Any ideas for once-per-revolution car ignition design?

Isn't this similar to the electronic tachometer idea that's being used to measure ground speed.

Mount magnets into the shaft with a HALL type pick-up and necessary electronics mounted to stable hardware. (Or one magnet and four HALL devices)

The impact on "sudden" changes in speed would purely depend on how "smart" your electronics interface is. If you only consider the presence of the magnet, then you are limited. You might be able to interpret "ignition advance" if you start to also consider the time between each signal. You might then want to consider the differnce between successive signals (to determine if it's getting shorter (faster) or longer (slower)) and then the rate of change of that signal.

If you have a simple magnet and HALL system, then you could manually "tune" the device by selecting/adjusting the position of the HALL device (just like adjusting the advance on an old mechanical distributor.)

With such a mechanical timer, the timing would be very strongly linked to the absolute position of the shaft and so would not be "out" in that sense, however my (very basic) understanding of modern ignition timing is that they try to have ignition such that the explosion propagation is optimum and that will be different at low and high revs.

An encoder (as used with servo motors) on the crank or cam could be considered.

Finer variations (advance and retard) could also be provided for.

GA from me!!

Thank you for your responses....

This is envisaged for a update to an old (60's) engine.
Personally I was thinking the system would work along the lines of:

On starting - the crank would rotate (actually be better kept with the distributor housing - keeping the advance/retard mechanical) until the hall-effect (or whatever) pulsed once.
This would start a clock running & say 60ms later a second pulse is registered. At this stage the electronics 'know' the engine is rotating at 1000 rpm & can drive four power transistors 15ms apart.

If the next pulse arrived 59ms later the electronics would spread the following four ignition pulses at 14.75ms.

Obviously this design takes no notice of whether the engine is speeding up or slowing down & so timing is probably going to be slightly wrong some of the time.

So, can any electronic engineers here give me pointers for a simple method of implementing this process? -By simple I mean with say 555 timers rather than programming a microcontroller. (I studied electronics at Uni gaining a good 'grounding' & now wish to use it practically. But I ended up going into IT instead...)

Further, do you think this configuration would be any less exact than an old mechanical points & distributor setup? That is: Should I bother?

Thanks Again.

Pertronics make a Hall effect sensor which fits in place of your points. Go to retrorockets.com to purchase. Silicon Chip Magazine make kits and plans for a series of high performance and CDI ignition systems including an advance mapping system. The kits are available through Jaycar Electronics.

I personally feel that you might just get the engine completely wrecked in the time you take to design the thing!! Look at Electroman's offering, it is much better...and practical too....

If the engine is 4 stroke, do not forget that you have two revs for each spark needed, not one as you appear to think. Sparking again at exhaust time is not a really big problem, but still incorrect really as you are simply wasting twice as much energy as really needed.....on a fast engine that could make the sparks weaker.....

Why not refurbish the distributor and bring it up to date as someone else suggested with a Hall effect sensor or similar?

"do not forget that you have two revs for each spark needed, not one as you appear to think"

Yes, I realised that mistake as soon as I''d posted! Woops!

The dizzy's running at a quarter of the engine speed & has the vacuum advance built in, so, yes I'll keep the pick-up (whatever it is) on that.

The reason I preferred one timing pulse (for actually every four engine revolusions) for every turn of the dizzy was that I only have to get one magnet in the right place & all four cylinders will fire on queue.

Conversely I'm likely going to waste loads more time (& expense) designing the electronics, so...

I think that answers my question: stick to well adjusted mechanics!

Oh, well - I think it was an interesting thought experiment if nothing else!

Unless someone has started designing 4 stroke engines in a completely new way, the distributor for a 4 stroke engine revolves once every 2 revolutions, that is a four stroke engine, not 4 times as you wrote.

There are also some very interesting 6 stroke engines around, one of which you can see here, but I am sure that is not what you meant!

Incidentally, a 6 stroke engine's "distributor" if it has one at all, revolves once every 3 engine revolutions if I remember correctly.....

The problems are due to people confusing the "stroke" in a "4 stroke" with movements of the piston....do not make that mistake.....the "strokes" here are:-

1) Inlet 2) Compression 3) Power 4) Exhaust - A better word might have been "phases" to my mind!!!

These 4 "Phases" happen for each cylinder in just two revolutions. You will find a perfect animation example here which should clear things up for you completely.

I think I need a refresher in engine basics! -I'd pictured the rotor cam with four high points equating to one bang per turn of the engine -which I know is wrong when I bother to think about it properly!

Anyway... I could produce an easy to adjust Hall effect points by fixing two small magnets to a bar centred on the crank pulley bolt. This should make it easy to line up the timing.

