Yes there is a primary - please see the video in post 21

Okay, I see your point. There is s primary and secondary coil on the magneto, but this is not what @strantor and @LowQCab were talking about. They thought the coil was activated by an external voltage source of something like 12V. In reality both are activated by the magnet, and the lower voltage coil is still hundreds of volts. If you tap this you will be removing energy from the spark, perhaps enough to stop the engine from running.

A separate Hall sensor will have no effect on your ignition and WILL work without having to deal with high voltages. Your scheme might or might not be feasible.

 

Okay, I see your point. There is s primary and secondary coil on the magneto, but this is not what @strantor and @LowQCab were talking about. They thought the coil was activated by an external voltage source of something like 12V.

No that is not what I was talking about, and that is not what I thought. Please keep your clarifications limited to things that YOU said.

 

No that is not what I was talking about, and that is not what I thought. Please keep your clarifications limited to things that YOU said.

I was talking about this post, which, I think, is the first mention of primary. Sorry if I misunderstood.

The first article was talking about engines with powered coils and the author was measuring the primary. I didn't realized unpowered ones were different.

 

I was talking about this post, which, I think, is the first mention of primary. Sorry if I misunderstood.

That was post #7 and we are now above 40 posts. At that time it wasn't yet clear what kind of magneto system we were dealing with. That was cleared up in post #21. Since then I have been referring unambiguously to the primary of the type of magneto that this thread is about.

 

There are no guarantees that there will be no chance of .........
Random Ignition misfires,
Ignition Timing variations,
Smoked Tachometer Electronics,
etc., etc,
when attaching any external Circuitry to a completely unknown, crude, Ignition-Coil-Controller.

Spend ~$40.oo and an hour of your time installing a Hall-Effect-Sensor
and the above potential-problems will never be created,
and you'll have a dependable, rock-steady, Tach-Signal.
.
.
.

 

It doesn't necessarily need blocking.
It would need to be a high-voltage diode. But try it with and without and see what happens.
Without any resistors across the diodes, you might get too much noise, so try with 100k across the top diode and see what the waveform looks like.

 

To use the pulse shown, first a current limiting resistor and then shunt diodes to clamp the pulse at whatever voltage the RPM measuring device requires..
Look up the application notes for an LM2917 IC,which is a linear tachometer IC designed to drive an analog meter. That note covers input clamping quite well.

 

It doesn't necessarily need blocking.
It would need to be a high-voltage diode. But try it with and without and see what happens.
Without any resistors across the diodes, you might get too much noise, so try with 100k across the top diode and see what the waveform looks like.

Does this look right?



The diodes are 1N4148 hard to see in the photo but I didnt have suitable TH diodes. The source is a Li ion battery sitting at 4 volt. There are 4 100K resistors from the signal to the diodes and another across the top diode. What should I expect to see on the output?

 

Does this look right?

View attachment 300402

The diodes are 1N4148 hard to see in the photo but I didnt have suitable TH diodes. The source is a Li ion battery sitting at 4 volt. There are 4 100K resistors from the signal to the diodes and another across the top diode.

Yes. Now see what the waveform looks like.

 

Yes. Now see what the waveform looks like.

it seems to have done its job,

 

View attachment 300409

it seems to have done its job,

Sorry but this is with ground connected duh...

 

Looks like a nice clean waveform.

 

No need for the 4 volts bias,just be sure the current limiting resistor valueis high enough.

 

Looks like a nice clean waveform.

Looking at this I am more comfortable with the dangerous voltages aspect but there is still the issue of two small positive pulses with one larger negative one. I still opine that using an opto with inverted input to catch the negative spike is the way to go. The two concepts could be combined. At these new lower voltages that your circuit provides, the diode in an opto could do the rectification and wouldn't need the additional series rectifier diode I described earlier.

 

I think that the two positive pulses are above the logic threshold, especially if it's a 555 with 33% of Vdd.
A bit more bias would reduce the positive pulses (i.e. reduce the resistor across the top diode)

 

This stuff is voodoo, first of all let me say thank you to all involved. Now the waveform as I understand it, I can use because the MCU is 5v tolerant (vcc+0.5v) and It goes below the threshold for a low signal.

It appears that there is NO negative voltage - voodoo - I'm sorry but can somebody explain like I'm 5 years old please what happens to the excess positive voltage, what happens to the negative voltage?

This maybe apparent if I understood the previous - what happens if the power rail is off? Will the signal try to power (or destroy) anything on the power rail?

 

Using a suitable tachometer circuit there will not be any requirement for an opto isolator, or very many parts at all. Choosing a hard to adjust tach circuit will assure lots of diagnostic experience in trouble-shooting. That would be the one and two transistor circuits that I have seen. Use the LM2917 and follow the application circuit.

 

This stuff is voodoo, first of all let me say thank you to all involved. Now the waveform as I understand it, I can use because the MCU is 5v tolerant (vcc+0.5v) and It goes below the threshold for a low signal.

Nearly all ICs will withstand up to 0.6V above and below the supply rails because that is when internal diodes start to conduct. That is why a Schottky is used, so that it conducts before the diodes in the IC.

It appears that there is NO negative voltage - voodoo - I'm sorry but can somebody explain like I'm 5 years old please what happens to the excess positive voltage, what happens to the negative voltage?

The combination of resistors attenuated the signal by about 5:1. The resistor across the top diode then shifted everything upwards. The negative voltage go shifted upwards and became zero. The positive voltage disappeared because the top diode prevented it from exceeding the power supply

This maybe apparent if I understood the previous - what happens if the power rail is off? Will the signal try to power (or destroy) anything on the power rail?

It will try to power the device. It has 500k in series so not much current will flow.
put a 5.1V zener across the bottom diode (and keep the schottky in parallel).

 

The answer about the purpose was "to read the RPM", back in post #39. Then later there is a mention of "an MCU".
Why use a anything other than an actual tachometer IC that is actually intended for that purpose???
The LM2917 is still in production after about 25 years because the application still exists and nothing better has arrived. The original application note is probably still available as well. No need to do the "Nuts and Volts" thing and use a processor board to tell if the switch is off or on.
AND, by the way, the voltage output of the tach circuit made for instrumentation applications, was linear within the resolution of my HP lab equipment when I first built the tach back in 1975. At least up to 4000 RPM on a V8 engine.

 

The answer about the purpose was "to read the RPM", back in post #39. Then later there is a mention of "an MCU".
Why use a anything other than an actual tachometer IC that is actually intended for that purpose???
The LM2917 is still in production after about 25 years because the application still exists and nothing better has arrived. The original application note is probably still available as well. No need to do the "Nuts and Volts" thing and use a processor board to tell if the switch is off or on.
AND, by the way, the voltage output of the tach circuit made for instrumentation applications, was linear within the resolution of my HP lab equipment when I first built the tach back in 1975. At least up to 4000 RPM on a V8 engine.

Hi.

I'm sorry if I was not clear about the MCU, however this circuit is part of a wider project that requires serial communication, ADC, and various inputs and outputs for mosfets, wakeup signals etc. I did not mention the wider project because I didnt think it necessary at the time, again, sorry