Burp tank 101 - Help from cooling system experts

[btoll]

I have a 351W with a basic "donor" coloing system and stock mustang burp tank. I have a CSR T-filler neck installed at the highest point in the system to help get rid of air bubbles. I have the "overflow" nozzle on the T-filler connected to the base of the burp tank and the top of the radiator has a plug sealing where the cap would normally sit.

After a long drive sitting at lights in traffic the engine was plenty warm and the fan was running and things were well within temperature and appeared to be working fine. But after I pulled in at home and things cooled off I came out again later to find a small puddle of coolant under the car.

It was hard to tell the source, but the T-filler was very damp and I’m guessing that it was seeping out from under the cap. Another key is that when I took the cap off it released a vacuum and things gurgled down in the hoses. So this led me to a few questions:

1. Why was there such a vacuum in the coolant system? Shouldn’t the burp tank connection allowed coolant to fill back up into the T-filler area preventing that vacuum? Because the cap seal is below the overflow connection in the T-filler is there some valve that is supposed to let fluid in for these vacuum situations?

2. What is the proper way to debug this kind of issue?

3. What is the basic functional model for the burp tank connection (Burp tank 101)? This would help me debug the problem.

Thanks,
Bret

[Dan Babb]

Having a vacuum in the cooling system is normal...it's what you want.

Simple explanation:
Pressure builds up in the engine until it gets higher than the cap rating.
The cap opens to release the pressure and air trapped at the high point, should go to the overflow tank.
When the engine cools off and the fluid compresses, it pulls in extra coolant from the overflow into the engine.
After a few cycles, you have no air bubbles left in the engine. You will still have a bit of a sucking sound if you pull the cap.


If you had fluid leaking from the radiator cap, that's a problem. Make sure the hose clamps are tight and that your cap is sealing properly...but since you heard the sucking sound, it probably is fine.

If you had the overflow filled up too high, then you'd have coolant spitting out from the top of that container when the coolant expanded and had nowhere to go but out (because the overflow was full). You want the overflow to be around 1/3rd full when the engine is cool (more or less depending on how much moves to the overflow tank when the engine is hot). It's a trial & error thing. Just make sure the overflow tank isn't empty...otherwise, you're sucking air back into the engine.

[d42davis]

You need to be more specific about what your "basic donor cooling system" consists of. If you have routed your T filler vent to the large plastic tank that straddles the top of the radiator, you have connected the vent to a system pressure point, attempting to send fluid back to the same pressurized system, so the cap is not relieving fluid and system pressure unless it is leaking around the cap. Since you have created a sealed system, with no vent to atmosphere, air can not get back into the system when it cools, and you develop a vacuum. Normally, the cap will relieve the vacuum by drawing air and/or water back into the system thru the overflow tank which is vented to atmosphere.

[Dave M]

Sounds like everything is hooked up correctly. Is the level in the burp tank going up and down with the heat/cool cycles?? Yes there is supposed to be a small valve in the cap to let liquid return to the system. Check to see if that is free/not stuck closed. As said check tighten all your joints and clamps. I had a couple small leaks after a few heat cool cycles when the car was first put on the road. After retightening, with the exception of a water pump problem after 7000+ miles, everything has stayed leak free.

[CraigS]

I wonder if your cap is malfunctioning. Apparantly this is something new after a bunch of no problem miles right? I actually haven't removed my cap until the car sat overnight but I agree there should be very little to no vacuum in the system as the vacuum should be pulling coolant back from the tank as the car cools. How long had the car sat before you removed the cap? I am also a little concerned that you saw dampness at the T-filler. Any overpressure should be directed to the overflow tank and no leaks at the filler. As I write this it makes me think there may be a kink or blockage in the hose from the T-filler to the tank. That would explain the leakage at the filler AND the vacuum in the system.

[btoll]

Thanks everyone for the ideas.

Criag, I was thinking along the same lines. If the hose from the overflow on the T-filler to the burp tank is clogged somehow, that would explain this.

