N54 Cooling - Temp Control Logic and what are your temps during extended track use?

scrllock

Sergeant
Dec 17, 2018
267
152
0
Ride
2011 335xi
Haha, no, though I've been fighting leaks for months, just recently replaced my rad with an oem one and trying not to second-guess it too much.
The concern is over-cooling, particularly since I drive the car at subzero temps on the highway. Lots of reviews of the csf unit with thermostat warnings, no heat, etc. Unclear to me if that's specific to the AT version however, or if any of the coolant logic discussed here could be used to negate this.
And yes, it works similarly, there's a "heater core" that transfers heat between the oem trans cooler and the radiator. The trans cooler kits I've seen plug it off and run independently, but I would probably try to have it in sequence with a t-stat to accommodate winter here.
 

Asbjorn

Lieutenant
Mar 10, 2018
854
602
0
European, based in China
Ride
Z4 N54 DCT
Haha, no, though I've been fighting leaks for months, just recently replaced my rad with an oem one and trying not to second-guess it too much.
The concern is over-cooling, particularly since I drive the car at subzero temps on the highway. Lots of reviews of the csf unit with thermostat warnings, no heat, etc. Unclear to me if that's specific to the AT version however, or if any of the coolant logic discussed here could be used to negate this.
And yes, it works similarly, there's a "heater core" that transfers heat between the oem trans cooler and the radiator. The trans cooler kits I've seen plug it off and run independently, but I would probably try to have it in sequence with a t-stat to accommodate winter here.

Please show/link to an example/log of a CSF equipped N54 that can't get up to temp in freezing ambient temps. Personally I have had no such problem, and since we have an electric water pump that can turn off or slow down to heat the engine, as well as a thermostat that can close the main radiator, I am not sure it is possible (even if the csf radiator was an actual significant upgrade, which is really isn't)?

Sounds more like examples of the thermostat failing?
 
Last edited:

scrllock

Sergeant
Dec 17, 2018
267
152
0
Ride
2011 335xi



Those are the ones I was able to quickly find that were firsthand accounts. I'm sure it's a little bit of exaggeration that's spread, but it gets down to -28C (if i'm converting properly) here sometimes, and a working heater is a must-have QoL item.
 

Asbjorn

Lieutenant
Mar 10, 2018
854
602
0
European, based in China
Ride
Z4 N54 DCT



Those are the ones I was able to quickly find that were firsthand accounts. I'm sure it's a little bit of exaggeration that's spread, but it gets down to -28C (if i'm converting properly) here sometimes, and a working heater is a must-have QoL item.

Very interesting!!

Alot of examples here where coolant wont go above 140-160F, and cabin heater doesn't work. The thermostat should be shutting off the main radiator at those temps, and 140F is plenty hot for cabin heating. So what is going on?

Now because the cabin heater is on, the water pump is running at low speed and the thermostat is off, sending coolant to the engine, heater and transmission heat exchanger connected to the AT core. Now someone mentioned that there's no separation between the AT core and main radiator on the CSF.


This might explain everything. It means that coolant still flows through the entire radiator entering in two places and then exiting through the transmission heat exchanger, even though the thermostat is closed. So the flow that should have gone to the cabin heater, actually enters the radiator instead and exits though the transmission heater. And the entire radiator is active, although at reduced flow, even when the car is warming up, and the thermostat is closed. This also explains why pinching the return line from the transmission heater helps, as more coolant is directed through the cabin heater instead of the entire radiator.

I will try and check my oem radiator to see how it was originally supposed to flow.
 
  • Informative
Reactions: doublespaces

scrllock

Sergeant
Dec 17, 2018
267
152
0
Ride
2011 335xi
That sounds like a pretty reasonable explanation given what I've read and the solutions people have proposed (most just tape over the kidney grilles it seems). Unfortunately no one that's tried it seems to have logged trans temps. I think @Payam@BMS might have done some with their transmission cooler, but they separate it from the heat exchanger and I think they said they only tested down to near-freezing, not sub-zero. I know this is a little off-topic given track use, but I'd be eager to find a good year-round solution that can keep the heater core and trans fluid adequately warm.
 

Asbjorn

Lieutenant
Mar 10, 2018
854
602
0
European, based in China
Ride
Z4 N54 DCT
That sounds like a pretty reasonable explanation given what I've read and the solutions people have proposed (most just tape over the kidney grilles it seems). Unfortunately no one that's tried it seems to have logged trans temps. I think @Payam@BMS might have done some with their transmission cooler, but they separate it from the heat exchanger and I think they said they only tested down to near-freezing, not sub-zero. I know this is a little off-topic given track use, but I'd be eager to find a good year-round solution that can keep the heater core and trans fluid adequately warm.

I found my old oem radiator. The AT and main cores are indeed separate. Some 6 rows or 13-14% of the radiator is used for heating and stabilizing the DCT.

Now we just need to confirm if the rumour about the CSF AT not having separate cores is true.
 

houtan

Lieutenant
Nov 2, 2017
611
293
0
Ride
135i N55 DCT; PS2
I found my old oem radiator. The AT and main cores are indeed separate. Some 6 rows or 13-14% of the radiator is used for heating and stabilizing the DCT.

Now we just need to confirm if the rumour about the CSF AT not having separate cores is true.
Based on the response i received from csf a few years ago, 7046 has separate circuits as well. I had emailed them asking about plugging up the holes to the AT heat exchanger if I got rid of it one day. This was the response. Not sure if anything has changed since then.
 

Attachments

  • 7A9CF45E-C1D9-4EC4-B840-99AE5856EC70.png
    7A9CF45E-C1D9-4EC4-B840-99AE5856EC70.png
    358.6 KB · Views: 74
  • Informative
Reactions: Asbjorn

Asbjorn

Lieutenant
Mar 10, 2018
854
602
0
European, based in China
Ride
Z4 N54 DCT
Finally got around to do a proper A-B test of the second aux radiator I installed behind the FMIC.

