'82 450 Igniter - Repair, Replace, or Make?

[Guest]

Wedding anniversary today, so nothing doing with the ignition stuff, however I did receive my package from Jaycar with a lot of resistors and a couple of LM741 op amps in it.

I also did take 5 minutes to put the oil in today as well so I can crank it next time I'm working on it.

Fingers crossed I'll get some time tomorrow night.

 

[bakalorz]

Wedding anniversary today, so nothing doing with the ignition stuff, however I did receive my package from Jaycar with a lot of resistors and a couple of LM741 op amps in it.

I also did take 5 minutes to put the oil in today as well so I can crank it next time I'm working on it.

Fingers crossed I'll get some time tomorrow night.

I've only been sporadically following this thread, but have some thoughts on what I read today ... FWIW

I think that if you are worried about the timing, an opamp is probably the wrong way to go. a pure transistor stage is likely to be MUCH faster than most opamps.

For that matter I don't think you can get the 741 to work at all.
(Well, If you had to, as matter of life and death, or to win a big enough bet, you probably could by playing games with level shifting and stuff ... but not without a boatload of agravation)
For an opamp you will have to set up a proper positive AND NEGATIVE supply voltage. There are single supply opamps that will work with the bikes ground being the negative rail and can operate near it. The 741 is NOT one of them, it would need something like 12 volts, ground and MINUS 12 volt supplies.

If you MUST use an opamp, something like the LM324 will work better.
it can operate with both the inputs and output being near ground.
(if you cant get the 324 easily, search for "single supply opamp" to find others that would be ok. after you find a possible candidate, find and read its datasheet by googling the part # and pdf for example "lm324 pdf" for this one)

All the above said, I think a single transistor amp is likely to work better, and be much easier to get working correctly.

Also, don't forget that a single transistor amplifier produces an inverted output by default (which seems to be what you need)

I suspect that with the correct biasing a single bipolar NPN transistor amplifier would work well.

If it were me messing with it, thats what I would use.

To set the biasing, feed the search term "transistor biasing tutorial" into google, one of those should work.

 

[Guest]

I've only been sporadically following this thread, but have some thoughts on what I read today ... FWIW

I think that if you are worried about the timing, an opamp is probably the wrong way to go. a pure transistor stage is likely to be MUCH faster than most opamps.

For that matter I don't think you can get the 741 to work at all.
(Well, If you had to, as matter of life and death, or to win a big enough bet, you probably could by playing games with level shifting and stuff ... but not without a boatload of agravation)
For an opamp you will have to set up a proper positive AND NEGATIVE supply voltage. There are single supply opamps that will work with the bikes ground being the negative rail and can operate near it. The 741 is NOT one of them, it would need something like 12 volts, ground and MINUS 12 volt supplies.

If you MUST use an opamp, something like the LM324 will work better.
it can operate with both the inputs and output being near ground.
(if you cant get the 324 easily, search for "single supply opamp" to find others that would be ok. after you find a possible candidate, find and read its datasheet by googling the part # and pdf for example "lm324 pdf" for this one)

All the above said, I think a single transistor amp is likely to work better, and be much easier to get working correctly.

Also, don't forget that a single transistor amplifier produces an inverted output by default (which seems to be what you need)

I suspect that with the correct biasing a single bipolar NPN transistor amplifier would work well.

If it were me messing with it, thats what I would use.

To set the biasing, feed the search term "transistor biasing tutorial" into google, one of those should work.

Thanks for that Martin.

I'm far too rusty with all my electronics theory to even pretend to remember how to set op amps up, so good to know whether I should or shouldn't go down that path. Speed of response was one question I had with them, but I ended up needing to order from Jaycar before I had time to look closely at the specs.

They're cheap enough that I'm not concerned if I don't use them, but I think I did see LM324's on the Jaycar site as well.

Either way, I have lots of BC547 transistors sitting around, I was just missing a lot of resistor values that I now have.

Thanks for the links too, that'll be very helpful along with Lou's web page on his GPZ and the CB HEI setups.

