Analog Oil Pressure Gauge on a 1998

[Chris_North]

From what I've been reading it should be possible to jump out the 20 Ohm resistor on the back of the instrument cluster to and replace the OP switch that closes above 8 PSI with a transducer from an earlier model (like 1982) that will output an analog signal to make a non-dummy OP gauge. Earlier today that's what I did, but I couldn't find any 20 Ohm resistor. I have read also that some people did not have the resistor for whatever reason, so I figured I would give it a shot anyhow. No luck. It's hard to tell for sure but it looks like the OP gauge reads in the same place as before or maybe slightly lower but does not fluctuate at all. It also pops right up as before and does not slowly build up like my mechanical gauge.

Is there something I'm missing? The part number for the OP sender I used is PS60. I remember reading that it was for an analog gauge, but now I can't seem to find that text. The body of the sender unit is much larger than the stock one (on the order of 3x), so I was figuring it was more than just an on/off switch, but possibly not. I haven't put my meter on it to test yet. These types of things just vary resistance to ground, right? Anyone done this before that can point me in the right direction?

 

[ab_slack]

Yep an actual transducer is much larger than the simple switch.

I haven't done this but was looking at doing this with our 99 Ranger last year. We ended up plumbing with a T and added an additional gauge instead of trying to get the dash gauge to read.

I don't recall what sensor I was looking at using. In my case I didn't want to pull the dash apart and found conflicting information about the resistor as well so I was trying to figure how I could get it to work as is even if I had to make a small converter circuit.

I gave up when I found to get the gauge to read the full range (without removing resistor) I would need to drive the signal negative and I didn't think it was worth putting a DC-Dc converter just for that. In hindsight it makes sense that with resistor in place it would need to go to a voltage less than zero since the switch closing provides zero volts and that is mid scale. So that is one problem you are against.

Also you can infer if 20 ohms yields mid scale, that less resistance is needed (or negative voltage) to get higher than mid scale. Any additional resistance causes the reading to go down. What is the resistance range of the sensor? Is that big compared to the resistor? If it is small it probably explains why the needle doesn't move much.

Does the needle jump up when engine starts or does it go up when ignition is turned on before cranking? If it is the latter, it suggests that the resistance range of the sensor is significantly smaller values than the fixed resistance used in the instrument cluster.

 

[a31ford]

To get a negative voltage, there is a really easy part it's called the ICL7660A (MUST be the A version to do 12 volts)

Below the first link is the pinout and hookup of the chip (note that they do NOT have an A version)

The second link is Intersil's and the DO have a A version, pinout for both is the same....

they are about 2-3 bucks at most electronics supplyhouses or online stores.

http://www.maximintegrated.com/datasheet/index.mvp/id/1017

http://www.intersil.com/en/products/power-management/isolated-power/charge-
pumps/ICL7660.html

Greg

 

[Chris_North]

Yep an actual transducer is much larger than the simple switch.

I haven't done this but was looking at doing this with our 99 Ranger last year. We ended up plumbing with a T and added an additional gauge instead of trying to get the dash gauge to read.

I don't recall what sensor I was looking at using. In my case I didn't want to pull the dash apart and found conflicting information about the resistor as well so I was trying to figure how I could get it to work as is even if I had to make a small converter circuit.

I gave up when I found to get the gauge to read the full range (without removing resistor) I would need to drive the signal negative and I didn't think it was worth putting a DC-Dc converter just for that. In hindsight it makes sense that with resistor in place it would need to go to a voltage less than zero since the switch closing provides zero volts and that is mid scale. So that is one problem you are against.

Also you can infer if 20 ohms yields mid scale, that less resistance is needed (or negative voltage) to get higher than mid scale. Any additional resistance causes the reading to go down. What is the resistance range of the sensor? Is that big compared to the resistor? If it is small it probably explains why the needle doesn't move much.

Does the needle jump up when engine starts or does it go up when ignition is turned on before cranking? If it is the latter, it suggests that the resistance range of the sensor is significantly smaller values than the fixed resistance used in the instrument cluster.

That's kinda why I figured this new sensor was definitely a transducer, but now I'm not sure. I will check it later today to see if I get any resistance changes. Pulling the dash apart was a deterrent, but there were some lights burnt out back there there were bugging me. It actually was a piece of cake to pull out and reinstall the instrument cluster. I didn't even have to get my hands dirty.

