Packard Forum Index » Post-War (1946-54)

Hey all! A member on the 22nd/23rd Facebook page shared a video of someone going through the cooling system on '38 282 engine. At about 7:45, he digs into the thermostat housing and points to the "modern" thermostat installed as "wrong," indicating that it'd be too restrictive. He shows a different (indicated as "correct") thermostat with a larger opening that would presumably allow better flow. After digging in the forum a bit, it appears that it would be a bellows thermostat, rather than a capsule thermostat, that he installed.

Curious if there's anything to this on our engines, in your collective opinions. Should we be running essentially a "high flow" thermostat? I know my engine has always run a little warm, despite new water pump, new coolant, flushed block/core, pulled water distribution tube, tracked down/installed upper rad splasher, ensured heat riser is functioning correctly, etc. I did send my distributor/vacuum advance out to be rebuilt to just wholly eliminate spark/timing/vacuum as a variable.

Or would the opposite be true - using a thermostat with a larger opening would actually result in a lower (slower?) flow rate and actually make the heating issue worse? I know this can be an issue when thermostats are removed entirely.

The link below goes right to his "assessment" of the wrong thermostat. He re-installs the thermostat at 14:07. Thanks in advance for your perspectives.

youtu.be/vn6k6ncN8KA?si=eGEFbPaQ2FMDOWAc&t=465

Packard used two types of radiator or flow control thermostats over the years. Originally they used what is called the bellows type and in non and low pressurized systems those work well. Problem is by the early 50s with the more powerful engines and in very hot climates or in cars with addition of AC they had to increase the pressure in the radiators to prevent overheating. Up to the stock 7# caps used until the early 50s all was fine but the increased pressure with the 12-15# caps they were starting to use actually works against the bellows type thermostats and delays their opening.

Packard went with the modern capsule type which is not pressure sensitive and is what the guy has as a replacement. When an older bellows type needed replacing capsule types could be used and eventually all bellows use was phased out. I didn't see a good enough view of the "bad" modern type but it is probably capsule too except the opening does look a bit small. It might be for a car that needs a serious restriction to control too fast a flow going thru the radiator but that is just a guess.

Packard issued a service counselor article explaiing the concerns and reasons for changing type.

I'm not particularly familiar with the cars of that era but is there a consensus as to why removal of the thermostat would make the cooling work worse? I have heard of this issue before but I can't picture why a missing or oversized thermostat would decrease flow rate. The only explanation I have heard is that a missing thermostat would make an increased flow rate which would diffuse worse because the faster flow would travel in a more efficient path. Though I'm not sure if anyone knows if this explanation is accurate.

Not sure about the more efficient path but I think there is sort of consensus that removing a thermostat removes any restriction and causes the coolant to flow so fast thru the entire radiator that it does not have a chance to drop much heat before it is circulated back into the hot engine. With the fast turnaround what little heat coolant did release gets replaced and more added until overheating fast becomes an issue.

Fast flow is almost the same issue as removing a heater and connecting the block outlet and inlet heater ports together. In that instance a significant amount of coolant completely bypasses the radiator and never gets a chance to cool so the engine overheats.

Removing the thermostat defeats important design factors of the cooling system. Besides regulating engine temperature, the thermostat provides a calibrated restriction to coolant flow, even when open. This causes some over-pressure in the block and head, especially at higher RPM, which prevents steam pockets and improves heat transfer. It also prevents excessively low pressure on the suction side of the pump at higher RPM, which can lead to air leaks at the pump seal and even cause cavitation/bubbles in the lower radiator hose and within the pump itself at operating temperature.The calibrated restriction also prevents forcing water out the radiator overflow on non-pressurized systems. The thermostat needs to be in place, and it needs to have the same flow characteristics as the original one when fully open. Removing the thermostat will not correct a clogged radiator or malfunctioning water tube.

All heard!

If I could reframe my original question(s):

- The thermostat used as a replacement in the video is in fact a bellows-type (you see it clearly at 14:05, and he says "bellows down" for the thermostat orientation).

- Acknowledging that Packard moved away from these types t-stats in pressurized systems as they learned about its flaws in this environment, leads to questions on flow characteristics:

- did the old bellows style stats have a higher flow rate than their capsule successors, or;

- do modern standard flow capsule stats have a lower, similar, or even higher flow rate those by Packard when they made the switch? (Speculating that the standard conception of flow rates may have changed over the years).

