Oil Pressure Sending Unit not available

D

Drdestructo

New Member
Joined
20 July 2018
Messages
8
Location
Hood River, OR
Reading the posts about low oil pressure indications and the service bulletin here : http://www.nsxprime.com/tsb-91-008/

Seems the part number 37245-PR7-002 is no longer available.

Anyone know of another unit that would properly drive the gauge. Splicing wires into a connector shouldn't be difficult so there must be some part that would do the job. I read zero pressure when idling hot.

Thanks,

Russ
 
OEM appears to be NTK 1S6780,

but at 2x the price of amayama.

Standard Motors PS-430 is not the same as it is simple switch.
 
Last edited:
MotorMouth93 said:
I know of several people who have had quick failures with the SoS sending unit mounted on the firewall, myself included, I wouldn’t recommend it. Mine died during engine break in, I almost shat a brick when the oil pressure gauge dropped to zero.
Click to expand...

I trust Chris will react to these failures.
 
I was nicely questioned on the veracity of my post by another NSXer, so I did a little bit of investigation.

The WVE-1S6780 is likely to be the exact OEM oil pressure sender. These appear to be existing stock (NOS) and cost is +US$300.

When looking up the support documents for the WVE-1S6780 https://wvebrand.com/part-detail/sales_drawings/PS430.pdf you get the original WELLS document. This document clearly shows the NSX oil pressure sender, indicates "Replaces OE # 37245-PR7-A02" and uses WELLS part number PS430.

NTK is now WVE

The oil pressure warning light is driven by a pressure switch located on the oil filter pedestal.

Oil Pressure Gauge in the instrument cluster wants to see a resistance range of 20-137Ω for a 0-8 bar range (0-116 psi).

[Edit: incorrect information removed]

(we need a knowledge base for PDF's like this, this is important stuff).
 
Last edited:
drew said:
I was nicely questioned on the veracity of my post by another NSXer, so I did a little bit of investigation.

The WVE-1S6780 is likely to be the exact OEM oil pressure sender. These appear to be existing stock (NOS) and cost is +US$300.

When looking up the support documents for the WVE-1S6780 https://wvebrand.com/part-detail/sales_drawings/PS430.pdf you get the original WELLS document. This document clearly shows the NSX oil pressure sender, indicates "Replaces OE # 37245-PR7-A02" and uses WELLS part number PS430.

The modern PS430 has a different package but appears to fit and function as original (including the connector). This part is less than US$20.


I personally would go with the newer form factor as it is smaller, lighter and a lot less expensive. (better, faster, cheaper). https://www.summitracing.com/parts/smp-ps430

NTK is now WVE

The Standard Motor Products PS-430 is now Made in China.

Appears to be used on the SAAB 9-3 2000-2003, 9-5 1999-2009, p/n: 55559824 or 9176660

(we need a knowledge base for PDF's like this, this is important stuff).
Click to expand...

New forum/site software will have a reference library function.
 
MotorMouth93 said:
Interesting info on the PS430. It seems to be interchangeably listed as a "sender" or "switch" so I wonder if its a variable resistance like an NSX OEM sensor or just a dummy light switch? Its cheap enough that I might order one to find out, the small form factor is nice.
Click to expand...

Perhaps save your money. I did a very cursory check, and as far as I was able to determine from the few SAABs that I checked that were listed in the application guide, none were equipped with oil pressure gauges. So as you speculated, it is likely a switch.

At that price, it would have been a nice find. There are now a lot of compact automotive class fluid pressure sensors available that are in the $30 - $40 price range. Unfortunately, all of the units that I have seen are 3 terminal devices with a 0-5v output, similar to a TPS or MAP sensor. It would be possible to use one of these devices by building an interface that converted the 0-5v output to a 'faked' resistance to drive the gauge. A Trinket controller and external PNP transistor would probably do that for less than $10. I think [MENTION=30613]Heineken[/MENTION] has effectively generated the code for that with his gauge tester module. The largest hassle would be finding a good place to install, wire in and power up the interface module.
 
As already mentioned, the NSX' oil pressure sensor is a resistor device. The dial on the gauge can be controlled by PWM together with an output driver that pulls the voltage from the cluster down to GND. The corresponding PWM values are known from the cluster tester development for each major mark on the gauge but can be interpolated linearly for intermediate values.

An Arduino nano would be the smallest device I can think of which supplies 5 V for a 3-pin pressure sensor and an ADC to read back the value. Like [MENTION=7701]Oldguy[/MENTION] said, the biggest issue is probably to give it a "roughly" automotive grade housing and define a proper place to install it.

There's one downside of the PWM control: It's a little noisy - depending on the position of the needle a slight hum or fizz can be heard - probably unnoticeable when the engine is running and similar to the noise floor hiss from the Bose speakers.
 
