I know koso sells a filter for their tach for the f2. JNS had to use it one his bike. If all else fails maybe order one up and see? It's on their website, I have it bookmarked at home I'll post the link up tomorrow. If nothing else maybe it'll help ya figure out what they did.

 

Uh yeah, Doc. Duh. The pulse rotor has nine teeth (not ten) and then a gap equivalent to about 3 teeth. That would create the nine sinewaves and a three sine wave wide space as it spins next to the pulse sensor.

So apparently the ECU turns this into an inverted square wave. Unfortunately my quick and dirty attempt to invert the wave was not successful. I will need to revisit it with a little more sophisticated setup than my resistors tacked to a transistor and clip leads setup.

@gt - the Koso is definitely a superior product feature-wise. I just studied their documentation and this would be a non-issue with the RX2N as it has the ability to be set for the number of pulses per rev and also can be set for up to 12 cylinders. I didn't see a filter listed, but I suspect this is more an issue that I will have to work out by modifying the actual signal pulse anyway. If I had known what I would end up spending on this entire project I probably would have just ordered a Koso, even though 4X the price. But I have sort of a bad habit of chewing on a problem until it either softens up for me or it goes crunch. So I will stay with this one a while and see if I can gnaw it into submission.

 

Kept fiddling with the tach signal inverter circuit and got it to invert the negative square wave from the stock tach output to a positive pulse. Unfortunately my battery was running down as I was cranking the motor quite a bit to observe the pulse on my scope, and I think possibly the voltage ran so low that it couldn't trigger the tach. Took a bit of fiddling with pullup and bias resistor values on the inverter to get a decent waveform. When I have a good charge on the battery tomorrow I will see if the tach sensor can find the positive pulse. If it works out I will publish the circuit, if not, I will keep gnawing on the problem.

 

I have munted a koso rpm tac and to make it work i had to put it in 1 as in positiv impuls and on 2 cyl sins ther are 2 coils. 1 for every pear of cylinders. hockt it on the back of one of them and it all works. If you stil do not get it to work properly try the negativ seting. Hope this helps

 

Thanks Emmanuel. I have tried that connection without success. The stock tach signal comes from the ECU output to the left coil pack, which is what I think you are describing. Unfortunately this copy of the Koso does not have all of the adjustments for positive and negative connections that a real Koso has. And that is why I tried to invert the square wave to positive with a transistor circuit.

At this time the only signal I can get the SS182 to tach to respond to is the pulses from the pulse generator. I do not yet understand why it does not respond at all to the square wave from the coil input, even after inverting it to positive. It doesn't seem to be a problem with the amplitude, as it responds well to the pulse generator signal and that is as small or smaller in amplitude than the square wave. I will spend some more time exploring this over the weekend. First I need to install new boots to seal a vacuum leak and get the carbs adjusted so that the bike will run well enough to let me observe the waveform for more than a minute or two.

I will also post some pictures of the different waveforms that are seen at the pulse generator and the coil pack/old tach input.

 

OK, a little more experimentation tonight leads me to believe that I might be able to get this thing to work with a simple pull up resistor on the tach input. Or, I might blow up my ECU...

I'll let you know what happens.

 

Having the engine running well has helped me see the pulse waveforms much more clearly. The tach output signal is already pulled up as it comes from the ECU and it is an inverted square wave between 12V and ground, so that was not the issue. Based upon some odd readings I got the other night I'm now guessing that the SS182 tach input may be referenced to 12V rather than ground and it's looking for a positive pulse, that is, one that goes higher than 12V.

Edit - I think I figured out which ebay source I bought this thing from a year or more ago, and on his listing he points out

Note：
The RPM signal is Positive signal.

Which is pretty vague, but I guess I am more or less on the right track.

 

This is going nowhere. The seller's only suggestion is that I ask a motorcycle shop to sell me a signal converter. But this thing is only responding to the pulse generator signal, no square wave pulse I have tried makes the needle budge, positive or negative, so I don't see how a signal converter would help. So it's a possibly functional tach for a 1 or 2 cylinder bike, but a four cylinder wasted spark bike is not going to work as there are twice as many pulses as it can handle.

And so I am at the juncture that I must admit I relish. Time to open the sucker up and see what makes it tick.

 

OK, I'm going to look at this from a slightly different angle. Part of the problem seems to be the frequency of the coil pulses. The pulse generator is giving me about 200Hz pulses at idle. If I understand the description of the way the tach reads pulses, 100Hz pulses would correspond to about 1500rpm on the tach if set for 4 cylinders, 3000rpm for 2 cyl and 6000rpm for 1 cylinder. If I remember correctly from last night's test session I'm seeing about a 2700 rpm reading on the 4 cylinder setting. So that seems to jibe if my idle is around 1350 rpm.

The ignition pulse I'm seeing at the ECU tach output (which is also driving the 1&4 cylinder coil input) is a much lower frequency than 200Hz. At 1350 rpm (what the motor is really turning at idle) that wasted spark coil should be firing once per rev -

1350 rpm is 22.5 revolutions per second. That does indeed correlate to my scope readings of the tach output pulse at around 22 Hz.

Ah hah! So the pulse generator output frequency is twice what the tach wants to see in 4 cyl mode and the frequency from the tach output is 1/4 of what the tach wants to see in 4 cyl mode. A 22.5Hz pulse would read about 337 rpm - maybe too low to register? So maybe if I switch to 1 cyl mode I might be able to get the pulse into some form that the tach can read and it would be properly calibrated.

Looks like I am back to playing with inverting that wave form.

BTW I did open it up. Looks like the thing is run by a microprocessor and might require reflashing an EEPROM to change the tach settings beyond the range available by the setting buttons (i.e. there is no easy way to switch it to 8 cylinders, which would allow it to work from the pulse generator output).

 

Any update on the installation? I have the same unit and I've been waiting to see if you made any progress before I attempt to install it on my f3. I wish I read your thread before purchasing the unit, as of right now it seems like the most important thing to me may not work with our bike.