The vacuum/weights advance-retard mechanism could stay on the distributor, but move a variable resistor instead. This instantaneous value of this resistor would electronically delay or advance firing.

Sheps,

I can only warn you to think about using a 555 in such an application. Remember that the 555 on one side has a digital output but on the other side is a pure analog circuit with a threshold being determined by an RC-combination. If your threshold voltage changes only some millivolts influenced by whatever (and there are lots of possibilities, believe me ...) you`ll get a completely different timing. Your automotive supply lines are full of spikes - they can go up to hundreds of volts (measured with a small time resolution scope) and noone can guarantee you that there will never be a spike which might fail trigger your device.

Another problem is, that the repetition constance is very poor because the capacitor is a very poor part, depending on temperature, aging, humidity etc. If you want an electronic solution then go digital, nothing else.

But have in mind what some of the guys have already recommended - the chaepest and best way for a reliable solution is a mechanically fixed sensor where the relation of sparking angle and mechanic is true.

Regards Uwe

Why not just utilise the stock dizzy, remove the points, add in a metal vane with 4 holes (if 4 Cyl) and a inner "Indexing" hole, and run optic sensors, or magnetic sensors to detect the leading edge/trailing edge, along with your indexing hole.

timing loops and the like get to be a pain.

Also how were you going to drive the coil pack? a Relay switching it? or have you got some driver chips that control the current flow into the Pri side of the packs? I think NSC make them still, I have the details in some of my old "Analog devices" books

Engine Management computers use the crank angle and engine speed pickup sensors, the ones I have used have a magnetic vane inside the dizzy, maybe you could get a modern Dizzy, and modify it to suit the older Dizzy your trying to adapt.

"Also how were you going to drive the coil pack?"

I plan to drive a wasted spark coil pack with two PNP power transistors - I've seen a number of web pages describing a circuit to drive a single coil - so I plan to just double up & fire alternatively: PNP1 fires cyl 1 & 3, PNP2 fires cyl 2 & 4.

A simple design including surge/spike protection (through inductor I think) is:

http://www.cs.berkeley.edu/~wkahan/TransIgn.pdf

Also

http://basilisk.chaosnet.org/motorcycle/diy_transistorized_ignition.html

seems to show nothing more than a PNP acting as the 'switch', but I'm wondering how long that would survive...

try summit racing & look at their crank trigger electronic ignition sets. they have adjustable timing curves in the computer that runs the coils & can run 1 coil per cyl

Electroman has it correct, and guest had some good info as well.

Most of GM's factory ignition systems rely on a crank signal to determine engine speed, and a camshaft sensor to determine ignition firing timing. Both systems feed into computer, and that determines how often to fire, when to fire, and which coil to fire. Some later systems have combined the two sensors system into a single system. This is done by using many teeth on the crankshaft trigger to get accurate engine speed change information. At one of the teeth there is a missing tooth. That "missing tooth" tells the computer that is the timing mark. The timing mark is the reference signal that controls timing events.

If you want to simplify the conversion, look into the MSD digital ignition systems. They rely on their "flying magnet" to trigger the system. It will work for 4,6,and 8 cylinders for sure. Possibly other numbers also. You can adjust the initial advance, advance rate, start of advance end of advance, max RPM, retard amount retard rate, and probably a few things I forgot. They have systems that will bolt right on to a small or big Chevy, a few Fords, and small Chrysler, I think. But the good part of this system is that you only have to mount the magnet to the balancer in the front, and then mount the pickup. It should work on anything you can dream of. They have a customer support that should walk you through any problem you have. They probably have the exact information needed to mount it to the engine you are planning to use it on. What are you putting this into? Good luck. Try www.msdignition.com go to products, then ignitions, then digital. Send photos.

Thanks Bob (& Electroman!)...

You said, "What are you putting this into?"

Rather embarrassingly, esp. after your mentioning big Chevies, etc - I am 'fooling' about with my poor everyday car - a 1965 Vauxhall with a measly 1000cc !!

Things operate a bit differently here in the UK -for a start our 'gas' is now around £1.20 a litre [don't know what that is in Dollars & US gallons, but Not Cheap!] !!

More the point I've just cut a big section out of the foot well which I discovered a couple of days ago is mostly rust! - So any ignition project is going to have to wait a while!

Still it's interesting to look at the options - even if it never comes to much!

I should also add, if this idea ever becomes practical I'm planning to do it on an absolute shoestring... like I do all my mechanicals! I've acquired a 'wasted-spark' ignition pack & this has started a few crazy ideas buzzing around my head!

