Ignition lead sensor

K

kando

Senior Member
Hi,
I need a bit of help if possible on the subject of sensing when the spark plug fires on a motorbike from the ignition lead. (four stroke or two)
I know that there used to be strobe lights that wrapped something around an ignition lead and flashed the light.

I suppose my question is what is the thing that went around the ignition lead and gave the signal and can I make the same signal for a Picaxe chip?

I need to move the thing from ignition lead to ignition lead so I need to be able to unhook it while the engine is going.

Thank you for any help.
:eek:
 
K

kando

Senior Member
Thanks for the link eclectic,
Flooby said :"For an inductive pickup you can use a "clam shell" ferrite choke designed to clamp around a wire"

Its the inductive clam shell pickup that I remember and I see they are still used on the strobe lights for ignition timing. Can you buy them separate?

I just need it to trigger a response in a Picaxe chip so that I can trigger another sensor. The picaxe may be too slow to react and I may not need it in any case as it is the colour from the cylinder that I'm interested in when the spark ignites the fuel and that lasts for a few milliseconds or so.

I suppose I could do without it and average the colour.

I hope this makes sense. It's an idea that has been bugging me for sometime and I need to try it out.
 
E

eclectic

Moderator
kando said:
Thanks for the link eclectic,
Flooby said :"For an inductive pickup you can use a "clam shell" ferrite choke designed to clamp around a wire"

Its the inductive clam shell pickup that I remember and I see they are still used on the strobe lights for ignition timing. Can you buy them separate?
Click to expand...
One of many
http://www.radioworld.co.uk/catalog/rff-213_ferrite_core_block_for_rg-213-p-4055.html?osCsid=ar04a3sn2ji6src777ddbme6f5

See any Ham radio store, or Ebay ....................


e
 
D

Dippy

Moderator
My old Gunson timing light from 20+ years ago had a simple coil with a little plastic strap with a pimple and a hole. It wrapped around the lead to keep it in place.
I don't know if they're available separately. Have a Google.
I'd consider modifying a plastic cable clip.

Are you on about the colour of the combustion?
I used to have another Gunsons thing which replaced the spark plug. It had a glass collar so you could see inside.
How is that related to the PICAXE being triggered?

Sorry, to me , it makes no sense what you are trying to do.
If you want help I reckon you'd better explain clearly what you actually trying to achieve.

btw: How are you going to interface the induced pulse to PICAXE?
What is your skill level with electronics?

Oh good old Eclectic...
 
K

kando

Senior Member
Dippy said:
My old Gunson timing light from 20+ years ago had a simple coil with a little plastic strap with a pimple and a hole. It wrapped around the lead to keep it in place.

Are you on about the colour of the combustion?
I used to have another Gunsons thing which replaced the spark plug. It had a glass collar so you could see inside.

Sorry, to me , it makes no sense what you are trying to do.
If you want help I reckon you'd better explain clearly what you actually trying to achieve.

btw: How are you going to interface the induced pulse to PICAXE?
What is your skill level with electronics?
Click to expand...
Thanks for that Dippy
Yes I used to have a Gunson's strobe light and Gunson's Colourtune which is what you are referring to.
It's the latter that I want to now go back to with an old motorcycle. You know the type... the ones with proper carbs, none of the injection electronic stuff of now-a-days.
I would like to just get a better colourtune using an colour sensor down the black tube to get a more accurate carb setting.
(did you know you can still buy the Colourtune kit?)

anyway so now you know what my idea is do you think I need a connection to the HT lead or is it better just to use the colours sensor and average the colour as the picaxe might be too slow to pick up every pulse from the HT lead?
 
R

rossko57

Senior Member
I don't see why you'd need to bother with ignition pickup at all; you're looking at combustion flashes and inbetween it are all dark.

I expect it would be beneficial to use some kind of light guide, bunch of fibres perhaps, so you can collect light at the quartz window (assuming you are using a colortune as base) which gets ruddy hot, but keep your sensor far away in the relative cool - and also not alongside the ignition conductor, where you might get pickup you wouldn't want!