Right now there is no needle fluctuation that I can see, from 20 PSI to 60 PSI. I would think that if this is truly a transducer I should see at least some needle deflection. The other thing would be that it jumps up to slightly below midway right after starting same as before. I can't recall if this is where the needle always pointed; it probably is.

I'm not sure I understand changing polarity. From what I can tell the OP gauge is just a voltmeter. +12V is fed to the gauge all the time, then when OP is high enough the sender closes and completes the circuit to ground. Since there is (supposed to be) a resistor inline it reduces the voltage to whatever and the gauge reads midscale instead of being pegged. That's what I gathered anyhow. If you inverted the signal, wouldn't the gauge needle move backwards?

Currently what I have is the mechanical gauge teed with the dash sender, which is fine and much better than the in-dash gauge could ever be. I just figured since it's there it might as well work, and I would rather use the spot the mech. OP gauge is in for something else.

 

[ab_slack]

I'm not sure I understand changing polarity. From what I can tell the OP gauge is just a voltmeter. +12V is fed to the gauge all the time, then when OP is high enough the sender closes and completes the circuit to ground. Since there is (supposed to be) a resistor inline it reduces the voltage to whatever and the gauge reads midscale instead of being pegged. That's what I gathered anyhow. If you inverted the signal, wouldn't the gauge needle move backwards?

You have shaken some of the cobwebs and I am remembering a bit more.

The gauge can be a volt meter or current meter. I had originally been thinking current and the resistor set the appropriate current for mid scale.

The problem with that theory is that 20 Ohms would be an awfully low resistance and current would be high unless a voltage regulator was also used to drop the voltage very low to start with. 20K Ohms would be more appropriate if a current meter.

It being a volt meter than the resistance doesn't matter really cause it is sensing voltage. So if one side of the gauge is tied to +12V and connecting the other end to ground, 20 ohms or not the gauge sees 12V. The resistance can be 500 Ohms and the gauge would still see 12V because there is virtually no current flow.

I looked up the transducer you are using and keep seeing it referred to as oil pressure switch or gauge switch without much info, but I found one listing that said 8psi 24-36 Ohms, 90 psi 8.5 to 17.5 ohms. That seems very sloppy to me, but it does infer a resistance change.

What I think is happening to you is that the gauge resistance is adding to the 20 Ohms but as I was saying, if it is configured as a voltmeter, it really doesn't matter what resistance you have in series. Depending on what kind of bridge the meter uses the total resistance won't affect the reading until at least multiple hundreds of K Ohms or very likely into the megohms.

So instead of 20 Ohms connecting the meter negative (positive tied to +12), you have maybe 36 Ohms at 8PSI plus the 20 Ohms for 56 Ohms, the meter still sees 12V and thus reads half scale.

I wonder what the resistance is at 0 PSI? Does it go fully open? Or does it go to a high resistance? Even if it went to multiple K ohms I expect it will still read half scale. But maybe, just speculating, it has both transducer and switch elements in it so when at 0 PSI the switch part is open? Again I am curious, if you just turn on ignition but not start the engine, what does the gauge show?

As to purpose of the negative voltage let me explain what is happening. The gauge is always connected on the positive side. When a pressure switch is used and the switch closes the other side connects to return. This yields a half scale reading with a voltage of roughly 13.5V (cause the vehicle is running). This means that the "voltmeter" used for this gauge needs roughly 27V to read full scale.

Since the gauge positive is connected to the vehicle positive, to get more than 13.5V across the gauge you need a negative supply relative ground/vehicle chassis. For a simple example, we can take an extra 12V battery and connect the (+) terminal to the chassis. The (-) terminal of that battery will now be -12V. If we were to measure between the +13.5V and the -12V with a voltmeter we would now get 25.5V. See where I am going with this? The negative supply is needed so that the gauge negative terminal can be pulled lower than ground in order to get a larger absolute voltage to the gauge to cause it to read larger than half scale. The polarity sent to the gauge was never reversed because the positive of the gauge remains connected to the most positive available voltage.

