My car doesn't warm up. ~5500miles on the rebuild. It started this when it got cold over winter.

My first thought was thermostat. Pulled the existing one out and bought a new one from Mazda. Stuck them both in a pot of water and turned on the heat. Both opened at the exact same time, 170-180ish I believe is when they started opening, fully open before 200.

Installed t-stat housing without t-stat and started the engine to verify that the pump was actually pumping the water. It was.

Installed the new t-stat (in the correct position), flushed and refilled the system with distilled water. After driving home from work, I stick my finger in the upper radiator hose with only minor discomfort, it's like hot hot water from the faucet. It's been like this for a couple months. I can pull off either coolant cap with no pressure.

The car runs top-notch, it's just that the hot-idle cam never comes off and it doesn't seem the coolant gets hot enough to open the thermostat.

It's all stock; ECU, intake, exhaust, cooling system....everything except the streetport. Money's been going into my bike, so the car's exhaust and fuel management is still on the to-do list.

Ideas? Should I be worried about it?

 

Pull the TB and remove thermowax. The coolant passages in the TB and thermowax get clogged easily, and you will have to dig all the crap out of them. This prevents hot coolant from ever getting to the thermowax, thus never releasing the fast idle cam.

Out of curiosity, what does the temperature gauge show?

 

My temperature gauge is inaccurate since I swapped dash's. At fully warm (before this problem started), the needle would sit on or just above the top of the double lines at the bottom of the gauge...if that makes any sense. Now it doesn't move at all.

BUT THEN ... on the drive home today after making this thread, it warmed up just freaking fine. Wtf??

 

So there are 2 problems here that I see. Number one, you have no idea what the temperature of the engine is. Even the old S4 gauge that goes from C------>H with no real quantified value attached to it is better than nothing. I'm not saying you need a digital coolant temp readout like I do, but you need something at least. Problem number 2, albeit closely related to #1, is that you really don't know whether or not the engine is fully heating up or not. Without knowing the temperature (relative as it may be) you don't know if it's not heating up, or if the issue lies elsewhere. My money still lies on a clogged coolant passage in the TB/thermowax.

 

I know the gauge just fine. The needle is just on the post incorrectly. With my old dash, the gauge would rise to about 1/4-1/3 when fully warm. Swapped the dash and it only rises a tiny bit.

Fast forward 6 months and the needle stopped moving at all. In conjunction with the needle not moving, I can remove the coolant caps with no release in pressure. Also after 15 minute warm up and 30 minutes driving I can stick my finger in the coolant and it's no hotter than what comes out of my kitchen faucet.

The gauge is not a problem, there are other methods of measuring coolant temperature. The fact that the coolant system is not pressurized is a dead giveaway that it is not reaching anywhere near operating temperature.

Removing the thermowax will lower my idle to normal. How would the thermowax prevent the coolant from reaching temperature? I can understand how it would prevent the fast idle cam from releasing. However, if the idle is staying fast, shouldn't the engine warm up faster?

 

This is weird but there's no magic going on...the engine has to be getting hot (after a 15 minute warmup and 30 minute drive), it HAS to.
The weirdness is why your gauge isn't reading properly and why the rad isn't getting hot.
Out of curiosity, does the rad core get warm? Are you sure the radiator is in good shape and not blocked?

IMO, the gauge IS a problem because you have no clue what temp it is displaying.
If the car was acting normally, I'd cut some slack here but it's not, so the first thing is to install a real gauge (VDOs are cheap and reliable) so you have some real numbers to work with.

Next, I'd disable the fan (unplug if electric, remove entirely if mechanical) and run the car.
It might take a while (if sitting in the driveway) but it should eventually open the thermostat and heat up the rad core.
If not, you have a problem with the thermostat, despite being new (and apparently, tested).
Or there's still a ton of air in the system.

Are you getting heat in the cabin?

 

I've never thought to feel the radiator. I imagine it's pretty clear and not clogged, it flowed water through very efficiently when I was flushing the old coolant out. I could always back-flush it just in case.

I get mildly warm air in the cabin, not what I'd consider heat.

Despite the gauge being wrong, I know where it should be when at operating temperature, and it's below that. I should've just said, "it reads cold" and avoided this derailment of the subject. I'll swap back to my old (bad tach) dash that at least has a properly working temp gauge. I could always move the gauge mechanism, but then I'd probably end up where I'm currently at with the needle reading wrong (must remove needle to remove mechanism) and everyone would still be fixating on that.

I'll then drive and see if it's still not coming up to temp (I doubt it will). Then remove fan (mechanical) and try again.

There's no air in the system, I've checked numerous times with various methods.

 

OK then, the first and most important thing to remember is that you DO have a problem.

The engine did not suddenly decide to violate the laws of physics, if it's running, the coolant is getting hot...certainly hot enough to open the thermostat.