Have you considered stuffing a standard oxygen sensor up the exhaust pipe on a stick instead? It does not give the same information, but is generally useful for similar purposes.
 
D

Dippy

Moderator
Ah yes, Colourtune with black tube (light shade) that falls off :) I remember it (nearly) well.

Are those symbols LDRs?
What is their response time?
Can you ReadADC in time? (3 channels in the space of a flash).
What about induced noises in analogue signal lines and power lines?
I wouldn't use that circuit.


You haven't said how you are going to translate an induced 'spark' into a PICAXE trigger...
You can't just shove a coil onto an input. Well, yes, you can.... but....
I take it you're not too experienced in electronics?


You're going to have to wait for the Big Boys to turn up.
At a complete guess I'd say that you may need to read up on ;voltage limiting and suppression, bypass and filtering, monostables and maybe even sample&hold.
I can imagine quite a lot of thoughts and suggestions so I'm going to run away :)
 
K

kando

Senior Member
Dippy said:
Ah yes, Colourtune with black tube (light shade) that falls off :) I remember it (nearly) well.
Click to expand...
Yes that's the one, it still works fine and I have four of them for the 4 cylinders of my motorcycle.
I just had an idea, rather than me peering down the black tubes, risking burning on exhaust pipes that the Picaxe (New found friend) could help out getting it more safe and more than me judging whether the left hand cylinder is brighter blue than the right hand cylinder. Giving it a bit of digital sameness so I could get the mixture slightly better on the four old classic carbs.

Dippy said:
Are those symbols LDRs?
Click to expand...
Yes
Dippy said:
What is their response time?
Click to expand...
No idea, theoretical at present
Dippy said:
Can you ReadADC in time? (3 channels in the space of a flash).
Click to expand...
I was hoping some here could tell me that. Perhaps I should take a RGB LED and try it that way or like the sweet sorter. I'm open to suggestions.

Dippy said:
What about induced noises in analogue signal lines and power lines?
Click to expand...
Now you have just gone past my knowledge level of electronics. I guess you mean other electrical interference from the HT leads. Can I not just surround my wires with a screened earth?
Dippy said:
I wouldn't use that circuit.
Click to expand...
I guess your saying, go away and have a think about it some more?

Dippy said:
You haven't said how you are going to translate an induced 'spark' into a PICAXE trigger...
You can't just shove a coil onto an input. Well, yes, you can.... but....
I take it you're not too experienced in electronics?
Click to expand...
I was guessing that a trigger could be done from an HT lead after all the old timing lights attached in the same way triggered the strobe light in someway.
You take it correct, about my electronics background. Nil. I learn all my stuff from you guys, tests and fails, and lots of reading. Great fun though!

Dippy said:
At a complete guess I'd say that you may need to read up on ;voltage limiting and suppression, bypass and filtering, monostables and maybe even sample&hold.:)
Click to expand...
Thanks for the heads up on more reading. One day I may get to be average at electronics (don't think I'll ever be good). :)
 
hippy

hippy

Technical Support
Staff member
kando said:
I just had an idea, rather than me peering down the black tubes, risking burning on exhaust pipes that the Picaxe (New found friend) could help out getting it more safe and more than me judging whether the left hand cylinder is brighter blue than the right hand cylinder. Giving it a bit of digital sameness so I could get the mixture slightly better on the four old classic carbs.
Click to expand...
It seems what you are after is a 'digital colourtune' which reveals the richness of the air/fuel mix on some display which makes it easier or more accurate than a direct visual inspection would.

I am sure that could be done and a HT pick-up may help with doing that or may be unnecessary.

The main requirement, if using a modified colourtune system, would seem to be knowing what the characteristics of the thing being measured is ( the burn and its colour ) and determining what sensors would be appropriate for measuring that.

The alternative would seem to be measuring the air/fuel mix some other way.

Having determined how to measure, the issue then becomes how to get a PICAXE to do the measuring. This may affect what the solution is in needing to find a sensor which does the job and can be used with a PICAXE.

