The Mustang Forum for Track Enthusiasts

Track Mustangs Online was built specifically for those who track their Mustangs!

Oil pressure gauge issue

captdistraction

GrumpyRacer
Messages
1,641
Reaction score
1,109
Location
Phoenix, Az
Since building a motor (and historically even), I've had a lot of problems getting a decent oil pressure gauge reading.

My current setup is two AEM brass 150psi oil pressure transducers/senders, one going to a AEM 150psi analog style gauge, and the other going to a Racecapture Pro MK3 data logger. The analog gauge also has a dedicated 0-5v output that I have going to the RCP logger as well.

What I'm seeing is once the car is out on track and heating up, oil pressure reading on the analog gauge (and what its sending to the RCP) is dropping significantly lap after lap, to the point where it was in the 40's at 7000rpm down a straightaway (very concerning indeed). Oil temps were climbing as well, however the RCP's readout from a similar sender did not show the same drop, and would read 80+psi as expected.

Here's my crayon diagram of how I have this setup:
1590865972155.png


and here's a data graph showing the drops: (2 laps, and can see the drop by lap 3 getting to be very significant) - this is reflected both in the logger as seen in this graph, and on the gauge face while I was out on track
1590866045444.png


Here's the RCP sender:
1590866120828.png


You can see some drop, but its to be expected with oil temps climbing from 220 in lap one, to nearly 290 in lap 3 (cooling is a separate issue, but this was at high altitude on a very hot day testing). I'd park the car after the 3rd lap.

So far, I've done the following:
Swap the sender and the gauge body out
Swap the RCP and Gauge's connections out (so the RCP would read the one sender the gauge was reading the session before)

And the behavior continued (with the difference of the pressure delta in the loop showing up, so in some tests it would show 40psi coming off the pump, but somehow 90psi going into the bearings, making little sense). I believe it has to be a wiring issue, but a quick test with a multimeter didn't show much that would affect the signal. I will order a replacement harness, however I don't think that's going to solve my issue as I couldn't find anything. having replaced each component at least once I'm at a bit of a loss.

There was some discussion around the bernoulli effect and some discussion about how noisy these senders are (which I've since added a 12" line to one to isolate it better from engine vibration). The other is well isolated with the oil filter block. I've not sent this to AEM yet for their opinion, but I don't like seeing oil gauge issues as that just gets in my head and sticks there as motors are expensive, though I believe this to be an obvious false positive.

Its costing me lap time and sleep, has anyone else run into something like this? Is there a better solution (I've been eyeing some compatible senders from bosch, autosport labs, and others that have the same scaling output)?
 

captdistraction

GrumpyRacer
Messages
1,641
Reaction score
1,109
Location
Phoenix, Az
for comparison, here's a session with the sender connections flipped:

1590866723978.png

1590866740709.png


In this case, despite the noise seen on the "oilpress" signal (straight to RCP), it somehow is picking up pressure through the oil loop as compared to the data from the aem gauge on OilPress2 (breaking known physics in the process). There's a TON of noise in that sender, but I've since isolated it to improve the signal. Using some smoothing algorithms showed it a tick under 80psi at the end of the lap, whereas the other sender was showing 42 psi. Based on temperatures, I can believe 80ish PSI in at 5800 rpm at 290* oil temps, but not the 44 number as I'd expect a camshaft to walk out of its caps or the phasers to start to loose control.
 

JAJ

TMO Addict
Messages
754
Reaction score
696
Location
In the V6L
I took a look at the gauge and sensor wiring and the oil filter adapter plate on the relevant supplier websites. One thing that I found - where you didn't say anything about it so it might be correct already - is the following point about the brown wire on the gauge wiring harness:

"WHITE – The WHITE wire should be connected to the Analog + input on the AEM EMS or the analog + input on a similar device.

BROWN – The BROWN wire should be connected to the Analog – input. If the EMS or similar device does not have a – input, the BROWN wire should be connected to a sensor ground. If no sensor ground is available, the BROWN wire should be connected to a power ground. Note: The BROWN wire must be connected in order to get correct readings from the analog output."


