My take on stock cooling systems!

[T_C_D]

Today the thermometer registered 91F while I was stock in a traffic jam for exactly 24 minutes. Traffic moved at about a consistent 5mph for most of the 24 minutes. Never faster. I was in my 1984 745i.

Before you say it has a superior cooling system let me give you some info about it. Car was in a front end collision last fall. The stock radiator, aux fan, turbo fresh air duct and after cooling fan were all destroyed. The radiator was replaced with a stock used 733i radiator. This radiator is exactly one row narrower than the 745i radiator. The aux fan and turbo cooling stuff was not replaced!

It does have fresh and properly proportioned green coolant/water.

Now the engine temperature gauge never exceeded just left of center (11oclock) on the gauge. It moved from 10 to 11 and stayed there. This is how a properly running engine with a good cooling system operates on these cars. It has always been my experience.

This was not a staged event, I just got lucky and decided to report my experience.

[turbodan]

I've never had any problems with my stock super eta cooling system. No problems in traffic or under full load for any period of time.

[Jeremy]

Out of curiosity, does the 745i use a different pressure cap than the other m30 equipped cars?

I'm still blaming my lack of an oil cooler for my moderately elevated coolant temps.

Jeremy

[cgraff]

Now the engine temperature gauge never exceeded just left of center (11oclock) on the gauge. It moved from 10 to 11 and stayed there. This is how a properly running engine with a good cooling system operates on these cars. It has always been my experience.

This was not a staged event, I just got lucky and decided to report my experience.

In a 'stock', properly operating cooling system on an E28 the temp needle should never really go above 12:30. (Tick marks defined as: beginning of blue, 9; end of blue, 10; middle 12; start of red 2, and end of red 3). With or without A/C, in 90 deg +, humid, NJ summers it has been my experience that I rarely get to 12, and only in heat soak situations with no airflow do I see 12:30. (Note I had brand new radiator, t-stat, pump, and fan clutch all at one time).

Also had the experience in a stock cooling system but with a 71 deg t-stat instead of 80 deg t-stat, the gauge basically never gets above 11:30.

In general, I'd recommend getting your t-stat to your appropriate weather conditions (i.e. northeast drivers should use 80 deg, southern drivers a 71 deg) one drives in, and change out the aux-fan switches to 81/91 instead of 91/99 deg ones as an extra measure of safety.

-Chris

[turbodan]

It does, between 1.4 and 1.5 bar. The 535 uses a 1 bar cap. I'm pretty sure that would only be your problem if you were boiling the coolant in the system though. Its not like it cools better at a higher system pressure, it just wont boil over as easily. My oil cooler is always warm on my m20, so it is contributing to cooling to a significant extent. Before it was turbocharged it wouldn't get hot enough to open the cooler thermostat most of the time.

[Duke]

It does, between 1.4 and 1.5 bar. The 535 uses a 1 bar cap. I'm pretty sure that would only be your problem if you were boiling the coolant in the system though. Its not like it cools better at a higher system pressure

A higher pressure system will keep the coolant from boiling. A 1 bar (14.7 psi) system will boil at a lower temp than a 1.5 (20 psi) system. Thats why all turbo cars should be running at least a 1.5 bar (20 psi) cap like BMW put on the 745i. I have a 20 lbs cap and rarely go above 190 F now.

[Jeremy]

Its not like it cools better at a higher system pressure, it just wont boil over as easily.

It cools better because the coolant doesn't "micro-boil" in the water jackets around the cylinders. Whatever term you want to use, the cooling system is removing a LOT of heat from the cylinders when you're on boost, which can cause the coolant boil locally in the jackets and overheat the engine without the temperature gauge reading significantly hotter.

M635 actually explained the concept pretty well at one point, but I can't find the post.

Jeremy

[T_C_D]

I have a 20 lbs cap and rarely go above 190 F now.

I heard you car was dumping coolant regularly at 5erFest. Didn't you hold the drive up several times?