When I said "donor coolant" I mean the setup from a mustang. So the radiator, shroud and the burp tank. The overflow hose used to go from the top of the radiator by the cap do the bottomr of the tank. It now goes from the similar location on the T-filler to the bottom of the tank.

I just rolled over 800 miles, but this was really the first good run in almost a year (paint and then many distractions and other issues). Very early on I had a leak around the thermostat housing, but this location is different. It's clearly not at the burp tank or radiator area, it's right under the T-filler and some issue with the capture tank matches most of the behavior.

In the expansion and contraction phases, I assume that the top of the radiator cap seals as much as the "pressure" spring part of the cap. Otherwise it would compromise the expansion tank system operation. But I assume if the expansion line was cogged or kinked it pressure would build up and the expanding coolant would work its way out past the top part of the cap. Then on contraction it would build up the kind of vacuum I witnesed.

Tonight I'll give it a good inspection and see if I can find a way to blow air through it or something.

Let me know if anyone has any other debug ideas. Thanks. Bret

[hind sight]

What is the psi rateing on the radiator cap?
Is the radiator cap a pressure/vacum cap, similar to what would be on the original donor set up?
The radiator cap has to be able to allow fluid/coolant travel back and forth between the over flow bottle and cooling system depending on temp/pressure.
Bad cap or restriction in hose between T filler and averflow are likely cause.
Does level in overflow tank rise and fall depending on engine temps?
Few checks to do

[btoll]

OK, I've done some checking and I'm afraid it won't be as obvious as I'd like. The overflow tube doesn't seem to be obstructed. I was able to connect a tube to the overflow spot and blow bubbles in the expansion tank.

I have a CSR T filler and matching cap:
Inline Radiator Hose Filler - 1.25 to 1.25 Inches - Cooling Products - CSR
Polished Chrome Radiator Cap - Cooling Products - CSR

The cap description says 14-18 PSI. I took the chrome piece off but the main cap had no writing, so I couldn't confirm anythin. The two rubber rings looked OK and it seemed like it should be working. I can see how on expansion the coolant should ooze past the main seal and be pressured down the overflow tube and then on contraction the center metal piece should pull down with the vacuum to allow coolant to ooze back into the system.

What is the next suggestion for debugging this? Fill it up and let it run until hot to see what happens? See if I can get a replacement cap (that will fit under the chrome piece) and see if that solves the issue? Anything else?

Thanks,
Bret

[d42davis]

What does you overflow tank look like?

[G-Pete]

Originally Posted by d42davis

What does you overflow tank look like?

Originally Posted by btoll

When I said "donor coolant" I mean the setup from a mustang. So the radiator, shroud and the burp tank.

[Jack Wojack]

NAPA has hexagon shaped thermostats. You want a Vented cap…it vents to the recirculating tank. You shouldn’t have fluid escaping from around the cap I would think it is either the cap or the mating surface on the filler has a bugger on it…for some reason the cap is not sealing properly. Is the smaller diameter seal able to move freely up and down in the bore of the filler neck?

Jack

[CraigS]

Does your cap look like mine?

Click to enlarge.
The small brass disc in the center is loose. It needs to move so it seals when psuhed against the rubber but can move about 1/8 inch away from the rubber so coolant can flow back into the filler as the engine cools.

[btoll]

Yes, it's the donor mustang tank. So it's the translucent plastic off-white tank with an air vent at the top and the expansion hose connected at the bottom.

The cap looks pretty much like the one Craig pictured. The lower rubber seal near the bottom has the smaller metal disc that can spring down like would happen under a vacuum during the cooling phase. The whole rubber seal area springs up and I would think during the expansion phase when the pressure is high it would let coolant seep out. On the top there is another rubber seal ring (can't see it on this picture, but it's probably there too). This should seal around the top of the filler area so the coolant goes down the overflow tube instead of being squished out the top of the filler port.

Mine does have that small brass disc in the center. It's not always loose, but when I compress the spring as it would be when installed the disc is loose and can move away from the seal.

