Multiplying Signal to Tachometer

Thread Starter

andrewleduc88

Joined Jul 22, 2020
4
Hello, I have a tachometer for a motorcycle, it's a dual fire tach and the motorcycle ignition coil is single fire, however it is single cylinder. So the tach is reading half RPM, if it idles at 1200 it's reading 600 rpm.

The motorcycle is a 2007 Buell Blast, it has the same engine as a Harley Sportster, except only 1 cylinder instead of a v-twin.
There are instructions for adapting this tach for 2 cylinders, but it doesn't work for one. In the picture its supposed to have two 100k resistors along with the diodes and its supposed to fix the problem.

However it doesn't fix mine.
Is there anyone familiar with this?
How do I adapt a dual fire tach with a single fire single cylinder?


BIKE IGNITION.JPG
I've looked into a frequency multiplier, I need a simple circuit to fix this not sure what to use. I found this
frequency-doubler-oscillator-transistor-circuit.jpg

I'm wondering what exactly to use?
like which number resistor capacitor ect..
I am not sure what the input or output mhz is supposed to be but it obviously is supposed to be between 0 and 8000 rpm. using a 12v signal from the negative side of the coil.

anyone know any about this? I'd really appreciate it.
 

Hymie

Joined Mar 30, 2018
848
When you applied the supposed fix, did the tacho still read half speed – or did it not work at all? The reason I ask is that the diodes may be the wrong way around if the output pulses to coils 1-2 and 3-4 are positive in relation to ground/chassis.
 
Just out of curiosity I did a little looking around. Be careful with adding frequency changing circuits. From what I could find on the Dyna S (maybe what you have... I don't know much in this area) the wire between the coil and module could have upwards of a couple hundred volt spikes.

What happens is the wire is grounded to create the magnetic field in the coil and when it is time to spark the ground connection in the module is disconnected. At that time the magnetic field collapses and makes the spark plugs spark. When that happens the wire is still internally connected in the coil and will spike to some extent... just how much is a matter of internal construction, but more than you want to touch. It looks to be pretty much the same as an older automotive coil setup and I can say for sure it's not pleasant!

Actually how many wires in your coil's electrical connector? If two be careful... if three nevermind (but check it out first).
 
Last edited:

jpanhalt

Joined Jan 18, 2008
10,109
A $0.50 optical coupler/isolator will solve the voltage issue.

What I am most concerned about is the signal the tach needs. Does duty cycle matter or is it just edges/time? And so forth, including voltage.
 
This thread has some screen shots of a tach signal https://forum.allaboutcircuits.com/threads/vehicle-tach-signal-noise-and-spike-help.41918/ -- only thumbnails left though.

The opto-coupler can be made to work, but there are some negative periods to deal with also. I am thinking with the diodes in the diagram it will trigger when the ground connection is lost. Duty cycle probably doesn't matter as long as it's long enough for the tach to pick it up.

Keep in mind these are somewhat educated guesses. I know how the coil itself works (if it is in fact what I think it is), but not so much details on tachometers.
 

jpanhalt

Joined Jan 18, 2008
10,109
I agree on the duty cycle. My experience is mostly with 2-stroke, petrol model aircraft engines. Before Hall sensors for crankshaft position became standard, a coil pickup was used (e.g., 3W engines). Of course, the models did not have real time rpm download at the time.
 
That's kind of the gray area. I don't know when exactly it happened, but some the newer coil on plug units for autos now run off a square wave from the ECU versus the old ground / unground the coil as I explained. The switching circuit is built in. The coil could possibly work that way too and If that is the case it will obviously be much easier, but going to need more information.

I'm also thinking the reason the proposed fix didn't work is a single cylinder engine would only have one coil. Maybe had both wires on same coil, but different wires which would cause it to not work.
 

Thread Starter

andrewleduc88

Joined Jul 22, 2020
4
When you applied the supposed fix, did the tacho still read half speed – or did it not work at all? The reason I ask is that the diodes may be the wrong way around if the output pulses to coils 1-2 and 3-4 are positive in relation to ground/chassis.
ya I tried putting them backwards. it's supposed to work if I had another signal, they make adapters specifically for this and it connects two into one. why have two if you could just have one? so that's why they have 2 because it uses two signals that's why it's still half.
 

