Hi,

Last friday, i've done my second oil change since i've had my engine rebuilt. roughly 3000 miles since the rebuild.

When i've had the rebuild done, the mechanic told me that i had blown one of the housing gasket, which lead to the oem radiator to crack due to the over pressure.

So in addition to fix the engine, he installed a big aluminium radiator (twin row) that he had modify first.

So after I changed my oil, I was surprise to see that the overflow tank was empty.
I filled it up to the full mark.
On a sunday, i had to take the car for a small road trip of around 150miles.

when i arrived, i let the car cool for around 1h30 or more and checked it.
the overflow tank was empty again.
i've check the radiator itself, it was almost full. i started the car for maybe 30s - 1minute and didn't saw any bubble so i put back the cap and drove home.

since then i haven't that the opportunity to check it.

what could it be ?
-leaking overflow tank (there is greasy stain behind it, but it could come from me doing a mess last time i filled it)
-radiator leak (one of the welding has been suspicious as there have been some coolant on it but it seams to have not reappeared since last time i cleaned it)
-punched line or loose collar between the tank, radiator, engine (but i don't see any stain...)
-blown gasket (not again....)
-anything else ?

PS : I forgot to precise that the original water temp gauge is always good (no over heat) for what it's worth.
also i have an aftermarket oil temp, during my last trip it stayed between 70°C and 80°C (highway, cold outside)

 

You can check multiple theories at once by doing this:

-Remove the overflow tube from the tank and put it in another container filled with water so the tube is submerged. The level is not relevant, only enough to ensure the tube end is always covered.
This eliminates your OEM tank from the equation.
BTW, that rubber hose from the rad overflow to the tank is critical to the whole process...you may as well replace it now and make sure it seals well at both ends and doesn't collapse under vacuum.

-Disable your efan or remove the thermoclutch...you don't want/need any external cooling, it just slows things down.

-Car stationary and hood open, start car and run it up to operating temp.
Another aside...your stock temp gauge is useless. S5 units are barely more than warning lights and the S4, while slightly more responsive, still doesn't tell you what the actual temp is and is very dampened compared to an aftermarket gauge.

-Turn off engine and watch.
While the engine runs, the coolant system is not under a lot of pressure but when the engine stops and coolant stops flowing, it heats up and expands. Ultimately it reaches 13psi and the radiator cap opens and the excess coolant is dumped into your overflow. As the engine cools, a vacuum is created in the now partially empty loop and the coolant is sucked back into the radiator from the overflow tank. It's basically like the cooling system was breathing.

This is also the time to look at your suspected bad weld. It could be a pinhole leak that is only "active" during that brief time the coolant is overpressurizing before the rad cap opens. Before you start this test thoroughly dry/clean the suspected area and dust a little baby powder on it...if it later leaks, you'll know.

That pretty much covers all the external possibilities...except the rad cap itself, which is cheap enough to replace now as well.

You'll see any radiator leak and the coolant burping into the overflow pretty quickly after shutting down the engine. It doesn't take long for the hot engine to start cooking the now stationary coolant.
The other part of the cycle, where the coolant is sucked back into the rad, can take a few hours depending on ambient temps and how quickly the engine cools down.

It's a start.

 

sorry, I'm never always sure to understand correctly but:
starting the engine, nothing should appear.
while the coolant heat up, it expand and raise the pressure in the system.

at this point instead of applying pressure on the radiator cap, the pressure/extra liquid would escape through the defect weld bead or leak. This leak would result in less pressure in the system and then shutting of the leak.

then the car could run fine as it has the normal amount of liquid at nominal temperature.

when the car cools down with the engine off, the vacuum works normaly and suck the liquid off the expansion tank.

would it makes some sense ?

on a oem rx7, how much raise should i expect in the expansion tank normally ? (i have a slightly bigger radiator than OEM, so more liquid, so i should have a bigger raise in liquid level is suppose if everything is ok)

also last time i took off the tank to get a tool stuck behind, there was a dead butterfly in it, a big one... no idea how it ended up here...

i'll try to do your check this weekend

 

While the engine is running you may not see any leak at the radiator and certainly, no coolant flowing to the overflow. It's not till you shut it down that the pressure really starts to rise and that's when things will happen.

 

ok, so basically the coolant heat up more when you stop the engine than when it's running ? is it because the waterpump is stopped and then the coolant becomes stationnary ?

(sorry for the stupid questions)

 

It's not a "stupid" question, so go easy on yourself.

Think about it though...
The radiator cap opens at 13psi so the normal pressure while running must be lower or coolant would just pump out the overflow until the loop was empty.
Yeah, it's the stoppage of the waterpump that will cause the fluid to heat/expand past the 13psi limit of the cap and that's when stuff happens.

When you do the test, get it good and hot...195° would be about right before shutting down.
Give the rad a chance to expand/contract, see if it exposes the leak.

 

i thought the operating temperature was enough for the liquid to expand, so the excess of liquid would be stored in the tank during driving and when you stop it would be suck back in order to avoid having air in the system.

I was almost right lol.

by the way, you talked about disconnecting the thermoclutch. I suppose it's not as easy as a relay or something, it's something on the end of the main shaft ?

 

The thermoclutch is the mechanical fan bolted to the waterpump snout.
Four bolts and it's off.
If you have an electric fan instead, just unplug it.

 

ah ok, i thought you could "disconnect" the thermo clutch from the main shaft....

I'll see how accessible it is, i might just leave the engine run a little longer lol...

 

It's just 4 10mm nuts that hold it on, use a wrench. While you "can" take it off with the fan shroud on, it is way easier if the fan shroud is not in the way.

The FSM says take the battery and tray out to get the fan shroud off, but I have never done that, just takes a little finesse.

 

Finesse ? I might have heard this a long time ago... Before breaking something....

With the thicker radiator, the shroud is closer to the fan, so to access the nuts, I will have to remove it... With a crowbar... Or finesse, I will have to chose.

 

You can leave the fan on if it's a real PITA, it'll just take a lot longer to get the engine hot.

You could cover the outside of the rad with cardboard to reduce airflow...that might help.

 

With fineese I mean it gets a little tight to get to the bolts on the driver side of the shroud is all.

Just take the damn shroud off.

 

lol yes i will remove the shroud. By the way, is it efficient ? in my opinion, it's the uggliest part of the engine bay...

 

Yes, the stock fan/shroud are efficient...at least in the sense of good airflow.
It is horribly inefficient in that it only crudely responds to coolant temp changes.

 

Just to add onto what clokker said, if you remove the shroud the stock fan wont draw **** for air through the radiator which will cause you problems.

If you want to ditch the shroud then get a good efan.

 

that's a mod i've been thinking of for a long time, but i don't have the courage to get into it...

 

Efans are easy...look here.

 

not bad at all... ford taurus fans must be tricky to find around here but i like the idea, moreover that there is already a unused thread on my radiator that would match the thermoswitch i believe... might be an easy summer project...

concerning the oem fan, did you just unbolted it or did you also removed the thermoclutch (if possible)

 

You unbolt it from the water pump pulley, clutch stays with it.

 

ok good, thanks

 

Ok i did the test and it seems it the little tube between the radiator and the tank that is faulty or the tank cap... not really a big problem apparently

 

Good for you.