I wished I'd taken a picture down the filler neck last night so I could post this. There is the top area seal point and then the seal area for the compression seal. The opening under the compression seal point is round for maybe 1/8" of depth and then it switches to an oval opening for the rest of the way down to the main passage. The width of this oval is 7/8" which exactly matches the diameter of the brass disc being discussed. At first I thougth maybe the disc was hanging up on this oval section and therefore not allowing fluid to return. But I think the 1/8" depth at the wider circle would be sufficient. And that would explain the vacuum, but not the coolant leak near the filler area (if it was near the expansion tank it would've made sense, but it wasn't). So I don't think that's the issue.

The best next step for debug I can come up with is to refill the system and then let it get nice and warm while watching it to see what's happening.

It's possible there is a leak somewhere in the filler neck as well which would explain this behavior or as Jack said that the mating surface of the cap or filler could have issues and not be sealing. Is there an easy way to hook up the air compressor to the system and test for leaks under pressure?

Any other debug ideas?

Thanks,
Bret

[AC Bill]

There is a hand pump type pressure tester designed for testing cooling systems. I don't think I would want to use an air compressor to test it..

Occasionally some of the cast T fillers have had issues, from what I recall. Could be a pinhole in the casting. Cast units also can be a little rough around the sealing point.
You might just want to try a new cap to start with. A fresh seal, and a new spring may solve the problem. Cheap too..

[Dan Babb]

Autozone has a 'loan a tool' program. Get the cooling system pressure tester and see if your system holds pressure. It's easy to check.

[hind sight]

At this stage maybe buy a replacement cap as they are not expensive and that will rule out a bad cap.
Don't forget everytime you remove the filler cap you will have to re burp the cooling system.
Allow engine to come up to operating temp then allowed to cool back down.
This may need to happen twice to get all air out.
Watch the coolant level in overflow bottle when cool then when hot and note the change(pencil mark on bottle)

[btoll]

Thanks for the suggestions guys. I won't have time to do any testing the next couple of days but I should be able to play with it on the weekend.

I think I'll start by filling it up and getting things nice and warm to see if I can find the leak point. That may answer a lot of questions. I may pick up a new cap anyway because I really think there's somethinb about the top seal on this cap (I just can't find the evidence). Depending on the results of those tests I'll see if I can borrow the pressure tester and pinpoint any other issues.

Thanks for all the help. I couldn't have gotten this far on the car without the help of the great folks on this forum.
-Bret

[CraigS]

When you are filling the system make sure to remove a temp sensor or heater hose from the manifold. This lets air escape as you fill the T-filler. Once coolant comes out of the hose or sender port, cap it off and fill a little more at the filler and install the pressure cap. From now on never remove the cap. All topping up as you cycle from cold to hot should be done at the overflow tank.

Click to enlarge.
In this pic you can see my heater hose to the right of the top rad hose. Just to the right of the rad hose you can see the small brass nut that holds the wire on the temp sensor. Either is a good air bleed.

[btoll]

Thanks Craig. The air bleed is a great idea. I did drill a small hole in the thermostat to help get the air out so I can see how having an air bleed on the manifold will help. I used the fancy shrink wrap-like hose clamps so it's a little hard to remove the heater hose. I think I can pull the temp sensor though so I'll give this a shot. If I can't get an air bleed on the manifold, do you have any suggestions?

Also, do I need to do anything to get the long overflow hose primed? For the short stock setup I can see that it would just self prime. But for a hose this long (probably 8') that could cause a problem. I could "suck" fluid up through the hose, but the thought of coolant in my mouth is very unpleasant. Maybe I could blow into the overflow tank somehow after blocking other hole at the top. Ideas? Is this needed?\

This is very sad because I just got it back on the road after paint and we're having the driest fall on record but my cobra is stuck in the garage :-(

Thanks,
Bret

[Mike Wilson]

if a mustang radiator.you need to make sure your radiator is plugged where the old cap was.