That leaves only a couple of places to look and despite your protestations that all is good, something clearly is not.

Unfortunately, you seem to have already checked all the things I'd normally look for, so I have no revelations to solve your issue, I think you need to start at square one and check everything again.

 

All is not good, that's why I'm here. I'm only insisting that the gauge is not the reason the engine is not warming up, nor is the thermowax.

The gauge reads incorrect, known issue. The thermowax worked fine last time the engine reached operating temp, no issue.

After 45 minutes of engine run time, the cooling system should be pressurized with ~190º coolant...it's not, it's around 110º measured with a meat thermometer from the upper radiator hose.

15 minutes of warm up should bring it to operating temp (it has in the past) and most definitely a 30 minute drive afterwards should...it's not.

-A new thermostat (verified working) did not fix it.
-A coolant flush did not fix it.
-Everything appears to flow more than adequately.
-Fixing a gauge won't raise my coolant temperature 80º (measured at the upper rad hose with a thermometer).

-I will fix the gauge issue regardless (at the expense of a tach, waste of time, imo)
-I will remove the fan

Any other suggestions?

 

What are you taking the temp readings from?
If you are relying in that gauge to tell you and you know that it is "screwin up" then you can't really tell anything at all.
Get an aftermarket gauge at try that.At least that can give you a reading that you can depend on.

 

I use the gauge to give me an idea from the drivers seat. Actual measurements are taken via meat thermometer in the upper radiator hose, as explained above.

 

Look I'm not inept at engines or the physics behind them. I've been wrenching since I was little, like 6-ish. Built my own go-kart engines when I was 12 for racing (17hp out of 5hp B&S block). I've been building car and motorcycle engines for about the last 12 years. This is my first rotary that I've successfully rebuilt on my own. Now it has issues, I went through the normal stuff that I know, I'm out of ideas so I'm asking if there's gremlins not normally associated with boingers.

Bad gauge is a known issue so temperature is measured manually, so the bad gauge can be disregarded.

 

If it hasn't been stated already, what does the radiator return hose feel like after the car has been running for a while?

 

The bottom hose? No clue, hadn't thought to touch it since the upper hose was only mildly warm.

Added to list of things to check, thank you.

 

If your oil cooler has had the thermostat bypassed, it could be cooling the oil all the time. That would decrease the cooling load on the coolant system. I seriously doubt it could be enough to keep down it at 110 though.

As stated before, get a real gauge that touches the coolant. Your meat thermometer on the surface of the hose isn't an accurate reading. It's something, but not good enough in this case.
Even IR temp guns on metal housings read low compared to supposed thermostat temps.

There is no way that the engine can run well without continually heating the coolant up to thermostat temps. Rotary or piston.

 

Where's that thermostat for me to check? That spring-thingy in the E-shaft that people put washers on, was replaced with a solid piece from Atkins rebuild kit. Though I thought that was oil supply to the rotors.

No, I remove the cap and put the thermometer directly in the coolant.

 

It's in the oil cooler itself, between the inlet and outlet. It would be pretty messy to check it, though. I should clarify. Removing the thermostat will allow oil to bypass the cooler, making it hotter. It would have to be modified so the thermostat is 'closed' and forcing oil through the cooler. The e-shaft one only helps heat the oil during warmup.

Ah, thanks for clarifying. Maybe run it without a thermostat and see if it will heat the entire system? I know you tested them, but just taking another variable out of the equation.

 

Perhaps I'll leave that for last, then.


That, with clokker's suggestion of removing the fan, should definitely bring the temp up. Hmm...I'll give it a go either later today or this weekend. Planning a wedding, building a motorcycle and out-of-state friends in town.

Thank you for the things to check/modify/test. I'll report back.

 

It ain't the oil cooler thermostat.

 

Would having no undertray contribute to this? The car came without one, though my understanding is that the undertray aids in cooling and running without one would cause the engine to run warmer than usual.

No bubbles in coolant, no dissappearing coolant, exhaust smells normal and does not smoke; I do not believe it's overheated and blown a coolant seal. That was the reason for the rebuild so I know the symptoms of which it currently has none.

 

I've never found the undertray to make much difference either way (I know, HERESY!) but you're correct...if anything, not having the tray would make it run hotter.

So, to recap:
You have a perfect running car that never reaches operating temp...ever.
Your gauge reads lower than normal, heat output in the cabin is poor and the upper rad hose does not get hot.

It's time we consider magic.

 

Could cold weather be the cause?

I changed the gauge cluster, so the gauge reads correct. Just as I'd been saying, it read cold, just barely above the double lines.

Now as the weather has warmed up, the car fully warms up as well.

Magic, I guess.

Though now there's a new issue, in a new thread.