[Duke]

I have a 20 lbs cap and rarely go above 190 F now.

I heard you car was dumping coolant regularly at 5erFest. Didn't you hold the drive up several times?

Nope, never had to add coolant. It would push some out into my ghetto overflow bottle and then suck it back in once the engine cooled. Once during the drive the cars temp spiked, I pulled over to check it (not holding up any traffic) and everything was fine (settled back to 180 F). Proceeded on with the drive with no issues. Was a mystery, thats for sure.

You would not have had to "hear" about it if you were there

[T_C_D]

I have a 20 lbs cap and rarely go above 190 F now.

I heard you car was dumping coolant regularly at 5erFest. Didn't you hold the drive up several times?

Nope, never had to add coolant. It would push some out into my ghetto overflow bottle and then suck it back in once the engine cooled. Once during the drive the cars temp spiked, I pulled over to check it (not holding up any traffic) and everything was fine (settled back to 180 F). Proceeded on with the drive with no issues. Was a mystery, thats for sure.

You would not have had to "hear" about it if you were there

The fact is that you still have cooling issues and all of the changes that you have implemented have not solved the problem. So it's difficult to gather any discernable knowledge from your adventure at this point.

What I do know is that a stock cooling system in proper working order is up to the task. 20psi cap may be better but certainly is not necessary. I do include the 71C thermostat with the kits fro a margin of safety.

Todd

[Duke]

The fact is that you still have cooling issues and all of the changes that you have implemented have not solved the problem. So it's difficult to gather any discernible knowledge from your adventure at this point.

I agree, Zane was very perplexed (we talked for hours) too and he has some great experience with FI M30s. Eliminating the HG and that leaves only a cracked head (very unlikely due to Paul Burke just going through it) or a cracked block (more likely due to the blocks sorted history).

Come August, the car will be essentially parked for 10 months due to me walking or biking to school and living on post. I am seriously considering pulling the engine and have Paul Burke check it out. Paul Burke's shop will be 40 minutes away from where I live. That will eliminate any question with the engine. With many months to do this and not needing the car, it makes it relatively painless.

The engine runs like a dream though. Just ask BNC.

As far as caps and the effect of different pressures, put a 13 lb (US M30) cap on your 745i and then see what the temp does. I am betting you will then see the benefit of the 20 lb cap.

[Boru]

Raising from a 1 bar to 1.5 or 2 bar radiator cap will have no effect on the coolant temperature... boiling point, yes, but not the temperature.

[shifty]

Paul Burke's shop will be 40 minutes away from where I live.

If you drive 200 MPH.

[shifty]

M635 actually explained the concept pretty well at one point, but I can't find the post.

Here you go. It came from this thread

Localized overheating occurs after coolant transitions from nucleate boiling to film boiling and the coolant ceases to absorb heat. Nucleate boiling providing very effective cooling when bubbles of steam coming off the surface are quickly swept away by new coolant. Once the transition to film boiling occurs, there are no "steam bubbles" but only steam. The coolant passes over this layer of steam but does not absorb its heat to a significant degree which results in the surface temperature raising. When this happens, there is effectively no coolant (or cooling) against the surface and temperatures raise precipitously causing conditions which may not show up on a "temperature gauge" or in a "bin file."

To say localized overheating of the cylinder head or too high a RA on its surface causes head gasket failure is technically incorrect; they do not cause failure but they do quite often accompany head gasket failure.

There was no excuse for this and Duke has to own the fact that the radiator should never have been in his car. He was cautioned after the last head gasket failure that the radiator and in fact entire cooling system should be inspected and replaced if and where necessary. This was not "so hard to diagnose" this was turbocharging 101.

