Cooling info. Should be a sticky

[Frank Mopar]

Engine Motor Cooling | Heat Load Exchange Temperature

 

[BillGrissom]

Engine Motor Cooling | Heat Load Exchange Temperature

Having an MS in Mechanical Engineering and having published papers on heat transfer, the misunderstandings of thermal and flow processes thrown about as facts by auto hobbyists and mechanics is amazing. I don't even try to argue with them because most don't understand the basic physics and can't grasp anything other than a relation with 1 causal variable (if A goes up, then B does this ...). Engine cooling has several independent variables, as this article discusses. Perhaps the strangest idea constantly passed is the mystical phenomenon this article addresses:
"Will it go through the radiator so fast that there won’t be time for cooling to take place? Not at all, from the expression, ..."
Strange as it seems, many hobbyists claim you must slow down the coolant flow to increase the engine cooling. To any engineer, that is bizarre. It even sounds totally non-intuitive, thus "mystical", and guess what?, it is wrong. You can find on the web a writeup by a Robertshaw engineer who also dropped his jaw when he heard that "slow the flow" concept. He did some research and found where it originated. Early cars, say a 1932 Ford, had weak springs on their radiator cap vents. At high rpm, say moonshiners on a run, the water pump pressure was so high that it caused the cap to vent and loose coolant. The solution was to add a flow restrictor to reduce the pressure. Of course, a stronger vent spring would have been better. The solution worked, but their explanation was strange and carried on. A good guess is that some geezer "experts" will try to argue with me and this article, but I can't be trolled by those without a degree.

 

[Frank Mopar]

Appreciate the expert feedback Bill G.

 

[mbaird]

Having an MS in Mechanical Engineering and having published papers on heat transfer, the misunderstandings of thermal and flow processes thrown about as facts by auto hobbyists and mechanics is amazing. I don't even try to argue with them because most don't understand the basic physics and can't grasp anything other than a relation with 1 causal variable (if A goes up, then B does this ...). Engine cooling has several independent variables, as this article discusses. Perhaps the strangest idea constantly passed is the mystical phenomenon this article addresses:
"Will it go through the radiator so fast that there won’t be time for cooling to take place? Not at all, from the expression, ..."
Strange as it seems, many hobbyists claim you must slow down the coolant flow to increase the engine cooling. To any engineer, that is bizarre. It even sounds totally non-intuitive, thus "mystical", and guess what?, it is wrong. You can find on the web a writeup by a Robertshaw engineer who also dropped his jaw when he heard that "slow the flow" concept. He did some research and found where it originated. Early cars, say a 1932 Ford, had weak springs on their radiator cap vents. At high rpm, say moonshiners on a run, the water pump pressure was so high that it caused the cap to vent and loose coolant. The solution was to add a flow restrictor to reduce the pressure. Of course, a stronger vent spring would have been better. The solution worked, but their explanation was strange and carried on. A good guess is that some geezer "experts" will try to argue with me and this article, but I can't be trolled by those without a degree.

Interesting . I must admit that I too have subscribed the Too Fast theory.
So once the engine is up to temp the thermostat does little ?

 

[BillGrissom]

...So once the engine is up to temp the thermostat does little ?

No, a T-stat is always needed to regulate coolant temperature.

The T-stat is widely misunderstood. It simply adjusts its opening to regulate the temperature at its bi-metal spring. For some reason, many mechanics seem to think it continually opens and closes as the engine runs. Perhaps they are thinking of a home thermostat which cycles your home heater on and off. That is termed "on-off" control or "bang-bang" in older textbooks. A car T-stat is a "proportional controller", using "negative feedback". My guess is that it reaches a steady-state opening position, rather than constantly oscillating open and closed, though I have no videos. The later is termed "feedback instability" and can occur in a proportional controller which has too much gain (too touchy, not desired).

One property of a proportional controller is that it can't regulate to an exact setpoint, except at one operating condition. If the heat load increases, the T-stat must open further to allow more coolant flow, in an attempt to hold the temperature setpoint. But, the only way it will open further is if its sensing spring gets hotter. Thus, it winds up at a higher steady-state temperature under a higher heat load (termed "proportional droop"). That is why you see the dash gage creep up when under load, like going up a steep mountain grade or on a very hot day. Of course, once it is fully open, it can no longer regulate the temperature. It is of course wrong to think that installing a 180 F T-stat instead of 195 F will keep your engine from overheating at say 220 F, since the 195 F T-stat should also be full-open long before then. Another reason for the dash temperature to vary is that its sensor is sometimes in a different place than where the T-stat regulates and coolant temperature varies around the engine.

