|
|
|
|
|
|
Registered
Join Date: Feb 2013
Location: Middletown, CT
Posts: 285
|
1992 968 Speedo Inoperative
(May 2024 note: coolant temp gauge resolved, see thread)
(June 2024 note: speedo resolved, see thread). New-to-me '92 968, just got it on the road today after some basic repairs and maintenance. I noticed that the speedometer is completely inop (speed and both odo) and the temperature gauge is pegged high all the time. Car seems to drive fine otherwise. I looked through the various 968 PDF workshop manuals I could find, none of them seem to offer anything on how to troubleshoot either of these. Can someone tell me where the coolant temp sensir is so I can check that it's connected and there's no wiring damage? Is the speedo cable- or sensor-driven? If the latter, where's the sensor? Also, no reverse lights, I've checked the bulbs. I presume the switch is on the transaxle somewhere? Any advice? Maybe some links to troubleshooting (or a manual that has this stuff in it)? I'm not real pleased with the WSM that's common out there...Thanks!
__________________
1992 968 Coupe 1974 914 2L street car 1974 914 2L race car Last edited by GregAmy; 06-29-2024 at 07:21 AM.. |
||
|
|
|
|
Registered
|
download PET catalog from Porsche: https://www.porsche.com/usa/accessoriesandservices/classic/genuineparts/originalpartscatalogue/
This parts catalog can be searched for parts names, or just be browsed. |
||
|
|
|
|
Registered
|
#18 is a back-up switch:
|
||
|
|
|
|
Registered
Join Date: Feb 2013
Location: Middletown, CT
Posts: 285
|
Thanks. I have the PET; useful for looking up parts to order, but not that helpful in troubleshooting to determine what parts I need to order...
__________________
1992 968 Coupe 1974 914 2L street car 1974 914 2L race car |
||
|
|
|
|
Registered
Join Date: Feb 2013
Location: Middletown, CT
Posts: 285
|
Thanks, that one's helpful. I can find that switch and connect those terminals to see if I get reverse lights (fingers crossed).
I think I've found the coolant temp sensor under the manifold. It's odd in that it has two regular crimp connectors on it; that's decisively not German. Removing any combination of those terminals, or connecting them together, does not change the temp gauge reading. Short of pulling out the speedo cluster, I'll presume it's sensor-driven and I'll start looking for errors there.
__________________
1992 968 Coupe 1974 914 2L street car 1974 914 2L race car |
||
|
|
|
|
Registered
Join Date: Oct 2002
Location: Boulder, CO
Posts: 1,205
|
Speedo is driven of the ABS module in a 968.
I do not recall if the 968 temp sensors are more-or-less in the same spots as the 944 but you can see the 944 ones here: https://www.clarks-garage.com/shop-manual/elect-19.htm 944Online shows one temp sensor for all late 944 and 968 which has the spade terminals as you noted.
__________________
1987 928S4 1992 968 cabrio 2009 957 Cayenne GTS |
||
|
|
|
|
|
Registered
|
GregAmy,
with 924-944's, there are two temp. sensors - one is for the computer (with two contacts), another (with a single contact - deeper under the intake manif.) is for the dashboard gauge. Both are just above upper balance shaft. for 968 - #29 and #30 on page 38 ?, illustration 101-10. |
||
|
|
|
|
Registered
Join Date: Oct 2002
Location: Boulder, CO
Posts: 1,205
|
If your ABS light isn't on it's probably time to check the connections to the cluster circuit board, see if the ears are broken off the wire harness connectors, if they are loose, dirty, etc.
__________________
1987 928S4 1992 968 cabrio 2009 957 Cayenne GTS |
||
|
|
|
|
Registered
Join Date: Feb 2013
Location: Middletown, CT
Posts: 285
|
I do have an ABS light on...I guess I need to get into that a bit deeper. I bought the Bergvil T-OBD but not able to get it to connect; now I'm told I should have bought the Durametric...sigh...
Per 944Online it looks like I have the right sensor in there (two contact?)...but no changes in the wires changes the temp display. A pegged gauge typically means a grounded sensor wire; that tells me it's either a gauge or wiring prob. I like the idea of pulling the cluster and checking the circuit board, make sure all is still connected. These two probs would be related to that. Clark's Garage is a doll of an info source. I presume most 968 tech is related to late 944 (such as removal of the instrument cluster?)
__________________
1992 968 Coupe 1974 914 2L street car 1974 914 2L race car Last edited by GregAmy; 03-20-2024 at 07:08 PM.. |
||
|
|
|
|
Registered
Join Date: Oct 2002
Location: Boulder, CO
Posts: 1,205
|
Yeah pulling the cluster, at least enough to check the connections, is the same as a late 944/S/S2.
