HT Ignition lead pickup...

P

pichutch

New Member
<preamble> just before I ask this, I have read a lot of warnings on here about interfacing with cars, and I'm very confident that what I want to do here will not effect the car while stationary or moving in any way, and if it does - it's my fault!</preamble>

Hi all,

I'm devising a circuit (in my head) that will use a PicAxe to monitor spark intensity via a pickup on the HT leads of my old car (there's not an ECU in sight, don't worry, honest).
What I'd like to know is if any of you have done anything like this, and if so what you used as a pickup? Some timing lights use a coil temporarily placed around an HT lead to get a trigger. I'm wondering if a hall-effect device might work in this way. The leads themselves are carrying 20kV pulses, so before I play about with bits of coil wrapped around a lead - and probably blow some fuses on my 'scope - I thought I'd ask if any of you have some sage advice.

In terms of speed, a PicAxe should be able to cope with 3 sparks/pulses a rev at 5500 RPM? That's... *counts* about 280 a second right?

Cheers,
Ben
 
D

Dippy

Moderator
Slightly tricky as the requirement isn't quite clear..well, to me anyway.

Do you just want to measure when a pulse gets over certain amplitude or do you want to measure duration as well to get some idea of the 'energy'. If you want to do this on every spark at every RPM (i.e. do some integration) then I reckon you'll struggle.

If it's a rev counter then maybe there is some sensor help in this:-
http://www.picaxeforum.co.uk/showthread.php?t=4176&highlight=ignition

Now that I've started the ball rolling you'll get some tips, it sometimes takes one to start the avalanche as everyone is waiting for everyone else to say something...
 
P

pichutch

New Member
Thanks a lot, that thread covers quite a lot of ground.

Here's a better idea of what I'd like to achieve (I'm making this up as I go along, seat-of-pants development!):

On several occasions, my old 6-cyl Triumph has killed a sparkplug during a long drive, or after standing for a few weeks. Believe it of not it's a pretty smooth engine and sometimes its not apparent that its only firing on 5 for quite a while, especially at high revs.

Also, dirty leads/plugs or hairline cracks can cause tracking and a weak spark.

So... seeing that PicAxes are my new favourite toy (since last week! :D) I thought that it'd be great to have a circuit and prog that monitors all 6 leads, giving an idea if they are pulsing at all, and then if they are all transmitting roughly the same size pulse. This should alert me to a dead plug or a badly tracking lead.

I'm not really worried about the timing or spark duration, I can detect that with other methods.
 
G

gengis

New Member
For an inductive pickup you can use a "clam shell" ferrite choke designed to clamp around a wire - your pickup coil needs a few turns of wire.

Capacitive pickup will work with just a wire wrapped around the HT lead with one end left unconnected.

Or monitor the Primary of the spark coil - the voltage there on a Kettering Ignition (standard points/condenser) is on the order of a few hundred volts. Doesn't do much good if your objective is to monitor a single plug of a distributed ignition though.

One interesting thing I found when building a tachometer - the spark coil doesn't produce a single spark per firing. It produces a decaying sine wave. The coil "rings" between the primary and condenser. I had to use a monostable to get just a single pulse per firing (with a transistor switched ignition on a Honda motorcycle). I assume a CDI system would ring a lot less.
 
D

Dippy

Moderator
Preamble: If I've understood it correctly.

A monostable is excellent for doing a tacho, but not too good IF he's trying to measure the pulse energy. Ditto opto-couplers.

But that's a good point Flooby, re: oscillations. So some sort of filtering plus integration would be required and I doubt if PICAXE is fast enough to do it precisely and in real time. Maybe an arrangement of RCs may help? Interesting project.

How good is your electronics pithcuch?
To get something good you will deffo need your 'scope. You can always protect the input to your scope with resistors and a handfull of various zeners.
Just remember not to balance your scope and breadboard precariously on the Air Filter box ... with dangling wires :).. ah, it brings back memories.
 
E

eclectic

Moderator
Last edited:
W

westaust55

Moderator
The split type ferrite core that is mentioned in the first link eclectic gave above are commonly available from old computers. (Had an armful of them but threw them out a few months ago after never finding a use over many years) :rolleyes:.

It is definitely current that you need to measure and not voltage to ascertain whether the plug is sparking and energy magnitude.

Another option is a hall effect device. Some of these are quite sensitive to small magnetic fields. Jaycar sell one (Cat No ZD1902) – the Allegro 3503 - which is quite cheap and that I have played with. They generate a voltage proportional to the mag field strength but may not be sensitive enough just strapping one against the HT lead.

The Philips KMZ51 is far more sensitive and can pick up variations in the earths magnetic field. This could be a good candidate in terms of sensitivity but more complex that the 3503 and surface mounted package.
 
P

pichutch

New Member
Wow this topic is picking up! I hope everyone doesn't get annoyed when it actually takes me ages to get around to building and testing this thing!

Flooby:
Great idea for the clam-shell choke, I think that's what Eclectic has just mentioned too. Do you have any recollection of how long the 'ringing' after each pulse lasted?

Dippy:
I guess the primary requirement of project is to see if there is ANY spark, I could leave the 'amplitude' part for later.
Regarding speed. I guess there is no reason why I have to measure all 6 sparks in order, afterall I'm not trying to measure the timing. I could make the circuit/prog cycle through all 6 'inputs' waiting a set amount of time for evidence of a spark then move on, and loop. Actually if I do it this way, I wonder if I can use ADC inputs to measure amplutude too and not have to worry about the ringing.