There is a way to avoid the setpoint offset with feedback controller, but requires "integral action", which I don't think any simple T-stats have. Some have extra parts for "flow balance", intended to make flow forces have less effect on the opening. Higher gain lessens offset, but as-mentioned, too much gain can cause oscillation. Anyone interested can google the terms above, or peruse an engineering textbook on automated control.

If anyone has fooled with Euro engines, you will find that most have an "alien" 2-headed T-stat. One disk regulates radiator flow, as in U.S. engines and the outer one closes off the bypass circuit as the engine warms. U.S. engines, at least classics, have a fixed bypass path which never closes off, even though doing so is likely better once heated up. I think the reason for the bypass is to avoid dead-heading the water pump when the T-stat is closed (higher pressures, risk of cavitating at high rpm).

 

[toolmanmike]

No, a T-stat is always needed to regulate coolant temperature.

The T-stat is widely misunderstood. It simply adjusts its opening to regulate the temperature at its bi-metal spring. For some reason, many mechanics seem to think it continually opens and closes as the engine runs. Perhaps they are thinking of a home thermostat which cycles your home heater on and off. That is termed "on-off" control or "bang-bang" in older textbooks. A car T-stat is a "proportional controller", using "negative feedback". My guess is that it reaches a steady-state opening position, rather than constantly oscillating open and closed, though I have no videos. The later is termed "feedback instability" and can occur in a proportional controller which has too much gain (too touchy, not desired).

One property of a proportional controller is that it can't regulate to an exact setpoint, except at one operating condition. If the heat load increases, the T-stat must open further to allow more coolant flow, in an attempt to hold the temperature setpoint. But, the only way it will open further is if its sensing spring gets hotter. Thus, it winds up at a higher steady-state temperature under a higher heat load (termed "proportional droop"). That is why you see the dash gage creep up when under load, like going up a steep mountain grade or on a very hot day. Of course, once it is fully open, it can no longer regulate the temperature. It is of course wrong to think that installing a 180 F T-stat instead of 195 F will keep your engine from overheating at say 220 F, since the 195 F T-stat should also be full-open long before then. There is a way to avoid the setpoint offset with feedback controller, but requires "integral action", which I don't think any simple T-stats have that. Some do have extra parts for "flow balance", intended to make flow forces have less effect on the opening. Higher gain will lessen offset, but as-mentioned, too much can cause oscillation. Anyone interested can google the terms above, or peruse an engineering textbook on automated control.

If anyone has fooled with Euro engines, you will find that most have an "alien" 2-headed T-stat. One disk regulates radiator flow, as in U.S. engines and the outer one closes off the bypass circuit as the engine warms. U.S. engines, at least classics, have a fixed bypass path which never closes off, even though that might be better once heated up. I think the reason for the bypass is to avoid dead-heading the water pump when the T-stat is closed.

Now you are talking PID controller. LOL

 

[yellow rose]

No, a T-stat is always needed to regulate coolant temperature.

The T-stat is widely misunderstood. It simply adjusts its opening to regulate the temperature at its bi-metal spring. For some reason, many mechanics seem to think it continually opens and closes as the engine runs. Perhaps they are thinking of a home thermostat which cycles your home heater on and off. That is termed "on-off" control or "bang-bang" in older textbooks. A car T-stat is a "proportional controller", using "negative feedback". My guess is that it reaches a steady-state opening position, rather than constantly oscillating open and closed, though I have no videos. The later is termed "feedback instability" and can occur in a proportional controller which has too much gain (too touchy, not desired).



One property of a proportional controller is that it can't regulate to an exact setpoint, except at one operating condition. If the heat load increases, the T-stat must open further to allow more coolant flow, in an attempt to hold the temperature setpoint. But, the only way it will open further is if its sensing spring gets hotter. Thus, it winds up at a higher steady-state temperature under a higher heat load (termed "proportional droop"). That is why you see the dash gage creep up when under load, like going up a steep mountain grade or on a very hot day. Of course, once it is fully open, it can no longer regulate the temperature. It is of course wrong to think that installing a 180 F T-stat instead of 195 F will keep your engine from overheating at say 220 F, since the 195 F T-stat should also be full-open long before then. Another reason for the dash temperature to vary is that its sensor is sometimes in a different place than where the T-stat regulates and coolant temperature varies around the engine.

There is a way to avoid the setpoint offset with feedback controller, but requires "integral action", which I don't think any simple T-stats have. Some have extra parts for "flow balance", intended to make flow forces have less effect on the opening. Higher gain lessens offset, but as-mentioned, too much gain can cause oscillation. Anyone interested can google the terms above, or peruse an engineering textbook on automated control.

