head facing vertically, & facing the 4 pin connection moulded plug ; from left to right the four (4) pins are as follows . . . . .

Pin 1:  +12 volts DC.

Pin 2: 0 volts/Ground

Pin 3:  Trigger pulse output.

Pin 4:  CAS pulse output. 

3.    It appears, that few if any modules actually produce a positive (+) going output pulse, for either the trigger or CAS pulse.

4.    This is easily checked, by either connecting a multimeter, or a LED test probe, to the output, & turning the dizzy slowly by hand.   If there is no positive voltage output, it is more than likely that the output is a “open collector” format, where the output pin is tied to ground, for the duration of the pulse.  If this is the case, the other end of an external “load” resistor can be connected to either +12v or +5V, from each dizzy optical output.

5.      The next question was . . . .  Is the pulse produced, when I.R. light passes through the punched disk, or when it is blocked ?

One would expect that when the I.R. passes through the disk’s punched hole, that a pulse would be produced. However, this is apparently not the case.

It is impossible to look into the narrow slot on the I.R. header reader, & see whether the light beam is blocked or open. The light beam is infra red, so cannot be seen visually. The only way, is to remove the punched disk completely, then slide a piece of thin “opaque material” into the slot, to block the light, whilst monitoring the output.  The answer; is in the common name for these devices, where the word; “interrupter” is used. ie: when the light beam is “interrupted”, the pulse is created / provided. This is the opposite to what maybe expected; that a pulse is provided, when the I.R. light passes through  slot in the thin stainless steel rotating disk.  Not the case !

One could assume, that the single CAS pulse & triggers pulses, are both produced exactly the same way.  If You don’t want to remove disk, then monitor the CAS pulse output & rotate the dizzy slowly, until the output pulse appears. As there is only one CAS pulse, per 360 degrees of dizzy rotation, it is easy to see the result.  You can pretty safely assume that both CAS & trigger pulses are both “light interrupted”  outputs. Not the case ! As there being only one CAS pulse per revolution, of the dizzy; that would result be a very long pulse.  The CAS pulse output, therefore has to be inverted, in post processing, to produce the C.A.S. & trigger pulses, in the exact  same format (negative going).  As far as I know, all commercial ECU, use & expect negative going trigger & C.A.S. pulses.  

6.      The stainless steel punched disk I am using with this optical reader,  (24-1) is purposely punched; such that the leading & lagging  edges of the slot for a single C.A.S. pulse, & the trigger pulses, do not coincide. (see photo of punched disk below) This is to prevent two interrupt  commands being generated to the micro-processor, at the very same time.  Micro processors, are very fast devices; but they are basically confined to carrying out one instruction/action at a time.  This is much harder to achieve, when there are 360 punched slots around the outer edge of the punched disk; as on my Nissan optical dizzy module.

The interface circuit I've built between the dizzy optical module & the Trigger & CAS board, also inputs a third pulse generated by a  Hall Effect sensor, mounted adjacent to the crank pulley. This interface board creates a single C.A.S pulse by logically "Anding" the CAS pulse generated by the disk above ; & the single crank pulse per crankshaft revolution.

Here are a few of the oscilloscope screen grabs of the pulses generated by the interface board, as fed into the Speeduino RPM2 & RPM1 inputs.

The vertical divisions on the oscilloscope screen are 2 volts; so the negative pulses depicted are 5 volts in amplitude.

The top trace is the single CAS pulse generated by the dizzy, & synchronised with the stable, wider pulse, derived by the crankshaft (not depicted).

The bottom trace; is of the negative going pulses, generated by the 24 slots around the disk.  Note that the "transition" of these two streams of pulses; do not occur at the same time.  This can be clearly seen in the following screen grab.

Here is the little interface circuit, I've built to process the Mitsubishi optical module outputs.

Working with CAS & trigger pulses, can always be fraught with noise issues, in the automotive environment under the bonnet of a car.

