nc_growler

Recently purchased 91 GT has a sometimes twitchy tach.

PO indicated that it started after he turned off the rev-limiter for a run, then turned it back on... (that made me a little nervous).

It tends to settle down, but there are times when it really jumps around during acceleration.

So... is it time for a new tach? Or is there a way to recalibrate it? Anyone BTDT?

Oh... And I wanted to say a huge "Thank You" to all of you. I spent the last 6 months learning a huge amount about the 928 here on RL. It's great knowing I've got such a supportive extended family.

And for the small group of you that I got to meet in person at SITM, I can't wait until next year when I can actually stay for the entire event!

Maleficio

I fixed my intermittent tach by going into the pod, pulling the cluster, and running in all the screws that had been backed out by the last guy that went in there (sabotage.......by European Autowerks).

nc_growler

Maleficio - Can you describe "intermittent"? Mine works great most of the time, but sometimes it gets "bouncy"...

Maleficio

Works, then hit a big bump, doesn't work. Hit another bump, works.

It wasn't twitchy, but my oil pressure gauge is twitchy, but has become more stable over the last two weeks after I switched to Royal Purple oil.

blown 87

Sabotage huh

nc_growler

So, mine works very consistently... But sometimes under heavy acceleration, it jumps around a lot... Not a very smooth looking rip through 2nd and 3rd to redline...

Maleficio

That's my theory. PO had them go in and replace all the light bulbs. If I'm not mistaken, the flex-board has to be removed to do this. There is absolutely no reason for eight or nine terminal screws to be backed out like that. All the other screws were fully turned in tight, but those critical terminal screws were barely holding on. It was too suspicious to me, especially knowing how much the PO paid those guys for very basic maintenance. They ripped him off on a regular basis.

I may be wrong. Perhaps it was sheer incompetence, or maybe the tech was exhausted? While these things may be possible, the fact that the only screws backed out were the terminal screws for the gauges makes me think it was deliberate.

Tails

Interesting problem.

The engine speed sender is located at the rear of the engine on top of the bellhousing. The engine speed sender/sensor sends a positive and negative signal to the EZK control unit via connector pin 23 and 6 respecitively and an output signal from connector pin 13, titled Speed Pulse to LH connector pin 1 (input titled "Speed from EKZ 13).

I cannot locate the output pin with any description relating to engine speed from the LH control unit that sends a signal to the digital dash and I cannot locate the connector pin on the digital dash on plugs 1,2,3 or 4 with reference to engine speed, however, in the booklet "Diagnostics of Instrument Cluster" it is noted that for Instrument Cluster K25 soft ware the diagnostic readout relating to" K3.1 Input Variables" give an Eng. Speed -K25- xxxxRPM readout in the three windows of the digital dash.

This looks like the output from the LH connector pin 12 (takes input and gives output signals) and is connected to pin 6 of plug 4 of the digital dash. This connector pin is designated as "Speed", hence the confusion as to whether it is engine speed or car speed. From the above I believe it is the engine speed.

This signal would be fed into the CPC within the digital dash giving an output signal to the engine speed analogue rpm gauge.

If all of the above is correct then the hypothis of correcting the "jumping around a bit" under hard acceleration I would suggest the following procedure/s:

1. Remove and clean the engine speed sensor/sender.
2. Check and clean the notches of the speed wheel.
3. Check the 3 pin connector for good connections. Beware the female connector of these 3 pin connectors usually fall to pieces when disturbed. I believe Roger from 928sRus sells spare connectors or they can be sourced direct from Bosch. There are numerous post on these on Rennlist. Reinstall and check operation. If OK then problem solved. If not OK proceed to 4. ( if you have PET then the Porsche part numbers for these connectors are listed in the 944 PRT, IIRC)
4. If you have an oscilliscope then you can check the square wave form output from the sender as per Test point 15: Engine speed signal from EZK in section D24 page 22 of the WSM. This section is titled LH-EZK - Diagmpsos / Troubleshooting for S4 and GTs. If this is OK then good, however if the rpm gauges still jumps around under hard acceleration proceed to point 5.
5. Remove instrument pod and remove instrument cluster and unplug plug 4 and clean pin 6 making sure that the female receptical is tight fit over pin 6 on reinstallation. WYAIT I would clean all the pins on all the plugs. A suggested method is by cleaning the pins with an ink erasor (rubber) ever so lightly. There again numerous post on how to clean these pins on Rennlist. Don't forget the "search function" is you best friend.
6. On reinstallation all should now work correctly, if not then you may have some "dry joints" in the digital dash's CPU. If this is so then you are getting into electronic circuitary and you may need an expert to check it out. Unfortunately no circuit diagrams that I know of exist, however, John of JDS Porsche or Theo maybe able to help.

