Hi,

Although I'm quite old I am new here. I found this forum after looking for books and electronic circuits on Amazon etc, as I couldn't find anything that I thought would help me. I'm not totally ignorant of circuit principals although my knowledge is severely deficient on how to assemble practical circuits out of semiconductor components. (I date from the valve era!) Part of my difficulty is that there is nowadays such a diversity of special-purpose building blocks.

My problem is that I would like to build a simple device to detect when a magneto is opening and closing its points. The device would be used to set the ignition timing on a classic motorcycle engine, when the crankshaft is turned slowly by hand to detect precisely when the points open. When the points are open the circuit resistance is about 4 ohms and when they are shut the resistance drops to about 0.3 ohms. As a practical point, it is quite possible that even when being rotated quite slowly, that some sort of voltage spike would be induced when the points open, after all that is the point of a magneto.

I had thought to use a circuit based on a Wheatstone bridge, where the corresponding arm on the opposite side of the bridge would have a resistance of say 2 ohms, and the two remaining arms would have equal resistances. One could conceive of a circuit where a centre zero milliameter linked the two bridges, a small 1.5 V battery drove the circuit. The equal resistances could have values of say 10 ohms, and I think it would probably work, with the needle flipping from one side to the other as the points open or close.

Having thought that, I then thought that it might be possible to put an LED in place of the milliameter, but I don't think they start to emit light at a particularly low voltage. So I suppose what I'm looking for is a circuit that can turn current or voltage on and off as the bridge voltage goes through zero. The circuit would also need a reliable means of disposing of overvoltage spikes (preferably in either direction). If the current or voltage being turned off was sufficient to drive one of those little 1.5 V sounders, that would be rather good, as when one is fiddling with magneto's one has to keep an eye on the magneto itself and audio information on the points would be really helpful.

I would very much appreciate knowing whether I'm thinking on the right lines in the first place and also knowing what sort of devices are the right ones to use. Thanks for any help.

 

Here is what I have come up with. I have no idea if it will work. It will need to be tested. Someone else may want to add some HV protection on the input. Also a 555 beeper can be added.

 

Is this something that will remain attached and functioning during normal use of the engine?

If you trust your points and coil, you can wrap about 7 turns of wire around the coil output insulation, through a 100Ω-ish resistor, to the anode of an LED, with the cathode of the LED going to ground. The LED will blink every time the coil sends the HV to the spark plug, which is very briefly after the points open (instant to human senses).

This would let you know if the engine is missing due to lack of spark, or other reasons, and give you some blinky RPM indicator too.

Or are you attempting something else?

 

Here is what I have come up with. I have no idea if it will work.

I think it would, but why not just let the points directly control the LED, eliminating R1, R2, D1, Q1?

I'm wondering why you chose R1 at 33Ω. That low value would cause a fair amount of current draw from the poor little battery.

 

I think it would, but why not just let the points directly control the LED, eliminating R1, R2, D1, Q1?

I'm wondering why you chose R1 at 33Ω. That low value would cause a fair amount of current draw from the poor little battery.

Because the OP says that there is about 4 ohms across the points (which I assume is coming from the magneto).

Yes, I tried to keep the current as low as I could. We're pulling about 270mA.

 

My problem is that I would like to build a simple device to detect when a magneto is opening and closing its points...As a practical point, it is quite possible that even when being rotated quite slowly, that some sort of voltage spike would be induced when the points open...

Others may disagree, but if this is only for hand cranking, I wouldn't worry about any pulse from the coil. It'll be really weak at low speed. If you wire up something fancy that you absolutely don't want to zap, then some protection might be warranted. But for something cheap, I just wouldn't worry about it.

 

Here is a revised circuit with input protection. This circuit works under simulation.

 

Because the OP says that there is about 4 ohms across the points (which I assume is coming from the magneto).

Yup, I was just about to amend my post when your's appeared. I was thinking he was simply detecting open or closed, not the 4Ω versus 0.3Ω.

Wouldn't the voltage across the points swing hugely, ~12v when open and ~0 when closed? Sorry, it's been a long time since I handled points.

 

Wouldn't the voltage across the points swing hugely, ~12v when open and ~0 when closed? Sorry, it's been a long time since I handled points.

No. He is cranking the thing slowly by hand. There is no 12V power applied.

 

Here is a revised circuit with input protection. This circuit works under simulation.

Makes sense. I believe almost any NPN transistor would work, if the OP has a parts box or a piece of junk to harvest a transistor from.

 

Makes sense. I believe almost any NPN transistor would work, if the OP has a parts box or a piece of junk to harvest a transistor from.

Yes, I just picked the first NPN transistor that came to my head.

 

To everyone who has contributed I offer my thanks. I'll build the circuit contributed bt Mr Chips and let you know how I get on. Thanks again!

 

You are welcome. Don't forget to post a photo.

 

Well, MrChips' solution might work, but that 33 Ohm resistor will be dissipating about 2.45 Watts of power; you'll need a resistor rated for at least 4 Watts, and even then you'll probably burn your fingers if you touch it - and the circuit will eat batteries like you're being paid to get rid of them.

This circuit is a tad more complex, but it is considerably more kind to your battery:

S1 is a SPST toggle switch; shown closed as I was running a simulation.

If you forget and leave this circuit connected and turned on, your battery should last about 10 days if the LED is off, and about 4 days if the LED is on. If you do that with MrChips' circuit, your battery might last 10 minutes if you're lucky.

The LM339 could be used if you can't find anything else (Radio Shack sells them) - but you'd be better off with an LM311. If you do use an LM311, you'll need to connect the output's emitter to ground. If you use an LM339, then you should ground all of the unused channels' inputs. An LM393 or LM2903 (both dual comparators) could be used as well.

Use a super-bright LED, color of your choice. I suggest using some sandpaper or a nail sanding board to roughen up the lens of the LED so that it will diffuse the light; most LEDs being made nowadays focus the beam in a very narrow pattern.

 

Thanks wookie. Much better!

 

Would it make any sense to power the simpler circuit from the engine's 12V battery, assuming one is nearby? I love comparators but I'm sure the OP would appreciate the simplicity.

I agree the 9V battery won't last long.

 

Dear SgtWookie, thanks very much for your circuit and your advice on the LED. I'll make this one. I've just looked up LM339, which I had no idea existed! Modern electronics is so rich in building blocks. Thanks to all who have contributed, I am indebted.

 

The LM393 is very similar to the LM339, but it has only two channels.
The LM2903 is very similar to the LM393, but it is rated for the automotive temperature range (hotter and colder than for the LM393).

The LM111/LM311 is somewhat different; as it is a single channel device, and has a separate ground for the output channel as well as a strobe input. The 111/311 can sink far more current than any of the other comparators listed.

Any of these comparators will work. You could also use an LM324 opamp if you had to, but you would be better off with a comparator.

 

The LM393 is very similar to the LM339, but it has only two channels.
The LM2903 is very similar to the LM393, but it is rated for the automotive temperature range (hotter and colder than for the LM393).

The LM111/LM311 is somewhat different; as it is a single channel device, and has a separate ground for the output channel as well as a strobe input. The 111/311 can sink far more current than any of the other comparators listed.

Any of these comparators will work. You could also use an LM324 opamp if you had to, but you would be better off with a comparator.

Hi I realise this posting is pretty old but has anyone made this device from Sgt Wookie and how did it perform please.