I've got a good brain teaser.

[Gary Gibbs]

I\'ve got a good brain teaser.

Ok folks. I have a good brain teaser for you.

Some relevant data. Engine is 351W (427 CID) with Ford Motorsport “Street Rod” short water pump driven CW at 1.25 times crank speed (Standard OEM ratio). I believe the thermostat is a high flow Robertshaw 180 degree model.

Recent trip to SAAC 29 included some open track in my GT40. Completed about 25 - 30 hot laps, etc. Was having one cooling system related issue. At no time before incident did the manifold coolant temp sensor register above 85 deg C. Coolant stayed between 80 – 85. At full throttle second gear (~6000 RPM) after the car was warmed up I ejected a freeze plug from the engine loosing all of the coolant in rapid fashion. I put in a new freeze plug and ejected it again after about 15-20 hot laps.

There was no evidence of coolant coming out of the radiator cap into the overflow tank. After some limited “at track” diagnostic work I had concluded that I either had a sticking radiator cap or a combustion leak into coolant. Later at home I installed a rubber expanding type freeze plug (tightened to the hilt). I did a cold compression check with all cylinders between 215 – 222 psi (normal). Did a warm (at TDC) leakage test with all cylinders between 2.5 – 4.5 percent leakage (normal). Did a hot compression check with all cylinders between 220 – 230 psi (normal). Thus it would appear there is no combustion leak.

Next I devised a method to monitor coolant system pressures and installed some gauges at lower left block drain plug, lower right drain plug and in expansion tank (vent input line). I measured pressures as follows for a warm engine:

GT40 at high idle ~1500 RPM:
Left block ~15-20 psig, Right block ~15-20 psig, Tank ~10 psig (rad. Cap closed).
GT40 at ~5500 RPM on street in 2nd gear:
Left block ~125 psig, Right block not measured, Tank ~8 psig (rad. Cap closed).
Pressures always came back down as one might expect if the pump was creating the pressure as opposed to a combustion leak.

For reference I measured a stock 1992 Mustang 5.0L at 800 RPM:
Right block ~12 psig
Stock 1992 Mustang 5.0L at ~5000 RPM sitting still with no load:
Right block 50 psig

Having never measured cooling system pressures this seems very high in the GT40. To try to get some useful information I looked at the SAE paper for the new Ford GT cooling system design I was surprised to see the cooling system drops while Engine 5846 RPM, Pump 7518 RPM, Radiator flow 71 GPM are: Engine 38 psi, Engine to radiator 19.6 psi, radiator 5.4, radiator to pump 2.5 psi, intercooler 5.8 psi. Assuming that the intercooler is in the main circuit that equates to a total pressure drop around the circuit of 71 psig which must be produced by the pump. Thus if a 16 psig radiator cap is used and located at the low pressure point of the system (pump inlet) then the maximum pressure would be 71 plus 16 = 87 psig. Seems ridiculous but less then the 125-130 psig that I measured.

Any ideas on what could be causing the pressure? Is it normal (I think not)? And what can be done next to check or correct it. Help greatly appreciated.

 

Re: I\'ve got a good brain teaser.

Hi Gary, I know that your motor is not like my 302, but I was told before I bought my CAV that 40s had a problem with over heating. I asked the dealer if it was ok not to put a thermostat in it. It was decided to leave it out since it stays so hot here in the Atlanta area. When it is above 90 degrees it warms up quickly and stays at about 180 degrees at driving speed and when in trafic I put the fans on and it will stay around 180 degrees. Maybe your thermostat is sticking? This may not have anything to do with your problem, but if it is cutting down on the flow maybe the preasure might increase. Joe /ubbthreads/images/graemlins/confused.gif

 

[Howard Jones]

Re: I\'ve got a good brain teaser.

OK I am going to add a little information but I really don't know for sure anyone else?
1. The GT40 427 must be making double the power the 5.0 mustang is at least. Correct? Heat generated must also be at least double.
2. The pumping distance (total) must be at least double that of the front mounted 5.0.
3. What would happen if you took out the thermostat and put in a round piece of alum with a hole in it. Start with 1 inch hole and work down to about 5/8 with an eye on temp. Shoot for 170 degrees or so.

I am guessing but I think that all that power is heating the water in the block quicker than it can be pumped out and cooled. That is a guess and thats all I have at this point.

 

[Chris Duncan]

Re: I\'ve got a good brain teaser.

Sounds pretty weird.

