The problem: My car (91 NA coupe) will not idle. It just dies unless I keep my foot on the gas. It also runs really rough below ~2000rpm.

Back story: I had a guy look at it, and he apparently forgot to put the clamps back up on these two hoses:


So when I redlined it (not the first time), they blew off, spewing coolant and steam (I didn't know what it was at the time). Gauges were fine. I stopped and looked to see what happened, noticed the hoses were off, put them back on, and started the car again.

Weird part: Idle was perfect! So I drove it again, this time it started over heating. I had to drive about half a mile nervously watching the temp gauge rise until I could stop. I went and bought 3 gallons of coolant (should've bought 2), and filled it up. It was nearly all gone. I waited a while to make sure it cooled off, and when I started it again, the idle problem was back!

Also, when I drove it home, it seemed to take way too long to warm up, and actually never did. That was about 5 miles. It usually warms up in about 2.

Why would my car only idle with no any coolant?

 

So basically your engine is burning coolant? u fill up but still there is less coolant in it after a drive?

 

No, this was a one-time thing. The coolant hoses blew off (because the clamps were not secured), causing the loss of coolant and the overheating somehow fixed my idle. I reattached the hoses correctly and filled it up with coolant and my idle problem returned.

 

So when u had the correct pressure and supply of coolant the engine didnt run well but when u didnt have the system correctly hooked up it did? It Cant be a freak case of coolant leak into the chambers cant it :S?

 

My problem is my idle, not coolant loss.

I had an idle problem. Two coolant hoses came off. Lots and lots of coolant spilled out. Idle problem was gone, but obviously the car overheated. I reattached the hoses and put in coolant. The idle problem came back.

I'm hoping this strange set of circumstances may help someone diagnose my idle problem.

Yes. The idle problem is related to either coolant or possibly just heat.

 

oops double post

 

Well if your engine is leaking coolant into the chambers u will have an erratic idle, so it could be connected, not that im saying it is

 

The hoses were hooked up correctly when the idle was "fixed", but the car had no coolant so it overheated. When I filled it back up with coolant and let the car cool down, the idle problem came back.

oh. that would not be good.

 

well, first priority in ANY case with the a rotary thats lacking power or is running unevenly is doing a compression check

 

Try turning the AC ON when at idle. What does it do?

 

AC? What's that? (Previous owner removed the belt, and I never bothered to put it back on)

 

I've read before about s5 na's having an idle issue and turning on the ac would casue it to idle fine. Which tells you somehting specific is wrong but I don't know what.


I'm thinking you may have overheated the engine and damaged something.

 

is nobody going to point this guy to "thermo wax removal"?.That is what it looks like the Previous owner of the OP's car did.
You need a Diagram showing Coolant flow.I would also check the Thermostat and the hoses.

 

Air Condition

 

So should I replace the thermowax? What else is heat related that would also affect the idle?

btw, it heats up fine now. I guess the new coolant just needed to be "broken in" or whatever.

 

I've noticed the car runs worse the colder the weather is. When it's really cold (20s-30s), it dies quicker and runs worse. It's also started bucking/jerking under slight deceleration. (If I let off the gas a lot/completely, no problem. )

Can I run the car with the intake tube disconnected from the throttle body so I can make sure the airflow meter isn't a problem and also check the throttle valves.

Could it be the thermostat? I'm going to check that and the thermo valve (that the same thing as the thermowax?) according to the FSM.

I guess it is possible that something is messed up just right, so that when it gets hot, it expands and closes up a vaccum leak or something.

The mechanic guy I took it to (the guy who left the hoses improperly connected and doesn't know much about rotaries) said it was the vdi valve and the tps, which doesn't make much sense to me. I replaced the tps (with a used one) a while ago, and the car ran ok for a while. He probably just checked error codes.

I am really stumped. Any help is greatly appreciated.

 

is the tps adjusted correctly?

 

I think so. I followed the FSM, but I don't have a checker lamp so I checked the resistance. I have not checked it for a while, though.

 

I recorded a video of my car to post on youtube for diagnosis, but I may not need to do that. ...

I took my car to a trusted local mechanic. He says rotaries are "stupid", but he still agreed to have a look. (He said his son used to have one.) He sprayed carb cleaner around the engine checking for vacuum leaks while I held the throttle steady. He found a huge leak. It's apparently this gasket:


I ordered one from mazdatrix as soon as I got home. I was excited until today, when I took the old one off. It looks fine. no rips or cracks or anything. I expected it to be torn or something. So I'm not sure a new gasket will help. I won't find out until it arrives, probably next week.

...

While taking off the the old gasket, I noticed that the VDI actuator rod (green) was not connected to the valve rod thingy (red/yellow).


I loosened the actuator and hooked up the rod, but noticed it was at an odd angle. I then looked at the manual closer and realized that the valve rod was backwards!

I don't understand how my car was functional like that.

Anyway, I got it hooked up correctly, but I didn't have time to put everything back together and take a test drive. When the weather here is nice again, I'll test it with the old gasket first, in case I have to take it all apart again.

One question: Should I use that gooey black gasket adhesive stuff or not?

oh, and I stripped out a bolt hole (blue spot in pic above) However, along with the other bolt, it seems tight enough, so I think it's ok. I also found a bolt just sitting on the engine block, stuck under a fuel line (i think). No idea where it came from, but it's obviously not vital since it's been that way for at least a few months.

 

gasket did not fix it. what a surprise.

Also, the vdi valve and/or actuator is not working. It doesn't move at any rpm.

Edit: I just learned that the VDI only operates under load above 5200 rpm, so it's not the problem.

 

After removing my lim, I'm a bit disappointed and becoming more skeptical that my vac leak is a hose. All my hoses look fine. They don't even look that old. I pretty sure they've been replaced before. I'll replace 'em anyway since I already have the new hose. I'm also putting on a new vdi valve gasket (it just had that rtv stuff on it - no gasket). If that doesn't fix anything, I'll probably replace the whole vdi manifold.

How likely is it that there's a leak in any of those metal vacuum lines?

 

I may also replace the primary injector grommets, just to be sure.

(sorry to bump, but i waited too long to edit. )

 

FIXED!!!



After a year of no idle and running rough, my car finally works!!!



The problem: missing primary injector air bleed sockets and mixing plates, causing a massive leak.
more details in this post

I think I may need a new TPS (the idle bounces a little bit), but other than that everything is perfect.

 

Glad to hear it... Those air bleed pieces are extremely important, glad you finally found it.