Careful use of shielded cables to transmit CAS & trigger pulses between the dizzy & interface board; & interface board & ECU, should always be installed; grounded at one end only. In addition, I have incorporated in this interface board; optical IR  couplers, at both input & output of this circuit. In addition the switching functions in both the 4093 CMOS quad Nand gate, & the MOC5007   opto-coupler, both incorporate hysteresis switching, which is a great way of reducing noise in the system. Hysteresis, simply is two levels of switching, with a dead band in between  the two switching points.  In other words, a 0-5V dc signal switching from zero to 5V, will not switch, until the voltage reads say 4.0 volts.  Conversely, a 0-5V dc signal switching /transitioning from 5V to zero; will not switch, until the voltage decreases to say 1.0 volt. For input voltages in between 1 & 4 volts, nothing switches. This is the hysteresis band, which is very useful in reducing noise in inherently noisy control systems. 

Cheers Banjo 

 

Edited November 24, 2025 by Banjo

[Big G]

Hey Banjo

Loving all the work you are putting into this. Have you seen the latest episode of Mighty Car Modes they chop down a dissy and put a module into it. They make it look so easy.

Cheers

G

[Banjo]

Hi Graeme,

                    No, I haven't !  I used to follow them religiously, in the early days, but haven't watched them, for a while now.

I'll certainly go take a look in the next 24 hours.  My exercise with the 24-1 punched hole S.S. disk & the Mitsubishi module; is to be able to compare the performance of both, & see if there is any perceptible bettering of performance with the high resolution 360 slot disk, with 2 degrees resolution over 720 degrees of rotation of a full engine cycle, as opposed to the 24 slot disk, with just 30 degree resolution over 720 degrees of rotation.

I'm always trying to find the limits of "things electronic", in things automotive applications, as breaking down "crossing the "Hay Plains", or similar long drive, could result in a very expensive tilt truck ride.  I've never had to be towed in my entire life;; so not planning on starting now.

Strangely, I came across an issue this morning with my setup, that has me baffled.

I had the Speeduino running on the bench with my motorized camshaft simulator.  I then  pulled the CAS plug input, & was surprised that the Speeduino ECU continued to run perfectly.  The readings on all the guages on the Tuner Studio dashboard, still read correctly.  However, when I shut down the ECU, & restarted it, the ECU would not run, until I plugged the CAS sensor back in.

There is no indication in Tuner Studio, that I can see; that indicates, that there is no CAS signal being received. 

So in real life, I'm driving across the "Hay Plains" & unbeknown to me the CAS signal Hall Sensor falls off, or becomes defective. I have no idea this has occurred. The engine continues to run perfectly.   The half way across the plains, I stop for a break & a drink, & let the engine cool a little (it's the middle of Summer !)  I hop back in to continue my trip.  The engine will not start !  It was running perfectly when I stopped. This seems unbelievable; as up until I stopped for the break, there was no indication that anything was amiss.  I need to add a bit of circuitry, that monitors the CAS pulse presence, whilst the engine is running, & alert the driver, if the CAS signal drops out.

So still not believing that this is possible, I ask Google AI the following question.

Quote

Why does a Speeduino ECU not fire an engine when the CAS pulse is disconnected, but will continue to run, if the CAS pulse is disconnected, whilst running ?

 Google AI did it's homework, for me, & Here is the answer it came up with.

A Speeduino ECU requires a Cam Angle Sensor (CAS) or Crank Position Sensor (CPS) signal to establish its initial synchronization and properly fire an engine. However, it can continue to run after the signal is lost because it uses historical data and prediction to maintain engine timing. This is common behavior for many ECUs, not just Speeduino. 

 

Why the engine needs the CAS signal to start

• Establish a baseline: Before the engine can start, the ECU must determine the exact position of the crankshaft and pistons. It uses the CAS signal to find a known reference point, typically a "missing tooth" on a trigger wheel, that corresponds to the engine's top dead center (TDC).
• Calculate ignition and injection timing: Once the ECU knows where the engine is in its rotation, it can accurately calculate the correct moment to fire the ignition coils and fuel injectors. Without this initial reference signal, the ECU cannot establish the engine's position and will not send any ignition or fuel commands. 