It's sorta possible to get the cluster out with the airbag wheel in place but it can scratch the cover if you don't put a bag over it or something. Pulling the wheel can easily cause the airbag light to come on and that requires a Durametric or similar to reset. Might be possible to avoid this if you disco the battery for a while and then ground the + cable, throw a pickle over your left shoulder and whistle "Edelweiss". I think it's the LF ABS sensor that sends the speed sensor, you can ohm out the sensors at the ABS module or just look for obvious wire chafing, etc. -Joel.
__________________
1987 928S4 1992 968 cabrio 2009 957 Cayenne GTS |
||
|
|
|
|
Registered
|
I slip a piece of heat shrink over the wire , toss the plastic bit and solder the crimp connectors. Takes a moment longer, but then they don't cause issues, ever. whenever one sees those crimped connectors , it usually means someone has been in there.. maybe crimped them incorrectly..
you can likely unplug the wires from the reverse switch , put your meter on it, on beep and hear if it the switch is ok, or use a test lamp. I'd inspect for bad connections due to oily wiring or vibration and movement of trans. maybe the wires got tie wrapped a bit too tight and pulled on the wires , or similar. you can check with others, or the schematic, but I "think" the temp gauge needs power to move provided this is true, if battery is unhooked and temp gauge is up then it must be a fault with the gauge. you might check with someone near a working car to confirm, but doesn't the gauge normally go to zero with the key off? Last edited by Monkey Wrench; 03-21-2024 at 08:57 AM.. |
||
|
|
|
|
Registered
Join Date: Feb 2013
Location: Middletown, CT
Posts: 285
|
Update...I do NOT have an ABS light on, it's the airbag light (which is reasonable, given the PO installed a Momo). I also have a Check Engine light but I don't think that's relevant.
I pulled the cluster this afternoon, gave it a good inspection, cleaned off the terminals with a rubber erasor and some light Sctochbright, cleaned it off and gave all a very light coating of dielectric grease, reinstalled and...no change. Coolant temp gauge pegged, no speedometer reading. I did notice a nice thumping banging noise and rear end vibration during the test drive...thought I'd lost a tire...then it suddenly stopped...did it Friday too...but I guess that's another topic. And yes, the gauge should be at zero with the key off. In fact, if I go back out there in an hour or so, I'm betting the gauge will be at the bottom, where it was before I removed the cluster. So I have no clue what's going on with this. I may have to spend the money on a Durametric so I can scan for inputs/outputs. I bought a Bergvil T-OBD but I can't get it to connect, and I don't think that's the right tool anyway...sigh...
__________________
1992 968 Coupe 1974 914 2L street car 1974 914 2L race car |
||
|
|
|
|
|
Registered
|
im kind of guessing but I picture the sender as bieng a variable resistor and thoguht maybe you can put it in a pan of water , connected to your meter and see a resistance change with the temperature change..
and likewise the temp gauge might be just basically a voltmeter, the face is calibrated for temp but it's actually probably basically working like a voltmeter , or possibly like an ammeter.. or measuring resistance to ground perhaps. so you might feed it some various voltages 12 V downward and see how it reacts.. ? the problem could be that the pointer is rubbing on something and mechanically stuck. I have a car with a fuel gauge that reads empty until I give it a "thwack" then it reads right.. so I know the fault is within the gauge, not the sender.. same thing there the guage reads current or voltage so its essentially a voltmeter or ammeter.. the way its painted is of course represnting " engine temoerature" i bet if you could connect to the two wires that control the guage and run them to your meter you could than watch the meter and see if the temperature sender is reacting.. you might see the meter representing voltage or maybe resistance to ground, but if you payed attention to what the meter says and the temp of the engine you could calibrate that somewhat.. If you boil the water it should be moving up there or even pegged. If you have them removed from the car I'd be tempted to put he sender in a pan of water and connect the sender to the gauge.. Maybe one side of the gauge is hot 12+ and the other goes through the sender to ground ( -) and then the sender can increase or decrease the resistance and maybe the gauge reads like a voltmeter and senses the voltage drop , basically reading the varying voltage.. it'll be something similar to that, the schematic might help. I think this would be analogue stuff so no need to worry about fancy digital measuring devices.. the one that reports to the ECU or ( DME) is probably an analog signal, even though the board may receive that and transpose that into some "digitized reaction" like maybe opening the idle bypass valve or similar.. perhaps remembering an overheating fault.. I think the gauge itself will be analog... a bad reading like an overheat reporting might save to the eprom ( or similar) and remember the fault,, then report that to you under diagnostics, as a a code.. if the car's electronics are that digitized.. my 85 porsche or 88 volvo dont have fault codes you can read but my 1990 volvo has a fault code ability.. they are just blinks of an LED that can be traced to one of 30 or so faults.. so the technology was developing just before the time of your car , by 92 maybe Porsche used similar diagnostics.. OBD,, OBD 2 maybe. my 88 ford can remember some fault codes.. all 3 are Bosche electronics.. if the sensor reports to the MCU, the MCU might adjust idle when it's cold or maybe throw up a warning when its too hot. maybe turn on dashboard idiot light - check engine or similar? perhaps control the fans turning on and off ? in the volvo I think if you disconnect he battery it will forget the old fault codes, and there may be a way to clear them out so you can read fresh the object is often to solve them until it stops reporting any.. Last edited by Monkey Wrench; 03-26-2024 at 03:23 PM.. |
||
|
|
|
|
Registered
|
so can we figure out what the sender does ( like changes in resistance) then just replicate that using some fixed resistors or a variable resistor..