Thanks for the excellent explanation TxGoat and Howard. It makes sense that the thermostat restriction would help increase coolant pressure in the block, exactly where it would be needed, since as I understand the water pump pumps into the block, but the thermostat is after the block. Back of the envelope calculation: water is some tens of times better at convective heat transfer than steam, so any steam bubbles forming from less-than-intended pressure would be an immediate detriment. This sounds to be a very reasonable possibility in causing an overheating condition.

Joe;
I’ve written pages and pages of a response, but they all suffer from too many words and too many numbers. I’ve run my ’48 in many cooling system configurations, and keep hitting being air-side limited with a coolant flow area anything above about 80% of a 1 inch orifice. The key is that area occurs at about 15F above the rating temperature of the thermostat. I believe most folks are expecting a coolant temperature the same as the thermostat rating temperature, but to open the valve it takes more temperature.
1.) Bellows type are for atmospheric coolant systems only. In a pressurized system the flow area becomes a function of both temperature and pressure, and I think pressure reduces the flow area . . . that was the basis of the switch over to the wax motor type of thermostat.
2.) In my experience running your engine without a thermostat is OK, but I would limit that operation when the air temperature is above 80F. Remember the upper radiator hose and/or the gooseneck exit will also act as a flow limiting orifice, and that area is close to the OEM thermostat. The origin of the belief that an engine will overheat without a thermostat is unknown to me. Based on my experience I would say it’s a myth. Perhaps a SBC at full song will result in cavitation the water pump impeller without an exit orifice at the gooseneck, but that’s not likely in a Packard I8. I’ve got a lot of hours without a thermostat and without exception that configuration produces the lowest coolant temperature . . . all else being equal.
3.) From a restoration ‘purist’s’ stand point it is the wrong thermostat, but going as far as stating a modern thermostat ‘won’t flow enough coolant’ is not consistent with the data I have logged on my ’48 288.
4.) I’ve found that a modern 2 ½ inch OD, 1 inch orifice thermostat will work, and if you are willing to add by-pass holes you can tune the coolant flow to match the expected coldest day you expect to drive your car . . . your trying to avoid a coolant temperature below some number, and as a result not being able to purge water in the oil.
5.) My testing has revealed the cooling system requires a flow area somewhere between 100% and 180% of a 1 inch diameter thermostat to have the same performance as no thermostat (likely limited by the gooseneck exit area, which is likely the area of the OEM thermostat.) Areas above 100% can be achieved with the use of by-pass holes.

dp

It would be interesting to compare wide open flow rates, at some specific pressure differential, between"modern" thermostats that will fit and the original style. I put a "modern" 180F stat in my '37 120 and it runs substantially hotter than it did with a 153F bellows type. It appears to have a more restricted flow capability, just looking at it. I drilled a 3/32" bypass hole in the flange to allow air to escape when filling the system. I suspect that my radiator is partially clogged, although water is not pushed out the overflow at higher speeds when the radiator is filled to the top. I have reason to suspect that both the car's temp gauge and my hand-held temp gun are inaccurate. The dash gauge can reach 200 F with no sign of boiling or afterboil in 100F weather. System is not pressurized, and coolant is plain water.
I got a reading yesterday of 245F on the crankcase area on the camshaft side, which I found implausible. Pouring water on the area showed no indication of boiling or excess heat. If the cast iron surface had been anywhere near 245F, it would have been immediately apparent. I would urge anyone to use the correct type thermostat in any application, and especially in old cars with non-pressure systems, and in any flat head Ford V8. The water tubhe on my engine has been replaced at some point,but I have not examined it. Water pump is a new Kanter unit that appears to be well-made and identical to the original. Elevation here is about 1200 ft. Radiator air passsages are clear. I can't see the top of the radiator tubes. The radiator does get hot all across and top to bottom.

As I understand it, the temperature rating of a thermostat is based on the temperature at which the valve STARTS to open. Full opening would probably occur at some higher temperature. I suspect that automotive thermostats are plus/minus 5 to 10% of spec. I have a 2008 Chevrolet V8 that quickly reaches 210F on the gauge and stays there, in any weather under any operating conditions. (Unless the fans fail to operate) I don't think is uses a traditional thermostat.