Another overheating thread

[twobjshelbys]

I've been watching this with great interest.

Overheating Issues, Perplexed

but I think this is not the same problem.

My CSX with a Roush 427IR has very erratic temperature. My problem occurs a medium amount of time after starting. I thought it was an air bubble because the top of thermostat housing to radiator link was above the radiator fill cap. I'd purge and fill and then it would lock up again. Erratic temperature swings from 78C to 110C, with some brief spikes to 120C. So I put on the expansion tank to get the high point above the engine. This morning when I check the tank it was dry - after having filled it almost to the top - and on the way to work it spiked up again. By the time I got to work the oil temp was about 100c and the water temp was down to 60c. That doesn't make any sense.

It did the same thing on the way home - although oil temp only got up to about 90 yet the water went up to about 90 and then back to about 80 (it's a 160F thermo) but was back around 60. I even turned off the aux fan - no radical temp change.

When I parked the car and let it sit for about 15 minutes both oil and temp gauges were at about 85C.

Tonight my best friend and mechanic checked the overflow tank and it was a little below the baffle plate but still not as much as I had put in. I'll check it in the morning when it is "cold" which will probably be about 65F ambient.

One of the things I've noticed are wildly erratic temp swings. It will move from 110C to 80C in several seconds. It makes me wonder about the temp gauge, but I have no way of verifying this.

I would expect the oil and water temps to be pretty close and change on driving conditions. The oil temp now will stay about 90 and go to 110, but the water jumps all over the place.

I'd thought about the lower radiator hose collapsing but the elbows are hard tubing and of course I can't see them when tese events happen.

When it isn't moving it is pretty good - gets to about 90 max and stays between 80-90.

Expansion tank shows the thermo opening. I drilled 1 hole into it - that what was recommended - but I see people putting up to 4.

Any ideas?

[olddog]

If the gauge and sender are working propperly, I would first make sure I burped this baby. I jacked the front wheels up a good foot off the ground and still had air pockets. You could hear it gurgle when you shut it off. Out of my desire to overkill this, I backed it over a bank that dropps about 8 feet in 20 distance. I got about another 3/4 gallon into it. Problem solved.

The fact that you keep adding fluid says it is either not full, boiling off, or leaking somewhere. First make sure it has no air pockets (I like the bank). It wouldn't hurt to check the oil and make sure no water is getting into the oil. Watch the exhaust to make sure no steam is coming out. You can try running it with the cap off to see if air bubbles is getting into the water from a compression leak, but I am getting ahead of myself. First step make sure there is no air pocket in the engine.

[elmariachi]

Based on my experience with that other thread:

- You have to rule the gauge and the t-stat out first. You can buy a little handheld IR gun from Summit Racing, several models under $50. That will tell you a lot about temps if you are doubting your gauge and don't have a backup.

- Remove the t-stat, jack her up and burp her real good. Then start it with the cap off and the tank 1/2 full and see if you have consistent flow. Then put the cap on and see if you have the same temp swings.

At this point you'll have a good handle on your gauge, the water pump flow and whether or not the temp swings are continuing. THEN you can start chasing whether its the engine or the cooling system or both.

And by the way, a buddy showed me his nifty little combustion gas tester AFTER I paid the local radiator shop $50 to test mine. He has been using it for a couple years with good results: http://www.lislecorp.com/tool_detail.cfm?detail=1332

[twobjshelbys]

Originally Posted by elmariachi

Based on my experience with that other thread:

- You have to rule the gauge and the t-stat out first. You can buy a little handheld IR gun from Summit Racing, several models under $50. That will tell you a lot about temps if you are doubting your gauge and don't have a backup.

- Remove the t-stat, jack her up and burp her real good. Then start it with the cap off and the tank 1/2 full and see if you have consistent flow. Then put the cap on and see if you have the same temp swings.

At this point you'll have a good handle on your gauge, the water pump flow and whether or not the temp swings are continuing. THEN you can start chasing whether its the engine or the cooling system or both.

And by the way, a buddy showed me his nifty little combustion gas tester AFTER I paid the local radiator shop $50 to test mine. He has been using it for a couple years with good results: http://www.lislecorp.com/tool_detail.cfm?detail=1332

Thanks! I've followed your saga hoping it was the same, but it doesn't seem so.

What I really need to know is what are the expected steady state temps? I reckon - only logic telling me - that the oil and water temps should track pretty closely - +-10*C or so, but that one shouldn't be really high and the other really low. The gauges tell me I'm cooling the engine with the oil cooler. I am going to find one of the IR temp spotters - it will tell me alot.

I was disheartened when I pulled the cap on the tank this morning and it was empty, but slightly encouraged when it was half full after sitting for about 1.5 hours. The cap was too warm to take off without a glove (or a cloth) but didn't steam when I opened it. Can there be that much expansion/contraction?

[twobjshelbys]

Originally Posted by olddog

If the gauge and sender are working propperly, I would first make sure I burped this baby. I jacked the front wheels up a good foot off the ground and still had air pockets. You could hear it gurgle when you shut it off. Out of my desire to overkill this, I backed it over a bank that dropps about 8 feet in 20 distance. I got about another 3/4 gallon into it. Problem solved.

