Fluctuating CLT: Head gasket, temp sensor or something else?
 

So I've noticed an issue I've been having lately with fluctuating coolant temperature readings. Take a look below:
https://www.miataturbo.net/attachmen...ine=1463869268

A little about the car: 93 with a 1.8 swap, begi reroute spacer (where the coolant temp sensor is located, I believe, the 2 wire one) with the front blocked off with a plate entirely. In an effort to make absolutely sure I don't have air bubbles, I first used my fancy new vacuum tester/filler tool to first pull a heavy vacuum and make sure it held (I think it started to creep up maybe 1 psi after about 20 minutes, I figured that was good) and then hooked up the filler hose and let her fill (supposedly prevents any air getting in at all by making the vaccuum pull the coolant). I did this all with the front end raised (never the back), however I was slightly disappointed when it didn't completely fill the radiator (it was mostly but not completely) so my next step was putting on my filler neck funnel thing for bleeding out air bubbles just to be double sure, put it up, warmed her up, thermostat opened, fans came on twice, after that I ran the fans a little long to really cool the radiator before putting the cap back on, and drove away to pick up my suzuki cappucino washer tank at the post office. This was the result. I can post the logs if needed, but I think you can see what's going on. I'm not overheating, fans kick on around 195 and it never exceeds 200.

Looks like a ground problem. What does the temp gauge say?

Agreed on the grounding issue - especially if your dash gauge is reading OK.

I would clean your sensor ground. I had to actually run myself a new ground wire when I had a similar issue.

Maybe I'm just not getting something, but I don't really see a problem.

I see a datalog that's zoomed way out (that's over 5 minutes of log right?) and I'm seeing coolant temps between 177 and 198. That seems like a pretty normal range, especially if you've got a cooler thermostat in there. Everytime you idle, the temps go up. Then you drive, temps go down. Am I missing something?

Couple of things. First, I'm *guessing* a ground issue is fairly unlikely given that the sensor has its own ground according to the wiring diagrams (also makes sense given that the sensor has 2 wires, not just 1). One wire goes to the ECU, the other to a grounding block (again, according to the wiring diagrams). So I'm guessing that's not the issue, though others can chime in with yes/no and why's. It's not out of the question given that the begi spacer appears to give a lifted ground in the sense that it doesn't actually touch the block, the bolts, or the neck (however it does touch the coolant, which does actually conduct some electricity, as evidenced by me checking my dead-times and touching my multimeter to my radiator, not that I was testing that way...). As far as the gauge on the dash goes, it's showing normal, but that stupid thing says "normal" all the damn time (it's biased to the low end of the gauge, even spanning 20F it barely moves if it all). By the time that thing even moves it's usually too late (if it's going hot, which it isn't, but my point being is that the dash gauge can't be trusted).

Second, it is relatively zoomed out but it was a short drive also. Here it is going from 196.9-184.6 in 0.6 seconds:
https://www.miataturbo.net/attachmen...ine=1463883819

I should also mention my idle kpa has always been a bit high. Somewhere between 38-42kpa. I saw in another thread Brain was asking someone who thought they had a bad head gasket what their idle kpa was and he thought 45kpa meant head gasket troubles. I've ordered a leakdown test kit that should arrive tomorrow. I've also uploaded my full log for this trip here:
https://onedrive.live.com/redir?resi...int=file%2cmsl

So the results of the leak-down test were good I think? So, first time I ran the leak-down test, I came up with 71/75 for cylinders 1+4, 61/75 for cylinders 2+3 (warm). Then I had to run to dinner at my parents. Unsatisfied with these numbers I did a little research and saw some people mentioning rotating the crank a couple of degrees either way to get a good seal and sure enough numbers jumped up to 71/75 on 2+3 (this was on a relatively cooled engine, but the numbers seemed to match what I was getting earlier which was about 61/75 at what I thought was TDC before shifting a few degrees). Is this cause for concern or just fine? When I had it leaking before it seemed to be into the crankcase through the oil filler, which was concerning. So there's a few ways to read this I think. One is I wasn't quite at TDC before and that's why my numbers were low and once I got there the numbers got higher OR, maybe I have cylinder wall wear at TDC and I need a rebuild. Thoughts? I feel like these results are good but I'm not sure... I did still hear some leakage through the oil filler, and even if I was only close to TDC on the compression stroke it still should have been sealing better than it was as no valves should be open anywhere around that time. Furthermore, if that is leaking at TDC because of cylinder wall wear there, are those numbers cause enough for concern to warrant a rebuild?

Just replace the sensor.

Temp sensors are too lethargic to record a change of 12F in half a second. So this is not a real change in temperature, but symptomatic of an electric or sensor problem.

Replace the temp sensor first. I had a similar issue a few weeks ago with one of my cars. I redid the ground wire and nothing changed, so I changed out to a new temp sensor and everything is good.

Reroute spacer isn't getting consistent ground for the sensor. Add a ground from one of the bolts holding the spacer over to the ground strap by the dipstick or to the top of the intake manifold.

It is a 2 wire sensor, the coolant block does not need to be grounded.