 

• Predictive timing: While the engine is running, the ECU receives a continuous series of CAS pulses. If this signal is suddenly lost, the ECU doesn't immediately shut down. Instead, it predicts the upcoming pulses based on the engine's most recent speed and acceleration data.
• "Keep it alive" mode: This predictive, or "limp-mode" function, is a built-in safety feature in many ECUs. It allows the engine to continue running for a period, though often in a degraded state, which can prevent dangerous situations like sudden stalling in traffic.
• Sequential vs. wasted spark timing: The type of ignition timing can also affect how the engine reacts. With a sequential setup, which requires a CAS signal for cam position, a lost CAS signal would cause a major disruption. With a wasted spark setup, which only needs crank position, the ECU has an easier time predicting timing based solely on the last known crank speed.
• Degraded performance: The predictive timing is not as accurate as live sensor data. While the engine may keep running, the ignition and fuel timing will likely be suboptimal, which can lead to reduced performance, rough idle, and misfires. 

 

What happens next

Once the engine with the disconnected CAS sensor is shut off, it will behave exactly as it would if the sensor were disconnected from the start: it will refuse to fire because the ECU has lost its synchronized state and cannot establish a new baseline for timing. 

 

_____________________________________________________________________

 

Google AI, advised it visited 10 off sites to detect the answer it provided.  I'll go & have a look at these sites, & see what they say.

 

Actually the Tuner Studio website was one website I saw listed.

 

As an aside, I asked Google AI a couple of weeks ago, a question related to what I'm working on.  I assumed Google AI  would not source it's info from forums, but only recognised websites. Imagine my surprise, when one of the sites it visited was RollaClub. Even more surprising, was that the piece of info it had sourced; came from a post I placed on this subject, back in 2022 !

 

Cheers Banjo

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Edited November 25, 2025 by Banjo
mistake

[Banjo]

Hi Graeme,

                   Thank You !  Well it took me a while to find this MCM video You mentioned.   The video, is less than a week olde !

https://www.youtube.com/watch?v=Z34a0estBNI

I believe this is it.  The bit about modifying the dissy, is right at the very end of the video.  It appears Marty got the bits He needs from a "Timmy" in Japan, from I6 Industries.

I stopped the video when He got to the point of holding up the toothed wheel. It appears to maybe have 36 teeth, with one missing (36-1).  The toothed wheel looks like it is aluminium; because of it's colour, but is probably iron. That means the sensor could be a VR sensor, or maybe a Hall sensor, with a magnet built in behind it.  Have to await the next episode to find out.  I gather it will be hooked up to, a Haltech ECU ?

I  have taken this a step further, because I what to synchronise the CAS signal, produced by the camshaft, with a crankshaft pulse, which will produce a pulse, that appears, as if it was produced by the crankshaft, that is "rock solid", timing wise.

Cheers Banjo

 

Edited November 25, 2025 by Banjo

[Banjo]

Many of You; like myself, spent yesterday afternoon, watching the final Supercar race, for the season; in Adelaide.

There was a lot riding on the result, & the big question was whether Broc Feeney could pull it off.

It was not to be, & He was devastated, during his interview after the race.  His car was not performing well, & He'd had a similar issue during the race the previous day (Saturday).

https://www.supercars.com/news/supercars-news-2025-broc-feeney-technical-issue-adelaide-grand-final-engine-sensor-triple-eight

https://www.supercars.com/news/supercars-news-2025-triple-eight-suspects-repeat-broc-feeney-saturday-misfire-engine-issue-explained-grand-final

It was suspected that the issue was a misfiring engine, due to the VR or Hall Effect crank angle sensor, which is fitted at the rear of the engine, & apparently picks up it's signals from a toothed ring, mounted in front & adjacent to the flywheel starter teeth around the edge.

I've been playing with Hall & VR sensors for some time now, & have never had a failure.  I did a bit of searching on the nett last night, & there was some interesting, if not confusing data.

