'81 Rabbit diesel odd coolant flow

[dabradford]

I just replaced the radiator, most of the hoses, thermostat, radiator fan temp sensor & relay. There is something strange now going on as the small coolant line between the top of the radiator and the overflow tank is carrying a steady flow of coolant into the top of the overflow tank. If the coolant line between the head and the top of the water pump were blocked, I think it would cause this, but it isn't. If I've put in some kind of incorrect thermostat that permanently blocks the port between that line and the pump intake, I think it would also do this, but I'm pretty sure the thermostat is right. This condition exists whether the engine is cold or hot. The engine is running too hot and the fan is coming on too late and shutting off too soon which makes me think think I'm not getting great circulation around the radiator and the area around the fan switch is staying cooler.
If I understand the flow correctly, the water pump pumps into the block and can pull from three different places: the bottom of the radiator (blocked by the thermostat when cold), the line between the head and the top of the radiator(open when cold) , and the line pulling from the heater core and the bottom of the expansion tank (which is always open to the pump).
So if the thermostat is mostly blocking the two ports it controls, I think it could give me this condition. The water pump would draw from the heater core/bottom expansion tank line and pump into the block, out the head and into the top of the radiator because the head to water pump line would be somehow blocked by the thermostat. This would create pressure in the small line from the top of the radiator to the expansion tank like I'm now seeing.
I have put in two different new thermostats with the same condition resulting, but it seems there are a few different seal arrangements around the thermostat, and I'm not sure I'm doing that part right.

Any thoughts or experience with this would be much appreciated.
Thnx,
David

[zagrot]

it is normal to have a strong stream of coolant entering the top of the coolant reservoir when the engine is running.

check that the thermostat is installed so the spring side is toward the waterpump (i'm not sure that there is room in the thermostat housing to install it backward so chances are that it is right). there are two gasket options for thermostats, one is the paper(ish) gasket and the other is the rubber seal that surrounds the outter perimeter of the thermostat flange. the early vw's use the gasket. check the thermostat by suspending it in a pot of water and heating it untill it begins to open and measuring the temperature of the water. the rating displayed on the thermostat is the temperature at which it will begin to open, it should be fully open within a few degrees above its rating.

there are two fan switches available, one turns on at a lower temperature than the other.

when you replaced the lower radiator hose did it have a long spring inside? this spring will prevent the lower hose from collapsing under strong suction. does the engine exhibit a tendancy to overheat when driving at highway speed or at idle? was this tendancy to overheat a prexisting condition or did it begin with the replacement of the hoses and radiator?

[dabradford]

Thanks for your help with this..
The thermostat is faced the right way, you're right that it won't fit in upside down.
One of the new thermostats came with an Oring, the other came with the same Oring and a slightly smaller quad ring. I'm wondering if the quad ring goes between the thermostat and the water pump mounting surface, thereby moving the thermostat sealing plate a little further away from the port it seal, but this seems an unlikely cause of my problems. The old seal is so mangled that it's alittle hard to figure out where it was, but I'll take another look.
The old lower hose didn't have a spring.
The fan switch is quite new and worked fine on the old radiator. The old radiator was leaking so that's what made me do all these changes.
I did chech one of the new thremostats with hot water and it was fine.
I just finished putting all this stuff in yesterday, and the overheating took place just while it was idling. Driving it didn't help with the cooling which is what made me think the circulation through the radiator wasn't great.
I keep coming back to the position of the thermostat as the thing that could cause all this, or wondering if I could have some sort of air lock that's messing up the flow.
Thanks again for your help.

[zagrot]

the small hose from the radiator to the coolant reservoir purges air form the cooling system so i doubt there is any air trapped inside, or at least not enough to be noteworthy. did you say the side of the radiator with the temperatue sensor stayed cold, or were you speculating? is the radiator new or used? what do you mean by "quad ring"?

did the thermostat you purchased have a disk on the end that went into the waterpump housing? maybe the mounting depth is critical. that is something i have never considered.

[dabradford]

did you say the side of the radiator with the temperatue sensor stayed cold, or were you speculating?
Slow to heat up after thermostat had opened

is the radiator new or used?
New Radiaor

what do you mean by "quad ring"?
A quad ring is a square or rectangular cross section Oring.


did the thermostat you purchased have a disk on the end that went into the waterpump housing?
It did. it was virtually identical to the one I replaced.

maybe the mounting depth is critical. that is something i have never considered.
I think it's abit of a stretch too. Moving the seals around wouldn't change it more than 1mm.