I also think I've forgotten to put a link in here to Lou's site, so here it is:

http://www.geocities.ws/loudgpz/

 

[bakalorz]

Thanks for that Martin.

I'm far too rusty with all my electronics theory to even pretend to remember how to set op amps up, so good to know whether I should or shouldn't go down that path. Speed of response was one question I had with them, but I ended up needing to order from Jaycar before I had time to look closely at the specs.

They're cheap enough that I'm not concerned if I don't use them, but I think I did see LM324's on the Jaycar site as well.

Either way, I have lots of BC547 transistors sitting around, I was just missing a lot of resistor values that I now have.

Thanks for the links too, that'll be very helpful along with Lou's web page on his GPZ and the CB HEI setups.

I also think I've forgotten to put a link in here to Lou's site, so here it is:

http://www.geocities.ws/loudgpz/

I looked at Lou's site and have a more detailed answer to what I would try.

Using Fig 1 on his HEI for points page:
connect everything to the right of the transistor as shown.
Use an R1 of 4.7k as shown.
Replace where he shows points with a resistor about 100 to 470 ohm. (we'll call this resistor R2)
Also connected to the base of the transistor: a capacitor with the other end of the capacitor connected to the pickup coil. At least 1 uf, bigger is probably better.

Note, while setting the operating point, disconnect the coils so they dont overheat.
To set the operating point (and dwell), adjust the ratio of resistors R1 and R2 ... for the coil to be off, the transistor should be turned on, with the base above about 0.7 volts.
Use your multimeter to measure the voltage at point G on the HEI module as you try various values of R2.
with R2 = 470, point G should be near ground.
with R2= 100, point G should be high (well above 1.65 volts anyway)
try various values of R2, Use the smallest one that still keeps G near ground (i.e. will result in coils being off when not getting signals from the pickup)
If this results in too much dwell, increase R2

To test for spark and correct assembly/bias, hook up coils, and briefly short R2 (i.e. short the transitor base to ground), when you break the short, the spark should occur.
If it passes this test, try it with the pickup and a magnet.


IF you really get a negative signal from the pickup(and it going positive again signals the spark) this should work.

If you really get a positive signal but still need amplification, a second transistor stage would be required.

 

[Guest]

I looked at Lou's site and have a more detailed answer to what I would try.

Using Fig 1 on his HEI for points page:
connect everything to the right of the transistor as shown.
Use an R1 of 4.7k as shown.
Replace where he shows points with a resistor about 100 to 470 ohm. (we'll call this resistor R2)
Also connected to the base of the transistor: a capacitor with the other end of the capacitor connected to the pickup coil. At least 1 uf, bigger is probably better.

Note, while setting the operating point, disconnect the coils so they dont overheat.
To set the operating point (and dwell), adjust the ratio of resistors R1 and R2 ... for the coil to be off, the transistor should be turned on, with the base above about 0.7 volts.
Use your multimeter to measure the voltage at point G on the HEI module as you try various values of R2.
with R2 = 470, point G should be near ground.
with R2= 100, point G should be high (well above 1.65 volts anyway)
try various values of R2, Use the smallest one that still keeps G near ground (i.e. will result in coils being off when not getting signals from the pickup)
If this results in too much dwell, increase R2

To test for spark and correct assembly/bias, hook up coils, and briefly short R2 (i.e. short the transitor base to ground), when you break the short, the spark should occur.
If it passes this test, try it with the pickup and a magnet.


IF you really get a negative signal from the pickup(and it going positive again signals the spark) this should work.

If you really get a positive signal but still need amplification, a second transistor stage would be required.

Thanks again Martin, this is great stuff.

John and I didn't go into this much detail while chatting on Sunday, but he definitely advised going down the same path as you're suggesting here, so it's good to see we have multiple people on the same wavelength here.

I sort of understand where you're going with this and I believe I should have suitable capacitors around too, but I'll check hopefully tonight.

I'll give this a go and report back on the results.

 

[Guest]

Really short on time tonight as I'm on call and am expecting to be called in about 2 1/14 hours to do some work and wanna get a nap in first.

