Tachometer Bounce 911SC/964 Conversion
 

Hi All,

I've read several threads on this and was wondering if someone might be able to give me a steer with this problem which seems to present to many users.

The car is a 1980 911 SC running a 964 engine and gearbox with independent throttle bodies and Bosch coil on spark with Motec M84 doing the control duties.

Tachometer "wandering" was noted from time to time over the last few months but this has now graduated to full on "bounce".

I do note that if I turn on my headlamps or my heated screen that the bouncing immediately stops.

I have logged the battery voltage through the Motec and also using test leads to battery etc and the alternator and voltage regulator seem to be fine with no excessive voltage noted. 10.5v immediately after crank, recovering to 13.4v within seconds and hitting a max of 13.75v over a 30 minute drive.

There is no evidence of the erratic RPM count on the Motec logs and the car is driving fine.

I have done my best to check the ground wires. They loop from instrument to instrument. All show good continuity to chassis when measured individually with engine/power off.

One other possible clue (although this has been a consistent issue for months) is that the oil level gauge is quite inclined to flicker rapidly and when not flickering it spends most of it's time wedged up at the max level. It does occasionally indicate a realistic reading but mostly not.

Thanks in anticipation for any feedback from fellow members. This forum seems to offer a wealth of technical advice and I feel sure it will be my point of reference ongoing even if it's "across the pond" from me.

Best regards,

Ciaran, Dublin, Ireland.

When I had my tach refurbished Helmut and he told me that it was simply a wear item. He told me that he added some silicone. I have yet to install and confirm if his repair was successful. On my 914 I had someone swap out the guts with a modern Bosch and it definitely fixed the bounce.

I install a whole new driver board in the tach.
The VDO will still bounce a small amount if you compare it to a autometer, but it's totally acceptable.

i think there are two issues being discussed here .... I dont think the OP is talking about the damping on the needle if i understand correctly ( that is what Harpo is talking about ). Harpos issue is wild bouncing after changing gears repaired by the silicone adding the damping efect to the neddle so it doesnt swing.

Sounds like OP is erratic in nature .. OP do you have a video of tis doing it in the car ? that may help but sounds like a grounding issue somewhere in the instrument panel.

Just a guess.
Common 12Vdc instrument circuits have a Zenier Diode for a steady reference voltage.
Suggest to check tach internal components next.

Thanks #IceMan, you are interpreting my issue correctly, this is not a damping issue. It's an electrical issue whereby at a steady state of say 2500 the needle could swing up to 5k and down to 0 for no apparent reason. The only thing I've noticed is that if I put a heavy electrical draw on the car normal service resumes and the needle sweeps nicely.

Regarding the Diode, #3rd_Gear_Ted, are you referring to a diode inside the tach or one somewhere in the instrument cluster vicinity?

After reading your initial post I believe you're seeing interference on the tach signal. The original tachometer signal from a CDI box or DME is a low-side switch that momentarily grounds the signal line vs. a pull-up resistor inside the tachometer. This happens for each ignition event. This setup can sink a reasonable amount of current and thus is pretty robust against electromagnetic interference. The resulting waveform is a square wave swinging between ~0V and ~12V with three pulses per engine rotation. At 2500 RPM the frequency of that square wave is 125 Hz.

I don't know how the Motec unit provides the tachometer signal. I can see two reasons for your issue:

(1) The wire from the Motec unit to the tachometer is not shielded enough and picks up electromagnetic interference (ignition noise, alternator spikes). This gets worse as grounding of the unit vs. grounding of the tachometer is not optimal.

(2) The Motec unit doesn't sink enough current on the tachometer output and that leads to distortion of the square wave signal where it doesn't get close enough to the 0V level.

You need to get an oscilloscope trace of RPM the signal with everything hooked up to see what's really going on.

Hijack alert - so, how do you add silicone to reestablish the damping for the needle?

Regarding the OP’s question - the issue does seem to be in the electrics.

Thanks #ischmitz

Your suggestions sound plausible. The system was working fine for quite some time (albeit with occasional weird "wandering" of the needle noticed and then, over a period of a few days, went to full-on bouncing.

The intriguing thing is that the bouncing stops as soon as I put a good heavy load on the electrics. I'm guessing this is somehow having an impact on the grounds and sorting out the signal. It does suggest however that the Motec itself continues to do enough and that the problem is downstream from there.

I'll trace back the tacho wire from the clock to the ECU and see if anything obvious occurs. Appreciate the suggestion!

Bit of an update ...

Thinking that perhaps the wild flickering of the oil level gauge had something to do with the tacho problem, I decided I'd investigate that. I disconnected the two wires at the sensor and needle went sharply to one end of the scale. Shorting the two wires sent the needle sharply to the other extreme. No flickering evident at all.

I took out the level sender and opened it up (thank you to prior contributors on this forum) and found that there was a break between the very first and second windings of the resistance wire within the sender. The result was that the sender was capable of operating at about 1% of it's possible measurement range and then it just went off scale. I dropped a wee bit of solder between the wires which were broken (out of reach of the sweeping part) and reassembled and it now works perfectly with no flicker at all. It probably reads from say 100% down to maybe 10% and then jumps straight to 0% below the 10% level but that's fine. I'll just replace it whenever it fails again.

Long story short, fixing this didn't help with the tacho at all.

Next up was to add a new quality earth wire to the tacho. This made no difference at all.

I had a bit of a look at the routing of the tacho sender cable to see if there was any obvious interference caused by routing but it got a bit complicated and tight and I gave up having found nothing.

Sometimes the problem is bad, with wild swings up and down and it's extremely distracting. I've just learned to turn on the headlamps to settle it down. Other days it performs perfectly and sometimes there's just a slight bit of wandering around.

I think at this stage I'm going to just sit and wait for it to get worse and then maybe it'll be easier to figure out where the problem lies.

Thanks for various suggestions anyway. Happy to have my oil level working :)

How is your charging system and battery condition? Any issues with it not holding a charge or low output?

I don't know the "why" but when the battery is bad in my car or when the ground strap had too much resistance to allow the battery to charge my tacho would bounce up to 5k and hold there for a moment.

+ 1 for charging system, voltage regulator is on it's way out
Ian

I'm not in a position to measure charging current but in all other respects everything seems to check out...

I have logged the battery voltage through the Motec and also using test leads to battery etc and the alternator and voltage regulator seem to be fine with no excessive voltage noted. 10.5v immediately after crank, recovering to 13.4v within seconds and hitting a max of 13.75v over a 30 minute drive.

Voltages also stable under load. Slight drop when lights / screen heaters are on but always high of 13v and never over 14v. No issues with battery holding charge or difficulty with starting etc.