I am not sure there is a ready made answer unless someone has already done the same, with PICAXE or something else, and there may need to be some research and experimentation.
 
D

Dippy

Moderator
I haven't got much time as I have nuclear reactor to build, so this is rushed and guessed....

I reckon a fast synchronised system is biting off more than you can masticate.
So, suspecting that are on the point of giving up, I'm going to think out loud....

Forget the sensing of spark. Forget synchronising, I doubt if it will be reliable even if you can handle the electronics.
Your sensors are probably way too slow. Investigate.
Run the outputs of the 3 sensors (via filters) into a op-amp integrator and simply measure the 3 output voltages.
This should give a running average of many flashes. It may need smoothing.
The 3 voltage levels will change as the colour changes.
You may have to do some multi-sampling and averaging but it could make life a lot easier.

You may have to compensate for the relative sensitivity of your sensors will colour and amplitude ; I doubt if they are similar or linear.
You will have to protect your circuit against transients in signal power line. Bypass/decouple and all that jazz.
Maybe even transient suppression if you are close to spikes and HT.
Ambient light may be an issue - that's another problem.

And, of course, plenty of sellotape to hold it all together :)

Job done. Right now I'm off to file a bit off that control rod :)
 
K

kando

Senior Member
hippy said:
It seems what you are after is a 'digital colourtune' which reveals the richness of the air/fuel mix on some display which makes it easier or more accurate than a direct visual inspection would.
Click to expand...
(exactly correct Hippy)
hippy said:
I am sure that could be done and a HT pick-up may help with doing that or may be unnecessary.
Click to expand...
I think it may also be unnecessary.
hippy said:
The main requirement, if using a modified colourtune system, would seem to be knowing what the characteristics of the thing being measured is ( the burn and its colour ) and determining what sensors would be appropriate for measuring that.
Click to expand...
Said so succinctly. Wish I could be the same but I ramble on. Yes, and I know the colours as the should be at different revs.
hippy said:
Having determined how to measure, the issue then becomes how to get a PICAXE to do the measuring. This may affect what the solution is in needing to find a sensor which does the job and can be used with a PICAXE.
Click to expand...
I believe a little more investigation is needed on my behalf and I will have it. (one is always optimistic)!

Dippy said:
Forget the sensing of spark. Forget synchronising, I doubt if it will be reliable even if you can handle the electronics.
Your sensors are probably way too slow.
Click to expand...
Too slow I agree and will be forgetting them for now until later test.
Dippy said:
Run the outputs of the 3 sensors (via filters) into a op-amp integrator and simply measure the 3 output voltages.
This should give a running average of many flashes.
Click to expand...
I was thinking 3 LDRs connected to three separate picaxe pins and to sense them one after the other into three different variables say: b1,b2 and b3 then combine in w3 then cut-out the unwanted colours, say anything over value 450 and under 150 or something to relate to the colours I want to find. Basically blues and yellows.

Why the Op-amp? is the ADC no good?

Thanks for the input guys it's really helpful.
 
K

kando

Senior Member
So I am now thinking 3 LDRs
4-5 LEDs for indicators or later could be OLED screen
Here is a bit of code I've knocked up with the subject in mind..
I'm thinking the ADC on B1,B2 and B3 and the indicator Leds on the C pins.
I think it may need a lot more work and I'm hoping I'm on the right track. :)
As I don't have to flash the colours because they are already being flashed by the cylinders. I believe it may need some smoothing as Dippy says but is it going the right way?

Code: 
[color=Green]; *******************************
;colour sensor 2
; Kando
; *******************************[/color]
[color=Navy]#picaxe [/color][color=Black]14M2[/color][color=DarkCyan]+[/color]
[color=Green];Read the LdS/cds sensors on B.3 B.4 and B.5  
;Set all outputs high [/color]
[color=Blue]symbol [/color][color=Black]RedVal[/color][color=DarkCyan]=[/color][color=Purple]b0[/color]
[color=Blue]symbol [/color][color=Black]GreenVal[/color][color=DarkCyan]=[/color][color=Purple]b1[/color]
[color=Blue]symbol [/color][color=Black]BlueVal[/color][color=DarkCyan]=[/color][color=Purple]b2

dirsB [/color][color=DarkCyan]= [/color][color=Navy]%00000111 [/color][color=Green]; set pinsB 3,4&5 to be ADC inputs[/color]
[color=Black]main: [/color]
[color=Purple]w3 [/color][color=DarkCyan]=[/color][color=Navy]0 [/color][color=Green]'This is accumulate the sum of all channels 