Lots of resistors and the possibly extraneous LM741's



I ducked down to do some cranking, and it was hard to get a consistent signal as I didn't want to crank for too long due to the noise.

Couldn't get a decent result with the black/white common wire grounded, but I got this with the scope just across the one coil:



Same type of waveform as cranking by hand gave me, but a lot higher amplitude at 89mV.

I really need to get some daytime time to get cranking without any need to worry about noise.

Martin, hopefully tomorrow night I'll have some more time to try your suggestions as that's what I really wanted to get to tonight but didn't quite have time.

 

[bakalorz]

Really short on time tonight as I'm on call and am expecting to be called in about 2 1/14 hours to do some work and wanna get a nap in first.

Lots of resistors and the possibly extraneous LM741's



I ducked down to do some cranking, and it was hard to get a consistent signal as I didn't want to crank for too long due to the noise.

Couldn't get a decent result with the black/white common wire grounded, but I got this with the scope just across the one coil:



Same type of waveform as cranking by hand gave me, but a lot higher amplitude at 89mV.

I really need to get some daytime time to get cranking without any need to worry about noise.

Martin, hopefully tomorrow night I'll have some more time to try your suggestions as that's what I really wanted to get to tonight but didn't quite have time.

Your scope capture looks like a positive waveform, so I suspect you don't want to invert it. so, if you need amplification you would use two stages and invert it twice.

attached is a schematic with both versions, the top one is the inverting amp,
the bottom one is non inverting.

The resistance of R2 that my simulator says is about right (assuming a generic transistor) is about 280 ohms FWIW

I also showed an optional R3, it shouldn't be needed, but if you cant get it to work right by adjusting R2 or R7, try adding it

 

[bakalorz]

Your scope capture looks like a positive waveform, so I suspect you don't want to invert it. so, if you need amplification you would use two stages and invert it twice.

attached is a schematic with both versions, the top one is the inverting amp,
the bottom one is non inverting.

The resistance of R2 that my simulator says is about right (assuming a generic transistor) is about 280 ohms FWIW

I also showed an optional R3, it shouldn't be needed, but if you cant get it to work right by adjusting R2 or R7, try adding it

trying a better attachment

 

[Guest]

Thanks again Martin, I'll endeavour to get something wired up tonight and start having a look.

One thing I know for sure is I need to invert the signal as the HEI module fires off the negative going signal. I have verified this by putting a 1.5v AA battery across the W to G terminals on the module.

 

[john82q]

Go with R3

Go with R3

Pete ,

Bakalorz's inverting amp is what you want, it s a more elegant version of what i was doodling the other day.

I am wondering about the HEI input, does it withstand 12V? I woulds use the r3 .

john

 

[bakalorz]

Thanks again Martin, I'll endeavour to get something wired up tonight and start having a look.

One thing I know for sure is I need to invert the signal as the HEI module fires off the negative going signal. I have verified this by putting a 1.5v AA battery across the W to G terminals on the module.

I got the impression that what you want is for it start running the coil on the positive part of the ramp, dwell while the peak is high, and then fire the coil after the peak as the ramp descends. so on the negative going signal, but using a positive amplification.

 

[Guest]

Ok, did what I could to get stuck into it tonight, minus more cranking to get a better signal read out. I may (slim chance) get an opportunity to try cranking it again tomorrow before I take my wife to an ex-work mate's funeral.

Anyway, I have 270 and 4.7K ohm resistors, BC547's, 135k marked green caps, and 22uF electrolytic capacitors. They're the only capacitors I have that seem suitable, and hopefully I'm right that 135k green caps are 1.3uF. Not 100% sure on the suitability to this application, but I didn't think electrolytics were right.



So, given that it appears at this stage that my pickup coils generate a small but positive signal when a spark should fire, I'm using the single transistor inverting circuit from Martin, and John, I did add R3 (4.7K ohm).





Yes, the connections do look dodgy, but they work. Every time I'm hooking a new circuit up, I'm continuity testing with the mulitmeter.

