PCP & PCM pins on Honda ECU - help!!

[Thrasherg]

well the replacement ECU plug arrived today, so with great joy I added a connector for pin 17 on the ecu (hopeful tacho pin) connected my scope and tested it. The pin has infinite resistance to either ground or Vdd (suggesting the pin is not connected to anything inside the ECU), I then started the engine and with my scope looked at the pin and again, nothing, no signal of any kind!! So its back to creating a pulse doubler with a couple of monostable..

 

[geezerbiker]

I think you could do this with 3 (555-type) monostables in series, each set for 3ms pulse width, then take the first and third outputs into two transistors in a wire-OR. You'll get two 3ms pulses out, separated by 3ms, for each (7ms) pulse input. However, depending on how your tach works, it may not like the narrower pulses, or the unequal pulse periods below 10,000 rpm (first pulse period is 6ms, next period is up to 120 ms). If I was doing this, I'd take a small processor (UNO etc) and generate the double pulse at various speeds to ensure the tach responds as you hope.

 

[Thrasherg]

I think you could do this with 3 (555-type) monostables in series, each set for 3ms pulse width, then take the first and third outputs into two transistors in a wire-OR. You'll get two 3ms pulses out, separated by 3ms, for each (7ms) pulse input. However, depending on how your tach works, it may not like the narrower pulses, or the unequal pulse periods below 10,000 rpm (first pulse period is 6ms, next period is up to 120 ms). If I was doing this, I'd take a small processor (UNO etc) and generate the double pulse at various speeds to ensure the tach responds as you hope.

That is exactly what I want to try, but with 2 monos tables, both set to 2ms, the first monostyle is fired by the rising input signal, 2ms later that monostable fires the second resettable mono stable which is also set to 2 ms, this creates a 2ms wide pulse that appear 2ms after the main pulse has passed. I am taking the input signal and the output of the 2nd mono stable pulse into a nand gate, this basically mixes the 2 pulses back together.. the question is if these 2 different pulses will work with the tacho! I will find out soon enough !!

 

[geezerbiker]

The two 2ms pulses will only have a 555 propagation time between them, as little as 200ns. If the combo produces 2ms, 200ns, 2ms, the tach may not recognize it as two pulses. You could try an R-C between them to delay triggering the second mono, or add another mono. It's just a question of how that tach was designed, and what timing it requires.

 

[Thrasherg]

The two 2ms pulses will only have a 555 propagation time between them, as little as 200ns. If the combo produces 2ms, 200ns, 2ms, the tach may not recognize it as two pulses. You could try an R-C between them to delay triggering the second mono, or add another mono. It's just a question of how that tach was designed, and what timing it requires.

I will get a schematic up soon, i think you have misunderstood what I am doing, my circuit should create a 2ms wide pulse that will be created 2ms after the master pulse from the ECU has passed. the master signal sits at 12 volts and dips down to Vss when fired, then returns to Vdd when completed, I am using that rising signal to fire a 2ms mono stable (this is just to create a 2ms delay after the master signal has returned to Vdd), the falling output of this mono stable then fires the second mono stable, which also creates a 2ms wide pulse. This second mono stable pulse is what I am using, it will be a 2ms wide pulse that occurs 2ms after the master pulse has returned to Vdd. I will combine the master pulse and the output of the second mono stable (I have to use the Qbar not the Q output) by passing them through a 2 input nand gate and the output of the nand gate then goes to the tacho input. Will post a schematic with the expected timing diagram..

 

[geezerbiker]

I understand now that you are using the 7ms pulse and two 2ms monostables. You are triggering with the rising edge to TH, rather than the normal falling edge to TR. For "normal" triggering, the low pulse must be shorter than the timing period. I couldn't find if your method requires the high pulse to be shorter than the timing period, and your input will be high for between 5ms (at 10K), to 113ms (at 1,000 RPM). You might find if this is a requirement by searching, or testing a 555. If so, you might need an R-C (or something) as an edge detector.

 

[geezerbiker]

Hi Thrasherg, did you get your tach working?

 

[Thrasherg]

Sorry, I forgot to come back, yes my tach now works perfectly, my circuit to double the number of pulses worked as expected, the fact that the 2 pulses are not the same shape did not seem to change anything!! The rev counter works exactly as expected. Very happy with the result.

 

[geezerbiker]

That's great to hear. Does it work to up to red line?

 

[Thrasherg]

That's great to hear. Does it work to up to red line?

Yes.. although Honda doesn’t give a max rpm or red line number for the crf250f, so I checked it worked up to 10k rpm, which I feel is higher than Honda would want this engine to be reved! So that is my limit, I now have the rev counter showing the redline at 9k rpm, but I know it can (for very short periods) rev to 10k rpm, but my wife never revs it that high..