The reason a restricted radiator contributes to localized overheating is that cooling occurs as a function of temperature differentials; the difference between coolant temperature entering the radiator and coolant temperature exiting the radiator. When coolant exiting the radiator is too hot, the cylinder head temperature is bumped up a notch and, in a turbocharged engine that may be all that's required to push coolant temperatures high enough to allow film boiling of the coolant. When this occurs, thermal hysteresis resists the return to nucleate boiling. For what it’s worth, whenever a turbocharger is installed, the cooling, oiling, fuel, and ignition system should be blueprinted.

[Skeen]

I guess I should just let you fix my car since apparently I don't know how to make a "proper stock system."

[T_C_D]

I guess I should just let you fix my car since apparently I don't know how to make a "proper stock system."

I dunno?

I have honestly never modified a stock BMW cooling system other than the thermostat.

[Skeen]

Me either. The only thing I can think of being a problem is HE placement, which might put more hot air through the rad on my car. The aluminum rad seems to be working fairly well, but I need to find a fan to put on it.

[Boru]

The point of this thread is that it is our (TCD's) experience that the stock cooling system, when healthy, has no problems maintaining normal coolant temperatures.
The twin turbo 635 has over 45K miles at over 1 bar boost and the temp has never gone over 210F. It's normally 180-185F on an actual temperature gauge... with numbers :o)
It has a 1 bar cap, stock mechanical fan and an aux. fan wired to the stock temp switches, 85C thermostat.
If your gauge is reading high, check the temp of the hoses, radiator, etc. with an IR thermometer or contact thermometer. Your gauge may be off. Check the circuit and sender. The gauge is a simple volt meter with a thermistor/variable resistor as a sender. Extra resistance in the circuit due to corrosion, etc. will effect the reading on the instrument.
Placing anything in front of the radiator (except an operational fan) will reduce air flow.

[cgraff]

It has a 1 bar cap, stock mechanical fan and an aux. fan wired to the stock temp switches, 85C thermostat.

85 C t-stat? :-/

If your gauge is reading high, check the temp of the hoses, radiator, etc. with an IR thermometer or contact thermometer. Your gauge may be off. Check the circuit and sender. The gauge is a simple volt meter with a thermistor/variable resistor as a sender. Extra resistance in the circuit due to corrosion, etc. will effect the reading on the instrument.

Very true....
Or...like you said get a gauge with #s on it and an appropriate sender.

-Chris

[rob6118]

Just a quick note here, I recently was playing around replacing all the coolant sensors in the thermostat housing including the housing itself as well. I replaced the lost coolant with a half bottle of pure bmw coolant and apparently that is enough with this cooling system to make it run warm I went from running consistent 10-11:30 temps to 11-1 temps.

This morning I ran out and put about 3/4 of a bottle of straight water by loosening the bleed screw and filling the reservoir and now I'm back to my 10 am temps with the a/c on and idling in the parking lot for 20 mins.

So fyi, water to coolant ratio in this car appears to be pretty sensitive. But when I replaced my radiator, water pump, and thermostat a year ago I did the exact 50:50 ratio and the cooling system ran probably the best out of all the cars I've ever had especially running in Florida's bumper to bumper freeway parking lot traffic with full a/c.

Rob

PS I also found out that the coolant sensors do not respond well to certain types of washers. I used some steel washers and apparently they didn't provide a suffecient ground to the sensors.

[Duke]

This morning I ran out and put about 3/4 of a bottle of straight water by loosening the bleed screw and filling the reservoir and now I'm back to my 10 am temps with the a/c on and idling in the parking lot for 20 mins.

I believe all you did is properly bleed the system. Anti-freeze will not have that much effect in the amount you are talking about. You just had air in the system.


EDIT - my bad, I thought you meant 1/2 a coolant tank not 1/2 gallon of anti-freeze.

[Boru]

85 C t-stat? :-/

Why? Okay, it may be an 80. It's been 7 years since it went in and I got it at NAPA... pretty sure it's an 85. I know it isn't a "low temp" t-stat.

[cgraff]

85 C t-stat? :-/

Why? Okay, it may be an 80. It's been 7 years since it went in and I got it at NAPA... pretty sure it's an 85. I know it isn't a "low temp" t-stat.

