Oil pressure gauge wiring question.

jmackro

Well-Known Member
Messages
2,452
Reaction score
721
Location
San Juan Capistrano, Ca.
They don’t. The 928/944 sender has the resistances as outlined in a post above. Low resistance(10-15 Ohms)=low pressure(0) and high pressure (5bar)=~180-200 Ohms.

Well that shoots down my idea that the gauge and sender may be incompatible with each-other.

I would still be curious what resistance (if any) makes the gauge register zero bar. E.g., what it would display with power on, but the wire to the sender disconnected. And what the gauge would display with the engine running and sender connected. It's possible that the gauge is simply bad or way out of calibration.

A shop like North Hollywood Speedometer could probably troubleshoot this gauge-sender pair, though their price might approximate the cost of a new set.
 

Bmachine

Well-Known Member
Site Donor $
Site Donor $$
Messages
3,525
Reaction score
1,778
Location
Northern California coast
I would still be curious what resistance (if any) makes the gauge register zero bar. E.g., what it would display with power on, but the wire to the sender disconnected. And what the gauge would display with the engine running and sender connected. It's possible that the gauge is simply bad or way out of calibration.
With power on but sender disconnected, the gauge reads 0. When I connect the sender it jumps to 5.
I will reinstall the sender today and test it with the engine running.
 

Luis A.

Well-Known Member
Site Donor
Messages
1,529
Reaction score
357
Location
Wisconsin
The way I tested mine was to use a potentiometer I had laying around that had an appropriate range, hooked it up to the wires that normally go to the sender. Then see what you get at various resistance points. I'm sure you can get one on Ebay from a USA supplier and have it in a few days. Conversely, if you place your compressed air supply agains the tiny hole on the sender while your DMM is hooked to it you can see the variations in resistance. Watch for nasty oil spraying out of the hole on your face...
 

Luis A.

Well-Known Member
Site Donor
Messages
1,529
Reaction score
357
Location
Wisconsin
With power on but sender disconnected, the gauge reads 0. When I connect the sender it jumps to 5.
I will reinstall the sender today and test it with the engine running.
This then sounds like the gauge requires a different sender. Plain and simple.
 

jmackro

Well-Known Member
Messages
2,452
Reaction score
721
Location
San Juan Capistrano, Ca.
This then sounds like the gauge requires a different sender. Plain and simple.

Agree. You have now verified that the gauge is capable of displaying something than other than just 5 bar, so it is probably OK. And, you have verified that this gauge displays a lower pressure reading with higher resistance (the resistance is essentially infinite with the wire disconnected), so it appears to work in the opposite manner from the data in the link that Arnie posted in #10. But where you would find the correct sender is an unknown to me.

Luis: Assuming Bo really has a Porsche 944/928 gauge, doesn't this contradict what you wrote in post #20? Is it possible that some 944/928 gauges work on the low resistance = high pressure reading principle?
 
Last edited:

Bmachine

Well-Known Member
Site Donor $
Site Donor $$
Messages
3,525
Reaction score
1,778
Location
Northern California coast
Can't work on the car right now so, in the meantime, I've ordered both a 200 ohms pot and a new standalone VDO 80 psi gauge. This way I will at least have a tool and a benchmark to test things out.
I also went on a 928 forum (crazy how these things always lead you down some rabbit hole that you never had any intention of exploring...) and at least one other guy mentioned having the same problem on his car:
"While on the oil subject: I just noticed that my oil pressure gauge is constantly pegged at 5 bar when the key is turned on. Come to think about it I don't know when this even started."
Unfortunately there was no answer to that question, apart from "Make sure the wires are not switched".
 

Bmachine

Well-Known Member
Site Donor $
Site Donor $$
Messages
3,525
Reaction score
1,778
Location
Northern California coast
Ok. Never mind on the “can’t work on the car now” part. I was just too curious and could not leave it alone. So I reinstalled the sensor, hooked everything back up and the gauge jumped to 5 bars as previously. Then I turned the engine on. And sure enough the gauge went to zero!. I measured the resistance from the sender and it was around 200 ohms with engine running so it really seems the sender is working correctly. It’s just that the gauge is working backwards.
 

Luis A.

Well-Known Member
Site Donor
Messages
1,529
Reaction score
357
Location
Wisconsin
Luis: Assuming Bo really has a Porsche 944/928 gauge, doesn't this contradict what you wrote in post #20? Is it possible that some 944/928 gauges work on the low resistance = high pressure reading principle?
While that looks like a 928 gauge (except the needle is white, not red/orange as in the 928), I can't vouch for how it has been mounted and wired in that instrument pod. Nor did I read that he tested the gauge by feeding a verified resistance value other than infinite/open, so can't say if the gauge is truly working as it should. But I can assure you that all transaxle Porsches (and many 911s) use that sender, up to the 928 GTS where they use a three-pronged sender.
 

Bmachine

Well-Known Member
Site Donor $
Site Donor $$
Messages
3,525
Reaction score
1,778
Location
Northern California coast
The electronics in those gauges is fairly simple. I wonder if there is a way to inverse the movement of the needle. Something like x1 = 1-x after normalizing our 10-185 ohms scale.

Maybe there is a little adjustable electronic converter/remapper I could add just before the gauge. It could also be used to calibrate the gauge to the sender.
 
Last edited:

Luis A.

