Jump to content
krohelm

Toyota Oil Pressure Sender

Recommended Posts

Toyota 4age with the factory oil pressure sender:  How do you read the pressure from it?  It's not the switch, it's the sender for the pressure gauge on the instrument cluster.

 

I wired it to +5v with a 100 ohm divider, and measured the voltage using AN4 thinking I'd just map resistance to pressure.

 

Here's what AN4 does while the engine isn't running:

5a6e5284d231f_EngineOff.png.36864395b6fa1fca6b1c5609d3e4ca1f.png

 

And with the engine running, revs changing dramatically, it just sits within a couple hundredths of a volt:

5a6e5288bed1c_EngineOn.png.cc0b6500f2736dfe7d66417d6c9de210.png

 

The sender is a Toyota 8352035031.  How do you measure the output?

Toyota Oil Pressure Sensor AN4 Log 2018-01-28 2;18;37 pm.llg

Share this post


Link to post
Share on other sites

These are a single wire sensor right? From a 90's toyota with a single wire its extremely unlikely to be pwm, and that engine off graph is probably supposed to be 5v constant but there may be a loose wire or something. The way you've wired it up should give you a 0-5v output as the resistance of the sensor changes, but without knowing the actual resistance values and what that sensor's max pressure is I cant tell you whether its going to be a full 0-5v or more like 0-2v. Is the dash gauge still wired up? you could approximate the voltage vs pressure numbers from this compared to your logged voltage.

Is that second graph when the engine is cold? Oil pressure stays pretty high regardless of revs until the oil warms up, then it tends to sit relatively stable above say 2k, but at idle revs it can drop off quite sharply. If the size of your resistor isnt quite right that 0.6v - 1.9v you see may be the extent of normal range.

Share this post


Link to post
Share on other sites

The engine off graph, there's definitely no loose wire.  Nothing is moving, it's on a lift, I've audited the wire all the  way to the gauge.  It's an 80's Toyota gauge haha :-)

Dash gauge is removed, the entire cluster has been replaced by a custom display I made from scratch...  I've read in the factory service manual that the gauge sends pulsed voltage to the gauge, and that to test it you put a light in series with the gauge, start the car and watch it flash faster as engine speed (oil pressure) increases.

In the second graph, the engine is warming up, and is approximately 170f.  The needle would normally swing quite noticeably between 900rpm and 3400rpm at warm temperatures.  (i.e., 1/4 to 4/5 at this temperature)

0-2v would be totally fine.  According to the 5v supply, 100 ohm r1 value and observed 1.56v on AN4, the sender sat at 47 ohms while the oil pressure swung at least 30psi.

Share this post


Link to post
Share on other sites

Is it a standard 1/8 NPT thread? might be easier to just buy an aftermarket sender that has a known resistance vs pressure slope.

If you want to keep that sensor... looking at your log again after that explanation, it does sound a lot like a pwm type implementation, similar to how a speedo works with pulses. In some cases you can read pwm signals as analog voltages anyway if the duty cycle changes. If its always 50% duty cycle but the frequency changes, the average voltage is always the same though, this looks like what you're seeing. Try connecting it to one of DI 1-6 instead and given you have an external pullup already, disable the internal pullup in the config. you can set these inputs to "frequency". See if that gives you some correlation between oil pressure and the numbers seen by the ecu. 

If it is the case that its sending frequency, i'm not sure how you can tell the ECU that the frequency it receives correlates to oil pressure. Maybe someone else has figured this bit out

Share this post


Link to post
Share on other sites

I am sure that type of sensor is open circuit when the engine is off, which will leave the ANV pin floating, you need to place a 4.7k ohm resistor across the output and ground.

The ADC in the link will still give 40 points of accuracy over 0.2v so over say 60 psi you will measure 1.5 psi steps.

Share this post


Link to post
Share on other sites
4 hours ago, cj said:

Is it a standard 1/8 NPT thread? might be easier to just buy an aftermarket sender that has a known resistance vs pressure slope.

If you want to keep that sensor... looking at your log again after that explanation, it does sound a lot like a pwm type implementation, similar to how a speedo works with pulses. In some cases you can read pwm signals as analog voltages anyway if the duty cycle changes. If its always 50% duty cycle but the frequency changes, the average voltage is always the same though, this looks like what you're seeing. Try connecting it to one of DI 1-6 instead and given you have an external pullup already, disable the internal pullup in the config. you can set these inputs to "frequency". See if that gives you some correlation between oil pressure and the numbers seen by the ecu. 

If it is the case that its sending frequency, i'm not sure how you can tell the ECU that the frequency it receives correlates to oil pressure. Maybe someone else has figured this bit out

It's not 1/8 npt, it's BSP thread...  You can do adapters, but darn it, there's already a sensor there and a wire to it :-)

I'll try a digital input and see what's there - but seems like the analog input would have shown some change with a wide pressure swing...  Definitely want to understand how to map frequency to oil pressure if this is the case!

3 hours ago, ClintBHP said:

I am sure that type of sensor is open circuit when the engine is off, which will leave the ANV pin floating, you need to place a 4.7k ohm resistor across the output and ground.

The ADC in the link will still give 40 points of accuracy over 0.2v so over say 60 psi you will measure 1.5 psi steps.

