[paradoxical3]

Hi everyone,

As someone who has run into what I believe are coolant-related power cuts on the track, I decided to take a chance and purchase the CSF Racing N54/N55 automatic aluminum racing radiator. There really isn't much info about upgraded radiators on this platform, so I figured I would do a comprehensive review.

When I first started tracking my car, I wasn't a fast enough driver to really stress the cooling system. But as I have progressed into the intermediate and advanced run groups the car has started to show weakness in it's cooling system, particularly during hot days when I am running high boost.

While oil temps are a known issue on this platform (less so for the N55 than the N54), I feel like coolant temps have been largely neglected. In fact, according to BMW's chart, our cars don't pull power due to oil temps until they are over 300 degrees. High coolant temps begin pulling power far earlier, in the 240f range. Worse, there is no easily accessible coolant temp gauge and there are no idrive warnings until you are in the danger zone.

There are two events in particular that made me realize how bad the stock radiator was. One was when I was on the track in Colorado at Pueblo Motorsports Park - the elevation here is much higher than my home track Mid Ohio, and the car felt sluggish after a few laps. This was the only time when the car ever displayed an iDrive message warning of high coolant temps (not oil) and put me into a full on limp mode.

The other was just recently at the National Corvette Museum Motorsports Park - this track is very fast, and has two things in particular that were very hard on the car. The first was a second gear corner after a very fast section. This is a worst-case combination of heat buildup during the straight followed by high RPM in 2nd gear and low airflow due to low speeds. The second is a ton of WOT. This track has a significantly higher percentage of WOT time than most tracks I have been on. I could notice the car significantly down on power after about 20 minutes (we were running 30 minute sessions).

As we all know, on automatic cars the transmission is cooled by the radiator core as well. This only makes things worse. Enter the CSF Racing automatic radiator, one of the very few radiators on the market that has the proper plumbing to be used with an automatic N55/N54 car. My hope is that this radiator will prevent limp mode, enable me to keep maximum power output through a hard-driven 30 minute session, and further cool the transmission.

I will do this review in two parts. The first is directly comparable temperatures between the stock radiator and CSF radiator in identical ambient temps. I will take readings both at idle cruise on the highway and after 5 consecutive 3rd gear through 5th gear pulls. My goal here is to demonstrate repeatable results and be as scientific as possible.

The second part will be track performance - I will be at Mid-Ohio this saturday and sunday and will be keeping track of coolant temps. The goal is to remain below the threshold when the car cuts power. For reference, the car hits about 140mph down the back straight - so it is making a lot of power (400+whp) and gets very hot.

The final interesting thing will be to see if the radiator has any impact on oil temps. I have hundreds of data points for oil temps at the track, and while a radiator shouldn't cool oil as much as a second dedicated oil cooler would, theoretically a cooler block will have cooler oil


Initial impressions of the unit:

This thing is light. On the scale it checks in at 8.62lbs, which I am guessing is significantly lighter than the stock radiator. I will weigh the stock rad tomorrow after it gets installed.

The radiator is packaged very well, in a sturdy cardboard box. There is a blue plastic fin protector that is securely wrapped around the unit. Upon removing the protector, I did notice several bent fins, but I believe these came from the factory. Overall, the bent fins were so minor as to not matter whatsoever in performance terms.

Welds: Some of the welds are not very pretty. That being said, they are perfectly functional and I believe CSF racing deserves a pass. It is extremely difficult to weld aluminum, and the price point for this radiator is low enough where I do not expect nor need pretty welds. I lightly pressurized the unit and did not find any leaks.

Finish: The polished finish is beautiful! They really did a great job on this, and it looks great. It won't matter much once it is installed, but it is a really nice piece to look at.

Cooling features: Upon first glance, one might notice that this radiator has several less rows than the stock radiator. This does not mean reduced performance, however. CSF racing uses a patented construction method called "B-Tube" technology. What this does is provide B-shaped tubes instead of the standard "O" shaped tubes, allowing a claimed 15% surface area advantage over a standard tube. This technology is proprietary to CSF racing and no other manufacturer uses it. Does it matter? Who knows - we will let the performance results speak for themselves.

One other note - I have had really pleasant interactions with CSF Racing while asking questions and ordering the radiator. They respond to emails very quickly and frankly I a grateful to CSF Racing for being willing to take a chance on our market. There really is a lack of radiator choices for automatic cars and CSF was willing to produce a seemingly high-quality one for our application.

Here are some pics to tide you over until data is in this weekend.

Data:

Stock radiator: 58f ambient, air conditioning set to 66, 6th gear cruise at 2500rpm: 108C consistently
Stock radiator @ cruise delta t: 93.6C
Stock radiator after 5 WOT 3-5th gear pulls: 110C

Upgraded radiator: 108C in same conditions as above
Upgraded radiator @ cruise delta t: 93.6C
Upgraded radiator after 5 WOT 3-5th gear pulls: 108C

Track max temp: 111c (99% of the time 108C) in 80f ambient
Pit-In temp: 92C, recovering to 108C after sitting for a while in 80F ambient

EDIT: Track review!