Eclectic:
That's a great application circuit and idea for the pickup, thanks!

My electronics knowledge isn't too bad. I did electronic engineering at college about.. uh.. 13 years ago and have used it once since! So im very very rusty. I know enough to understand why I need a monostable and roughly how to build one and set timing, but not enough to see why zenners will help protect my scope (a very old one a nice tutor let me have when the college got new storage scopes). So I'm going to look up zenners now.

thanks.. uh.. 'People' (can't assume you're all 'Guys'!)
 
D

Dippy

Moderator
Well, your first steps are picking up the signal from the leads (clam inductor ideal, choke and cable-tie if desparate). And then looking at the signal on your scope.
The zener (one 'n'... it could save hours searching with correct spelling) was just an idea to limit your signal to a safe amplitude for your 'scope and later for your circuit.

Detecting 'any' spark should be quite easy and the monostable will be good. Note: if you use a 555 or similar as a monostable it will extend your signal pulse, whereas an edge-triggered monostable (eg 4098B and derivs) will just do it's pulse output on the edge pos-going or neg-going input.
Maybe also have thresholds on input to discriminate between bad and good ?

But all this, and 'ring' duration, needs to be seen and measured on your 'scope. These things will determine the detail of the rest of the electronics.

And, I'll just get this in now; Assuming your PICAXE+circuit will be powered from the car supply, ensure you have suppressors and voltage regulators that are butch enough for large surges and spikes. Older car electrics can be a bit sparky. And be aware that cables and circuit boards can pick up noise/spikes when you don't want it. But that is all later.... first things first, get scoping :)

Good luck.
 
P

pichutch

New Member
Ok thanks for summarising. I'm busy this weekend watching fast old cars hoon about at the Silverstone Classic, but will get cracking with this next week.

Regarding a steady stable 5v power source, I was going to build a small supply based on a linear regulator, about 1 amp, and fuse it. BUT... I see that switching regulator IC's are coming down in price nicely. Anyone used one to power pic circuits?
 
D

Dippy

Moderator
Yes, but only for circuits where I needed 500mA or more and simply for the efficiency. But I've never used one where the power is supplied by car electrics.
Do you really need anywhere near 1 Amp? Or was that a wet-finger-in-the-air figure?

I thought you were just using a PICAXE, maybe some LEDs, maybe an LCD and some odds and ends?
If less than few hundred mA then a linear is cheaper, smoother and less effort. AND, be aware that some tiddly S/mode ics have a relatively low max input voltage. Many will be fine, its just something to keep an eye on. i.e. READ the Data Sheet of whatever device you are considering. Some like Simple Switcher from NEC or their Hysterical PFET contrller have a healthy spec. which I would certainly consider on circuits >500mA. But careless circuit design or blinkered/tight-fisted component selection can result in noisy circuits.

From the car electrical supply point of view: I would also (in addition to a low val fuse) have a diode, a small choke/inductor, ceramic capacitor and an appropriate transient suppressor on the power supply.. before your regulator, whichever type you choose. For peace-of-mind choose a regulator with Vin >25V when using a good suppressor circuit or considerably higher if your suppression is weedy. There are 'automotive' rated devices but I forget names/numbers and values. I'm sure someone can remember.
 
P

pichutch

New Member
Yea it was a wet finger hardly raised guess, with the thought that even though this project will use about 120mA absolute MAX (I reckon it'll be about 70-80 if I use some of this fancy newfagled best thing since sliced bread PWM on the LED's), I will probably want to power a couple of other circuits.. well.. maybe, so might as well build something big enough for future 'in car' picaxe projects.

Ok go easy on my and please remember I'm VERY rusty with electrical/electronic theory: I'm going to need some more help on why/where/how to use a choke and suppressor. I was also wondering why you suggest using a diode as I'm not using any inductive loads. I see that 'Transient voltage suppression diodes' are readily available from the usual suppliers, will using on of these kill two birds with one diode?
And for 'capacitors' do you mean over an above the smoothing and filtering caps that are part of the 'usual' 7805 style circuits?

Thanks :)
Ben
 
D

Dippy

Moderator
"And for 'capacitors' do you mean over and above...."
- yes. Ceramics are good for this.

The diode was just for polarity protection to save embarrassment, time and your pocket.
The transient supressor will shunt away nasty spikes which may get superimposed on your car 12V power line.

A regulator will not stop a HF spike and may even be damaged by one.

Remember the old in-line chokes? You might have one on your old car radio? You know - inductors... impedance, frequency etc.

So, something like: +CAR ---- Fuse ---- Diode --- Choke/Inductor ---- Transupp(to ground) --- Cap(to ground) --- regulator circuit -->etc.
(Thought: is your old car pos or neg earth??)

I'm sure if you use that Google thing you'll find some example circuits and a better description than I can give. The better your suppression then the safer your circuit and the better it will perform. It may seem OTT but better safe than sorry.

Also, right next to your PICAXE +V pin have a cap or two down to ground. Common practice is something like a 100nF ceramic // with a few microfarad electrolytic or tantalum. No need to overdo it though.

Good luck.
 
P

pichutch

New Member
Hi, quick note to say that the water pump has failed on the car I was hoping to do this for, so I won't be able to test pickups etc for a few weeks until I can get around to replacing the pump and rad. Thanks for all the help so far though!
 
You must log in or register to reply here.
Top