Making this work, from what I had (and my transducer had low ohm resistances like yours) it was going to require not just a negative supply which could be done with a small DC-DC converter chip like the one a31ford mentioned, but also some additional circuitry that would put a small amount of current thru the pressure transducer (like 10ma or 20ma) to get a small voltage out of it and then scale that voltage up such that 5psi would output something like +8V to the wire going to the gauge to read at the bottom of the normal range and 60 psi would generate -11V to read at the top of the normal range.

While the circuit to do that wouldn't be terribly complex, there was going to be a pain to build. Then there would be the issue of where to locate. Close to the sensor where less chance of picking up noise but then there are all the environmental stuff requiring it be well sealed and/or potted. Put it up in the dash which probably makes more sense but then two wires still need to go out to the sensor (chassis return may affect reading) and then one wire going to the gauge. When needs to be found in the instrument cluster or a third wire running out the sensor and connect to the existing wire (may be the easiest).

I just didn't want to mess with all that.

I don't know if there is some other sensor that simplifies this, but it seems to me that no matter what there is the problem getting the gauge reading more than half scale without additional voltage from a negative supply unless other mods are done in the instrument cluster.

I don't know what others have done that have done this. When I was looking at it I found people in the process of trying, speaking about what they heard, and even mention of knowing someone who did, but didn't come across anyone who had and what they did.

Maybe there would be demand for a small circuit board that would convert this type of transducer into a signal that would work with the dash pressure gauge?

 

[RonD]

From what I've been reading it should be possible to jump out the 20 Ohm resistor on the back of the instrument cluster to and replace the OP switch that closes above 8 PSI with a transducer from an earlier model (like 1982) that will output an analog signal to make a non-dummy OP gauge. Earlier today that's what I did, but I couldn't find any 20 Ohm resistor. I have read also that some people did not have the resistor for whatever reason, so I figured I would give it a shot anyhow. No luck. It's hard to tell for sure but it looks like the OP gauge reads in the same place as before or maybe slightly lower but does not fluctuate at all. It also pops right up as before and does not slowly build up like my mechanical gauge.

Is there something I'm missing? The part number for the OP sender I used is PS60. I remember reading that it was for an analog gauge, but now I can't seem to find that text. The body of the sender unit is much larger than the stock one (on the order of 3x), so I was figuring it was more than just an on/off switch, but possibly not. I haven't put my meter on it to test yet. These types of things just vary resistance to ground, right? Anyone done this before that can point me in the right direction?

I think the external resistor on the oil pressure gauges was removed in the mid-'90s
The newer gauges use a 3 wire setup.
When power is supplied to the gauge it passes thru a resistor circuit(two resistors in series) to a ground which will show 0 on the gauge, so 2 wires, 12v and ground.
When engine is started and pressure is above 5psi the oil pressure switch closes and becomes a less resistant ground, this is the 3rd wire.
It divides the resistor circuit(only 1 resistor is being used now) and gauge needle shows 1/2 depending on voltage.

Not sure you could easily change this externally, you for sure only have 1/2 the needle travel to work with.
You can't lower the resistance because the resistance is fixed internally, the 3rd wire is either a ground or it isn't.
A higher resistance could get the needle to go down a bit, until the other ground was better, at 0.
So finding or making a sender that got higher resistance as oil pressure went DOWN might work.

12volts--------20ohms-----/------30ohms---------ground
Ground----Sender---------/

I don't know the resistor ohms just using this as example.
So at 50ohms needle is at 0 pressure
At 20ohms needle is at 1/2
So you would have a 30ohm window(20 to 50) to adjust gauge between 1/2 and 0
25 ohms would be just under 1/2
45ohms just above 0

 

[ab_slack]

I think the external resistor on the oil pressure gauges was removed in the mid-'90s
The newer gauges use a 3 wire setup.
When power is supplied to the gauge it passes thru a resistor circuit(two resistors in series) to a ground which will show 0 on the gauge, so 2 wires, 12v and ground.
When engine is started and pressure is above 5psi the oil pressure switch closes and becomes a less resistant ground, this is the 3rd wire.
It divides the resistor circuit(only 1 resistor is being used now) and gauge needle shows 1/2 depending on voltage.