There were statements that the Supercars used VR sensors; while other articles stated Hall Effect. 

Maybe, there is a mixture of sensors utilized ?  You would assume; that as Broc's team had experienced the fault during Saturday's race, that they would have changed it out automatically, overnight.  Maybe regulations made that messy or not practical ? 

My reading this morning, indicates that some Hall Effect sensors are "temperature sensitive". 

This guy in the UK, found that that is; & can be an issue.

https://www.youtube.com/watch?v=kAOlPEnuTMw 

I've always fitted mine down on the crankshaft pulley at front of the engine. Altezzaclub, has advised me that rally cars usually don't like fitting them on the crankshaft, as if the engine is mounted north/south, then the crankshaft pulley, is in the danger zone; as sticks/stones/dust & debris, often finish up is area.  However the Supercar engines all being V8s, there would be a lot of heat, in the area the sensor is fitted, from the exhaust headers & pipes; which no doubt glow red during the race. Could not find a photo or picture of it on the nett, but presume, is is probably similar to this picture.

I was interested to find in my reading, that SuperCars all run in "waste' spark mode, rather than "sequential".

This surprised me, as I would expect you would need sequential operation, if you were tuning each cylinder to the "n"th degree, to get the most out of it.  It also stated that they are not allowed, uder the regulations, to fit a CAS sensor or signal.  They still need a signal, to determine the position of the crankshaft, so the teeth or openings on the ring adjacent to the flywheel ring gear, must have one or two holes or tabs missing.

No doubt there will be more discussion on this particular event yesterday, in the days to come, & some of answers to the questions I've posed & pondered, will be answered.  Please add to this topic, if You are aware of an answer to some of my unknowns.  Most VR or Hall Effect sensor pulse outputs in ECU controlled engines, use "pattern recognition" expectation, & can signal or substitute, if a pulse does not appear.

I'm interested, as I'm not sure, at this point in time; whether the Speeduino ECU I'm using is sophisticated enough, to have those timing pulse fault detection features.  I will pursue. I'm also guessing, if Supercar race engines only run i waste spark mode; that the flywheel mounted punched hole or tab gear is in a  60-1, 60-2, or 72-2 format.

I'd like to be able to start my K series engine in waste spark mode; & then switch to sequential once it is running.  This makes starting quicker, as it only needs one revolution of the crankshaft, for the ECU to know it's position, whereas waste spark, can require up to 2 revolutions, to gain the same positional information.

Please add, answers You have to questions I've raised, above. A very interesting subject. 

Cheers Banjo

 

 

Edited November 30, 2025 by Banjo

[Banjo]

My KE30 2 door Corolla, has always been a very fast starter, since I installed my very first electronic basic ECU; which was a Jaycar project, from back, nearly a decade ago. Many people who have seen my car, have remarked how quickly it starts.  It still has that Jaycar ignition system installed, (but not for much longer). The triggering is an olde K Series Denso dizzy with the springs & bobs weighs all removed, & the shaft "locked up".  The points were removed, & in their place, I fitted one of those Hall Effect sensors, with an ignitor built in.  I described it back in this thread, that Graeme started, way back in 2017, when His daughters Rolla was going through points regularly.

https://www.rollaclub.com/board/topic/73743-electronic-distributor/#comments

My setup was possible, because our ignition switch, has a start & a run position. In the start position, the starter relay to drive the starter motor solenoid, is activated, & the "ballast resistor" in series with the coil is effectively shorted out.

The rotor button directs the HT from the coil to the very next cylinder's spark plug, in the "timing format".

The Jaycar module was not in circuit, during the starting process, & the timing was the static timing, of 10-12 degrees BTDC.

I would like to do a similar thing with my Speeduino ECU.  However, not running a rotor button & distributor cap & leads, the best You can achieve is start the engine in "wasted spark" mode. I believe some modern day cars use this technique.  However, I'm not sure whether the Speeduino ECU has the capability of carrying out this action.  I'm currently awaiting an answer on that question. However, it may well be possible, using the start position on the ignition key switch, to close a relay, that mates up the paired cylinders in "Waste Spark" mode.  (1 & 4) & (3 & 2)  I'll try a couple of experiments on the bench here, & see if it is possible.

