How do I cut pulse signals in half?

[Rick4004]

I have an electronic tachometer on an engine that reads the pulse from one of the ignition coils. RIght now, the coil is fired once per engine revolution. I have a new elecronic ignition I would like to install, but it fires the coil twice per revolution, so the tachometer will read 2x actual engine rpm.
The ignition manufacturer says I can "adjust the value of the resistor in the tach", but what does that really mean?
Bearing in mind I am not an Electronic Engineer, can someone explain this to me, or offer an alternate method of getting the tach to read every second pulse?

Thanks

Rick

 

[VEBill]

"...coil is fired once per engine revolution..."

Is it a 2-cylinder (4-stroke) engine !?!?

"...from <~>one<~> of the ignition coils."

Ah, a GM motor with a coil for each two cylinders ??


Many commercial aftermarket tachometers intended for cars will have a 4/6/8 (cylinder) switch inside. The switch simply calibrates the analog meter - trivial.

If you adjust the same resistor value, you could make the tachometer work with just about any ignition-equipped engine (for example 1, 2, 3, 5, 10, or 12-cylinders).

This approach is far easier than dividing the incoming pulse train by two (a flip-flop plus ~much~ signal conditioning).

 

[Rick4004]

It is a Honda 4 cylinder, 4 stroke, motorcycle engine. The standard ignition system is 2 sets of points, firing 2 coils, with 2 leads on each coil. So standard, there is one "Dead" spark on the exhaust stroke of each cylinder.

The tachometer is also a Honda tach, so there is no switch to adjust it.

Can you elaborate on "adusting the resistor value"?

Rick

 

[VEBill]

In a generic aftermarket tachometer, it would be the resistors associated with the 4/6/8 switch making them very easy to pinpoint on the circuit card. For your example, see if you can trace out the resistor that might be in series with the meter coil. You would need to roughly double the value (to halve the meter indication), but the exact value might depend on the coil resistance as well, so be prepared to select-on-test if you can't predict it from a circuit analysis. (There might be a calibration potentiometer inside, worth looking for... hopefully with enough range.)

Another approach to consider would be to connect the tach to the original points (presumably still with the original pulsing) before - rather than after - the new ignition module that provides the double pulsing to the coils. This assumes that the new module 'loads' the points with voltages similar to the original circuit (perhaps not likely). It is most likely that the new circuit uses the points as a low voltage switch (~5 volts, not 12 volts). Then again, it is quite possible that the tach might accept the lower voltage signal as a valid trigger.

It is also possible that the tach would ignore the double pulse if the timing was sufficiently fast that the second pulse could be interpreted as point bounce. Then again, this approach might only work over a limited rpm range.

If I were in your shoes, I'd work backwards through the options listed above.
1) Try it to confirm it doesn't work properly as is.
2) Rewire it to the points, instead of the coil.
3) Modify the tachometer resistor value.

For option three, see if anyone on the

http://WWW has

information about the correct value to make it easier. You'll need to Google the subject using model numbers (motorcycle, ignition module) and the words 'tach' or 'tachometer', and 'resistor'. You might have to bang on Google a few times with various search terms to see if anyone has crossed this bridge before or not.

 

[Comcokid]

Standard tachmeters made by US manufacturers - Stewart Warner, Teleflex, Datcon, Faria, Medallion, etc. tend to use the Cherry Semiconductor CS-289 and later variations on this device made by Cherry or Motorola. (Earlier versions were the CS-189 and National LM1819 - both long obsolete). This chip is designed to drive a air-core type meter movement (sin-cos type). The selector switch selected a resistor which set the tach for that range. Sometimes there is a trimmer for exact calibration.

However, Japanese manufacturers many times used di-arsonal movements and completely different circuits. Sometimes, Japanese products sold in the US use tachs custom-branded from the US manufacturers, but usually only in smaller markets where volume is low.

The input to a tach contains a first stage that is a filter and/or transient limiter (if pickup is directly off a coil). Usually just a R/C followed by a zener. This then drives a transistor or schmitt-input device to square the input and avoid extra pulses. The switch theshold is usually 4 to 10 Volts.

If you need exactly a divide-by-two, you could attempt to do a little circuit - RC filter into a transistor or comparator - and then into a flip-flop configured to divide-by-2 - and then into a transistor (level-shift) into your tach. However, this will take a little playing around, some basic electronic knowledge, and possibly a scope to do it.

 

[VEBill]

The several that I've seen (the inards) had nothing but super-cheap discrete components - no chips. These were the $20 'Canadian Tire' (automotive Wal*Mart) type. The Honda is likely similar.

Typically some basic input filtering (C + R + diodes), a one-shot (not a chip, perhaps a JFET?) triggered by the ignition pulse, and perhaps a filter on the meter to still the quivering needle at low rpm.

YMMV.