Is the brown wire on the gauge connected to ground or to the (-) input on the datalogger? Now, this may not be relevant - it says "correct readings from the analog output", implying that the gauge reading should be correct in any case.

You're absolutely sure that the reading isn't real - right? A blocked filter could be the problem. The filters are supposed to bypass when the pressure differential gets too high, but maybe this one isn't.
 

captdistraction

GrumpyRacer
Messages
1,641
Reaction score
1,109
Location
Phoenix, Az
flipping the sender orientations eliminates the filter as a concern (since a low reading on one side then somehow high back at the bearing galley inlet?). On the oil filter block, it has a gauge input on both the inlet (before filter) and outlet (after sides). I'm currently reading it after the filter, but the readings were the same either way (within margin of error for each other). I can be wrong, but I would expect there to be some loss from the reading near the filter vs the reading back at the block inlet (since the cooling loop and core should account for some pressure drop). However, I can't imagine any scenario where pressure would increase downstream towards the block return.

Brown wire is noted as the dedicated ground on the far right of the diagram and is using the same gauge ground as the other 7 analog gauges I have on my logger (they use a common ground, and outside the AEM gauge input as noted, they all share a +5 vref signal)
 

xr7

Overdosed on TMO
Messages
188
Reaction score
155
Location
Minnesota
Have you thought about using a mechanical gages temporarily to confirm diagnosis? My only thought wiring wise, your diagram shows three ground wires sharing a common ground. Might be an issue? I have ran into several bizarre electrical situations that turned out to be ground problems. Twisted pair wiring is used in some situations with sensors. Sometimes we ran independent grounds directly to the battery to isolate systems for testing. Sure interested to see what ends up being the problem. Good luck.
 

captdistraction

GrumpyRacer
Messages
1,641
Reaction score
1,109
Location
Phoenix, Az
Have you thought about using a mechanical gages temporarily to confirm diagnosis? My only thought wiring wise, your diagram shows three ground wires sharing a common ground. Might be an issue? I have ran into several bizarre electrical situations that turned out to be ground problems. Twisted pair wiring is used in some situations with sensors. Sometimes we ran independent grounds directly to the battery to isolate systems for testing. Sure interested to see what ends up being the problem. Good luck.
they call for a common ground for the analog outputs, but resistance measuring and ground loops are something I’ve put a lot of thought to. I might need to check the gauge body’s ground, it’s in a harness I hadn’t checked so a possibility.

the mechanical option is something I’m considering seriously just for my sanity
 

xr7

Overdosed on TMO
Messages
188
Reaction score
155
Location
Minnesota
they call for a common ground for the analog outputs, but resistance measuring and ground loops are something I’ve put a lot of thought to. I might need to check the gauge body’s ground, it’s in a harness I hadn’t checked so a possibility.

the mechanical option is something I’m considering seriously just for my sanity
I hear you, electrical gremlins can drive you nuts.
 

JAJ

TMO Addict
Messages
754
Reaction score
696
Location
In the V6L
flipping the sender orientations eliminates the filter as a concern (since a low reading on one side then somehow high back at the bearing galley inlet?). On the oil filter block, it has a gauge input on both the inlet (before filter) and outlet (after sides). I'm currently reading it after the filter, but the readings were the same either way (within margin of error for each other). I can be wrong, but I would expect there to be some loss from the reading near the filter vs the reading back at the block inlet (since the cooling loop and core should account for some pressure drop). However, I can't imagine any scenario where pressure would increase downstream towards the block return.

Brown wire is noted as the dedicated ground on the far right of the diagram and is using the same gauge ground as the other 7 analog gauges I have on my logger (they use a common ground, and outside the AEM gauge input as noted, they all share a +5 vref signal)
I figured you had it wired right, but I've overlooked things like this in the past, so I put it out there. AEM doesn't say it in so many words, but from the way they describe it, the "analog" gauge is actually digital and it's fitted with an internally isolated power supply that runs the gauge electronics. Those electronics are electrically isolated from the red and black wires that are the +12V and chassis ground return. The brown wire is the isolated internal negative lead - usually referred to as "common" - it's not "ground" unless you ground it. The gauge itself doesn't care if the brown wire's connected to anything, but any external signal tracker needs both wires hooked up to be able to work properly. I had a digital gauge on my 2011 GT and it was built that way.