P.S.  When an engines stops, it usually comes to a spot in the 720 degree rotation, that is in one of two spots. I asked Google AI about this, & this is what it came up with . . . .

Quote

When an automotive multi-cylinder engine is switched off, the engine is most likely to become stationary when at least one piston is near top dead center (TDC) on the compression stroke or just starting the power stroke [1]. This position creates the greatest resistance to the engine's rotation. 

Here’s why:

• Compression Resistance: As the engine winds down, the remaining momentum must work against the compression of the air-fuel mixture in one or more cylinders.
• Maximum Force: The peak force required to compress the mixture occurs right before TDC on the compression stroke [1]. The pressurized gas acts like a spring, pushing back against the piston and bringing the crankshaft to a halt more effectively than any other point in the cycle (intake, power after the initial combustion, or exhaust).
• Balance of Forces: While multiple cylinders are in various stages, the cumulative force of the cylinder(s) undergoing maximum compression is usually enough to overcome the remaining inertia of the crankshaft and flywheel, stopping the engine at that specific point. 

Cheers Banjo

 

 

 

Edited December 1, 2025 by Banjo

[altezzaclub]

Those Toyota TPS have an idle wire that you could use for waste spark activation if the ECU can handle it. Have it go onto waste spark when the idle wire is live and sequential as soon as you move the throttle.

You'd need a piggy-back board I expect as the ECU I've seen have you pick one or the other in a menu.

[Banjo]

Yes,

       The Toyota TPS sensor, which apparently was utilised on Camry & Corolla vehicles, definitely has 4 wires. I know I played with it a bit, when I received it, after buying it on line; but couldn't quite work it out. That did not concern me, as I wasn't going to use it, anyway.   However, I did wire it up, with 4 wire cable, so I have the idle wire, that I could play with, & add a piggy back board, as it is already on the 4K-U engine.

Not sure that will do what I want; as if I use that to "start the engine", it will drop back into "waste spark mode", every time, You pull up at a set of traffic lights, & take your foot off the accelerator.  I really need it only during the engine "starting process".  If You research it, a lot of countries will not allow vehicles with their engines running in waste spark format.  Apparently, this is because of emission regulations, as waste spark, produces slightly higher emissions than sequential operation.

Apparently, there were other factors that came into it; especially for the Americans, with a preponderance of V8 & 6 cylinders engines. The dual HT output coils used on GMH vehicles here in Australia, were widely used, & very popular. Simon / TAZ used them on his K Series engine. No rotor button, or dizzy cap with individual spark plug leads.

I've always believed in sequential control, as You can tune each cylinder slightly differently, & pick up & control 
"knock" in an individual cylinder.

Apparently, cost comes into it also. No CAS signal required from the camshaft.  You don't even need a separate crank angle sensor.  Just use a 24, 36, 60, or 72 toothed crank wheel, & remove one or two teeth, to become your crank angle sensor.  Super simple; super cheap ! 

Initially, I will be using my ECU, as ignition only.  My technique, whilst starting the car, has always been the same.

Never have to use the choke, here in Qld. Before turning the ignition key, two quick pushes of the accelerator pedal, to activate the pump squirt function in the carby, to enrich the first suction cycle of the cylinders.  Then I press the accelerator slightly (to open the throttle butterfly; & turn the ignition key to the start position.  Perfect starts, every time !

I'll have a play with the TPS, later today, but I don't think it would work for me, in my situation.

With COPs, it should be easy, as I'd only be redirecting the Ignition trigger outputs 1 or 4, to fire COPs & 1 & 4 together.

Signals on ignition trigger outputs 2 & 3, would switch fire COPs 2 & 3 together.  Remember, with COPs, I'm only talking about redirecting 5V DC trigger signals, as each COP, has it's own ignitor within/onboard.