Try taking the belt off the water pump and see if you get the same pressure rise. That would eliminate the slight possibility it's leaking compression.

If it still is do a leak down and remove the coolant fill cap and see if it's blowing bubbles.

It sounds like your bypass bleed hole in the Tstat is plugged. A little chunk of dried silicone sealant could do that. Also some builders enlarge the bypass hole.

Either that or youv'e got an incorrect head or intake gasket, something is not allowing circulation within the block. Don't these engines circulate within the block even when the Tstat is closed?

I believe the PSI drop pressure and the system pressure (cap rating) are two different things.

125 is extreme the water pump must have really close pump vane to body clearance, if that kind of pressure gets into the rest of the system it could blow anything and everything up including the radiator, they're not even made for that kind of pressure.

 

[Frank Catt]

Re: I\'ve got a good brain teaser.

Gary, interesting you have gone to such depths in your investigations into this problem. This sort of thing used to happen regularly until we gave it some thought. As Howard has suggested, use a disc restrictor instead of a thermostat, I usually use an old thermostat and remove the centre core, just leaving the hole at approx 1 inch dia to slow flow through the block. Obviously this restrictor can be adjusted to get the right running temparature you want by changing the hole size. Secondly it is in my opinion very important to fit water/air bleed tubes from the rear of the inlet manifold/cylinder heads to the expansion tank, but not to use any larger pipe that -3 as this could cause a loop system for water flow if too large ( even -4 is too large !) This has been recently discussed on the Forum. Is your problem one of localised steam pressure ? this can cause huge pressure variations withour being apparant in the system generally, and the steam disappears very quickly in contact with the coolant nearby, I seem to remember reading somewhere that this caused a race team to lose a a whole season because they just did not realise that this could even happen inside an engine.Fran Hall must have huge knowledge in his engine testing experiences with GM. Frank

 

[Gary Gibbs]

Re: I\'ve got a good brain teaser.

[ QUOTE ]

I am guessing but I think that all that power is heating the water in the block quicker than it can be pumped out and cooled. That is a guess and thats all I have at this point.

[/ QUOTE ]
Thanks for the help and input. The engine shows no signs of overheating. The pressure under the radiator cap actually drops by a couple of psi under full RPM. I think if the system were locally steaming you would see a rise in pressures everywhere in the system and thus the system would gradually rise in total pressure and start opening the radiator cap and push out coolant. In this case the pressure rises like a tachometer. Up with RPM and rapidly down when the RPM drops. Always back to the same point.

 

[Gary Gibbs]

Re: I\'ve got a good brain teaser.

[ QUOTE ]

If it still is do a leak down and remove the coolant fill cap and see if it's blowing bubbles.

It sounds like your bypass bleed hole in the Tstat is plugged. A little chunk of dried silicone sealant could do that. Also some builders enlarge the bypass hole.

Either that or youv'e got an incorrect head or intake gasket, something is not allowing circulation within the block. Don't these engines circulate within the block even when the Tstat is closed?

I believe the PSI drop pressure and the system pressure (cap rating) are two different things.

125 is extreme the water pump must have really close pump vane to body clearance, if that kind of pressure gets into the rest of the system it could blow anything and everything up including the radiator, they're not even made for that kind of pressure.

[/ QUOTE ]
Thanks for the help! When I did a leak down test I did look in the radiator cap for bubbles on each cylinder and saw none. I took the worst cylinder for leakage and left the air pressure on it for an hour with the radiator cap on and saw zero pressure in the cooling system with gauges.

I am not sure if the thermostat itself has a bleed hole, but there is a full time bypass hose of 5/8" running from the thermostat housing to pump.

I have now wondered about the head gasket(s) being on backwards which I think would restrict flow from block to head. I need to take a look at book photos as I do not have a disassembled engine to look at. The water does not circulate significanly through the engine with the thermostat closed except the small amount that goes through the bypass hose.

Yes there is a difference between psi drop and cap rating. What I was attempting to say /ubbthreads/images/graemlins/smile.gif Was that in the main cooling circuit of the Ford GT the coolant pressure rises 71 psi through the pump and then proceeds to drop going through each component listed then returns to the pump to get its pressure built back up. Now the radiator cap can set the "DC" level of the pressures relative to the outside of the system. If the cap were left off then that point in the cooling circuit could not measure above zero psi or coolant would come out the cap. In the extreme case were the coolant cap opens that means that that point in the circuit would be a maximum of 16 psig. Thus all other points around the circuit would be 16 psig higher. The highest point being just downstream of the pump (the block) at 71 + 16 or 87 psi. My feelings are the same concerning the 125 psi. No wonder the freeze plug wants to come out.