Some interesting new developements this AM. I started it up and the same overheating eventually started to happen while it was idling. The temp switch side of the radiator was slow to heat up (to the touch) after the thermostat had opened. (We make biodiesel here so I have a nice supply of digital thermometers that when stuck in the expansion tank give me a slightly reduced idea of the engine temp) The fan would cycle, but it came on way too late and shut off too soon leading me to believe I wasn't getting good circulation through the radiator. I took a clamp and squeezed off the hose between the head and the water pump and made sure the heater valve was closed. My thinking was that this would force more circulation through the radiator and perhaps pull through any air locks. Lo and behold after about 30 seconds the temp dropped to where it should be and the fan started cyling the way it should. I took off the clamp.
I took it for a drive and all was well for @ 10-12 miles at 30-50 mph when it started to heat up again just like before. I baby'ed it home, put the clamp on it, and in about 30 seconds it was fine again. Before I clamped it the second time, I was wondering about the impeller/shaft slipping possibility, but clamping the hose shouldn't be a cure for that.
A re-occuring air lock seems like a bit of a stretch... Any other thoughts? Thanks many times!!

[zagrot]

your new data does make the situation more intriguing. it would be of intrest to be able to know the pressure difference in the inlet side of the radiator when you clamp the hose between the water pump and the head with the thermostat open or removed entirely and the heater valve closed (could be done by terminating the small hose to the reservoir with a low pressure gauge). i suspect that a substantial rise in pressure would place the radiator at fault, or at least indicate some type of blockage between the radiator inlet and the water pump inlet. do you think there might have been some foreign object in the radiator or the hoses upon instalation (packing material, paper, ect...). not that i find this likely.

maybe something else, does the lower radiator hose become hot before the opposite end of the radiator (i mean really hot)? after rodding one of these beasts (something i do not advise anyone try) i found that the only thing keeping the flow of coolant from going directly from the radiator inlet to the radiator outlet is a divider soldered in the middle of the passenger side radiator tank (cap, end, whatever) about half way down oriented horizontaly; in relation to how the radiator would sit in the car. the divider was surrounded on three sides by the tank and on the fourth side, that contacted the radiator core, there was a piece of what appeared to be rubber tubing cut down the side and slipped over the divider to provide a seal when the tank was mated to the core. that is for the version with the metal tanks (ends, caps, whatever), if you have one with plastic tanks replace "soldered" with molded and you'll probably get a fairly close description. anyway that rambling has brought me to this: maybe that devider is faulty, and when you clamp the hose there is nowhere for the increased flow to go except through the radiator core (where it should be going anyway). i also find this unlikely, but i thought i'd mention it because it came to mind.

somehow my thoughts keep coming to the fact that when you clamp the hose between the head and the pump things work the way they are supposed to. following the logic that fluids in a closed circuit would follow the path of least resistance it makes me wonder if the disk on the end of the thermostat really is supposed to completely block the upper passage on the water pump housing.

this is not the first time that i have heard of an instance involving overheating after changing a thermostat, the problem is that when doing batch repairs on a car no one says "i'll change one part, reasemble the whole deal and see if it still works." i'm beginning to think that there is a popular manufacturer of thermostats, whoe's name probably probably begins with an "s," that has a part number for the vw diesel but doesn't really correspond to vw's specifications. then again, maybe i'm reading too far into the whole thing. you said the reapirs were conducted because the previous radiator was leaking. if the engine did not overheat when using the previous thermostat maybe you might try reinstalling it and giving a test run. let me know what you think.

[dabradford]

Thanks for all the good thoughts.
My first concern when all this started was a missing divider in the radiator, but I poked around and it was there..
However, this is what happened today:
After running fine for @ 45 minutes today it started to heat up again and this time pinching the hose didn't help. So I drained the system, took out the thermostat ( the second new one). I put it, the other new one, and the old one ( which had been working fine) in a pot of water with a digital thermometer and heated it all up. Here's where things got interesting. As the temp passed 80C, the old one started to move and finished its full stroke to 38mm OAL before the others moved 1/4 of their stroke. The water was almost at a boil before the 2 new ones reached 38mm at which point the old one was at 42mm. The distance from the mounting surface in the water pump to the port the thermostat disk covers is 38mm. All the thermostats were rated for the same temp. So your theory about different makes of thermostats being passed off as having the correct specs when they are a little different could be my problem.
So i put the old one back in and everything seems OK......so much for the air lock theory I was so fond of....

[zagrot]

if things continue to work correctly could you let me know in a few days? i'm currently not in a position where i have access to parts for testing, but that does not change the fact that i know someone who would be very interested in what you observed when testing the three thermostats.