OK, everytime I shut the car off, let it cool, and start it again i need to add coolant (buzzer goes off).

I know the coolant is being forced out of the hole at the top of the resevoir. I thought that it was either that the system wasn't sucking the coolant back in from the overflow when it cooled or it was a dreaded blown o-ring . Well I drove the whole way to RX7Store (roughly 300miles) and the coolant light never came on. But then when I pull in, shut it off, and restart it later to leave it needs another couple cups of fluid. Then the whole way home is fine.

Changed the plugs, they weren't wet, no sweet smell or smoke.

I have a pettit AST, replaced the line going from the AST to the overflow, filler cap (greddy unit), and AST cap (NAPA part). Now some may say that I should use an OEM AST cap, but I tested the cap and it holds the 13psi it's supposed to.

I also pressure checked the system. The tester said to let it sit for 2 minutes and check. It moved the width of the needle and that's it. I am going to go get the tester back from autozone and let it sit overnight as I have read in other posts, but does anyone else have any ideas?

Does anyone have a coolant hose diagram? What hoses will affect the vacuum? Some have replaced their overflow bottles but I don't see how a crack in the overflow bottle would affect the vacuum.

Any help would be much appreciated, the car is running great, but I can't take it out because of this stupid problem.

 

If you do not see any signs of coolant under the car other than whats comming out of your overflow. Most likely you have a failing coolant seal.
Let the car cool off, remove the cap at the filler neck (not the ast), start the car let it idle till operating temp. If you see a steady steam of very small bubbles and or coolant backing out of the filler neck get prepared to do a rebuild.

 

I highly doubt it's the o-ring, as stated above there is no smoke, no coolant on the plugs, no sweet smell at startup, and I drove it 300+ miles with no loss of coolant.

I believe the problem is occuring when the coolant cools and contracts.

 

An o-ring failure won't necessarily leak coolant into the combustion chambers. In my case the leak was in the other direction and exhaust escaped into the coolant system which overpressurized it and forced the coolant into the overflow tank. Do you have black goo on the overflow tank dipstick?

 

ok but if this were the case, the longer it was ran the more evident the problem would be. I could run it all day long and never lose the coolant, but when I "heat cycle" the car, it'll lose it right away. I could leave it parked and let it idle up to temp, shut it off, then when it's cool it's a couple cups low.

Someone please help, and know that I HAVE searched, and HAVE narrowed the problem down. I just don't know the solution.


Thanks for the suggestions though.

 

4CN,

Your engine behaved EXACTLY like my did. You should really check for bubbles in the coolant by putting one of those oversize prestone funnels in the filler cap neck (not the ast cap) and bring the car up to temperature at idle and look for a continuous stream of bubbles. My coolant looked like it was at full boil at 50C. My car still ran ok because the exhaust gasses could escape through the overflow tank as long as the coolant system wasn't pressurized but when I let the car cool off and restarted the buzzer always went off. Hopefully you have a bad coolant hose somewhere sucking in air when your car cools off but you should really do the bubble test.

 

Take it to a qualified emissions repair shop and have them check for excessive hydrocarbons present in the cooling system (exhaust leaking into cooling system). There are about 13 coolant/heater hoses on this car, it's still possible it can be one of them.

 

I'll check that first thing tomorrow (fingers crossed), but if exhaust gases were getting into my coolant:

A.) I'd see some dirty looking coolant (mine is nice and green)
B.) constant driving for 300miles should make this type of leak pretty apparent


*also what type of bubbles should I be looking for? and the proper procedure is to put the funnel in the filler neck, put some fluid in it, start the car and let it get up to temp? when at normal operating temp would I see lots of apparent bubbles if I have a bad o-ring?


Another thing, when I compression test the system, if it does in fact leak down, this really doesn't narrow the problem down any further since it could be a line or the o-ring correct?


BTW: I put the tester on, and ran the car up to temp watching for the pressure to build rapidly (to signify combustion gases entering the coolant), it didn't. Nor did the needle "bounce" which would be a sign of a smaller comb. gas leak.

 

anyone have a diagram showing them all?

 

If exhaust is leaking in to the coolant system what is the damage that it can cause other than having to fill up your coolant? I'm having a small coolant problem and you guys just scared me to death that it might be a coolant seal. I just got done rebuilding the engine.

 

BTW: What rating cap on the AST? (not that this will mean anything but just curiosity)

 

How many miles since last rebuild?

 

Air or exhaust in the cooling system will cause the engine to run hotter because air does not transfer heat nearly as well as coolant and the water pump isn't designed to pump air. Hot spots caused by air or exhaust in the coolant system only compound the problem of a failing o-ring. Pressure testing the cooling system overnight is also a good way of testing for a bad o-ring. If there's no pressure in the morning and coolant pours out the spark plug hole then you have a pretty obvious problem.

 

The cap on the AST is 13lb. The mileage on the engine is unknown but really should not be the basis for diagnosis anyway. I've gathered more meaningful information which seems to be being overlooked. I think some people read "overflowing coolant" and instantly say "coolant seal" without reading the symptoms.

I've seen that this occurs alot in previous coolant overflow threads I've read.