'Set the red channel low turns it on. [/color]
[color=Blue]readadc B.3[/color][color=Black],RedVal [/color]
[color=Blue]serout [/color][color=Navy]0[/color][color=Black],[/color][color=Blue]T2400_4[/color][color=Black], [/color][color=Blue]([/color][color=Black]#RedVal[/color][color=Blue])  [/color]
[color=Green]'Repeat for Green Channel [/color]
[color=Blue]readadc B.4[/color][color=Black],GreenVal [/color]
[color=Blue]serout [/color][color=Navy]0[/color][color=Black],[/color][color=Blue]T2400_4[/color][color=Black], [/color][color=Blue]([/color][color=Red]","[/color][color=Black],#GreenVal[/color][color=Blue])  [/color]
[color=Green];Repeat for Blue Channel [/color]
[color=Blue]readadc B.5[/color][color=Black],BlueVal [/color]
[color=Blue]serout [/color][color=Navy]0[/color][color=Black],[/color][color=Blue]T2400_4[/color][color=Black], [/color][color=Blue]([/color][color=Red]","[/color][color=Black],#BlueVal,[/color][color=Navy]13[/color][color=Black],[/color][color=Navy]10[/color][color=Blue]) [/color]

[color=Purple]w3 [/color][color=DarkCyan]= [/color][color=Black]RedVal [/color][color=DarkCyan]+ [/color][color=Black]GreenVal [/color][color=DarkCyan]+ [/color][color=Black]BlueVal [/color]

[color=Blue]if [/color][color=Purple]w3 [/color][color=DarkCyan]> [/color][color=Navy]450 [/color][color=Blue]then [/color][color=Black]nocolor [/color][color=Green];Cutout at high intensity [/color]
[color=Blue]if [/color][color=Purple]w3 [/color][color=DarkCyan]< [/color][color=Navy]150 [/color][color=Blue]then [/color][color=Black]nocolor [/color][color=Green];Cutout at high intensity [/color]
[color=Blue]if [/color][color=Black]RedVal[/color][color=DarkCyan]<[/color][color=Black]GreenVal [/color][color=DarkCyan]and [/color][color=Black]RedVal[/color][color=DarkCyan]<[/color][color=Black]BlueVal [/color][color=Blue]then [/color][color=Black]red [/color]
[color=Blue]if [/color][color=Black]GreenVal[/color][color=DarkCyan]<[/color][color=Black]RedVal [/color][color=DarkCyan]and [/color][color=Black]GreenVal[/color][color=DarkCyan]<[/color][color=Black]BlueVal [/color][color=Blue]then [/color][color=Black]green [/color]
[color=Blue]if [/color][color=Black]BlueVal[/color][color=DarkCyan]<[/color][color=Black]RedVal [/color][color=DarkCyan]and [/color][color=Black]BlueVal[/color][color=DarkCyan]<[/color][color=Black]GreenVal [/color][color=Blue]then [/color][color=Black]blue [/color]
[color=Blue]goto [/color][color=Black]main 

nocolor: [/color]
[color=Green]'sertxd ("nocolor",13,10) [/color]
[color=Purple]pinsC [/color][color=DarkCyan]= [/color][color=Navy]%00000000 [/color][color=Green]; set pinsC 0 1 2 to low [/color]
[color=Blue]goto [/color][color=Black]main 

red: [/color]
[color=Purple]pinsC[/color][color=DarkCyan]= [/color][color=Navy]%00000001 [/color][color=Green]; set pinsC 0 to high pinsC 1 2 to low [/color]
[color=Blue]goto [/color][color=Black]main 

blue: [/color]
[color=Purple]pinsC[/color][color=DarkCyan]= [/color][color=Navy]%00000010 [/color][color=Green]; set pinsC 1 to high pinsC 0 2 to low[/color]
[color=Blue]goto [/color][color=Black]main 

green: [/color]
[color=Purple]pinsC[/color][color=DarkCyan]= [/color][color=Navy]%00000100 [/color][color=Green]; set pinsC 2 to high pinsC 0 1 to low[/color]
[color=Blue]goto [/color][color=Black]main