So, I disconnected the coils and hooked it up, and noticed that instead of about 12.5v being available, the circuit appears to be taking it down to around 11.7v.

Also, I discovered that with R2 being 270 ohm (no 280 ohms in my big pack), the base of the transistor was only at about 0.63v, so I upped R2 to 330 ohm which took it to about 0.7v. I'm guessing I may need bigger still due to the voltage drop to 11.7v.

So, I tried shorting R2 out to ground to generate a spark, but it doesn't seem to do much at all, and in fact after that's done, C on the HEI module stays at about 1.6v and doesn't return to 11.7v.

I hope what I'm saying here is making sense because I'm quite confused by it all still at the moment.

My biggest confusion right now is how the biasing of the base of the transistor works with the pickup coil.

What I mean is... we've biased it to about 0.7v to keep it turned on, which in turn keeps the coil turned on.

When the positive signal comes in from the pickup coil, won't that attempt to turn it "more on" and therefore not turn the coil off and cause a spark?

I realise I'm going to be missing some fundamental piece of electronics knowledge when I ask that question, but my head's just spinning round in circles trying to work it out.

I need to go read through those transistor biasing links again and see if I can soak some of that info in I think.

 

[Guest]

I got the impression that what you want is for it start running the coil on the positive part of the ramp, dwell while the peak is high, and then fire the coil after the peak as the ramp descends. so on the negative going signal, but using a positive amplification.

Hmmm I see what you're saying... most likely I'm conveying information poorly.

What the manual says to do to test the ignitor works is to connect a multimeter on the lowest resistance range from the black/white (common) pickup wire to either the green/white or brown pickup wire that is the input to the ignitor.

Connecting the probes should fire the spark, so the positive signal should fire the spark.

Does that make sense?

If the negative signal should fire the spark, I imagine they would say removing the leads would do that, which is what John and I saw in his manual for the 850 ignitor test.

 

[Guest]

Oh, and of course the other important part is that the negative going signal is what causes the HEI module to turn the coil off and fire the spark, which is the exact opposite to the pickup coil.

 

[john82q]

I might be wrong (often are) but I was under the impression the IC in the HEI module looked after the dwell based on the frequency of the triggering.

That being the case its input just needs to be held above the theshold (say 1.5 v for arguments sake)and you need is a negative going pulse to make it fire.

data sheet for the chip in the module.
http://www.datasheetarchive.com/MC3334-datasheet.html

john

 

[Guest]

I believe that's exactly right John.

Voltages are mentioned earlier in the thread, above 1.65v to turn it on and below 1.37v to turn it off if I remember correctly.

My tired and worn out brain is not understanding how the signal from the pickup can turn the transistor on to generate the negative signal required...

 

[bakalorz]

I might be wrong (often are) but I was under the impression the IC in the HEI module looked after the dwell based on the frequency of the triggering.

That being the case its input just needs to be held above the theshold (say 1.5 v for arguments sake)and you need is a negative going pulse to make it fire.

data sheet for the chip in the module.
http://www.datasheetarchive.com/MC3334-datasheet.html

john

from looking at Lou's website, I got the impression that the IC adjusted dwell using the W terminal (which we aren't using in this configuration)

If you only use the G terminal, I think the only think that matters is if the voltage is above or below the threshold.

You still end up getting variable dwell because the faster the bike runs, the bigger (and wider) the pulse from the pickup.

one possible way to reconcile using a negative pulse with the Oscope trace would be to switch the two leads from the pickup, this would make the trace go the opposite way. That would let you use the inverting amp.

Two things I forgot before:

You could replace R2/R7 with potentiometers (500 ohms or 1k or so) to make finding the right value easier (and potentially adjust dwell as its running)

for the non-inverting one, changing R7 works the opposite way around ... raise R7 to turn on the coil, and lower R7 to turn it off.
to test after its built, parallel R6 with an additional 4.7k resistor to turn on the coil, spark should occur when you disconnect the additional paralleled resistor.
(on the non-inverting amp you can't just connect the base to 12 volts, that would smoke the transistor)

 

[Guest]

from looking at Lou's website, I got the impression that the IC adjusted dwell using the W terminal (which we aren't using in this configuration)

If you only use the G terminal, I think the only think that matters is if the voltage is above or below the threshold.