*I will pressure check and champagne test tomorrow. If pressure is lost but there are no bubbles, then can we finally count out the coolant seals being bad, and move onto the coolant lines?


ANYONE HAVE A COOLANT FLOW DIAGRAM?

 

I haven't seen any diagrams in the factory manual or any site. When I replaced my engine I replaced every one. I can get you part numbers of all of them if need be or you can go to Mazda and have them print out some schematics. I'm not sure if there is a "common" hose that leaks the most other than the obvious upper and lower, but the long bypass hose under the radiator is one to look at and the supply at the turbo. under the intake are a couple also, those get a lot of heat. Another question,, heater core? You have no coolant smell at all? even after you shut it off hot or run the heater??

 

nope, no coolant smell

 

From the FAQ:

Cooling system diagram:
http://www.autosportracetech.com/RX-7/coolingsystem.htm



If your system holds pressure overnight, then it sounds to me like the hoses are OK. Check to be sure that your overflow tank level isn't too low: the tube must be immersed in fluid to suck coolant back into the radiator.

What the water/glycol mixture in your coolant? You might be boiling fluid after shutdown, due to heat soak near the turbos. Shut the car off, pop the hood, and listen near the turbo area. If you can hear a bubbling sound, you might want to change your coolant mixture. I've heard that switching to Evans coolant eliminates this completely, due to its high boiling point.

Good luck,
-s-

 

just thought about something else:

How can the system ever pull coolant back from the overflow, does the lower seal on the cap allow fluid to enter the system but only exit once the pressure rating of the cap is exceeded?

 

It can do this because there is a check valve (the little disk in the center of the spring-loaded area) built into every pressure relief cap that lets fluid flow freely (less than 2 psi vacuum) back into the system.

 

4CN Air,

Great thread. I'm following this closely because I've had ALL of the same symptoms and have performed all of the tests that you've described so far. When attaching the pressure tester to my fill cap and running the car, it would SLOWLY climb to about 13 psi before the fans kicked on at 86C, then drop off to around 7-8 psi with the fans on. Throttle cracks to about 4K drop the coolant pressure about 1 psi as well. I've also attached my monster Yellow funnel and run the car. I witness maybe one or two firm round bubbles periodically, but nothing like a steady stream or "constant champagne" flow as many have described. One thing I did try was slightly tweaking the tabs on my fill cap so it would fit more snugly, but that has not cured the problem either.

Something no-one seems to be able to tell me is how much leakage is ACCEPTABLE from the coolant system tester over WHAT period of time. Is the engine supposed to hold the 15 psi for several hours? Overnight? The shop manual just says to pump it up and inspect for pressure loss, but it fails to specify any kind of acceptable range.

Hoses replaced for me so far include (AST to overflow, U/L Radiator, U/L turbo coolant hoses, hose from AST to radiator. Also replaced water pump gasket, installed new thermostat and filled with fresh coolant).

Go-forward plan is to replace small 90 degree coolant line just left of my alternator and coolant hose that comes off my water pump and feeds the heater core as well as the hoses that connect to the firewall which also feed the heater core since mine look original and are ROCK hard. Last hoses that I will replace will be the hose that runs from under the upper IM to the back of the throttle body, and finally the small hose that runs from under the throttle body and connects back to the block.

I've also checked for wet carpet in the car, but can not identify any wet spots. Anyone know if there are coolant hoses inside the car that connect the heater core to the inside firewall? That may also need to be looked at. I figure even if I have an O-ring going out, at least I'll have all new hoses to use for my next install. Good luck, and I'll let you know when I find my problem.

 

I have no clue, I replaced some lines today and it's still doing it.

Pressure tests shows a drop off after 20 minutes, but if I go by the directions included it's fine (it says to wait 2 minutes).

I did the funnel trick to check for bubbles, and got a bunch, but I had just replaced a bunch fo lines and refilled, so a lot of them were probably the car burping.

The bubbles weren't steady though.

What kind of place would be able to check my coolant for combustion gases since really this is the only conclusive test. I live in an area where emissions testing is not required, which is a good thing most of the time, but not when looking for a garage with an emissions machine.

 

Could there be something wrong with the overflow bottle that causes this to happen? I don't see how a crack or leak could really cause any of this since it wouldn't result in a loss of vacuum as long as the end of the plastic "straw" is immersed in fluid.

 

I'm done lying to myself, it's the o-ring. I was excusing the slight "cloud" at startup, but my friend says he def smells anitfreeze in it. Also did the bubble test again, I have bubbles.

This is great, the car needs painted, and now it needs an engine! Guess single turbo is going to have to wait a LONG time.

Well I'm going to give the fd3s.net o-ring fix a try. Cross your fingers.


Thanks for the help guys.

 

Congratulations,

You've gone through the classic stages of o-ring grief.

1) Denial
2) Depression
3) Anger
4) Acceptance



I tried the o-ring fix on fd3s.net and it did work for about 6 months for me which gave me enough time to save some money for a new motor. If you send your core in Ray Crowe at Malloy Mazda will sell you a reman for $2300 or a brand new motor for $3300.

 

If I am getting bubbles at operating temp at the filler neck is there any other explanation than a bad o-ring?