I'll try it here on the bench, with the Speeduino ECU & four spark plugs.  I'll let You know how I fare.

If You  haven't worked me out yet; I just "love experimenting", & seeing whether ideas are "goers"; or "dead ducks".

You'll never know, unless You try it out. 

Here is what AI dug up, about the pros & cons of Waste Spark, & Sequential ECU control:

Quote

 

What are the pro & cons of using waste spark operation in automobile engine ECU control ?

 

The primary pros of a waste spark system in ECU control are its 

simplicity, lower cost, and high reliability compared to older distributor systems. The main cons are reduced performance tuning potential and increased spark plug wear compared to modern, fully sequential systems. 

 

Pros of Waste Spark Operation

• Simplicity and Cost-Effectiveness: The system uses fewer coils (one for every two cylinders) compared to a coil-on-plug (COP) system which uses one coil per plug. This reduces the number of components, making it less complex and cheaper to manufacture.
• High Reliability: By eliminating the mechanical distributor, the system removes a common point of failure caused by moisture, dirt, and mechanical wear.
• Accurate Timing Signal: The system typically uses a crankshaft position sensor for timing, which provides a more direct and accurate signal to the ECU compared to older, camshaft-driven distributor systems.
• Reduced ECU Complexity: Waste spark ignition only requires a crankshaft position signal (not a camshaft signal), simplifying the computational needs and required outputs from the ECU. This was a key advantage when engine control units were less powerful. 

 

Wasted spark system - Wikipedia

* Advantages. Perhaps the most significant advantage of the system, compared to a single coil and distributor systems, is that it ...

 

Wikipedia

 

 

The beauty of crank-fired Wasted Spark Ignition - DIYAutoTune

May 17, 2023 — Keeping it simple and powerful with Crank Fired Wasted Spark. The key benefit of wasted-spark lies in the simplicity: ...

 

DIYAutoTune

 

 

 

Cons of Waste Spark Operation

• Increased Spark Plug Wear: Because each plug fires twice as often (once for a useful combustion spark, once for a "wasted" exhaust spark), the electrodes wear down more quickly. This necessitates the use of more expensive, durable double-precious-metal (platinum or iridium) spark plugs to maintain a reasonable service interval.
• Limited Individual Cylinder Tuning: The ECU fires two plugs simultaneously, so ignition timing adjustments (e.g., in response to knock detection) affect both cylinders equally, even if only one has an issue. This limits the ability to individually optimize performance or efficiency for each cylinder, a feature available in sequential (coil-on-plug) systems.
• Potential for Reduced Spark Energy at High RPM: At very high engine speeds, there might be insufficient "dwell time" (coil charging time) between the two consecutive sparks for the coil to fully recharge, potentially leading to a less powerful spark during the compression stroke.
• More Difficult Diagnostics: The lack of moving parts makes the system more reliable but can make diagnosing specific ignition problems without specialized diagnostic equipment more difficult. 

Here is what AI produced regarding emission standards requirements.

Quote

Which countries in the world only allow new vehicles with sequential engine controls to be utilised, as waste spark engine control produce higher emission outputs ?

 

There are 

no specific countries that explicitly ban new vehicles based solely on the use of waste spark engine control systems. Instead, the high emissions standards in most developed nations effectively mandate the use of more advanced, precise technologies like sequential injection and ignition systems. 

Here is a breakdown of why this is the case:

• Sequential control as standard technology: In modern automotive manufacturing, especially since the early 2000s, sequential port fuel injection and direct injection have become the de facto standard due to superior emission control, better fuel economy, and improved performance compared to older methods like batch-fire or waste spark systems.
• Emission regulation rather than technology mandates: Countries like those in the European Union, the United States Japan, and China
   set stringent, legally binding emission limits (such as Euro 6 standards) that new vehicles must meet. While the regulations do not specifically name "waste spark" technology as banned, it is technically infeasible to pass these strict modern emission tests using such an outdated system due to its inherent lack of precision in fuel and spark timing.
• Technological evolution: The superior control of sequential systems, which fire injectors and spark plugs individually as each cylinder requires, ensures optimal combustion and efficient use of catalytic converters, which are essential for meeting contemporary pollution reduction targets. 

