4.9 volts into coolant temp sensor but...

Hey electron lovers,

I have 4.9 volts into coolant temp sensor (2008 Honda Fit with 120k miles) but when I used a LoadPro to simulate a load voltage drop in the circuit it dropped to to 82.1 mv.

How does that make sense if the circuit is working most of the time? The fault code appears every few weeks.

I'm testing because am getting low coolant temps signal about 9 months after putting in an aftermarket thermostat and water pump.

Those results look about right for a 5V pulldown ECT signal. Could repeat the test with a scan tool hooked up just to confirm the ECM sees the ECT signal being pulled to ground if you want to be thorough.

Have you got freeze frame for these low temperature codes? If so take a look at ECT to see if it's just low or pegged low.

I agree with Matt. Thermistor circuits are designed to be pulled low.
The Load Pro really isn't for signal circuits or reference circuits, it is for loading a battery positive voltage where a component is going to consume amperage.
Be careful where you push that button, there can be unintended consequences.

Noah wrote:

Be careful where you push that button, there can be unintended consequences.

Noah wrote:

I agree with Matt. Thermistor circuits are designed to be pulled low.
The Load Pro really isn't for signal circuits or reference circuits, it is for loading a battery positive voltage where a component is going to consume amperage.
Be careful where you push that button, there can be unintended consequences.

Oh wow...shows you how little I know about electrons! But man its fun learning. ok so know I know, Loadpro isnt for signal circuits!

So on these signal circuits I can still test for "ghost voltage"/open and shorts to ground but NOT voltage drops?

Matt T wrote:

Those results look about right for a 5V pulldown ECT signal. Could repeat the test with a scan tool hooked up just to confirm the ECM sees the ECT signal being pulled to ground if you want to be thorough.

Have you got freeze frame for these low temperature codes? If so take a look at ECT to see if it's just low or pegged low.

I could do that! I have Autoenginuity and I can watch live data and load the circuit and see what the engine computer is seeing.

thanks for this informative video! I love this kinda stuff.

...but I'm not using a power probe. Im using a LoadPro.

You can still use the load pro as a a regular meter lead, but you WILL pull a reference or signal voltage to ground by pushing the button. Which could be helpful if you are watching for a data pid to change state, or could end up activating components.
I am reminded of my first week with a load pro before having been introduced to Scanner Danner. I quite mistakenly started the engine of a SAAB while "testing" the ignition switch. A co-worker could have been injured, tools could have been pulled into the drive belt, etc...

Noah wrote:

You can still use the load pro as a a regular meter lead, but you WILL pull a reference or signal voltage to ground by pushing the button.

Splendid example showing the importance of knowing your tools, the circuit design and both limitations and capabilities oof them.

That's why we are here. to learn.

That is what I wanted to do. See if the computer picked up the changed so I know the circuit is good.

The other option according to Dan is to jump the connector and see if the temp drops down to the -40 or whatever the manufacturer min temp is.

Keep getting this intermittent p0128...once a week on highway. Might be the aftermarket thermostat, but that was installed 8 months ago.

Keep getting this intermittent p0128...once a week on highway. Might be the aftermarket thermostat, but that was installed 8 months ago.

It isn't unusual for cooling system codes to surface in cold weather. Thermostat will appear to be working fine but ECM ain't happy with it.

Took a look at aftermarket code description and it mentions two ECTs, either of which can set P0128. And Rock Auto do catalog two sensors for this application. One for the thermostat housing which is likely ECT1 and the one you can see in OBD-II datastream. Another in the radiator which is probably ECT2 and may only be available in the OEM data. If you can access PIDs for both sensors comparing them KOEO after sitting overnight would be a good starting point. If one of them is skewed that could set the code.

www.dtcdecode.com/Honda/P0128

Correct. ECT 1 next to thermostat and ECT2 in radiator.

My friend who I am helping with this issue just casually tells me yesterday...oh, the cabin heat hasn't been as strong since we put in this Gates thermostat and water pump!!!!

I wanted to strangle him. Why didnt he tell me this before all of his chasing of ghosts started???

We took out the Gates and put in a Honda OEM thermostat ...and now he says the heat is is incredible. I'm guessing this is the cause.

OEM versus Gates 33340S "heavy duty". Aftermarket thermostats...me no like.

Honda on left, Gates on right.

I bet that fixes it. I honestly have only seen 2 instances of a 128 that wasn't the thermostat.
Once on a 98 Corolla, where that code points to an oxygen sensor problem for whatever reason, and once on a Chrysler 300 where the the temp sensor was bleeding coolant into the connector.

Did I mention I wanted to strangle my buddy for withholding that info?:evil: