5v Reference Circuit problem (maybe) (FIXED)

Customer brought his 2011 Mazda 3 with a 2.5 engine because his fans are on all the time (it has a module type fan with 3 wires) and has had it in several shops, and now its my turn at it.

Codes :

P0102 Mass airflow circuit low reference
P0113 IAT sensor 1 circuit high
P0117 ECT circuit low input
P0118 ECT circuit high input
P061B Internall control module torque calculation performance
P0685 ECM/PCM power relay control circuit/open
P2184 ECT sensor 2 circuit low
P2185 ECT sensor 2 circuit high

Among other ABS U codes (wich I can post if you guys think its relevant).

When I look for live data I noticed that ect sensor 2 (wich is located in the radiator) was showing 55 F (3.4 volts), and the ect 1 sensor (wich has been changed twice) was showing 186 F (0.67 volts), IAT was showing 64 F (2.49 volts) and the outside temperature sensor was showing 62 F (I couldnt find a voltage PID). So I unpluged ALL of the thermistors and this is what they read:

ECT 2 -40 F ( 4.99 volts)
ECT 1 -40 F (4.63 volts)
IAT -40 F (5 volts)
Ambient sensor -40 F

So I go and test rigth at the sensors and basically it was the same readings (a bit lower on all of them because of the multimeter). So Im thinking of high ressistance on the ECT 1 signal wire, but with PCM disconected the wire from the ECT to the PCM is 0.2 ohms and OL Megaomhs to groung and power, so its not the wire.

So I get everything connected and acces live data again with the ECT 1 disconnected and monitor the voltage PID wile I use my Power probe hook to inject 5 volts to the circuit just to see what hapens and my ECT 1 voltage PID shows 5 volts (again, signal wire ok right?)

So I think to myself that there has to be a sensor that shares that 5 volt reference internally (5 voltage ref 1 and 2 maybe?) and is partially shorted. I now procede to unplug all of the sensors that use a 5 volt referance (Throthle, MAP, fuel tank pressure sensor, ac pressure sensor, egr, etc.. and everthing shows the same voltage.

Im now check powers and grounds at the PCM and everything is ok.

Now what?

Im at a point were Im about to call the computer but that has been tried in another shop already.

Hi georG, I moved your post to the repair questions section.

Do you happen to have a diagram for the fan circuit?
I'd pull one for you, but I'm mobile at the moment...

Noah wrote: Hi georG, I moved your post to the repair questions section.

Do you happen to have a diagram for the fan circuit?
I'd pull one for you, but I'm mobile at the moment...

here it is

One more thing I forgot to mention, when I touch the ECT 1 signal wire with my multmeter or my power probe 2, the TAC makes a clicking sound. This doesn't happen with any of the other thermistors or sensors that use a 5 volt reference.

I should also add that the fan will turn on with the car running because PCM is "seeing" 200 + F after a few seconds, so I don't think the problem will be in the fan circuit.

check out this video, have you performed these tests??

cheryl hartkorn wrote:

check out this video, have you performed these tests??

Yes all of that's been sorted out. Like I said earlier, the fan comes on because PCM is "seeing" 200 + F after a few seconds, and then it comands the fan on, so my problem is that the 5 volt feed from the pcm is lower than it should be. This is why I've been chasing a "short" on the 5 volt regulator.
I know there's no way to have a "short" on the 5 volt signal wire of a thermistor, but I'm "guessing" it's internall, but the problem with that "theory" is that none of the other sensors (including thermistors) have any drop on their 5 voltage references, wich is why I "guess" this PCM might have two 5 voltage regulators, but I'm really just guess working at this point., and that's what brought me to ask here.

So... HELP !!!!

I already checked the wiring, and it's not shorted.

This is what I've checked :

-Unpluged ALL the sensors, valves and actuators to see if my ECT 1 voltage goes to 5 ( it doesn't).
-Unpluged PCM, unpluged ALL the sensors, valves and actuators, and ohm tested ALL and every single wire that comes from them to the PCM (ohmed check to power and gound).
-Powers and grounds for PCM are in spec.
-Introduced 5 volts to the ECT 1 signal wire while monitoring the voltage PID, and it shows a steady 5 volt and my power probe hook doesn't show a short.
-After clearing the codes, none have returned (I think only ECT low voltage did, I would have to check).

What am I missing before I call the computer?

nice work, with the unplugging all the sensors. I wouldn't of had the patience I would of cut the reference wires at the pcm and see if I got the 5 volts back then if so started tracking the wiring. its hard when your not there working on the car. maybe try putting a potentiometer in place of the ect sensor that reads 200?

cheryl hartkorn wrote: nice work, with the unplugging all the sensors. I wouldn't of had the patience I would of cut the reference wires at the pcm and see if I got the 5 volts back then if so started tracking the wiring.

Imagine cutting ALL the sensor's that share a 5 volt reference, I think that would be more dificult, dont you think?
Any suggestions?

Based on all your testing, I'm wondering if there's high resistance inside the PCM (between the 5V regulator and the connector pin) that's causing the readings you're seeing. Besides a short, which you've shown doesn't exist, this is the only explanation I have. :blink:

Just for giggles, I'd be curious to know if the 'skewed' temp sensor reading followed the circuit. i.e. wire the ECT1 harness to the ECT2 sensor. If the ECT2 sensor read, say, 70 degrees when hooked up correctly, then ECT1 should read the same when connected in reverse. But, based on everything you've done so far, it probably won't. That'd help to verify a circuit problem.

If you wanted to be double sure of your wiring, you could also load it with a test light. Not that there was anything wrong with your PP Hook test, just offering a verification test.

Otherwise, I think you're looking at a PCM. To be clear, the fan does shut off with the key, right? And stays off at lower (measured) coolant temps?

Tyler wrote: Based on all your testing, I'm wondering if there's high resistance inside the PCM (between the 5V regulator and the connector pin) that's causing the readings you're seeing. Besides a short, which you've shown doesn't exist, this is the only explanation I have. :blink:
Otherwise, I think you're looking at a PCM.

I think so too, but what are the ods of another PCM having THE SAME problem (because the other shop already tried a used computer)?
So Im looking for others to chime in with tests that I might have missed.

Tyler wrote: Just for giggles, I'd be curious to know if the 'skewed' temp sensor reading followed the circuit. i.e. wire the ECT1 harness to the ECT2 sensor. If the ECT2 sensor read, say, 70 degrees when hooked up correctly, then ECT1 should read the same when connected in reverse. But, based on everything you've done so far, it probably won't. That'd help to verify a circuit problem.

I will try that tomorrow.

Tyler wrote: To be clear, the fan does shut off with the key, right?

Yes.

Tyler wrote: And stays off at lower (measured) coolant temps?

It doesnt take that long for the car to reach high temperature (wronfuly tough), but yes.



Thanks to everyone

GeorG wrote: I think so too, but what are the ods of another PCM having THE SAME problem (because the other shop already tried a used computer)?
So Im looking for others to chime in with tests that I might have missed.

Ah crap, I missed the part where another PCM had been tried. :blush: My mistake!

Now that I think about it, ECT1 and 2 probably don't agree with the engine cold and the key on, do they?

Tyler wrote: Ah crap, I missed the part where another PCM had been tried. :blush: My mistake!

Jajajaj, thanks Tyler, its great that you think its the PCM because I also think its a PCM problem, but like you, just knowing that another PCM has been tried makes me double think my own conclusions. Like the Danner always says, its tough calling a computer.

Are you sure it is not a failed water pump? What are the ohms readings of the ect1 when its cold and when fans kick on?

borntoroll wrote: Are you sure it is not a failed water pump? What are the ohms readings of the ect1 when its cold and when fans kick on?

This is the second ECT that has been tried. Like I said, the other shop already tried all the obvius, plus, with ECT disconected we should have near 5 volts on the signal wire, not 4.6

Ok, suppose PCM sends 5v through a 100k resistor. The question: what will be the voltage measured with your 1megaOhm impedance voltmeter?

borntoroll wrote: Ok, suppose PCM sends 5v through a 100k resistor. The question: what will be the voltage measured with your 1megaOhm impedance voltmeter?

Yep it drops a bit like I mention in the begining of the topic

GeorG wrote: So I go and test rigth at the sensors and basically it was the same readings (a bit lower on all of them because of the multimeter).

But comparabitly (with the other thermistors) it is substationally lower.

What am I missing guys.

HELP!!!!!!!

Check ohms

borntoroll wrote: Check ohms

Could you be more specific to were I should check ohms please? :huh:

cheryl hartkorn wrote: nice work, with the unplugging all the sensors. I wouldn't of had the patience I would of cut the reference wires at the pcm and see if I got the 5 volts back then if so started tracking the wiring. its hard when your not there working on the car. maybe try putting a potentiometer in place of the ect sensor that reads 200?

I still like Cheryl's idea of snipping the 5v ref wire at the PCM and measuring voltage there. I know you've exhaustively checked for shorts on this wire, but I think it's relevant. If it happens to jump to 5v, great (wishful thinking).
Other than that, I'm thinking that logically, a poor PCM ground could cause a low voltage condition on that circuit, but you say you've checked powers and grounds at the PCM.

How about the thermistor? Is it possible that it could be incorrect for this application? Maybe a cheap white box unit?I wonder if both ECT sensors have the same resistance at the same temperature, or if they are interchangeable. I'm thinking particularly of the double step thermistors.

I was also wondering if you have a way to measure the actual temperature around the sensor location while the ECT is reporting nearly 200.
Unfortunately, I have little service info for this car. The diagram you posted blows away anything All Data has to say.