Yellow:[/color]
[color=Purple]pinsC[/color][color=DarkCyan]= [/color][color=Navy]%00001000 [/color][color=Green]; set pinsC 4 to high pinsC 0 1 2 to low[/color]
[color=Blue]goto [/color][color=Black]main[/color]
 
R

rossko57

Senior Member
kando said:
Why the Op-amp? is the ADC no good?
Click to expand...
You're looking for flashes of unknown duration, with bigger gaps of darkness inbetween.
You could try synching up to ignition, but that gets complicated quickly.
Or, integrate - capture and hold the value of the pulses from the sensor, process at liesure.

Were you thinking LDR with coloured filters? I think response times are in the 100mS league - too slow at first glance ... on the other hand, that solves your integration problem; light pulses will be averaged and smoothed.

Phototransistors (or purpose built colour sensors) are quick enough to capture individual flashes. Not needed if you are integrating. Or it could allow processing of individual flashes, if the processor can keep up.

Probably worth thinking about what the human eye+brain can glean from a colortune.
Color - basically on a scale from yellow to blue - will just two sensors determine the color here?
Consistency - at tickover, your eye can pick out misses and irregularities ("dollopy fuelling") - do you seek to reproduce this ability? This would drive a need for speed.
Change - similar to above, like blipping throttle to observe throttle pump effect etc.
Brightness/intensity is I think irrelevant? That would be handy for electronic calibration, as all you'd be interested in is relative colour.
How would you envisage "output" here - some arbritary number representing the colour? Just reproducing the colour on an RGB LED?
 
K

kando

Senior Member
Thanks Jim Perry and Rossko57 for the replies,
Jim, the colour sensor is very good value but has LEDs to light up the sensor in the middle so it uses its own light source to flash at another surface and the bounce back is calibrated. Here I need just the sensor as the light source is created by the cylinder igniting. (Good thing for another project though).

Rossko57,Smoothed ,Yes that sounds like what I think I need.I don't think just reproducing the colour on and RGB LED would be any better (other than getting my face a little further away from the hot source) Than looking at it in the normal way. I need to have a better solution than LEDs which is why I thought later an OLED screen which could show numbers or something when its correct.

There are only three tests and the basic colour is based on how you open the throttle and close it at certain revs.
(I can't get the colour picture uploaded at the moment...don't know why. Later)
Blues and yellows are the colours.
 
G

Goeytex

Senior Member
Back to the ignition lead sensor ...

Attached is a circuit that I have used on Lawn Mower Engines for a tachometer/ rev limiter. R1 and R4 can be Pots to allow for adjustments. R3 can also be a 10K.

L1 is 4 - 6 turns of solid copper hookup wire wrapped around the HT lead. You may need more or less depending upon the HT voltage.

The PC817D has a current transfer ratio of 300 to 600 so it works well with low current signals.

EDIT: I changed the One shot to a 74HC221 non retriggerable type so that the output pulse could be shorter than the input trigger. Also changed the output pulse width to 500us. Make R4 a 100 K pot to adjust the output pulse width (if needed.)
 

Attachments

Last edited:
neiltechspec

neiltechspec

Senior Member
I would have thought an exhaust gas analyser would be the way to go.

Try a Gunson Digital Gas Tester (available from Halfords for £80).

Another option is a Wideband Lamda sensor - albeit an expensive way to go when the control circuitry is added as well.

Neil.
 
D

Dippy

Moderator
I'm getting confused (it must be my age)...

