CLT Monitoring, getting warm/cooling off fast?
 

Getting the sensors right, for context:
https://www.miataturbo.net/megasquir...1/#post1139952

Now I'm trying to monitor some really peaky CLT values. I've attached a quick snip of a graph for reference.

Right smack dab in the middle, you can see the CLT/MAT climb. Great, it's boosting, getting hot -- normal. Uhoh, getting close to CEL time. Running out of revs, shift ....

My real question: Wow, in a matter of 3/4 of a second, the CLT drops ~20*F. Really?

I knew I was pushing the limit of my 1.6 cooling gear, but it seems fine most of the time. Was hoping to get some autox in this weekend, but would like your input.

Views, but no thoughts. Likely bordering on "deal with it/upgrade like the rest" territory.

Not to be hasty, but I am still tentative about my sensors actually reading out properly. Observing this kind of behavior (rapid fluctuation) doesn't help.

That looks weird, are you sure you don't have any air locks in the coolant?

It would be interesting to see voltages too, it could be dipping?

It's very much possible, I recently dumped the system and filled with distilled and Waterwetter. I'll get it up to temp again soon and get the front wheels up to burp it.

Went back and checked the voltage, it's essentially as flat as anywhere else on the chart.

I've been discussing this off-line, so to speak, and a valid point was brought up.

It doesn't seem likely that a nearly instant drop of 20* would be measurable by such a simple sensor. Note that MAT does the same thing as CLT.

Looking for more suggestions on what to chase here.

Well all this "simple sensor" has to do is measure temp, what were you expecting it to do? ;)


However your CLT trace is too suspiciously identical to your MAT trace. Are you sure you don't have a short between those 2 sensors?

Agreed on the suspicion.

I don't know how this (the short) could be, though? The MAT is two wires from pins 1/6 on the old AFM connector, to the sensor.

The CLT is the 1.6 sensor in the BP location.

The only questionable thing I have is what I did with the two ground connectors that were left stranded when the 1.6 was pulled. I believe they originally grounded on the fuel rail, but it's since been replaced w/ a 99-00 rail and no connectors are present. Thus, I moved those connectors to a single ground wire I terminated on the back of the head.

Ran new connectors/wire for the MAT.
Re-did under hood grounds to back of head (removed eyelet for engine lifting -- just a bolt, ring terminal, and the head surface).

Sharp drop in CLT continues. About a 30*F drop in a very short amount of time.

CLT and also MAT seem to be following the TPS to a large degree. Looks more like TPS than MAP to me. It HAS to be sensor ground.
If you're absolutely certain that your sensor ground is wired properly and not cross-linked to power ground somewhere in your harness, or connected to the engine, then something may be wrong on your MS board.
Seeing that you have a 1.6, you must have wired in your TPS yourself. Maybe something's wrong with that....
Have you tried disconnecting the TPS?

Or maybe something with your reference voltage is screwed up? But I can't really see how that would happen.

Yep, 1.6 chassis but 1.8 (MSM) under the hood.

OK, now sensor ground vs. power ground, you've got my interest as this is a new distinction that I wasn't aware of. Have something I can check? As mentioned, I'm using the stock grounding locations aside from the pair of small ring terminals on the passenger side in the engine bay.

I haven't touched the TPS in the course of troubleshooting.

Your sensors must be getting their ground from the ECU and must *NOT* be hooked up to anything else. OneGND wire going from the ECU to all the sensors. No other connections.
Otherwise current flowing through the ground wire will cause a voltage drop across the wire and cause false reading.
Not sure if it would cause this much of a difference though.

I see, I've definitely seen it worded that way...

Being that I'm using the body harness for everything, is this a concern?

Wut? How does that even work?

It's just a 1.8 swap w/ with a turbo, really. Using 90-93 MS2 from Reverant.
The only new wire I had to run, concerning the MS, was from the IAT to the AFM connector pins.

So what are you using the body harness for then?
Sounds like you're using the stock 1.6l engine harness hooked up to the Reverant MS2.
Again: Make sure your grounds are not cross-connected. I don't know your schematic. On my '99 there is a dedicated ground going from the ECU to the sensors (MAT, CLT and O2). It's black/brown.

Easy check:
- Look at the schematic and figure out what your sensor ground is
- Disconnect your ECU
- Measure the resistance from sensor ground on the harness connector to chassis ground. If you see a low resistance, you have the grounds wrong.

Yeah, I'm using body/engine harness as interchangeable terms... which is incorrect.

Anyway, I'll check in to your suggestions, thank you.

Resolution:

1. Referred to http://www.mellens.net/mazda/Mazda-M...991_wiring.pdf

sensor ground is black w/ light green stripe

2. Observed prior referenced ring terminals under hood, the ones that I grouped together on the engine head. Only ONE of four wires was black w/ light green. The other three were black.

3. Snipped sensor ground out of ring terminal, as it was grouped with a black wire. Terminated back at the MS sensor ground. Put the black wires back on the engine head.

4. Drove, monitored, success.

Congratulations. Enjoy your MS-enhanced ride :-)