Tell me how this can be!!!

[warren3200gt]

someone asked me to take a look at their tacho issue. Bike is a 650 with points ignition. Tach is a straight 3 wire jobbie. Live, earth and signal.
As expected due to wasted spark tach reads 50% of actual revs. No problem I think, take a signal from both coils with diodes inline to stop feedback. Double the pulses equals double the reading I think. But no, it actually halved the reading further.
How the hell does that happen?

[kev edwards]

You've looked, foobared, now give it back, , dark art is those electronics.

[warren3200gt]

Kinda heading that direction Kev, alternatively, swop the pos and neg. It'll either work and double the reading or go bang. Etherway it would be a win.

[needaz1100r]

someone asked me to take a look at their tacho issue. Bike is a 650 with points ignition. Tach is a straight 3 wire jobbie. Live, earth and signal.
As expected due to wasted spark tach reads 50% of actual revs. No problem I think, take a signal from both coils with diodes inline to stop feedback. Double the pulses equals double the reading I think. But no, it actually halved the reading further.
How the hell does that happen?

Do the revs on the tacho increase as bike revs increase?

I've had to use one of these before.

https://www.ebay.com/itm/141931549970

[warren3200gt]

Yes it does. Just confused about with one coil wired it halves the actual revs, which is to be expected, but with 2 coils connected i.e., doubling the pulses, it halves it even further. Made no sense.
However, after further investigation and adding a 1m ohm resistor inline not only did it smooth the pulse reading it corrected the double halving.
Kinda goes against everything I know about lectrickery which I admit isn't much!