Alan Moore may like to chime in here as well.

Hope this helps and keep us informed.

Tails 1990 928S4 Auto

jcorenman

It was great to meet you in NC, nice car!!

The rev-limiter is part of the EZK code, accessible via Sharktuner but that's it. So that story seems unlikely, unless the PO had a sharktuner...

The tach signal is also generated by the EZK as Tails indicates (along with the spark signals), but if it were flaky at the source (e.g. intermittent crank-position sensor) then this would effect engine running also, and the engine would be missing. I assume it is running OK?

Which points to the tach (or the connections to it) as the likely culprit. Getting the instrument cluster out is a bit tricky but less than an hour as I recall. The knee-bolster has to come out (for airbag cars) but not the steering wheel. I used Schocki's write-up here (it only took him 15 min's, I am slow).

Once you get the cluster out you can remove the backside PCB with a few screws and get to the back of the instruments themselves. Check for loose or corroded connections, but beyond that I don't know what else can be done in terms of diagnostics-- it would not be easy to power it up on the bench.

Cheers, Jim

John Speake

A processed RPM signal is fed from the EZK to pin 1 of the LH ECU. This is the signal that tells the LH to energise the FP relay.

The rev counter is fed from that same signal.

blown 87

I will buy forgot or got in a hurry, but no mechanic is going to leave stuff loose that he is not going to get paid for to do a second time.

How do you figure they ripped him off?

Some folks may think I rip them off, but they get what they pay for, many times more than they pay for.
When you are paying some one 80-120 bucks a hour it adds up quick.

I had a lady get pissed at me yesterday because I need to get paid for my work, she had one shop replace a alternator four times in the last six months and it broke down in front of my shop, blocking access to my shop actually.

She was very mad that I would not diagnose her car for free, because the other shop did not charge her for it. (she was just to stupid to understand they did, but hid it in the bill)

I told her to take it back to them, maybe the fith one would work better than the other four.

BTW, I know what was wrong with the car, just had to prove it with a volt meter, BTDT

Or is this just mechanic bashing?

nc_growler

Tails - thanks for the entire read-out and collection of the manuals... Very helpful indeed. If I end up having to do detailed troubleshooting, that's the reference!

Jim - I like your approach - Reasoning? The PO had pulled the pod and recovered it with a replacement from Budd. That recover job appears to be coming loose and the cover is bagging a bit. I was planning on pulling it to check that work, and that would be a good opportunity to check out the connections on the tach while I am in there.

I'll take pictures and post my progress next weekend when I've got two days set aside.

Tails

John,
I note that the FP relay is activated from pin 20 on the LH connector plug, so can you confirm that this signal from the EZK is processed internally within the LH and passed on to the digital dash via LH connector pin 12 to plug 4-pin 6 as described in my post #8 to activate the RPM gauge?

Thanks,
Tails 1990 928S4 Auto

John Speake

Yes and no.....

Yes, FP relay is energised by LH pin 20, when it receives processed RPM signals from EZK pin 13 on pin 1 LH. This is the tach signal.

Now trace that signal into the CE panel on W11, see it go to the kickdown relay and then out on square
C88 to "L27" on Sheet 4. Here it goes to the diagnostic connector and then onwards to digi-dash plug 4pin 7 (i.e Not pin 6, which I think is a road speed signal)

LH Pin 12 goes to plug 2 pin 4 and is K line diagnostic.

nc_growler

John - Hopefully it doesn't come to that level of tracing to figure this one out...