Buy yourself the Pertronics hall effect unit and install it in the dizzy with a new set of Magnacor wire suppression leads. Dunno where you get the leads in UK but you can still order the Pertronics unit over the net at Retrorockets.com per my comments yesterday. The system cost me under $200AUS for my Rover V8 so should cost under 50 pounds for you. You will find the leads by themselves will make a marked improvement in running.

The Rovers used to use a Buick-Olds V8 design from back around 1962. It was sold as 215 cubic inches. If that is the engine you have, the conversion would be easy, as Petronics has found great acceptance in converting engines that were pre electronics to reliable quality ignition systems that are almost invisible. How hard will it be to find a Petronics for that Vauxall? Mallory used to make a Uni-lite system that installed in place of the points in older distributors. They were good, and reliable. I am not sure if they are still in production.

That must be the Engine, it was sold in Europe as a 3.5 liter V-8, and 215 Cu.Inches translates to about 3.5 Liters!!!

It proved to be very "Tuneable" and a lot of race cars were made using that engine.....I believe 500 BHP could be made available if needed!

Also, people who bought Lotus Europas with its heavy cast iron 4 cylinder French (Renault 16) engine with very little power, quickly found out that the Rover V-8 fitted remarkably easily into the limited space available.

The biggest problem was that the rear suspension had to be modded to get the car back onto an even keel as the Rover V-8 engine was considerably lighter!!!

The lighter weight also made the handling even better.

All in all a very nice car to drive if I remember correctly, but it must be around 40 years ago now when this was current!!

Buick sold off the 215 (3.5L) before the first fuel crunch here. They started to unload that V6 that they replaced the 215 with just as the fuel crunch started. They had already stopped installing it in all but the smallest of cars ( clones of the Chevy Nova, called Apollo, and the Skyhawk clone of the Chevy Monza) They were the upgrade engine in Jeeps from 1966. Then the gas went scarce and they developed the Odd fire crankshaft, and used it as their only in house engine for thirty years.

I know a store that had an aluminum manifold for that 215 that used 2 Rochester two barrel carburetors on it. He couldn't give that thing away. It was on his wall about three years ago when I last was there. If you were to look really hard there were Turbocharger installations from Oldsmobile on the 215 that relied on a liquid injection system under boost. This was before the redesign of the GM mid size car line in 1964. The Buick, Olds and Pontiac used the same basic small body from 61 to the 64 replacement. Buick and Olds used the 215. Pontiac cut the side off of their 389 engine to make a slant 4 that used all internal parts from the 389

yep thats it. Bob below gives a good rundown of teh US side. Teh Olds engine was actually somwhat different to the buick the olds had fully machined combustion chambers whereas the buick and rovers are caste, though I might have it the other way round. Des Hammil has a good book with total history in it, including the Range rover versions with crank mounted oil pump and no Dizzy. Many parts are a straight swap to the buick 300 V8 and the 90 degree V6 which came from the alloy engine design.

On Topic, Vauxhall use standard GM/Delco dizzies so I'm certain they will have one available for your particular unit

Oh Leyland Australia also made a 4.4 version of the Rover engine for their P76 and Terrier trucks

Ok. Now that the the rusted area is out of the way it is time to shoehorn that 454 Chevy in where it belongs. Don't worry about gas. After the axle splits, it will get better gas mileage sitting on jack stands.

Back to the ignition. On this side of the pond, the Chrysler ignition pickups from the seventies were easy to find, and could be used to trigger any coil by using the easy, reliable Chrysler control box. Why not try to make some easy to find pickup, from the same manufacturer as the ignition pack, fire the ignition pack? Set it up on a wood work bench. You need a 12 v battery, the ignition pack, the pickup of choice, whatever ignition coil is called for, and plug wires and plugs, or plugs directly to the coils if that is the design. If everything is grounded, just passing a screwdriver past the pickup should fire the coil(s). Do you have the wiring schematic of that ignition pack? If not, you need to get it.

Please do not use the word redneck.

In parts of the world there are people that race lawn mowers. I have been thinking about converting a Briggs and Stratton to run a Chrysler ignition. once it works on the fixed timing setup then convert to the digital control and get the adjustable timing features. And besides if you work at it you can make the door bell button trigger the system, and deliver a 40,000 volt shock to that traveling salesman at your door. That alone makes this project worth the time. Good luck.

Two suggestions:

1. Use the points to drive a solid state circuit thereby reducing any significant wear on the points.

2. Make a shutter with a slot for each pulse and use a led/photo transistor to drive the circuit to fire the coil.

A retrofit kit is available for some engines, but I could not say you could get one for your specific engine.

Good Luck!

Thanks Bill.

Something like this seems the simplest method to use points with low current:

http://www.cs.berkeley.edu/~wkahan/TransIgn.pdf