I'm thinking maybe you can save the work of messing with pulling the guage cluster if we can just replicate the expected input and prove that the gauge works,, I massuming this is just one wire and the guage senses the senders variable resisance to ground.. If thats the case and it's a one wire senseor then just try connecting some various resistors throught that wire , then to ground.. one little thing that might mess things up , the engine ground, Can you take a black jumper cable from the negative battery post, clip the other on the engine itself and retest the guage? if your engine ground is bad that should prove it, if its that, the temp gauge should start working correctly and then if so, you'd just have to clean up the ground-to-body strap. that strap is probably part of the return path and since it't very low amperage, it might be dirty enough for it to have that teeny bit of resistance, enough to mess it up.. |
||
|
|
|
|
Registered
|
sometimes my mind ponders things, I was thinking of the sensor mentioned for the temp gauge , maybe its screws into the head and you mentioned something about that crimps had been used indicating a previous owners efforts in the area,,
could it be possible someone used plumbers heat tape or a pipe sealant or never seize or maybe even just rust or corrosion and could that be insulating the sensor from the block? I'm thinking that sensor can be checked with a meter.. and the gauge should be able to operate from the wires that lead to it without removing the gauge. maybe if the sensor has a metal body you could check if that has a good connection to the block? Id want find out for sure if the problem is the sensor or if it is the gauge as removing either might be necessary but I doubt the fault is both.. Some electrical troubleshooting could prove the fault in either direction without taking stuff all apart needlessly. removing the cluster might not be all that hard but I'm always afraid of breaking any brittle plastic bits.. it might not be so hard to study the schematic a little and maybe feed the guage some varied voltages or varied resistance to ground ? ( or what it expects) to see it operate rather than pulling out the whole cluster ? Last edited by Monkey Wrench; 03-28-2024 at 09:56 AM.. |
||
|
|
|
|
Registered
|
sorry if I wrote too much and confused things..
what Im thinking after re reading i the previous post ( above ) from wwdwgs - he said the temp sensor for the gauge is a 1 wire sensor and you mentioned two crimped connectors so it made me think you are examining the wrong sensor. if it's a 1 wire sensor, it must be sensing the resistance of that 1 wire to ground because how else could it possibly work? I'm thinking maybe its pegged because that wire rubbed on something and shorted to ground.. if you can disconnect the same wire, maybe from the gauge will it return to 0? if you ohm out between that wire and ground is it close to zero ohms? if other things aren't working maybe its a problem with a ground or a hot to the cluster? bad fuse or corroded wire near the fuses perhaps? maybe a ground point was unhooked and not reconnected? seems a little strange the cluster had multiple failures, ok bulbs can burn but why would the temp gauge and speedo both fail together? I'm thinking if it's the car's wiring itself, the new cluster might prove the same symptom, less a couple hundred.. mine is early 85 I think later ones had a different dash, that's probably the division so careful if you get an "85" because I think there are 2 different clusters and that year and the changes were mid year, I think.. maybe just get a photo before spending.. |
||
|
|
|
|
Registered
|
Quote:
I check PET catalog where this type of senor has two prongs. Therefore, this sensor gets ground from one of the wires. Probably brown. Maybe, the temp. sensor wires are switched if the PO crimped same size terminals. One should be narrow.
|
||
|
|
|
|
Registered
Join Date: Jul 2004
Location: Upstate New York
Posts: 3,276
|
Quote:
The square dash cars do not have the idiot light, thus only one pin needed on the sender.
__________________
Good luck, George Beuselinck |
||
|
|
|
|
Registered
|
Thanks, George.
|
||
|
|
|
|
Registered
Join Date: Feb 2013
Location: Middletown, CT
Posts: 285
|
Quote:
One of the crimp conector's insulation is red (the other is blue) and there's a light red paint mark on the smaller sensor terminal...and it's wrong. Now, on to the speedo...no ABS errors but without a Durametric I can't monitor an signals. Any other easy tests I can do to isolate it? - GA
__________________
1992 968 Coupe 1974 914 2L street car 1974 914 2L race car |
||
|
|
|