[CraigS]

I think you could rig a temporary small funnel to fill the long hose from the T-filler end. Just hold it up higher than the rad tank,fill, and push back onto the T-filler. The only other tip I can think of is that while warming the car up, I always run it at about 15-1700 rpm. My gut says faster movement throughout the cooling system may help move bubbles along to the top. Make very sure that the overflow tank never empties as the engine cools so you don't suck any air back into the system and prolong the process.
One other general thought is you need to figure out how to post pics here. I use Photobucket as my host. If you don't have a digi camera, a cell phone is fine. The pic I posted of the cap was from my cell. I found that instead of entering and saving a phone number in my phone contacts list, I could enter and save my email address. So I take the pic in the garage,save it in the phone, and then email to myself. I walk inside to the home PC, open the email,save the pic to 'my pictures' and then upload it into Photobucket. I can take the pic and have it on here in about 3-4 minutes.REALLY helpful. Good luck, I hope you get out on the road soon.

[rich grsc]

Check to make sure you have clearance in the T-filler.

[johnnybgoode]

I have had numerous cooling system leaks over the past three year that I know can be very frustrating. Here's a list and solutions that seemed to have worked:
1. Coolant on top of the timing cover. Was a leaking water neck, switched to a Billet Specialties with an O-ring seal and has been dry ever since. I also took this opportunity to swap out the thermostat for a 180 vs the 192 that was in the car, I also drilled the hole in it as well.
2. Leaking out bottom of overflow tank (Summit). Changed out the overflow hoses and clamps and made sure the clamp on the bottom of the tank was tight. This one was very allusive and only showed up as small stains on my polished passenger footbox.
3. Leak out the WP snout. Replaced the water pump with a high flow AL pump vs the old rusty iron one. I then found I had a leak out the back of the pump. I warmed the car up and laid on my back for most of a Saturday to finally see coolant dripping off the very back edge of the balancer. I also saw coolant running down the lower rad hose which was really throwing me off since I saw leaks on the paper on the floor that were almost a foot apart (was time for a beer!). Before getting ready to pull the whole thing apart again, I re-torqued the WP bolts to 200 in/lbs and put an additional T-clamp (now have 2) on the lower pump/rad hose and cranked them down as hard as I dare.
I have driven the car through several heat cycles and the leaks seem to have stopped so I topped up the Forte de-gas tank right to the top (I left it down about 1/4" while all this was going on) and filled the overflow tank 1/3 full from the top with a turkey baster. So far so good as there are no leaks and the water temp is rock solid now with the fans turning on at 95C in traffic and never needing to come on when on the highway. Car seems to run very happy between 80-90C and has no problem in stop and go traffic. Not sure if any of this help but thought I would pass on the WP bolt re-torque since this saved me a ton of rework. I had originally torqued the bolts to 180 and they were still tight but the extra little bit after the inital heat cycles seemed to have fixed the prob. (I hope). Scott
Cooling System 001.jpgCooling System 003.jpgCooling System 005.jpgCooling System 007.jpg

[dv/dt]

It would be interesting to get a pic of your T-filler from the same angle as rich grsc's pic. You need to make sure that the pipie fitting for the overflow line did not screw in so far that it sticks out in the inside of the filler and possibly jamming up the rad cap. If it does then you need to shorten the fitting so it doesn't.

If you plugged the rad cap hole in the radiator you need to make sure that it is plugged of below the old overflow tube there (or the tube itself is plugged off).

[btoll]

Thanks for the tips Criag. They all sound like excellent advice. I like the idea of filling the overflow tube from the top like that. I really can post pictures, I've just been feeling lazy and hoped it would work without pictures. I've got one here as requested by dv/dt.

Mike, dv/dt, I have a NAPA expansion plug in the normal radiator cap hole, which closes off completely at the lower hole so fluid never makes it up to the radiator's overflow hole. Also, as you can see in the picture the overflow tube fitting on the T-filler doesn't stick out into the radiator cap space.