I have just returned from two days of hammering on the car at Mid-Ohio in 75-80F temps (which means even hotter on the track) and I am 100%, completely satisfied with the radiator. The radiator completely cured my coolant limp mode issues.

To give an idea of the type of load it is under, I am running:

N55 engine with Pure Turbo Stage 1 @ 18psi with 40% e85 and the rest 100 octane
FBO mods (downpipes, intercooler, chargepipe, intake, etc, etc)
JRZ suspension
M3 subframe bushings, sways, guide rods
Nitto Nt01 r-comp tires

I run the car hard and hit about 140mph down the back straight. I purposely ran the entire 25min sessions hard without cooldown laps except for traffic to see what the radiator can do. It stays at 108C (which is the temperature the car's ECU maintains on the street or just at idle) the vast majority of the time. Very occasionally, if I hit traffic after a 3-6th gear pull towards the end of the session and have to slow down to 40mph and get stuck behind a car, the temp will jump to 111C and stay there for a minute or so. But the minute I start moving again the temperature drops back down to 108C. It is clear that the thermal recovery time of the CSF radiator is quite fast.

This is significantly more effective than the stock radiator. For the first time, I could drive my car hard without even thinking about what the coolant temp was. That is priceless.

Given the cost of track time, this radiator pays for itself by allowing you to get the most out of your car during extended sessions or hot ambient temps.

I completely recommend it.























image hosting site

[sirdaft1]

Subbed! Looking forward to seeing results w/ actual data! And are you running the same antifreeze or a water/water wetter combo?

[Cloud9blue]

You can actually access the coolant temp through the diagnostic menu, under menu 7, after you unlock it.

http://e90.wikifoundry.com/page/BC+hidden+menus

From what I found it with mine, the coolant temp takes quite a while longer to reach the usual 94C during typical street driving in cooler weather (>40F). I run the standard 50/50 anti freeze with distilled water the last two time I was on track, not a hit of overheating at all. The temp hovers below 100C according to the JB4 dash hi-jack.

From my opinion, CSF radiator is a much more effective cooling upgrade than the PPK radiator, particular for AT cars. And the prices on these things are quite reasonable too!

[paradoxical3]

Originally Posted by Cloud9blue

You can actually access the coolant temp through the diagnostic menu, under menu 7, after you unlock it.

http://e90.wikifoundry.com/page/BC+hidden+menus

From what I found it with mine, the coolant temp takes quite a while longer to reach the usual 94C during typical street driving in cooler weather (>40F). I run the standard 50/50 anti freeze with distilled water the last two time I was on track, not a hit of overheating at all. The temp hovers below 100C according to the JB4 dash hi-jack.

From my opinion, CSF radiator is a much more effective cooling upgrade than the PPK radiator, particular for AT cars. And the prices on these things are quite reasonable too!

Yes, that is how I am measuring the coolant temp. It is interesting to me that your cruise temp is 94c..wonder why it is so much lower than mine. Your 94c cruise is on both your stock and CSF rad?

[Cloud9blue]

Originally Posted by paradoxical3

Yes, that is how I am measuring the coolant temp. It is interesting to me that your cruise temp is 94c..wonder why it is so much lower than mine. Your 94c cruise is on both your stock and CSF rad?

It takes a LONG time to reach 94C that last time I logged it. Like 30 mins of highway driving and only after a few WOT pull. Before that, it hovers at around mid 80C most of the time. But again, I also have aux. radiator installed (check my mod list).

I bet it has something to do with the thermal management logic in the DME. But who cares about the street temp, as long as the oil temp gets hot enough for it to work properly (+180F), it is good enough for me (you might lose a mpg or two with coolant than usual coolant temp though). Track temp is where it really counts.

[paradoxical3]

Just got back from the track and updated the first post with my experience.

In short, the radiator is awesome. It completely cured my coolant temp limp mode issues and enabled me to run full boost lap after lap after lap.

[Cloud9blue]

Good data. And you are still running with stock oil cooler correct? What's the oil temp like?

[paradoxical3]

Originally Posted by Cloud9blue

Good data. And you are still running with stock oil cooler correct? What's the oil temp like?

Yes, stock oil cooler. That is my next upgrade, I am having a dual core system fabbed with an in-line tstat.

Oil temps were high when really pushing the car, around 270-275. This is higher than they have been before, most likely because previously the coolant limp mode would not allow me to run full power for long enough to heat up the oil.

One note is that my oil cooler is beat to hell - it has been running without the grill for 50k miles and a ton of track days, so the fins are all bent and loaded with melted rubber and debris kicked up from the tires. It could probably use a good pressure washing.

[Cloud9blue]

Originally Posted by paradoxical3

Yes, stock oil cooler. That is my next upgrade, I am having a dual core system fabbed with an in-line tstat.