Post#11: "Are those symbols LDRs?"
Post#12: "Yes"
Post#14: "Your sensors are probably way too slow."
Post#15: "Too slow I agree and will be forgetting them for now until later test."
Post#16: "So I am now thinking 3 LDRs"
Err..umm... that was a quick test!

Op-amp integrator:
(Example).
http://www.electronics-tutorials.ws/opamp/opamp_6.html
So, over a period of time (depending on component value) it sums-up the signal (from analogue sensors) - just like your eye,

It doesn't replace the ADC, it feeds the ADC.

The potential problem here is sequential reading. I'm guessing 3 x READADC is too slow for the flash period. (I'm not sure).
So, how do you get around this?
Well, an integrator or sample & Hold can give you time.
AND the 3 sensors would be working in parallel not sequentially.

Consider using a photo-diode based sensor, I reckon you'll have trouble with COTS Photo-Tr as they haven't got much dynamic range as they saturate easily.

Goeytex has provided the basis for a trigger to the PICAXE.
So, Trigger + Read or Trigger + (Start integrator / S&H) + Read. Or free-running integrator... I dunno.
Without experimenting I don't think you're going to get a solution served up.

I think I'd follow Neil's suggestion ;)
 
O

oracacle

Senior Member
this all sounds like a great project and all that, but there is a method that is cheaper, more accurate and still used by tuners even with the modern age of injectors and alikes - it normally refered to as hot plugging or a plug chop. its easiest with a rolling road but i have done it to a few bikes on normally roads, that combined with a decent balancing prior to riding should give the best results providing you know what colour the plugs should be.

I am also hoping that you have a full understanding of what changes will need to be made, simple things like having the correct float hight is very important altho seemingly insignificant when reading the manual for your machine (tuning a bike can also a require a change here), other things like needle hight, main, mid, primary jets and mixture settings, which should be kept the same accross the bank of carbs (a mismatch can cause and overstrain of the crank)

my 2 peneth worth is that humans are still doing this as it something that cant be done very well by machines (there would be a comercial version if it was possible), no matter how clever we make them.

however i do wonder what changes there may be with a chnage from normal 95 octane to 97 octane fuel (my bike loves the 97 with its 12 to 1 compression ratio, yeh 140psi+ per cylinder)

even with all that said, the other route that maybe simpler than colour, and is still used in the tuning world is the exhuast temperature, an IR temp sensor on the manafold as close the to the head as possible should give an indication of differences in the combustioncycle of each cylinder (you may want to look into thing like EGT).
 
K

kando

Senior Member
Dippy said:
Without experimenting I don't think you're going to get a solution served up.
Click to expand...
I think this says it all so armed now with some information and reading to do that's what I will be doing.

Thank you all very much for the help so far. I will get to do this after I have moved house..so some months probably. (more time to read).
 
K

kando

Senior Member
Goeytex: Thanks for the diagram and info. Very useful, can even use this on another project. I'm presuming the 1ms pulse line is going to a Picaxe pin. does it need any resistor or straight in?
 
G

Goeytex

Senior Member
Goeytex: Thanks for the diagram and info. Very useful, can even use this on another project. I'm presuming the 1ms pulse line is going to a Picaxe pin. does it need any resistor or straight in?
Click to expand...
Straight in is ok. However, be sure to discard the original circuit and see the edited circuit. I drew up the first circuit from memory and mistakenly used a 74HC123 that cannot be configured as non retriggerable.

The 74HC221 is natively non retriggerable and is what I actually used. This way the output pulse can be adjusted to be shorter than the input pulse. With R4 as a 100K POT, the output pulse can be adjusted from ~ 30 us to about 1.5ms.
 
K

kando

Senior Member
Goeytex said:
Straight in is ok. However, be sure to discard the original circuit and see the edited circuit. I drew up the first circuit from memory and mistakenly used a 74HC123 that cannot be configured as non retriggerable.

The 74HC221 is natively non retriggerable and is what I actually used. This way the output pulse can be adjusted to be shorter than the input pulse. With R4 as a 100K POT, the output pulse can be adjusted from ~ 30 us to about 1.5ms.
Click to expand...
Thank you. Have now adjusted the shopping list!
 