 

[Gary Gibbs]

Re: I\'ve got a good brain teaser.

[ QUOTE ]
I usually use an old thermostat and remove the centre core, just leaving the hole at approx 1 inch dia to slow flow through the block. Obviously this restrictor can be adjusted to get the right running temparature you want by changing the hole size. Secondly it is in my opinion very important to fit water/air bleed tubes from the rear of the inlet manifold/cylinder heads to the expansion tank, but not to use any larger pipe that -3 as this could cause a loop system for water flow if too large ( even -4 is too large !) This has been recently discussed on the Forum. Is your problem one of localised steam pressure ?

[/ QUOTE ]

I don't think it is localized steam, but who knows. The engine seems to react the same even at medium throttle as long as you get the RPMs up to 5k or so. The pressure gauge would make a great tachometer. Pressure rises as quickly as the engine does on run up and falls just as quickly (tenths of a second). I would think that either a combustion leak or localize steam would take a while to drop down and that it would raise the pressure of the whole system which would be visible at the tank pressure gauge.

I have only fitted a single -3 vent line from the front of the intake and want to add -3 vents to the back of the intake when the intake is next off. I downsized the line to a -3 to prevent the tank from getting pressurized.

I am considering another round of tests adding two more pressure taps. One just downstream of the thermostat housing and one at the inlet of the radiator. Then do the same run up WITH and WITHOUT a thermostat installed. This could tell me were most of the pressure drop is occuring (in Engine? do to thermostat restriction?)

 

[Gary Gibbs]

Re: I\'ve got a good brain teaser.

Having not been involved in any cooling system design before the details of this stuff is a bit new to me. But what I have always found interesting is the fact that the systems are designed by controlling the coolant flow to the radiator by restricting the coolant flow from the engine (pump) and not diverting the flow back to the pump inlet. It would make better sense to use a "thermostat" to direct the "full flow" to either the radiator or back to the pump inlet thus not restricting the flow and driving up pressure. Thus the thermostat is not a resrictor just a diverter. As the engine needs more cooling it just sends more of the "full flow" to the radiator. I think there are some OEM cars that do it this way (not sure which but uses a funny dual disk thermostat) but most do it the old way.

I am wondering if I do not have a high flow thermostat or am spinning the pump to fast (at 25% overdrive, OEM style).

 

[Gary Gibbs]

Re: I\'ve got a good brain teaser.

/ubbthreads/images/graemlins/smile.gif /ubbthreads/images/graemlins/smile.gif
By the way. If any of you are dying to add a pressure gauge to a "working" GT40 in the block drain hole and run around the block I would love to see the data. You need about 7 feet of hose a gauge that goes up to ~100 psi and the proper fittings 1/4" NPT for block.

 

[Lynn Larsen]

Re: I\'ve got a good brain teaser.

Gary,

As I read through this, my mind kept saying that there is some kind of internal restriction either in the block, heads or intake manifold. I have a feeling that you may be on to something with the comment about the head gaskets. Other thoughts were a casting imperfection, possibly something dropped into the cooling system, could be at one of many possible locations, which has lodged somewhere in the block and/or heads. Faulty thermostat should be the easiest and first thing to play with as many have indicated.

Fixing the symptoms can be done with screw in freeze plugs. Many racers install these to stiffen the block. You may want to think about this as a cure to your immediate problem. If it turns out that these 125psi pressures are just the nature of the beast and the engine is running at nominal temps, this may just be the solution you need.

Lynn

 

[Gary Gibbs]

Re: I\'ve got a good brain teaser.

[ QUOTE ]
Gary,

Fixing the symptoms can be done with screw in freeze plugs. Many racers install these to stiffen the block.
Lynn

[/ QUOTE ]
So this can be done to a block after the fact? This is a Dart block with siamesed bores so I am not sure how that effects flow. These blocks are used in applications up to 2000 HP so...

Another difficulty is that there is little to no evidence (data) on what is "normal."

Thanks for the nput.

 

[Peter Delaney]

Re: I\'ve got a good brain teaser.

Gary, I had a similar problem courtesy of the 1st "engine builder" - all went ok until we loaded it up on the dyno, it got hot very quickly, then on no load, came back ok. After a few cycles, blam - it blew a hose. We then noticed some steam coming out of the exhaust pipes when the engine was shut down : tear-down time !