Not sure you could easily change this externally, you for sure only have 1/2 the needle travel to work with.
You can't lower the resistance because the resistance is fixed internally, the 3rd wire is either a ground or it isn't.
A higher resistance could get the needle to go down a bit, until the other ground was better, at 0.
So finding or making a sender that got higher resistance as oil pressure went DOWN might work.

12volts--------20ohms-----/------30ohms---------ground
Ground----Sender---------/

I don't know the resistor ohms just using this as example.
So at 50ohms needle is at 0 pressure
At 20ohms needle is at 1/2
So you would have a 30ohm window(20 to 50) to adjust gauge between 1/2 and 0
25 ohms would be just under 1/2
45ohms just above 0

I don't know the exact wiring internal other than only one wire goes out to the switch.

I remember trying to play with resistor values between the wire and ground and see what I could do. I really don't remember the details there other than thinking it wasn't making much sense. But maybe I was able to make it read from half scale down by going larger resistance.

Just observing that if the resistances I looked up for the pressure transducer Chris is using are correct that if his gauge always remains in the center it is probably small compared to resistances around the gauge.

I do distinctly remember feeding voltages in and mapping the response of the gauge and needing to drive it negative to get greater than half scale. Which makes sense even if there is already a divider in the dash.

 

[Chris_North]

I've been drinking a little so bear with me, haha.

I'm pretty sure I've found the problem with my setup as it is and that is that this supposed transducer is really just another switch. Engine off OP sender to ground I get 75k or so Ohms (basically open circuit). As soon as the engine turns on I get 0 ohms. I also did a current check to see what figures we're working with here and only came up with .071A through the sender but oddly .092A when bypassing the sender and connecting directly to ground. When connected to ground the gauge needle moves up very slightly as well. This is weird because like I said I get a resistance of zero from sender to ground.

ab_slack I understand where you're coming from a bit more now. I think you're right that the gauge is probably a current gauge. At 14V 20 ohms should be about .7 Amps, so that's not really high. Much higher than what I got, but not higher than I think would be allowable. I see what you're saying about the voltage but i don't see any way to create more than 12V DC out of a 12V DC source with out a complex circuit or another battery in series, which as you said is not worth it to make the OP gauge work. Unless, also as you mentioned, there was a company that produced a simple in line fix for cheap.

RonD I'm not doubting you but I don't understand the purpose of the third wire/resistor. If there is no ground, doesn't the gauge read zero? Why wire it to an additional resistor just to have it read zero? In any case I think you're right. Even with the right sender without eliminating whatever causes the gauge to read halfway the best i could hope for is a reading halfway when i would have really high OP or slightly less than a 1/4 when I have normal OP.

At this point I might try a different sender but i think without figuring out some way of cutting out the resistor I might as well just give up and stick with the mechanical aftermarket gauge. I've had an idea to buy and modify a better cupholder/center console and mount additional gauges inside that, but I've never followed up on it.

 

[ab_slack]

I've been drinking a little so bear with me, haha.

drinking bears, okay gotcha. I hear that is difficult to do without getting fur in your mouth. And just how many of those bears did you drink I wonder?

Like I am one to talk since I regularly type the wrong word all the time and bears and beers yes they are pretty close.

I want to say, RonD always has great information so don't discount what he says. I've learnt lots from what he has posted. I can actually think of several reasons why there may be a divider in there like he describes.

I'm pretty sure I've found the problem with my setup as it is and that is that this supposed transducer is really just another switch. Engine off OP sender to ground I get 75k or so Ohms (basically open circuit). As soon as the engine turns on I get 0 ohms.

Was the 75K across the sensor with the sensor disconnected? Or was the wire going back to the gauge still connected? I would expect a pure switch to measure infinite when open. If it was with the wire connected, the voltage present on the wire may have confused the ohmmeter and 75K may not be meaningful.

ab_slack I understand where you're coming from a bit more now. I think you're right that the gauge is probably a current gauge. At 14V 20 ohms should be about .7 Amps, so that's not really high. Much higher than what I got, but not higher than I think would be allowable.

Actually .7A is very high. If alternator rating is 140A that be a half a percent. Yes, not much, but for a little gauge? Also 0.7A would mean there is close to 10W of power being lost. That seems like quite a bit to me.