You still end up getting variable dwell because the faster the bike runs, the bigger (and wider) the pulse from the pickup.

one possible way to reconcile using a negative pulse with the Oscope trace would be to switch the two leads from the pickup, this would make the trace go the opposite way. That would let you use the inverting amp.

Two things I forgot before:

You could replace R2/R7 with potentiometers (500 ohms or 1k or so) to make finding the right value easier (and potentially adjust dwell as its running)

for the non-inverting one, changing R7 works the opposite way around ... raise R7 to turn on the coil, and lower R7 to turn it off.
to test after its built, parallel R6 with an additional 4.7k resistor to turn on the coil, spark should occur when you disconnect the additional paralleled resistor.
(on the non-inverting amp you can't just connect the base to 12 volts, that would smoke the transistor)

I think you're both right, in that yes the dwell on the HEI module is adjusted by the bias on the W connector, but there is also dwell in relation to the pickup coil. I don't fully get the whole dwell thing, but I do see that both the W and pickup coil come into it.

Martin, when I hooked it up last night, I was connecting W to 12v like the points setup on Lou's page showed, so was this actually wrong and I should've left it disconnected?

The other thing I'm looking at now is the unconfirmed GS450 ignitor schematic.



I can't read the component values on there, but it seems they use a 1K resistor combined with maybe a 3.8K resistor, a zener diode, and maybe a 0.22uF capacitor to couple the pickup coil to the base of the transistor, and a 4.7K resistor is also used to connect to the collector.

I do actually have some zener diodes, but I don't know what type they are or if that actually matters.

Is it worthwhile trying this setup at all? I can't get my brain around how that should work though.

There is also the attached schematic the German fellow did that uses a 47K resistor in conjunction with a diode to couple the signal to the base of the transistor, and this one I can actually sort of comprehend.

I'm not trying to second guess any of your ideas here either guys, I'm just trying to get my brain around it all and at the moment I'm not succeeding at all.

I also now am thinking that my perceived need to invert the signal is actually not quite right.

For the positive going signal to fire a spark, it needs to turn the coil off, which means the transistor needs also to be off. Does that sound right?

So going just by the German fellow's circuit, normally T1 will be off and T2 will be on, meaning the coil will therefore be on.

When the signal generator generates its tiny positive going signal and therefore raises the voltage slightly on the negative side of D1, this turns T1 on, thereby turning T2 off, and therefore firing a spark by turning the coil off again.

I imagine this is also the exact same theory behind your two transistor setup too Martin, so probably that's what I should've been trying.

Sorry for my confusion here and if it's making you guys confused too...

 

[Guest]

Martin I just realised if I go with your second schematic, I'll be one short on 4.7K resistors, so will it reduce the current too much if I use a 5.6K for R8 or should I go with 3.9K? These are the values I have on hand and all are in 1/4 watt rating.

 

[bakalorz]

Martin I just realised if I go with your second schematic, I'll be one short on 4.7K resistors, so will it reduce the current too much if I use a 5.6K for R8 or should I go with 3.9K? These are the values I have on hand and all are in 1/4 watt rating.

I pretty much used 4.7k everywhere because thats what Lou used. if R8 is too small, the resistors (and transistor if WAY too small) will overheat. If its too large, it may cause problems providing enough drive to the HEI module. my offhand guess is that <2k is too small ... too large would depend on the module ... It could be 10k, 20k, 100k or infinite ... no way to know without testing, but apparently Lou didn't need it, so you can skip it if you need to. I put it in for insurance, and it makes it possible to test the circuit without having the HEI module connected. It might even be bad and make the circuit not work if it drives the HEI TOO high....
I would probably make it 5.6 k or skip it. If you are making two modules though (for two cylinders) make both the same so they work symetrically