Erratic rpms Rotax 912/Garmin G3X

Here are some thoughts that might help you.

The amplitude of the output signal of the tacho sensor increases with RPM.

I don't know anything specifically about the Garmin tacho input circuit but I would imagine that it has to try to cater for a wide variety of tacho output signals. The fact that others are talking about a resistor-diode module I suspect that Garmin have decided that the Rotax signal is too large. The diodes are probably being used to clamp the amplitude of the signal and the resistors are being used to limit the current through the diodes.

If it was me I would put an oscilloscope onto the connections feeding the Garmin and see what the signal looks like at different RPM. Without that everything will be a guess unless someone else has come across this problem and they can offer up a fix.

If your problem is at higher RPM it could be that the resistor-diode module is not doing its job. It could be that the signal from the tacho pickup is distorted giving rise to additional pulses that we be counted by the Garmin giving a false reading.

The tacho signal is a low frequency so virtually any oscilloscope would suffice and allow you to visualise the signal. I've attached the waveform shown in the heavy maintenance manual. You can see that at 500rpm the signal amplitude is +16V/-20V. The older versions of the manual had an error where the showed the wrong waveform with the wrong load resistance. At 6000rpm the amplitude can be as high as 75V depending on the load resistance.

I didn't quite understand what you were saying about the RC tacho. Do you mean that you tried to use an optical tacho off the prop? Don't forget that you would have to factor in the gearbox ratio. It would be easier to buy a cheap digital tacho, the type that has the single wire that you wrap around one of the HT leads. Something like this: https://www.amazon.co.uk/backlight-Inductive-Tachometer-Maintenance-Motorcycle/dp/B08CGPMKYM/ref=dp_fod_2?pd_rd_i=B08CGPMKYM&psc=1

Let us know how you get on.

Thanks for taking the time to offer comments Kevin!  First, I don't think I've known someone with an o-scope since 1987!  Back when analog was still king.  So it's not like it's a problem at higher rpms, it's that after several minutes in cruise, the rpm indication displayed on the Garmin starts becoming inconsistent within say 1000 or more rpms - meaning I think I'm at 5000, but it flickers between 4800, then 6100, then 5400, then 6000, and so on. 

Sorry about any confusion about the RC tachometer.  What I tried to convey is I have a small optical teach sensor I used for RC planes, and I am aware of multiplying the gearbox ratio, and it works on the ground when I aim the plane into the sun, and have the door open and aim it at the prop.  (idle).  But at cruise rpms, in the enclosed cockpit, I select 3 blade setting, and aim it through the window, fly into the sun, or any direction really; and unfortunately it will not pick up anything.

That's a great idea to try a small remote tach as a second opinion.  I actually have a Tiny Tach I use on my Fokker DVII.  That's an easy thing to try - then see the reading in air, multiply the gearbox ratio, then I know what I really have.  Thanks so much!

Bert

OK, here's another thought. If it is only happening in the cruise, i.e. with maximal air flow, could it be that there is a break in the tacho sensor wiring and air is pushing on the cable causing to to be intermittent? This would generate noise which would be interpretted as a higher RPM by the Garmin.

You can easily check for this if you measure resistance across the 2 terminals of the tacho sensor where it plugs into the diode-resistor module and then manually flex the cable on the engine side of the firewall.

Is the sensor mounted securely and the gap set correctly?

At that point, if it were me I would connect up my Fluke hand-held scope and go for a flight in the hope that the mystery would be explained :-)

Thanks Kevin, it's just two wires coming out of the generator, after 2 feet into a molex plug, then from there two wires info the fuselage where it feeds into the circuit board then into the GEA24 engine information box.

Should be easy enough for me to check those wires.  Maybe even squirt some WD40 into the molex connector female side.

And to clarify, it's not really just at cruise - meaning once it starts being erratic, it stays that way regardless of being in cruise or doing touch and goes.  But this should be easy for me to look closely at the wires in the firewall forward area.  Thanks!

Bert

I had the same problem on a VDO tachometer. After many tests the correct solution was what is shown in the diagram.

It has now been working without any anomalies for two years.

I don't know if it works on a Garmin, but it's easy to try ...

Tell us the solution when you find it.

Jacques