The fact that you keep adding fluid says it is either not full, boiling off, or leaking somewhere. First make sure it has no air pockets (I like the bank). It wouldn't hurt to check the oil and make sure no water is getting into the oil. Watch the exhaust to make sure no steam is coming out. You can try running it with the cap off to see if air bubbles is getting into the water from a compression leak, but I am getting ahead of myself. First step make sure there is no air pocket in the engine.

I have no gurgling. Oil is clean - nothing that looks like water, although I think I still have some fuel from my previous fuel injection system problem but that's nothing a tune won't fix.

I don't think there are leaks (small bubbles) but there definitely was a big one and the tank helped alot.

I'm not sure of the water temp gauge. It looks like the connection is a tap tube that picks off some water. I've thought about replacing that tube but getting under the dash to make the connection to a gauge requires a midget that can stand on his head.

[sparks]

I would first of all put a 180 deg balanced t-stat in. Get rid of that 160 deg t-stat. Start w/the easy fix first.

[vector1]

with otc thermostat you'll see temp swings, my experience. i changed to stewart or robert shaw high flow unit, no more temp swings. some come with 3 3/16" drilled holes, what i use. i use 160 unit, see 180 temps max so far.

[twobjshelbys]

Should I expect the expansion tank to be dry (empty) when the engine is cool? I filled it up to just above the baffle plate and got some blow off at the cap as things heated up. I expected it to do that to some degree and that it would find steady state. It was empty yesterday morning, but I left it alone and checked it again when it was cool enough to open the cap, and there was a little (but not alot) in the tank. This morning at cold it's empty again.

Does it do any good to put an additional recovery (like the bag I've seen on Mustangs) on the cap tube?

[elmariachi]

Once you get all the air out, you should be able to maintain a consistent coolant level in the tank. If you look on my thread I posted a video of my tank gurgling like a volcano, at idle spitting out hot coolant. This is what happened immediately after I installed a hi-flow t-stat. So I went back to a $6 NAPA OEM stat with 4 hols drilled. I also had to bump up to a 16 lb. radiator cap to stop the boil-overs at idle. Of course, once I solve the hot idle issue (hopefully with this new fan) I would bet I can go back to a stock 14 lb pressure cap.

You can install a recovery system, but it needs to be closed, with the right cap capable of allowing fluid to move in and out of the coolant recovery tank. But you need to figure out if the coolant loss is all attributed to boiling over, or if you are losing it somewhere else.

I'd want to stop the boilover first (and the source of the high temps causing it) before addressing a recovery tank.

[vector1]

if your cap is on the pressure side (between block outlet and radiator core) the pressure will be more and you will see fluctuations in your water level if you don't have a can to catch the overflow and send it back when it cools, put on a 20+ cap and a hi flow thermostat and a catch can and your problems will go away. don't believe your expansion tank is the same as a recovery tank.

[twobjshelbys]

Originally Posted by vector1

if your cap is on the pressure side (between block outlet and radiator core) the pressure will be more and you will see fluctuations in your water level if you don't have a can to catch the overflow and send it back when it cools, put on a 20+ cap and a hi flow thermostat and a catch can and your problems will go away. don't believe your expansion tank is the same as a recovery tank.

Here is a picture of the setup



I think the cap is a 16lb. So the catch can gets the tube that is out of the top of the cap? Right now it's just draining to the ground. I've seen the catch setups used on mustangs that are a plastic bladder - that could fit in there nicely. How much capacity should I get?

Also, where is a source of a high-flow thermo? I've been to all of the usual places and all they have are the cheapo ones.

And while we're at it, it has/had a 160*F. Is 180*F better? (This is a Roush 427IR, basically a 351W).

Thanks!
Tony

[vector1]

my 16 lb cap would lose fluid, i would use a 20+ lb. cap, your pressure side can develop around 15 lbs of pressure from the flow by itself, once you break the seal of the cap, you lose cooling capacity and must wait for the system to cool before you regain it.

the hose from under the cap should be routed to the catch can, 1 qt. should work. the other hose goes to what, i am not familiar with the fe but maybe should be plugged.


the thermostat can be had from speedway motors or stewart components, look for the high flow unit, i would start with 160.

nice rig, bet it scoots.

make sure the hose under the cap is routed to the bottom of the catch can so it can recoup the water if any is lost, catch can must have breather hole also.

[olddog]

I would get an over flow tank (catch can) set up properly. Then see if you are still loosing water. If it is indeed loosing it, you will have to figure out where it is going.

[twobjshelbys]

Originally Posted by olddog

I would get an over flow tank (catch can) set up properly. Then see if you are still loosing water. If it is indeed loosing it, you will have to figure out where it is going.

But isn't that what the surge tank is supposed to accomplish? Do I really need both of these?

Last night I filled the tank and let the engine warm up until the thermostat opened up and started to flow. It burped twice and then flowed nicely. I put the cap on and went for a drive. It still spiked to over 110C and stayed there for a while. By that time the oil temp was at 95C and then the water temp went down to 65C and seemed to stick. I was running about 4500RPM so I wonder if the lower hoses are collapsing?