Two problems found :

- One head gasket was installed back-to-front : restricted the water passage between block & head.

- Head bolts not tensioned down correctly - allowed combustion pressures to bleed into the water system. This could have been tricky as a normal compression test would have probably come up ok, but under combustion, things are different.

I know its a pain, but if all else fails, it might be worth lifting the heads to check gasket orientation (after checking the head bolts for correct torque).

Kind Regards,

Peter D.

 

[Gary Gibbs]

Re: I\'ve got a good brain teaser.

Thanks for the info. After the first pop at the track I retorqued the head studs. Using the leakage test you do subject the engine to 100 psi continuous (with no noticable bubbling in the tank), but this is less then the ~~600 - 700 psi peak combustion pressures.

 

[Gary Gibbs]

Re: I\'ve got a good brain teaser.

Glancing at a couple of small block rebuild books it would appear that installing the head gasket backwards would result in the about the same total coolant flow. You are shifting the high flow to the front of the engine and low flow to the rear. So I would expect that the block would see the same net restriction and pressure, but the rear cylinders would overheat.

 

[Lynn Larsen]

Re: I\'ve got a good brain teaser.

Gary,

Have you talked to Dart yet? I just have a feeling that they would be interested in what is going on here and may decide to get involved so as to avoid any negative publicity about their block (be it real or percieved.) Just a thought.

Good Luck!!
Lynn

 

[Gary Gibbs]

Re: I\'ve got a good brain teaser.

I plan on calling Dart (block) and Ford Racing (pump) on Monday to see what they know and recommend, etc.

 

[Andy Besic]

Re: I\'ve got a good brain teaser.

Gary
I'm not an engine builder but have gone through similar trials with my GTD.
I'm running a K code 289 Hipo with iron Hipo heads, 11:1 compression, screw in freeze plugs and dual point mechanical distributor with high initial advance.
Every time out on a long "aggressive" drive I would eventually puke fluid past a 16lb radiator cap even though my water temp sender read 85 deg C. After doing many of the tests you have done, I tapped -6 fittings into the rear of the manifold and ran them into a T and then in a single -6 to the expansion tank. I
I was worried that this might by-pass too much coolant past a closed T-stat so I have taken a more careful warm-up procedure. I have put 1100 miles on the fix so far 700 of which have been track days and have been very satisfied with the results.
If I can create steam in a stone age 289 I would think with the amount of "groceries" your 427 is processing that you could produce steam almost instantaniously.

Just my humble 2 cents.

 

Re: I\'ve got a good brain teaser.

Gary,
I think you are correct about the fast rise and fall in preasure. If you have no air bubbles or a very very small amount there is no room for steam to form which would cause the preasure to fall off slowly. Without air the preasure can increases rapidly and decreases rapidly. The head gasket installed incorrectly seems a likely thing to look at.
Joe

 

[Gary Gibbs]

We have new data!

Before I took the data I talked to Keith Craft tech support, Dart, and Ford Racing. Only Dart seemed real eager to put some time into it and in their experience only one application had they seen pressures up to 100 psi and thought 70 - 80 would be a max. This is not something they measure though nor have they seen many freeze plug issues.

Ok folks we have new data. This data was taken in the driveway with neighbors wondering what the **** was going on next door! Car in neutral, warm with fans on. Obviously the t-stat is open enough to keep the car cool under these conditions, but would expect it to be more open under full power, etc. Pressures are ordered (P1 - P4) such that they should decrease as you go around the circuit to radiator and back.

P1 = in block pressure at lower left rear drain plug hole.
P2 = in front of intake about 3 inches upstream of t-stat.
P3 = in coolant pipe about 1 foot downstream of t-stat toward radiator.
P4 = vent line from radiator very near tank (considered tank pressure which is fed directly into pump inlet).


RPM, P1, P2, P3, P4
1500, 20, 25*, 15, 11 (RPM not real steady during measurement)
2000, 30, 25, 13, 11
3000, 50, 35, 11, 11
4000, 70, 60, 10, 10
5000, 100, 95, 13, 9

So under these conditions the block and intake are showing only a 5 - 15 psi difference thus probably not head gasket. The BIG drop occurs through the t-stat! at 5000 RPM the pressure drops 82 psi through the t-stat and 1' down the pipe. Obviously it looks like the culprit and I will take data with t-stat removed tomorrow maybe including a road test to see differences.