The automotive grade hardware is already there. Cars with EGR deleted could theoretically repurpose the ECU I/O to read an aftermarket 3 pin oil pressure sensor (EGR valve lift plug on the front valve cover is 5v, gnd, and signal) and generate a PWM output (from the EGR valve PWM) to run the gauge. I've thought about attempting it but have been too busy with other ECU projects. It would be nice since the gauge could be calibrated to actually be accurate and the factory pressure gauge location is only a few inches away from the EGR lift sensor plug.

The code would be fairly simple, basically just pulling the sensor voltage from where its kept in RAM, feeding it into a lookup table mapping voltage to PWM duty cycle, then write that value into the EGR PWM register. Just linking a few existing APIs together.

Now that I'm thinking about it, I might try to test this theory tonight.
 
Last edited:
So now I'm in agreement with [MENTION=7701]Oldguy[/MENTION] and [MENTION=33247]MotorMouth93[/MENTION]


The Standard Motors PS-430 is a simple switch. This is for the cluster indicator lamp.

The NTK, WVE, Duralast PS430 is a ranged sender for the gauge.

Seems the part number has been corrupted over the decades with corporate M&A's.
 
Last edited:
Heineken said:
As already mentioned, the NSX' oil pressure sensor is a resistor device. The dial on the gauge can be controlled by PWM together with an output driver that pulls the voltage from the cluster down to GND. The corresponding PWM values are known from the cluster tester development for each major mark on the gauge but can be interpolated linearly for intermediate values.

An Arduino nano would be the smallest device I can think of which supplies 5 V for a 3-pin pressure sensor and an ADC to read back the value. Like @Oldguy said, the biggest issue is probably to give it a "roughly" automotive grade housing and define a proper place to install it.

There's one downside of the PWM control: It's a little noisy - depending on the position of the needle a slight hum or fizz can be heard - probably unnoticeable when the engine is running and similar to the noise floor hiss from the Bose speakers.
Click to expand...

If you are willing to deviate from the 'library' PWM routines, you might be able to adjust the PWM carrier frequency so that it is above the audible range or at least away from any resonant frequencies in the meter (which could shorten the life of the meter mechanism). EM coupling into the audio system was a problem that did not occur to me.
 
I've been meaning to test whether I can hijack the EGR PWM output for my own purposes for a while so I figured this was an excellent opportunity to do so, I've thought about adding an oil pressure monitoring function to the ECU (limp mode and/or CEL if oil pressure drops below a predefined RPM vs pressure curve for too long, for example) so this would be a cool additional feature.

I started by depinning the EGR valve control circuit (red wire on the 6 pin connector on the emissions box) adding a weak 10k ohm pullup to 12V, and hooking it up to my oscilloscope. Its an open drain PWM output so it toggles between shorted to ground and open circuit, so the pullup was needed to for it to be easily visible on my scope. This happens to be perfect for driving the oil pressure gauge which points higher as resistance to ground decreases. Turns out its a 25hz PWM so quite slow, but probably good for a device as slow as the oil pressure gauge. Slow enough that there's zero discernable noise from the gauge.

Once I had confirmed that I could manipulate the PWM duty cycle on the scope, I wrote a small function to pull a value from a predefined ROM address and write it into the EGR PWM control register, then hooked this function into the main background loop of the ECU code, then from there I can just edit the value in ROM through TunerPro and see the changes in real time.

It worked pretty much perfectly, playing around with it I found that I could walk the gauge through the full range by varying the PWM duty cycle. A little more code could capture the voltage from an oil pressure sensor, use it to look up the desired PWM output from a table, and send it to the gauge. Something similar could easily be done with an arduino or any other small development board like that, or you could take it a step further and use a digipot to use an actual resistance instead of a PWM output.

I might do this myself just as a party trick, but ultimately I wanted to see if I could use the EGR output to control something like, say, a nitrous solenoid valve, and it appears that it would be quite possible. The frequency is even very close to the ideal frequency that most nitrous solenoids seem to work well with.
 
Last edited:
That is nice work and much faster turn around than I ever achieved on any idea I have ever had.

I am not particularly in the know about digital pots; but, as I recall they typically have a feature of a minimum resistance (wiper resistance - 75 ohms on a MCP413 chip) which may bump into the normal operating range for the sensor and if they are a 5 volt class device they don't deal well with voltages above 5 volts on the 'resistor' terminals. They also typically use a serial interface so you would have to implement that in the ECU which would be a double yuck factor for me. If the PWM output from the ECU can sink sufficient current to drive the gauge that is a particularly nifty solution if you have deleted the EGR.
 
Thats a good point with the digipot - if they have a minimum resistance that probably wouldn't allow the gauge to reach max - the PWM duty cycle was like 97% (meaning grounded 97% of the time, open circuit/high impedance 3%) to get the gauge to reach the highest tick so resistance would likely need to be very near zero. A small Arduino or TI launchpad board with a slow PWM output hooked to the gate of an N channel MOSFET or (BJT if you prefer) would probably be the best approach to driving the dash gauge. Analog electronics are not my forte so it would be interesting to see if changing the PWM frequency would affect the response time of the gauge, as it was pretty slow with the 25hz signal.
 