In summary, the market and regulatory framework in major automotive markets worldwide have naturally phased out waste spark systems by requiring emission performance that only advanced technologies like sequential control can achieve. 

 

 

 

 

 

 

Cheers Banjo

 

 

 

 

 

 

Edited December 10, 2025 by Banjo

[Banjo]

Good news on the "start engine" in Wasted Spark mode; & drive in sequential, there after.

It appears that a Speeduino ECU, can only be changed from Wasted Spark to Sequential; or visa versa; by changing the mode of operation in the project file setup screens, within TunerStudio. There appears to be no way to change it, "on the run/fly".

Well I'm pleased to advise, I have my Speeduino, running on the bench, in sequential mode; but can switch it in & out of "wasted spark" mode, at will, on the run. This was done, without making any modifications to the Speeduino, whatsoever, or changing it's setup tables.

The olde dissy cap, with four High Tension leads & plugs, never really had an issue, with timing or starting, as the rotor button in the distributor, will always fire the next spark plug & cylinder, in the sequence, irrespective of where it is, in the two revolution cycle; or where in the cycle, the engine finally came to a halt.

With COPs, & no rotor button, & four (4) COPs under direct control of the ECU, there is a better way of possibly getting the engine to fire earlier; (in the first rotation of the crankshaft), while cranking.  Start the engine in Wasted Spark !

The solution was quite simple. Take the four (4) ignition outputs from the ECU, (IGN1, IGN2, IGN3, IGN4) & run them through a set of logic gates, which pass them through as "Sequential", or convert them automatically to  "Wasted Spark", as a result of an external signal, supplied by a switch on the dash, possibly by the extra wire on some TPS sensors, or instigated by the start position in the ignition barrel.  When You hold the ignition key in the spring loaded position, whist cranking the engine; this signal can apply the extra signal to the logic gates to create two outputs for each sequential input.

ie: 

When the ECU supplies a signal to fire  cylinder no: 1, then the logic gates are set, so that COPs 1 & 4 fire together.

When the ECU supplies a signal to fire  cylinder no: 3, then the logic gates are set, so that COPs 3 & 2 fire together.

 When the ECU supplies a signal to fire  cylinder no: 4, then the logic gates are set, so that COPs 4 & 1 fire together.

 When the ECU supplies a signal to fire  cylinder no: 2, then the logic gates are set, so that COPs 2 & 3 fire together. 

The Speeduino ECU has four little LEDs, indicting sequential or wasted spark mode.  In Sequential mode, the ECU simply fires the four (4) Ignitions outs one after another. IGN1; IGN2; IGN3; IGN4.  It's up You, or the installer; to have to wire the four outputs to the appropriate COP, or external Coil Ignitor.

I've hooked up four (4) LEDs, on the output of the small logic gate PCB, & I'm sitting here watching the LEDs on the Speeduino ECU board, in individual sequence, & that replicated on the logic board LEDs.  However, as soon as I provide the additional logic signal to the small logic gate PCB, the four (4) LEDs on small logic board,; instantly change to 1 & 4 on together, & 2 & 3 on together.  There is no delay; it happens instantly.

Here is the logic circuit, if anyone wants to replicate this.

No micro-processor, or programming required. Speed or frequency is not an issue. There are just 4 off "integrated circuits" (ICs) involved, that can be purchased at any electronics store, like Jaycar; for about $5-$6 total.  The whole thing is powered by 5V DC, which will come from the ECU. A small 12V relay, powered by the start position on the ignition key switch, should be used to switch the +5V signal, to the logic board. I might add an opto-coupler there, with hysteresis built in; so there is no chance of "chattering" contacts, from a relay.

 

Just have to wait now, until this is all in the KE-30, & I can try it out, on the road.

Cheers Banjo

 

 

 

 

 

Edited December 11, 2025 by Banjo