Tonight I might just try to duct tape a liter coke bottle to the inside somewhere and see if that helps.

I also turned off the inside fan and it didn't change temp.

I'm going to leave it with my mechanic on Friday so he can drive it to see what it does, fully drain the system and pressure/flow test the radiator.

[elmariachi]

Originally Posted by twobjshelbys

But isn't that what the surge tank is supposed to accomplish?

No, that external tank was a feature of the FE engine platform and was designed to be an extension of the radiator and a place to add coolant. Yours was likely added to help it look more CSX-original. Is it needed? It is if its the only place you can put a radiator cap. You have it installed, its functional, it looks cool and I doubt it has anything to do with your problem. But if you are bent on catching the overflow, then you are also going to have to add a recovery system if you don't like it draining out on the ground and having to constantly refill it. But I would suggest you solve the overheating issue first and then decide about the recovery system. If you can make your system stop overheating, it won't boil over and you won't need a recovery system.

This is provided that you think that the coolant loss you are seeing is just the boil over and not a leak elsewhere.

[elmariachi]

I just re-read your original post and see that you added the expansion tank. Maybe other Roush guys can comment on your setup to confirm you have it right.

[twobjshelbys]

Originally Posted by twobjshelbys

But isn't that what the surge tank is supposed to accomplish? Do I really need both of these?

Last night I filled the tank and let the engine warm up until the thermostat opened up and started to flow. It burped twice and then flowed nicely. I put the cap on and went for a drive. It still spiked to over 110C and stayed there for a while. By that time the oil temp was at 95C and then the water temp went down to 65C and seemed to stick. I was running about 4500RPM so I wonder if the lower hoses are collapsing?

Tonight I might just try to duct tape a liter coke bottle to the inside somewhere and see if that helps.

I also turned off the inside fan and it didn't change temp.

I'm going to leave it with my mechanic on Friday so he can drive it to see what it does, fully drain the system and pressure/flow test the radiator.

I did the coke bottle experiment:





And it shows some promise. There might have been another air bubble too. Tonight while I was putting in the coke bottle overflow tank I started it up until the the thermo opened, filled up the expansion tank to the top and put a 20lb cap on. I had put a little water (I marked it with a sharpie) in the bottom of the coke bottle so the tube would be a vacuum. Then took it for a drive.

This time it was behaving much better. I suspect there was still an air bubble yesterday and that today. Water temp only briefly (when I was going 65 in 3rd at about 4500RPM) went above 100. Oil temp was also about 110. The temp did go up but only when I was running high RPM. After I slowed down it went down slowly. By the time I was going street speed back in town it was back at about 80C, and the oil temp went back down too.

I don't have a good fast road - too many cops - so I was running in 3rd at 65 (speed limit is 60 but too many slow cars) at 4000RPM during the load test.

I still see some fluid weeping in the upper right corner of the radiator and I'm Positive it is not from the vent pipe. I'll drive it for a few days and see what happens but I'm still going to get the radiator pressure tested.

Tomorrow morning I'll check the cool engine against the sharpie marker. If this is all right I'll try to find one of the bladder type expansion tanks (like they used on the old Mustangs) and tuck it in the passenger side above the pipes.

[elmariachi]

So you are saying either:

1. The 10-12 ounces of coolant that is leaving your system (and now going in that plastic bottle) is enough to make your temperatures fluctuate to such erratic levels; or

2. You had more air in the system that needed to be burped out; or

3. You needed more radiator cap; or

4. Some combination of the above.

I am surprised anything can get past that 20 lb cap, but maybe that's common for that engine platform.

[vector1]

i think one of the high flow thermostats will help with your temp fluctuations and your oil temp will likely run 20-30f hotter then your coolant.

sounds like your making progress.

[twobjshelbys]

Originally Posted by elmariachi

So you are saying either:

1. The 10-12 ounces of coolant that is leaving your system (and now going in that plastic bottle) is enough to make your temperatures fluctuate to such erratic levels; or

2. You had more air in the system that needed to be burped out; or

3. You needed more radiator cap; or

4. Some combination of the above.

I am surprised anything can get past that 20 lb cap, but maybe that's common for that engine platform.

1. It's a 1 liter bottle so it is shifting out more like a quart.

2. I might have had more air. When I started the engine to warm it up the expansion tank was empty. I let the engine warm up again and slowly added water (no more experimenting with 50/50) until I could see it flowing again but then it go too turbulent so I turned the engine off and then filled and capped the tank. There was a little water in the bottle to provide a vacuum. Then I took it for the drive.

3. I did put a 20lb cap on an yes, it still blew some by.

This morning when it was cool it had not pulled back into the expansion tank as I expected. That could be because the cap is just too much. The bottle was about 1/2 full and the tank was just below the baffle line so there was no loss in the system. I may try the 16lb cap again to see if it'll pull back.

I'll go on the search for a bladder-type overflow I guess.

Hard to believe the engine produces that much heat but I guess it does.