You probably know all that already, but writing it down helps me think things through, so bear with me. Since the gauge is isolated, it should read correctly regardless of what's going on around it. Have you tried unhooking the RCP from the gauges and seeing what happens? If the gauge still claims that the oil pressure is dropping, then you're stuck with one of three possibilities - bad gauge, bad sender, low oil pressure. A cheap mechanical gauge is probably in order to rule out the latter so you can focus on figuring out the former.

Of course, if you hook up the RCP again and the problem comes back, then just go stepwise through reconnection until you find the wire that's the problem

Question - does this only happen when you're on track where the oil temperature gets really high? If it can idle all day without this happening, then it's not a classical electrical problem like a ground loop. If it's only when things get hot, it could be a sensor that's got a thermal problem or there's heat building up somewhere in the electronics.
 

JAJ

TMO Addict
Messages
754
Reaction score
696
Location
In the V6L
Another test that might be helpful would be to put both senders on the same pressure tap with a 1/8 NPT tee fitting. Do they read the same?
 
  • Like
Reactions: Coz

JAJ

TMO Addict
Messages
754
Reaction score
696
Location
In the V6L
I took a look at the RaceCapturePro Mk 3 setup guide. Unlike the gauges, it's built on the basis that the internal electronics are not isolated - the black ground wire that's tied to the chassis is the "common" or zero volt reference for everything it's logging. So, all analog signals it processes have to be referenced to the chassis. That means that all of the brown wires on all of the AEM gauges need to be tied to chassis ground (the black wire on the AEM gauges) as well.

Having thought this through a couple of times now, and if the oil pressure is actually fine, then the problem probably lies in an open wire somewhere in the wiring harness. I had this situation last week - I was wiring in an Xineering blip module and one of the wire taps I used actually severed the wire I tapped, leaving an open connection that was surprisingly hard to find because the wire and the tap looked ok.
 

xr7

Overdosed on TMO
Messages
188
Reaction score
155
Location
Minnesota
I took a look at the RaceCapturePro Mk 3 setup guide. Unlike the gauges, it's built on the basis that the internal electronics are not isolated - the black ground wire that's tied to the chassis is the "common" or zero volt reference for everything it's logging. So, all analog signals it processes have to be referenced to the chassis. That means that all of the brown wires on all of the AEM gauges need to be tied to chassis ground (the black wire on the AEM gauges) as well.

Having thought this through a couple of times now, and if the oil pressure is actually fine, then the problem probably lies in an open wire somewhere in the wiring harness. I had this situation last week - I was wiring in an Xineering blip module and one of the wire taps I used actually severed the wire I tapped, leaving an open connection that was surprisingly hard to find because the wire and the tap looked ok.
If the wire tap you are referring to is a Scotchlok or similar product don't use them, Scotchloks and wire nuts belong in the trash bin. My Dad's first 5th wheel trailer was littered with these types of connectors. Wire nuts used to wire the electric brakes under the trailer!!!!!!!!! Rewired the hole mess and problems went away.
 

blacksheep-1

TMO Addict
Messages
2,269
Reaction score
1,771
If the wire tap you are referring to is a Scotchlok or similar product don't use them, Scotchloks and wire nuts belong in the trash bin. My Dad's first 5th wheel trailer was littered with these types of connectors. Wire nuts used to wire the electric brakes under the trailer!!!!!!!!! Rewired the hole mess and problems went away.
agreed. Scotchlocks are belched from the pits of hell

all soldered
Kogwjqwl.jpg


 

TMSBOSS

Epic Contributor
Messages
4,982
Reaction score
2,043
Location
Illinois
Hard to beat a system with all connections soldered.

Just stay away from the word gun.
 

302 Hi Pro

Boss 302 - Racing Legend to Modern Muscle Car
Messages
1,766
Reaction score
235
Location
Northeast
Just stay away from the word gun.
TMSBoss Man:

I thought you would never type those words!

Well, It struck me as funny when I saw that & hope you see the humor in that as a guy who has shot (give or take) 100k rounds.
 

TMO Supporting Vendors

Top