G

Goeytex

Senior Member
If you build the circuit make sure the coil polarity is correct ( Left hand rule). Start with 4 turns and include the 10K series resistor but nothing else.

With your scope ( 10X probe) You are looking for a 10 to 15 volt peak signal. It will look ugly. The BAT85 diodes break down at 30v. Once the number of turns has been determined to get a nominal 15v peak then the rest of the circuit can be added.

Alternatively you could use 1N4004 diodes on the front end. These are good to 400 volts and are good enough for this application.
 
K

kando

Senior Member
Goeytex said:
Alternatively you could use 1N4004 diodes on the front end. These are good to 400 volts and are good enough for this application.
Click to expand...
Thanks for the info: Was just trying it out in circuit wizard. No Bat 85s but had 1N4004 in its place seemed to work in the circuit.
Just one quick question as circuit wiz doesn't allow turns of wire just an inductor in Henrys. What henry would that be? (if not too much to ask) I could work it out I guess... more reading!
 
G

Goeytex

Senior Member
Just one quick question as circuit wiz doesn't allow turns of wire just an inductor in Henrys. What henry would that be? (if not too much to ask) I could work it out I guess... more reading!
Click to expand...
Less than .1 microhenries. And if you are modeling the circuit you will need to add a very small amount of series resistance and a few picofarads of parallel capacitance.
 
G

Goeytex

Senior Member
Pongo said:
What polarity are you assuming for the spark voltage/current?
Click to expand...
I would assume the center electrode of the spark plug to be negative in relation to the end electrode. Since the pickup wire ( inductor) and circuit up to the OPTO is isolated from the chassis, the "polarity" or direction of the induced current in the pickup coil is determine by the direction that the coil is wrapped around the HT lead.
 
Pongo

Pongo

Senior Member
Goeytex said:
I would assume the center electrode of the spark plug to be negative in relation to the end electrode.
Click to expand...
So would I - but it's probably counterintuitive and necessary information for anyone applying Mr. Fleming's rule.
 
K

kando

Senior Member
Thank you guys,
It takes me a little while to read the Flemming stuff and any other bits that have been suggested I read. All necessary stuff too. So it will take me a little time to get up to speed!
Pongo, quite glad you mentioned that. It makes me think. I assume that the centre electrode of a spark plug (the one with ceramic wrapped around it) was the positive and the bent metal bit overhanging that with the gap that you adjust to be the negative. is this not so? or am I completely being on the wrong track with what you are talking about?
 
D

Dippy

Moderator
Well, that's interesting. And I'm sure we're giving Google a pasting! Give it a go Kando.
I've learnt something today, thank you gents.
 
A

AllyCat

Senior Member
Hi,

Goeytex said:
If you build the circuit make sure the coil polarity is correct ( Left hand rule).
Click to expand...
It was (almost) 50 years ago, but I was taught to use the Right Hand Rule because "Motors (in the UK) use the Left and GeneRIGHTers use the Right". But there isn't really any "Motion" with a transformer anyway. :confused:

However, with "binary" decisions like this and so many "unknowns" (direction of coil windings, etc., etc.) and remembering that auto electrics used to use "Positive Earth" (it had been claimed to reduce burning of the Contact Breaker points), the chances of actually "working it out right" is so close to 50% that I'd just "suck it and see". Also the "signal" may approximate to a decaying sine wave, so both polarities are available.

Furthermore, in many cases, I wonder if the above circuit would actually detect the current (only mA?) or the voltage (>>10 kV) in the HT cable. Add a few pF of "stray" capacitance between the cable and the diodes and you have a classic "Voltage Pump" circuit (a "Voltage Doubler" or "Peak-peak detector" with a high load capacitance). ;)

But as suggested much earlier, why not just use the light detector(s) as your trigger?

Cheers, Alan.
 
K

kando

Senior Member
Why is everything so complicated.. I'll give it a try.. well after I have read some more!;)
Thank you guys..
 
You must log in or register to reply here.
Top