The 71mA you measured is much higher than needed for a gauge, but I suspect it is biases to provide some current for contact reliability purposes. I don't know all the details, but on a physical contact like a relay or switch (in this case a pressure switch) too little current can result in contact failing early because not enough energy to burn thru corrosion so running higher current can be more reliable. 71ma doesn't seem unreasonable in that case.

One question, did you measure the voltage across the sensor when the engine was off? Assuming the sensor is functionally an open circuit the voltage there would give some idea if there is a divider someplace else. If it is the full battery voltage, it suggests that there isn't a divider. And the 71ma suggest approximately 200 Ohm series resistance.

Without knowing for certain what the gauge wiring is, what resistances are around it or what current it requires, much of this is guesswork and I can only fall back on my memory of that playing with our 99 Ranger where I found I would need negative voltage..

Now I did do some poking around and did come across this http://www.rogueperformance.com/oilsw.html

That seems to suggest the sender you used with resistor pulled provides some variable reading. Maybe only up to half scale, it wasn't clear, but the sensor you used is on the list. I don't know how if it is just a switch function.

 

[Chris_North]

drinking bears, okay gotcha. I hear that is difficult to do without getting fur in your mouth. And just how many of those bears did you drink I wonder?

Like I am one to talk since I regularly type the wrong word all the time and bears and beers yes they are pretty close.

Haha not drinking beer (okay well that too) but I meant bear with me as in share the burden.

I want to say, RonD always has great information so don't discount what he says. I've learnt lots from what he has posted. I can actually think of several reasons why there may be a divider in there like he describes.

I absolutely agree, that's why I said I didn't doubt him. I was just thinking aloud.

Was the 75K across the sensor with the sensor disconnected? Or was the wire going back to the gauge still connected? I would expect a pure switch to measure infinite when open. If it was with the wire connected, the voltage present on the wire may have confused the ohmmeter and 75K may not be meaningful.

Nope, wire disconnected but sender connected to the block. TBH I'm not entirely sure it was kiloOhms, it may have even been megaOhms. I think my meter had a k, but I just took it as open. I have seen very high resistance as opposed to true open circuit before. It could have just been some oil or moisture between the insulator on the sensor. It could as you said be that the sender is really a transducer, but given how i got zero as soon as the engine turned over I personally don't think so.

Actually .7A is very high. If alternator rating is 140A that be a half a percent. Yes, not much, but for a little gauge? Also 0.7A would mean there is close to 10W of power being lost. That seems like quite a bit to me.

The 71mA you measured is much higher than needed for a gauge, but I suspect it is biases to provide some current for contact reliability purposes. I don't know all the details, but on a physical contact like a relay or switch (in this case a pressure switch) too little current can result in contact failing early because not enough energy to burn thru corrosion so running higher current can be more reliable. 71ma doesn't seem unreasonable in that case.

One question, did you measure the voltage across the sensor when the engine was off? Assuming the sensor is functionally an open circuit the voltage there would give some idea if there is a divider someplace else. If it is the full battery voltage, it suggests that there isn't a divider. And the 71ma suggest approximately 200 Ohm series resistance.

Without knowing for certain what the gauge wiring is, what resistances are around it or what current it requires, much of this is guesswork and I can only fall back on my memory of that playing with our 99 Ranger where I found I would need negative voltage..

Now I did do some poking around and did come across this http://www.rogueperformance.com/oilsw.html

That seems to suggest the sender you used with resistor pulled provides some variable reading. Maybe only up to half scale, it wasn't clear, but the sensor you used is on the list. I don't know how if it is just a switch function.

You're right, .7A is an awful lot for just a gauge. I was thinking more along the wiring aspect of things, but that's not really accurate.

I didn't measure voltage across anything, though I would also be interested to see what comes up. Like I said though this is getting to be really complicated for something that even if I can get to work still won't be as accurate as the mechanical gauge I already have. The only reason I really wanted to get the stock unit to work is because it was there and it would look a lot neater than an aftermarket gauge bolted to my dash.

 

[RonD]

Gauge would be off if there was no ground, which would be fine I suppose, but I would think with key on you would expect a gauge to respond, even if it's just to show 0 pressure.
Like a Cold temp or Empty fuel tank