Check the log, it isn't floating - it cycles.

+5v -> 100ohm -> AN4 and Sender -> Ground
If Sender is Open, AN4 should be 5v.  If I put 4.7k ohm to ground, AN4 should read 4.9v if the sender is open, right?  What does that gain over just 5v without the extra divider path to ground?

Share this post


Link to post
Share on other sites

With the 100 ohm pull-up, the digital input does not fire.  I've been told the transition point for digital inputs is ~1 volt.  Look up at that first graph to see it doesn't really cycle that low with a 100 ohm pull up.  Would be great if that were configurable!!

The graph above could indicate a 66 ohm resistance when "on" or "low," so I'd need a 600 ohm pull-up to get to 0.5v when driven "low."

With a greater resistance on the pull-up, the modulating behavior is not observed anymore, but the input is still rock steady while running (despite great rpm swings at operating temperature) as observed by AN4.  Either the ECU's +5v is overloaded and cycling with a 100 ohm load (seems unlikely) or the current offered through a 600 ohm resistor is not sufficient to charge whatever's in that pressure sender bulb and get the PWM on DI-1.

Share this post


Link to post
Share on other sites

Can you draw a simple diagram of how you have this wired? is your resistor a pullup to 5v, a pullup to ground, or in series with the signal wire? 

Have you also tried confirming the behaviour in the manual? get 3x AA batteries and tape them together with some wires to make a nearly 5v battery pack, wire positive directly into the sensor with a bulb in series and ground the negative side to the chassis, and check you actually see the pulsing behaviour you expect. See if you get pulses or a dimmer/brighter bulb.

Share this post


Link to post
Share on other sites

A very fair request!  Here's what I've tried:

image.png.23deb19d2f3544d6630d1400b9fec934.png

This was my first try, still seems like it should work...  I've varied the resistance from 100 to 460.  Via math, my sensor's resistance seems to sit around 66ohms regularly.

 

image.png.dee21ee7f53a15cc9dcc9f68e1ce8662.png

This should work if it's doing something PWM-y, but there's no PWM activity from the multimeter (should read like 0.75hz with engine off) but if it's using the power incoming from the ECU to drive the sensor, maybe the built-in pullup is too restrictive.

 

image.png.67e34a658e7add21e35351b849235d1e.png

Tried a couple different resistors to get the original voltage graph to below 1v, but in no case have I observed PWM (even via multimeter) while this sensor is hooked up to a digital input.

 

Am I wiring this wrong?  :-) That would be great.

Share this post


Link to post
Share on other sites

The diagrams all look correct for a 5v pullup. If none of them are working, and you'd rather spend time than money in replacing the sensor, I think you need to start testing the behaviour of this sensor to confirm what's written in the manual.

Can you measure its resistance while the car is off? Do you see any fluctuations like in the graph when its got 5v supply?

Can you measure its resistance again with the car running - but the sensor still dicsonnected from the ecu.

Do you have a multimeter that measures frequency or duty cycle, if so can you power it from the 5v line only (no pullups, no signal wire), and measure across the sensor to see what if any frequency type signal is happening.

can you either wire it from the 5v line with a bulb in series as described in the manual, or build a battery pack from 3x AA batteries to test the same thing? 

Do you still have the original dash you could wire up to confirm the sensor still works and how it behaves? old toyota dashes ususally have pretty obvious traces on them so you can figure out which of the 3 or 4 pins you need to liven up pretty easily.

Share this post


Link to post
Share on other sites

Spent a long time measuring this sensor.

  • The resistance does not change when wired in a way that is consumable by an analog input.
    I should give up on analog input.
  • Even if I can produce PWM from this sensor, I have to link it somehow to "Oil Pressure."

I can measure frequency in voltage, but this crazy thing seems to operate via using the voltage to operate the sender, and "send" on the same line by consuming some amount of current periodically.  I.e., it toggles a potential increase across a powered line.

Found the testing procedure in the FSM, it wants 12v.  I gave it that and looked - I got a similar periodic blip in voltage, but it was only slight, nowhere near enough to divide down to get a reliable 1v digital boundary.  It's hard for me to measure an in series resistance wave form, or graph an amperage waveform.  I think I need to give up on using this sensor.  It's going to be a real hassle finding one that'll fit this space!

Share this post


Link to post
Share on other sites

run a BSPT adapter elbow out of the block to a line with a decent sender mounted away from the PS/AC bracket, or grab a defi/greddy pressure sender which is already BSPT and will screw straight into the block behind the bracket

 

Share this post


Link to post
Share on other sites

I've only seen their NPT sender, and those Honeywell 150psi senders don't have a BSP option according to the spec sheet.  I'll check in with my local fittings & hydraulic supply shop to see what we can come up with.

Thanks all!

Share this post


Link to post
Share on other sites
3 hours ago, krohelm said:

I've only seen their NPT sender, and those Honeywell 150psi senders don't have a BSP option according to the spec sheet.  I'll check in with my local fittings & hydraulic supply shop to see what we can come up with.

Thanks all!

The NPT sensors were labelled wrong that are actually BSPT I reported this a while back. The dirrerence is only 1 thread over an inch so an easy mistake to make.

Share this post


Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

Loading...

×
×
  • Create New...