IMG_20190823_170030.jpg


I have a ball valve on the hose going to it, making it easy to do back-to-back testing. Here's the results form the test, it seems to reduce peak coolant by 5F:

Weather at the time of logging:
Day, clear
26C / 79F indicated in cluster
25C / 77F indicated on weather app
36% rH humidity on weather app
1011mb pressure on weather app
almost no wind
not far from sea-level.

First test: All coolers on + WI:
Peak iat 111F 44C
Peak dct 217F 103C
Peak oil 262F 128C
Peak coolant 223F 106C

Second test: As test number 1, but with the 2nd aux radiator turned off:
Peak iat 111F 44C
Peak dct 210F 99C
Peak oil 262F 128C
Peak coolant 228F 109C

First test:

 

Asbjorn

Lieutenant
Mar 10, 2018
854
602
0
European, based in China
Ride
Z4 N54 DCT
Second test

 
  • Like
Reactions: Torgus and MDORPHN

gmx

Specialist
Dec 8, 2017
69
29
0
Ride
BMW 1M

119C peak oil temp after 3 consecutive laps on this circuit. Same config, same oil. Coolant was at 89C down the straight. Ambient temp around 22-27C peak. Track temp would've been pretty high as it clear with the sun blazing and the tyres didn't like it and the lap times showed. My fastest, relatively mistake free runs were the morning and very last session on my last lap in the afternoon. Humidity was 71% in the morning and 53% by the last session.

Not my best lap but only recorded session. No logs since I lent my cable to another N54'r chasing power reduction issues throughout the day.

Next year/season is R comp time, avg RPM should be higher. Let's see the data then.
 

Asbjorn

Lieutenant
Mar 10, 2018
854
602
0
European, based in China
Ride
Z4 N54 DCT

119C peak oil temp after 3 consecutive laps on this circuit. Same config, same oil. Coolant was at 89C down the straight. Ambient temp around 22-27C peak. Track temp would've been pretty high as it clear with the sun blazing and the tyres didn't like it and the lap times showed. My fastest, relatively mistake free runs were the morning and very last session on my last lap in the afternoon. Humidity was 71% in the morning and 53% by the last session.

Not my best lap but only recorded session. No logs since I lent my cable to another N54'r chasing power reduction issues throughout the day.

Next year/season is R comp time, avg RPM should be higher. Let's see the data then.

Amazing to see that although it got warmer, your coolant improved while oil got at tiny bit worse since last time. Maybe you were just pushing for longer? Oil takes some time to get up to temp while coolant stabilizes more quickly. Anyway, 90C coolant in combination with 120C oil is really a dream come true for any track driven N54s. Some logs and exact information about ambient conditions would be great next time as you said.

Also really impressed with your driving. What kind of power does your car produce? Since your engine temps are so awesome, in your case I would definitely be more concerned about oil pressure.
 

gmx

Specialist
Dec 8, 2017
69
29
0
Ride
BMW 1M
Yes oil pressure, and MSL's CNC'd billet housing thermostat for easier daily driving is next. I'm still thinking of a logging setup that would accomplish my needs. 0-5v oil pressure sensor, RaceChrono + 10hz ext GPS. Overlay with the MHD logs - somehow. I don't like the idea of overlaying things. Sounds inaccurate apart from video which is just for show imo. I also don't want to extend the budget and scope to vbox, aim, Motec etc.

The tune is custom around OEM turbos peaking at 16psi and taper to 13psi by 6000rpm and 12psi by 6500rpm.
 
  • Informative
Reactions: Asbjorn

berns

Corporal
Jan 15, 2018
175
266
0
Ride
'09 135i
Full session at Buttonwillow in 80*F ambient. Probably about 7 sustained hotlaps with laptimes averaging about 1:55
Garrett GTX3076r with a 16.8psi peak and CES Motorsports kit
E40 fuel
Twin Setrab 19-row oil coolers with RedLine 5w40
CSF Intercooler
CSF Radiator with distilled water only



GP5j8qql.jpg


rPGL8h8h.jpg
 
Last edited:

Asbjorn

Lieutenant
Mar 10, 2018
854
602
0
European, based in China
Ride
Z4 N54 DCT
Nice lap! It seems those 1-series really like to dance with all that torque and stiff suspension.

To me the log confirms that you are mostly radiator limited in terms of cooling. But it seems you could easily do twice the number of laps, ie half an hour, without ever overheating at that pace and power level.

Your oil and coolant temps were really low in pit lane, were you running MHD track cooling?
 

Asbjorn

Lieutenant
Mar 10, 2018
854
602
0
European, based in China
Ride
Z4 N54 DCT
Stumbled on this relevant cooling overview on another site: https://f87.bimmerpost.com/forums/showpost.php?p=23595167&postcount=223

Interesting how far they went to separate oil and water temps.

Lots of good info there.

He later added two M4 aux radiators and a booster pump to drive them afaik. However the latest track videos he posted still showed pretty high coolant temps on track. Not limp mode high, but still way higher than Barry set out to achieve in this thread.

I tried to contact him by mail back when I was adding an aux radiator booster pump myself, but never heard back.
 

rac

Sergeant
Nov 14, 2016
341
204
0
Australia
Ride
135i ST
Anyone had experience with aftermarket electric water pumps (from other platforms perhaps)? I'm considering ditching the stock pump as they are getting pricey locally to me, and I've had issues with cracked housings etc. Meziere, Davies Craig and others appear to have options that might be more durable. Compact size with sufficient flow rate is probably the criteria given our tightly packed engine bays. Running syvecs, not worried about control. Some of them have their own control units in any case.