Never heard of an 85 t-stat in the M30s of the era (1980s). Only the 71, 75, and 80 deg t-stats, of which the 80 is stock, and therefore would be carried by all parts suppliers.

On another note, the newer engines (M50s, M52s, M54s, S52s, etc) use 88, 91, 92 deg t-stats.

-Chris

[Bill in MN]

FWIW, I flushed my system the other day and decided to see what would happen if water only (no AF) was in the system.

Ambient temps were in the low 80's. With copious amounts of boost (12psi range), engine temps were sub 190. Leaving it idling for about 20 mins. brought it to 205 and it stayed there. The idle temps are pretty much irrelevant, though. I'm also not using the stock gauge. Where this would be on the "clock", I don't know.

Time allowing, this weekend I'll go back to 50/50 and see where the temps come in at.

[Rich Euro M5]

Just a quick note here, I recently was playing around replacing all the coolant sensors in the thermostat housing including the housing itself as well. I replaced the lost coolant with a half bottle of pure bmw coolant and apparently that is enough with this cooling system to make it run warm I went from running consistent 10-11:30 temps to 11-1 temps.

I've noticed the same thing after replacing the HG on my '86 535i. I used Zerex G-05 and mixed it per the coolant manufacturers recommendation of 60/40, coolant / H2O. After the repair I noticed the temps running higher than previously. I purged again and added straight distilled water to the reservoir and the temps dropped down to what they had been before the repair.

This makes perfect sense since straight water has better heat transfer characteristics than a glycol/water mixture. According to the Zerex temp charts (15 psi cap) a 70/30 mix is good for -84F to 276F, 60/40 -62F to 270 F, 50/50 -34F to 265F. Therefore for typical Houston Texas year round temps I could probably run a 40/60 blend with a 15 psi cap and improve cooling efficiency.

Rich

[Boru]

Never heard of an 85 t-stat in the M30s of the era (1980s). Only the 71, 75, and 80 deg t-stats, of which the 80 is stock, and therefore would be carried by all parts suppliers.

-Chris

Lets compromise at 82.5

[Duke]

This makes perfect sense since straight water has better heat transfer characteristics than a glycol/water mixture. According to the Zerex temp charts (15 psi cap) a 70/30 mix is good for -84F to 276F, 60/40 -62F to 270 F, 50/50 -34F to 265F. Therefore for typical Houston Texas year round temps I could probably run a 40/60 blend with a 15 psi cap and improve cooling efficiency.

Rich

I see no reason not to run straight distilled water with Red line "water wetter" for corrosion protection in the spring/summer and fall.

[turbodan]

Raising from a 1 bar to 1.5 or 2 bar radiator cap will have no effect on the coolant temperature... boiling point, yes, but not the temperature.

Thats what I'm saying.

[Rich Euro M5]

This makes perfect sense since straight water has better heat transfer characteristics than a glycol/water mixture. According to the Zerex temp charts (15 psi cap) a 70/30 mix is good for -84F to 276F, 60/40 -62F to 270 F, 50/50 -34F to 265F. Therefore for typical Houston Texas year round temps I could probably run a 40/60 blend with a 15 psi cap and improve cooling efficiency.

Rich

I see no reason not to run straight distilled water with Red line "water wetter" for corrosion protection in the spring/summer and fall.

I could do that, but then I would have to drain and refill the system in the winter. Since I don't care to mess around with the cooling system I'll run a mixture. I'm still running an 80C stat, so I could install the 70 C stat and the lower temp fan switches and get all the cooling I need.

BTW: Humidity plays a big part in cooling system efficiency. High humidity equals high density altitude and less cooling from the airflow passing through the radiator.

Rich

[Duke]

I could do that, but then I would have to drain and refill the system in the winter. Since I don't care to mess around with the cooling system I'll run a mixture.

You are also not FI. It is well worth the 30 minutes to do it for the FI application.