When you click on links to various merchants on this site and make a purchase, this can result in this site earning a commission. Affiliate programs and affiliations include, but are not limited to, the eBay Partner Network.
For years, my '84 F250 has had a mysterious slow coolant loss. Well, loss may be the wrong word. The coolant was collecting in the overflow tank, so I knew where it was. Never had any overheating issues, just the recurring coolant escape.
I had pressure tested the cooling system and cap several bazillion times, all with no discrepancies. I tested the system cold. I tested the system hot. I pressure tested several caps submerged in a bucket of 200 degree water (the caps were submerged, not me), wondering if the heat was making them vent early. I had tried so many caps on my truck, I might as well have rigged up a cap dispenser next to the radiator.
When the truck was working hard, such as hauling our camper during the summer, I could expect to find the overflow bottle full by the end of the day. I had literally given up trying to solve this mystery, and just kept the cooling system full and the overflow bottle empty. Even though I'd check the cooling level daily when on a long trip, the amount of coolant loss was a crapshoot. One day there'd be very little. Occasionally the overflow bottle would spill over. Most times it was somewhere in the middle. What was changing?
So then I gets to thinking. What was the one thing I had never tested? Despite testing the system and cap separately, I never tested them together. Could the junction between the cap and radiator neck be at fault? That's a bingo!
Here's how I found the problem. I added a cooling system flush fitting (the kind that connects to a garden hose) to the 'downstream' heater hose. That's the hose that connects to the water pump, not the cylinder head. Here's the type of fitting I'm talking about:
Then I modified my pressure tester by adding a garden hose female fitting on the end, and connecting it to the fitting freshly installed in the heater hose:
The system is rated at 13psi, but as soon as I reached 10psi or so, I heard a bubbling coming from the overflow tank. I stuck the overflow hose in a small bucket of water to make it easier to see any bubbles. Oy vey, it looked like Lawrence Welk had turned on the bubble machine!
A cap which tested perfectly fine by itself was venting way early. I wiggled the cap slightly and it sealed up properly. I could nudge it and the bubbles would come back. I had finally found the leak and the reason why it varied from day to day. Every time I reinstalled the cap after checking the coolant, it wouldn't always seal the same.
Curiously, this was with a genuine MotorCraft cap. I even had a spare MotorCraft cap, still in a sealed package. Same as the first one. Look at it sideways and it would leak. I had an old Standard brand under the seat as a spare, and it was the most reliable of all.
I can see no signs of distortion or damage at the radiator neck. I even tested for a good sealing surface using some Prussian blue dye, and there were no bare spots. I wonder if the radiator neck if flexing under the pressure of the cap. (The neck is stamped sheet metal, not cast.) Heck if I know what's going on, but I'm glad to have finally found the problem.
My ThoughDuJour™ is to just replace the radiator. It's an aftermarket brand over 20 years old. That's a long life for a radiator, isn't it? Any recommendations on brands? My truck has the 351W, 8600 GVW, factory AC, and the super cooling package. It's a stick, so there's no transmission cooler.
I assume you used the Prussian blue where the cap hits the top of the neck. But what about at the bottom of the neck? If the top and bottom aren't parallel with each other it won't Seal as it is supposed to seal.
As for a radiator, I don't have a favorite brand. But I like the all-aluminum ones instead of brass as I believe they cool better.
I assume you used the Prussian blue where the cap hits the top of the neck. But what about at the bottom of the neck? If the top and bottom aren't parallel with each other it won't Seal as it is supposed to seal.
I only did the Prussian blue check at the bottom of the neck. The top seal isn't nearly as critical, since that area is vented to ambient via the overflow tube.
I bet you're right about the two surfaces not being parallel. I checked for out-of-round and flatness, but not if parallel.
By the way, if using Prussian blue, resist the urge to scratch any itch. I do not care to discuss how I know this.
I only did the Prussian blue check at the bottom of the neck. The top seal isn't nearly as critical, since that area is vented to ambient via the overflow tube.
I bet you're right about the two surfaces not being parallel. I checked for out-of-round and flatness, but not if parallel.
By the way, if using Prussian blue, resist the urge to scratch any itch. I do not care to discuss how I know this.
The top seal is critical. If it's not sealing, then when the engine cools it will not suck the coolant back into the radiator. You need to make sure you have a recovery style cap and that both seals are good. You probably would have never known about your problem if the cap sealed properly to suck the fluid back out of the overflow tank. You need to make sure the hose leaving the radiator neck is in good shape, and it stays submerged in the overflow tank so it will not suck air.
When the engine is worked hard and the coolant gets hot, it expands and fills the recovery tank. When the engine cools it sucks the fluid back into the radiator. This keeps the radiator full to the top. The older cars and trucks that did not come with a tank just left a air space in the top of the radiator for expansion. You could not fill the radiator to the top on those.
The top seal is critical. If it's not sealing, then when the engine cools it will not suck the coolant back into the radiator.
No argument there, point well taken. I would have been more accurate to say the top seal is not critical for maintaining the cooling system pressure.