Thread Starter

andrewleduc88

Joined Jul 22, 2020
4
A $0.50 optical coupler/isolator will solve the voltage issue.

What I am most concerned about is the signal the tach needs. Does duty cycle matter or is it just edges/time? And so forth, including voltage.
I believe voltage stays the same. it reads rpm by pulses which has to do with duty cycle or frequency. there is something that will solve my problem but it is 50 bucks. it's a universal tachometer converter. you can convert the input and output to whatever you want. 3:1, 1:3, 2:1, 1:2. what I need is 1:2 ratio. from 1 pulse to 2 because the tachometer is 2:1.
 

Thread Starter

andrewleduc88

Joined Jul 22, 2020
4

Alec_t

Joined Sep 17, 2013
11,538
You may be lucky, but I doubt the circuit in that link will work with the variable (over a 15:1 frequency range) and relatively low (10Hz-150Hz if 1 pulse per rev) frequencies involved with ignition pulses. For each pulse input to that circuit there will be two output pulses, but they will be separated by a fixed delay (of less than 1 micro-second for the component values shown, though that could be increased by increasing component values).
I suspect the commercial circuits use a micro-processor to do the frequency conversion. An ATtiny or 8-pin PIC would be up to the job.
 

jpanhalt

Joined Jan 18, 2008
10,109
I agree. The duty cycle and spacing may affect the tachometer and mentioned that concern. The TS has assured us that is not the case (post #10, "... it reads rpm by pulses..."). Perhaps, the tach just looks at edges, but I still have doubts. Many operate as analog frequency-to-voltage converters (see last paragraph).

As for rpm range, we do not know what the TS wants. I suspect most motorcycles idle a bit faster than cars -- maybe 1000 to 1500 rpm -- and rev higher to 8000 to 11000 rpm. The ratio may be closer to 10:1, but that is still quite large.

In any event, if one uses 15000 rpm as a limit, that is only 250 Hz or 4 ms between pulses. The doubling could be done easily with an MCU at 16 to 32 MHz virtually revolution by revolution. Duty cycle would be close to 50% (or whatever one wanted it to be). Averaging a couple of revolutions might help, but also may not be necessary.

A solution that has been discussed on AAC before is to use a frequency-to-voltage converter followed by a differently scaled voltage-to-frequency converter (e.g., the well know KA331/LM2907/LM2917 and related chips). I suspect that is what the "universal converters" referenced by the TS may use. Their disadvantages are analog and settling time.
 
I believe voltage stays the same. it reads rpm by pulses which has to do with duty cycle or frequency. there is something that will solve my problem but it is 50 bucks. it's a universal tachometer converter. you can convert the input and output to whatever you want. 3:1, 1:3, 2:1, 1:2. what I need is 1:2 ratio. from 1 pulse to 2 because the tachometer is 2:1.
It may well be worth the money.

How about a picture of your coil? Can you measure the resistance between pins and post details? I did a little more checking and it seems from what I can find so far you will have a voltage spike to deal with.
 

Alec_t

Joined Sep 17, 2013
11,538
Their disadvantages are analog and settling time.
I see the settling time being a bug-bear for analogue. At 600 rpm the pulse interval is 100mS. Any integration/smoothing would likely need a 1 sec time constant. That's a significant lag considering the way bike engines get accelerated.
 
Last edited:

shortbus

Joined Sep 30, 2009
7,854
What it says in post 1 -
Hello, I have a tachometer for a motorcycle,
There are instructions for adapting this tach for 2 cylinders, but it doesn't work for one.
Not all tachs are useful on a bike. He just said he had a tach for a motorcycle, not that it was made for one or even the model/brand of tach. Then there are ones that will work with points only and not electronic ignition. So we're left to use our crystal ball. And when $50 is too much to spend on a fix I'm wondering how much the tach cost.
 
Top