Well-Known Member
Site Donor
Messages
1,529
Reaction score
357
Location
Wisconsin
The electronics in those gauges is fairly simple. I wonder if there is a way to inverse the movement of the needle. Something like x1 = 1-x after normalizing our 10-185 ohms scale.

Maybe there is a little adjustable electronic converter/remapper I could add just before the gauge. It could also be used to calibrate the gauge to the sender.
I still would test the gauge to verify its function without the sender, or the wiring to it. I once had one of those gauges acting weirdly and it was a pinched wire, intermittent and with varying resistance, to ground between the sender and gauge.

Take the 200 ohm potentiometer and hook it up in lieu of the sender. What do you get? Then do the same directly to the gauge, without using the in-car wiring. What do you get?
 

jmackro

Well-Known Member
Messages
2,452
Reaction score
721
Location
San Juan Capistrano, Ca.
I still would test the gauge to verify its function without the sender, or the wiring to it. I once had one of those gauges acting weirdly and it was a pinched wire, intermittent and with varying resistance, to ground between the sender and gauge.

Take the 200 ohm potentiometer and hook it up in lieu of the sender. What do you get? Then do the same directly to the gauge, without using the in-car wiring. What do you get?

Both good suggestions

Bmachine said:
I wonder if there is a way to inverse the movement of the needle. Something like x1 = 1-x after normalizing our 10-185 ohms scale. Maybe there is a little adjustable electronic converter/remapper I could add just before the gauge. It could also be used to calibrate the gauge to the sender.

I think you're heading down another rabbit hole there! Simply verifying how the gauge works (as Luis suggests above) and then finding a proper sending unit would be a lot simpler.
 

Bmachine

Well-Known Member
Site Donor $
Site Donor $$
Messages
3,525
Reaction score
1,778
Location
Northern California coast
I went on a 928 forum and they confirmed that the gauge should work “normally”, ie 10 ohms means no oil pressure and 180 means 5 bar. So the sender is good.
The logical next assumption is that the gauge is wonky. But when a gauge (or anything for that matter) is out of whack, it will either not work at all or show some crazy behavior. Not simply showing the right info, except precisely backwards. It is really puzzling.

@Luis. Re the color of the needle: yes, the gauge had the Porsche orange needle when I bought it. We took it out and replaced it with the white one from the coupe gas gauge.
 

frogisland

Well-Known Member
Messages
294
Reaction score
99
Location
Cheshire, UK
This might be a very simple fix - have you checked that the sensor is actually grounded where it screws in to the pipe? My race car build had what i take as the same problem; pressure needle to far right at battery on. Once i connected a temp ground to the base of the sensor the needle went across to zero, started engine and correct pressure reading.

I was fortunate that i have two sources (running in series) for my pressure warning light (back of head head and into remote filter at warning/pressure sensor), at battery on, the warning light went out when the head warning was disconnected which made me think there was a ground issue across at the warning/pressure sensor itself. Connected a temporary ground and all ok.

Edited to add; further look into this, its probably a weak ground rather than no ground,tested with multimeter, grounding to sensor was weak. Mine is in aero fittings, the anodising on them was preventing good contact. Removed where necessary on the threads and all working as it should :) The other sign was the slightly weaker warning light on the dash. Brighter with better ground path.
 
Last edited:

Bmachine

Well-Known Member
Site Donor $
Site Donor $$
Messages
3,525
Reaction score
1,778
Location
Northern California coast
Thanks for following up, Luis. I have not found the final answer yet. However I have a suspect.
In order to test the validity of the wiring and sensor, I bought a new VDO 80 psi cockpit gauge. I hooked it up with the same setup in the car and sure enough it works exactly the way you expect it to. Showing Low teens at warm idle and between 40 and 55 at high revs.
So the issue is definitely either with the Porsche gauge or with the conversion that was done by a friend of mine to incorporate it in the coupe gauge cluster.
That part I still have to investigate. It involves taking that gauge apart which I have not had time to do yet.
 

deQuincey

Quousque tandem...?
Site Donor
Messages
8,371
Reaction score
2,389
Location
BIO - 43°15'46.5"N 2°56'03.7"W
Thanks for following up, Luis. I have not found the final answer yet. However I have a suspect.
In order to test the validity of the wiring and sensor, I bought a new VDO 80 psi cockpit gauge. I hooked it up with the same setup in the car and sure enough it works exactly the way you expect it to. Showing Low teens at warm idle and between 40 and 55 at high revs.
So the issue is definitely either with the Porsche gauge or with the conversion that was done by a friend of mine to incorporate it in the coupe gauge cluster.
That part I still have to investigate. It involves taking that gauge apart which I have not had time to do yet.


it looks like a very interesting project
 

Bmachine

Well-Known Member
Site Donor $
Site Donor $$
Messages
3,525
Reaction score
1,778
Location
Northern California coast
I have been absorbed by home renovation projects so I have not been able to do more testing. However, I did find out a couple of interesting things related to these two pics, both labeled as coming from the 928:

928 oil clockwise.jpg 928 oil counterclockwise.jpg


First, it looks like the 928 did come with two different style oil pressure gauges. One reading clockwise and one reading counter clockwise. Aha! (Unless the second one is mislabeled and actually from another car...)


Second, my gauge was the same as the one in the first picture. And isn't it odd that the oil pressure in that pic reads full pressure but the voltage is at 0? If there is full pressure, it means the engine is running so the voltmeter should read at least 13volts. It could simply be that the volt gauge was disconnected of course...
 
Last edited:
Top