Once the coolant does vent above a certain pressure, as designed, that upper seal is critical for the overflow and recovery function, just as you were saying.
The main issue I've been having is the lower seal is so erratic, based on how just nudging the cap was enough to make it vent. Some days it probably made a good connection and held pressure just fine. Other days I bet it was venting far too early. Even after confirming the overflow bottle was empty in the morning, by the end of some days the level went over the top.
I'm pretty stoked to have finally found the problem. I'm going to check if the two sealing surfaces are parallel, but I think I'm going to replace the radiator no matter what. Now if I could just find a reasonably priced model without plastic side tanks.
It says "3 row 100% welded to 55 MM thickness for desert cooling. This is a polished upgrade for offroad, towing and extreme climates." Should be good.
It says "3 row 100% welded to 55 MM thickness for desert cooling. This is a polished upgrade for offroad, towing and extreme climates." Should be good.
For years, my '84 F250 has had a mysterious slow coolant loss. Well, loss may be the wrong word. The coolant was collecting in the overflow tank, so I knew where it was. Never had any overheating issues, just the recurring coolant escape.
I had pressure tested the cooling system and cap several bazillion times, all with no discrepancies. I tested the system cold. I tested the system hot. I pressure tested several caps submerged in a bucket of 200 degree water (the caps were submerged, not me), wondering if the heat was making them vent early. I had tried so many caps on my truck, I might as well have rigged up a cap dispenser next to the radiator.
When the truck was working hard, such as hauling our camper during the summer, I could expect to find the overflow bottle full by the end of the day. I had literally given up trying to solve this mystery, and just kept the cooling system full and the overflow bottle empty. Even though I'd check the cooling level daily when on a long trip, the amount of coolant loss was a crapshoot. One day there'd be very little. Occasionally the overflow bottle would spill over. Most times it was somewhere in the middle. What was changing?
So then I gets to thinking. What was the one thing I had never tested? Despite testing the system and cap separately, I never tested them together. Could the junction between the cap and radiator neck be at fault? That's a bingo!
Here's how I found the problem. I added a cooling system flush fitting (the kind that connects to a garden hose) to the 'downstream' heater hose. That's the hose that connects to the water pump, not the cylinder head. Here's the type of fitting I'm talking about:
Then I modified my pressure tester by adding a garden hose female fitting on the end, and connecting it to the fitting freshly installed in the heater hose:
The system is rated at 13psi, but as soon as I reached 10psi or so, I heard a bubbling coming from the overflow tank. I stuck the overflow hose in a small bucket of water to make it easier to see any bubbles. Oy vey, it looked like Lawrence Welk had turned on the bubble machine!
A cap which tested perfectly fine by itself was venting way early. I wiggled the cap slightly and it sealed up properly. I could nudge it and the bubbles would come back. I had finally found the leak and the reason why it varied from day to day. Every time I reinstalled the cap after checking the coolant, it wouldn't always seal the same.
Curiously, this was with a genuine MotorCraft cap. I even had a spare MotorCraft cap, still in a sealed package. Same as the first one. Look at it sideways and it would leak. I had an old Standard brand under the seat as a spare, and it was the most reliable of all.
I can see no signs of distortion or damage at the radiator neck. I even tested for a good sealing surface using some Prussian blue dye, and there were no bare spots. I wonder if the radiator neck if flexing under the pressure of the cap. (The neck is stamped sheet metal, not cast.) Heck if I know what's going on, but I'm glad to have finally found the problem.
My ThoughDuJour™ is to just replace the radiator. It's an aftermarket brand over 20 years old. That's a long life for a radiator, isn't it? Any recommendations on brands? My truck has the 351W, 8600 GVW, factory AC, and the super cooling package. It's a stick, so there's no transmission cooler.
This baby is a thing of beauty. I'm so excited, I could cut glass with my nipples. I've taken a few preliminary measurements and it looks like a direct fit. I'll post more details once I get it installed.
Sadly, I will have to wait a few days to install the new radiator. I've got new rubber mounts on order, and they haven't arrived yet. I've gone to the UPS website, and repeatedly clicked on my tracking number to make the package move faster, but it hasn't helped. Apparently there is some kind of shipping vortex in Hodgkins, Illinois, where packages check in but can never leave.
And the day my package is due to arrive, I have to (forgive me, I'm breaking out in a cold sweat just thinking about it) go to a wedding. Suit, tie, the whole nine yards. Not mine, I should clarify. On a perfect summer weekend, just made for wrenching. As I get older (54), I've come to realize I only have a finite number of summer weekends left on this planet. How many of those will be taken away from me by weddings?
That's why I'm proposing that men should be specifically excluded from attending weddings. Unlike 101% of women, no man has ever said, "Woo-hoo! A wedding! I can't wait to attend." Women will have a better time at a female-only event, just like baby showers, batchelorette parties, etc. Think of the health benefits alone, as never again will a woman dislocate her shoulder or tweak her knee trying to drag a reluctant man onto the dance floor.
But in the meantime, unless I can quickly shift societal views on wedding attendance, I am doomed. The sword of Damocles is hanging over my head....
05-26-2017, 08:48 PM