Last edited:
MotorMouth93 said:
I know of several people who have had quick failures with the SoS sending unit mounted on the firewall, myself included, I wouldn’t recommend it. Mine died during engine break in, I almost shat a brick when the oil pressure gauge dropped to zero.
Click to expand...

Sorry to interrupt the conversation, but i'd love to know more about these failures, like what in particular failed (was it just the pressure sensor itself?) and did the others experience the same exact failure? Has anyone blown an engine because of it? and has everyone had a working oil pressure gauge before the SOS unit that never dropped to zero?

I just installed the SOS sending unit about 100 mi ago. In general, cold start idle shows a normal gauge reading, but after driving it around to normal op temp, I do see zero on the gauge at idle. It still jumps up when driving, but drops to zero whenever I come to a stop. However this is the same way my gauge would behave with the OEM (leaking) sending unit before the SOS unit, so I've been figured the gauge may just be bad. Would love to know more info about the SOS unit and failures in general to know if I should remove it and purchase a new OEM unit (or whatever it is that you guys are discussing here, it's all mostly above my head tbh lol)
 
Going to 0 psi would normally cause the oil pressure switch to illuminate the oil pressure warning light on the dash. That said, the pressure switch contacts usually open up if the pressure is much above 5 psi so its possible that the gauge could read 0 with no warning light if the gauge is a little out of calibration. The oil pressure switch and the sensor are also at different locations on the engine so it is 'theoretically possible' to have good pressure at one location and poor at the other.

If your new sender and your old sender are exhibiting the same behaviour it is possible that a bad electrical connection between the sender and the gauge can cause it to drop to zero. However, I would be inclined to get a mechanical oil pressure gauge with a length of tubing and temporarily connect it to the port where the sender is normally connected. That would allow you to confirm that your oil pressure at that port is actually good before chasing some electrical gremlins. If the mechanical oil pressure gauge exhibits the same flakey behaviour as your dash gauge, then you likely have some bigger issues to address.
 
I have had a zero and/or close to zero reading on my oil gauge since I bought it in 2003. I have just learned to live with it, I check the oil, I listen for trouble, and, yes, every now and then I think about getting a new sending unit, figuring that is the first place to start, and then I forget about it since I rarely ever have time to drive it. Last year I did 200 miles. I would love to have it working properly though. But it really never has. So...
 
Chris_Lum said:
I just installed the SOS sending unit about 100 mi ago. In general, cold start idle shows a normal gauge reading, but after driving it around to normal op temp, I do see zero on the gauge at idle. It still jumps up when driving, but drops to zero whenever I come to a stop. However this is the same way my gauge would behave with the OEM (leaking) sending unit before the SOS unit, so I've been figured the gauge may just be bad. Would love to know more info about the SOS unit and failures in general to know if I should remove it and purchase a new OEM unit (or whatever it is that you guys are discussing here, it's all mostly above my head tbh lol)
Click to expand...

I had trouble with my gauge for the longest time as well and replaced it with the SOS unit a couple of years/2k miles ago. It did react very similar to the OEM sender when new. Down the road the blue wire broke at the resistor so I wired it back up with out the resistor in line. The readings are now much higher on the gauge at all rpms and load. I believe that resistor is only 5 ohms. Talking with Chris at SOS, you can run this sensor without the resistor safely. I now also have a canbus GaugeART and it reads exactly what the ECU reads instantly. I'm covered either way but I certainly feel better seeing a higher oil pressure on the OEM dash gauge.
 
drew said:
Where is this "blue wire"?

Did you remove the resistor at the gauge?
Click to expand...
The Blue wire is the sensor wire that plugs into the OEM Wire Loom stock location. It comes with the SOS sending unit. Here is a picture. You will notice this sending unit is sitting on top of a "T: with an AEM sensor branched off of it. This is wired directly to my standalone ECU where the GaugeART canbus pulls it's data.
.SOS Oil Press Sending unit.jpg
 
Last edited:

Similar threads

asylum
Oil pressure sensor / sending unit
Replies
8
Views
150
Old Guy
Old Guy
Chris_Lum
DIY: SOS Oil Pressure Sensor Upgrade (91-99)
Replies
0
Views
213
Chris_Lum
Chris_Lum
warrenw
Oil pressure sending unit
Replies
1
Views
560
drew
drew
1oldpoop
  • Locked
***SOLD*** 1991 NSX Red & Black Automatic with Black interior
Replies
2
Views
1K
vh_supra26
V
Wild Turkey
Electrical: Radio display not dimming, buttons not illuminating when headlights on
Replies
9
Views
1K
Wild Turkey
Wild Turkey
Back
Top