Oil temps were high when really pushing the car, around 270-275. This is higher than they have been before, most likely because previously the coolant limp mode would not allow me to run full power for long enough to heat up the oil.

One note is that my oil cooler is beat to hell - it has been running without the grill for 50k miles and a ton of track days, so the fins are all bent and loaded with melted rubber and debris kicked up from the tires. It could probably use a good pressure washing.

Dual oil cores might be unnecessary in your case, as with the single turbo N55 engine runs a bit cooler than the N54. My oil temp has not raised much above 250F from the last two times I took mine to the track, granted the ambient temp was only around mid 50F. And this is with just the single core Dinan setup.

I think the coolant issue has long been overlooked on these engines. Limp mode doesn't trigger until 300F for the oil temp (not that is a good thing...), but most of the track data suggests people experience limp mode long before the oil temp reaches 300F, indicating it is the coolant temp that is causing the issue.

There is a reason why 335is and 1M come with an auxiliary radiator instead of an additional oil cooler from the factory...

[paradoxical3]

Just had an experience at Putnam Park yesterday that really hit home how great this radiator is.

I was in the middle of a hot lap, when just after a 3-4 gear acceleration zone I saw a red flag from a wrecked S2000. I quickly came to a stop and the car was absolutely baking. To give you an idea of how hard I was running, I could see smoke from my stoptech BBK and performance friction PFC01 pads coming over my hood. The coolant temps were 105 when I was accelerating, but once I stopped they climbed up to 113C peak. While I was sitting there, the coolant temp never once climbed above 113C and slowly returned down to 109C with no airflow whatsoever besides the fan.

I have no doubt that with my stock radiator I would have hit coolant limp mode here.

[Raptor04]

How much did the radiator cost you and where did you buy it from?

[Cloud9blue]

Originally Posted by Raptor001

How much did the radiator cost you and where did you buy it from?

http://www.e90post.com/forums/showthread.php?t=1089057

You can either buy them at full price with any of the online seller or contact CSF directly (see the thread I linked for info) to see if they would still honor an intro discount.

[Cloud9blue]

paradoxical3

How is the fitment on yours? Mine fits perfectly, but there is another forum member complaining that his mounting slot for the radiator shroud on the passenger side was off by quite a bit.

[paradoxical3]

I bought mine direct.

Fitment was perfect on mine...no complaints. It is a bit of a tight area to get it installed but all of the connections lined up exactly.

[whosdady]

I've noticed my car running cooler since I installed this as well... I also like the increased coolant capacity.

[Cloud9blue]

Originally Posted by whosdady

I've noticed my car running cooler since I installed this as well... I also like the increased coolant capacity.

You fixed your fitment issue?

[whosdady]

Originally Posted by Cloud9blue

You fixed your fitment issue?

It fits, these are hand made and one of the slots is too far over. (It doesn't touch) Ravi will be sending me out a new one in a couple weeks. He didn't have anymore in stock. He said they sold fast... Besides the slot not fitting perfectly, I'm very happy with it and Ravi has offered nothing less than excellent customer service!

[kimoboy]

I would like to know will it lower ATF temp.

[wgknestrick]

Am I the only one failing to see a significant benefit to this? 2C at WOT is not much of a difference at all. This could probably be achieved with better fans. You are still right on the edge of thermal throttling. It does not appear that the radiator is the weak link (assuming this radiator can shed more BTUs than stock).

[Cloud9blue]

Originally Posted by wgknestrick

Am I the only one failing to see a significant benefit to this? 2C at WOT is not much of a difference at all. This could probably be achieved with better fans. You are still right on the edge of thermal throttling. It does not appear that the radiator is the weak link (assuming this radiator can shed more BTUs than stock).

Fan won't help anything once your car starts moving at speed faster than 30mph... I hope you realized that.

Highway WOT pull is not a good indicator of effectiveness of an radiator, as there are just too many variables... Honestly, there is no good way to test it aside from strapping the car down at the temperature controlled dyno room and do a half of dozen multigear WOT pull there. Then compare the temperature data between the stock and CSF unit. A blog did a similar test recently in a comparison test against a CSF unit, which is how I know about this company in the first place.

http://www.motoiq.com/MagazineArticl...-vs-Masiv.aspx

All I know that my coolant temperature stays below 100C at the track (road course, 20 min session) last month, with ambient temp around 10C. But I also have the PPK/335is auxiliary radiator and Dinan oil cooler as well.

[Cloud9blue]

Originally Posted by kimoboy

I would like to know will it lower ATF temp.

Don't think there is an easy way to log that without hooking the flash tune only car up with a BT cable. But it should help since the engine coolant is cycled through a heat exchanger that cools the transmission fluid.

[kimoboy]

I meant will be cooler in the CSF radiator then in the stock one.

Originally Posted by Cloud9blue

Don't think there is an easy way to log that without hooking the flash tune only car up with a BT cable. But it should help since the engine coolant is cycled through a heat exchanger that cools the transmission fluid.