Johnny, it does sound like you've had your fill of issues. I've already worked through the thermostate housing issues and have that one sorted out with the correct O-ring. I think the WP is OK (at least it's not this problem). The thermostat is 180 with the extra hole to help get the bubbles out.

Rich, what are the red areas inside your T-filler? My T-filler is oval like yours and I've wondered if the skinny section of that Oval was preventing the brass circle from dropping down on the cap and allowing coolant to return. I'll need to watch for that to make sure that as things cool it returns from the expansion tank back into the system.

Do any of you see any issues with the T-filler or the cap? I was playing with the cap in place tonight and it seems to rock a little eaiser one way than the other (front to back with the car rocks a little but side to side is solid) and I wonderered if that is an issue (or maybe it's just a side effect of the cut-out sections for the cap tabs).



Any other thoughts? I should have time tomorrow to try filling the system and running the engine for a while. My fear is that nothing will happen with it just sitting in the driveway, forcing me to take it on a drive to get any leaking action. But I guess I'll see tomorrow :-).

Thanks again for all the consultation.
Bret

[btoll]

After looking at my picture again I really started to wonder about that oval section and the brass disc on the cap. You can see that there is a round indentation where the brass disc is supposed to go and then the ledge with the oval opening is down below that.

The diameter of the brass disc is 7/8", but the width of the oval opening is only 3/4", so the disc will definitely not drop down into the oval section of the opening. Now the depth of the ledge is about 3/32" and the height of the disc is around 1/16". That means that there is around 1/32" worth of travel for the disc to go down in order to return fluid to the system during the contraction phase.

Do you think that is sufficient? If not, why did they build the T-filler like this and sell it with this same cap? Do differen caps have a different disc size so that maybe I can get one that would work?

I can see how this would prevent coolant from getting back into the system which then might lead to air bubbles, but why would there be a leak around the cap/filler area? Do you think this could be the problem?

Thanks. Bret

[CraigS]

I think you and Rich have come up w/ the source of your problem! We can't say "Bingo" yet but I think we may be saying it soon.

[rich grsc]

Bret, the red area in my photo was making contact with the small disc on the bottom of the cap. I ground a bit of relief so the cap would seat correct. Looking at your filler, it looks like it has already been done, but it may be too small. The disc needs to set into that area.

[btoll]

Thanks Rich, Craig. Rather than just filling it up and runnig the engine as a test I decided to go ahead and grind away some of the aluminum there to make sure the small disc on the cap had enough space to drop down and return fluid to the system. 1/32" seemed a bit small to me.

With these fancy "shrink wrap" like clamps I wanted to avoid taking off the T-filler so I stuffed a rag down inside and put grease in there hoping it would catch all the shavings and then hit it with the die grinder with carbide bur bit. The grinding went fine, but the shavings did not cooperate. I hit it with the vacuum but as the system was not open there was no place for the air to come in and run across the shavings so it didn't get much from down below. I pulled the rag out and got most of them but there were still quite a few shavings left so I started using wads of masking tape on the end of needle nose pliers to get them out. Unfortunately I lost a blob of tape in there somewhere and I really wasn't able to get all the shavings out. So I have to take the whole system apart.

I have the T-filler out now but need to take the front wheel off and get the whole top tube disconnected which means I need to drain more fluid. Then another trip to NAPA to pick up some more fancy clamp things, a bunch of cleaning, putting stuff back together, filling, and I'll be ready for a test.

Here's a picture of the T-filler after the grinding. I found a washer that is the same size as the brass return seal on the cap so I could use it to test fit for increased drop and clearance. Looks like it definitely should work now. I just need to finish everything else.



Not sure I'll have time for all this today, but I'll see. It may have to wait until next weekend. :-( Thanks for all the help. Bret

[shand dill]

Btoll, i am experiencing the same issue. However my system works and then pucs the next time i go for a run. Has your machining of the oval proved that the your system was spilling over due to the incorrect seating of the cap. I have employeed a lever action type of cap.
shand.