If anyone has fooled with Euro engines, you will find that most have an "alien" 2-headed T-stat. One disk regulates radiator flow, as in U.S. engines and the outer one closes off the bypass circuit as the engine warms. U.S. engines, at least classics, have a fixed bypass path which never closes off, even though doing so is likely better once heated up. I think the reason for the bypass is to avoid dead-heading the water pump when the T-stat is closed (higher pressures, risk of cavitating at high rpm).

Which is why I was told by either one of the Stewart Components guys or someone at Evans said the thermostat should be full open by the temperature it’s rated at. IIRC one of them said in certain conditions the thermostat will close and then open. I suspect that is maybe at highway speeds on days when the ambient temperature is pretty low.

 

[famous bob]

No, a T-stat is always needed to regulate coolant temperature.

The T-stat is widely misunderstood. It simply adjusts its opening to regulate the temperature at its bi-metal spring. For some reason, many mechanics seem to think it continually opens and closes as the engine runs. Perhaps they are thinking of a home thermostat which cycles your home heater on and off. That is termed "on-off" control or "bang-bang" in older textbooks. A car T-stat is a "proportional controller", using "negative feedback". My guess is that it reaches a steady-state opening position, rather than constantly oscillating open and closed, though I have no videos. The later is termed "feedback instability" and can occur in a proportional controller which has too much gain (too touchy, not desired).

One property of a proportional controller is that it can't regulate to an exact setpoint, except at one operating condition. If the heat load increases, the T-stat must open further to allow more coolant flow, in an attempt to hold the temperature setpoint. But, the only way it will open further is if its sensing spring gets hotter. Thus, it winds up at a higher steady-state temperature under a higher heat load (termed "proportional droop"). That is why you see the dash gage creep up when under load, like going up a steep mountain grade or on a very hot day. Of course, once it is fully open, it can no longer regulate the temperature. It is of course wrong to think that installing a 180 F T-stat instead of 195 F will keep your engine from overheating at say 220 F, since the 195 F T-stat should also be full-open long before then. Another reason for the dash temperature to vary is that its sensor is sometimes in a different place than where the T-stat regulates and coolant temperature varies around the engine.

There is a way to avoid the setpoint offset with feedback controller, but requires "integral action", which I don't think any simple T-stats have. Some have extra parts for "flow balance", intended to make flow forces have less effect on the opening. Higher gain lessens offset, but as-mentioned, too much gain can cause oscillation. Anyone interested can google the terms above, or peruse an engineering textbook on automated control.

If anyone has fooled with Euro engines, you will find that most have an "alien" 2-headed T-stat. One disk regulates radiator flow, as in U.S. engines and the outer one closes off the bypass circuit as the engine warms. U.S. engines, at least classics, have a fixed bypass path which never closes off, even though doing so is likely better once heated up. I think the reason for the bypass is to avoid dead-heading the water pump when the T-stat is closed (higher pressures, risk of cavitating at high rpm).

I built a hot 406 sbc and put it in a 92 gmc short wide , and installed a 31x19 crossflow , w/ two rows of 1 1/4'' tubes and a ''big'' elec fan (lotsa room ) , it cycled off and on constantly after warmed up , u could watch the temp gauge as it opened and closed .

 

[67Dart273]

One simple thing OldManMopar posted some time back and I'm CONVINCED this happened to me. When my 67 was together, and I put the 360 in, it ran hot Rad shop claimed it was clean as a whistle. OldManMopar (Steve) says that old radiators can crack the fins loose from the tubes from many thermal cycles. This then still flows lots of water, it simply doesn't radiate much of anything.

"I just tried it" swapped in an unknown used 2 core 73 rad I had out of a slant, basically same rad as came out, and the temp dropped like a rock. I was shocked

 

[Frank Mopar]

I built a hot 406 sbc and put it in a 92 gmc short wide , and installed a 31x19 crossflow , w/ two rows of 1 1/4'' tubes and a ''big'' elec fan (lotsa room ) , it cycled off and on constantly after warmed up , u could watch the temp gauge as it opened and closed .

Are you saying the cycling is due to the fan coming on and off or the thermostat opening and closing?

 

[famous bob]

One simple thing OldManMopar posted some time back and I'm CONVINCED this happened to me. When my 67 was together, and I put the 360 in, it ran hot Rad shop claimed it was clean as a whistle. OldManMopar (Steve) says that old radiators can crack the fins loose from the tubes from many thermal cycles. This then still flows lots of water, it simply doesn't radiate much of anything.

"I just tried it" swapped in an unknown used 2 core 73 rad I had out of a slant, basically same rad as came out, and the temp dropped like a rock. I was shocked

U are wrong , I was refering to bill grishoms statement of the thermostat not opening and closing

Are you saying the cycling is due to the fan coming on and off or the thermostat opening and closing?

thermostat , had a manuall switch on the fan , didnt run it on the strip or on the hi way.
Was due to the big efficiant alum rad .