[kern417]

Originally Posted by NISFAN

If what you are doing is increasing the coolant conditioning and capacity (CSF chargecooler radiator), then surely the most important parameter is the temperature of the coolant in the circuit? Ideally you would measure pre chargecooler and pre radiator fluid temps to see the delta T, as well as average circuit temps.

If the coolant temps entering the charge cooler are lower, then you know you are giving the chargecooler a better chance of reducing MIAT.

But measuring the fluid circuit could also show up a true bottleneck that isn't being monitored, like is the coolant being recirculated fast enough? Maybe the stock water pump isn't up to the job? Maybe the chargecooler core itself isn't as effective as it could be?

It could be that you are installing a much bigger radiator thinking that will surely work? when actually it's the stock coolant pump that's letting the system down?

I'm really interested in the results you get.
Personally I'll be surprised if a bigger cooler does do anything for street driving. From what I know the stock unit has a full frontal form factor radiator. Compare that to the N55's IC core that is what? 550mm x 170mm? Couple that to continuous maximum flow of coolant regardless of engine load, rather than instantaneous volume related cooling on an A2A (N55) system. I can't see the radiator core as being the limiting element. It will quickly return temps from a 5-15 second pull as seen on street, and if you are on the autobahn at high load, then you'll be having enough air flow to be able to use a pocket sized radiator.

The drag racers amongst us would likely want a larger cooler with a manual fan switch to cool the core between runs.....oh and a water spray bar.

Originally Posted by OzBMR

Updated my earlier posts, MHD confirmed that the IAT as measured by their data logging is from the manifold sensor after the intercooler and that the charge air temp reading is from the sensor in the charge pipe, between turbo and throttle body.

The stock water to air intercooler design is pretty impressive to be able to reduce the charge air temp from 178 C (352F) to 53C (127F) in such a compact format.

kern417 what are you hoping to gain from the upgraded intercooler heat exchanger? More consistent timing?

I was thinking of making a "hypothesis" post so it makes more sense, but wasn't sure who really cared lol. But this is what i know, and what I hope to gain.

The stock cooling system is designed for daily driving in mind, not racing. More specifically, the pump doesn't even turn on until it's needed. During normal driving, the coolant is just a giant heat sink. When you start a pull, you see the IATs drop initially. That's because the pump turns on and sends a surge of cold coolant through the IC. However, when doing a multiple gear pull, you'll see IATs begin to rise again. After the pull, the pump will only run until the IATs are within a range set by the DME (I believe no more than 20F over ambient).

Also keep in mind that I'm in a unique situation on a custom tune that is pushing the stock turbo. So I see IATS drop briefly, then a 20+F rise in temps over a 40-140mph pull. It'll recover a bit after the pull, but temps are still elevated over ambient. So my 2nd pull starts at a higher temp, and ends at a higher temp.

I didn't experience those characteristics on OTS maps.

My goals/expectation with the bigger radiator are:
1. Increase the length of time that cool coolant is pushed through the IC during a pull
2. Decrease the peak IAT gain after a 100mph+ pull
3. Improve the system recovery after a pull

I do get some timing pull. The way my tune is setup, it doesn't show as cylinder-specific timing corrections so I can't show that in my logs, but it does show overall adjustment. I'll also show curves that illustrate the rise in IATs when doing back to back pulls, and the IAT recovery when cruising/stopped.

That being said, if you're hoping for any other gains or concerned about any other issues it could bring, let me know now so I can log before and after data. The cooler is arriving today so it should be installed by this weekend.

[ivanj308]

Are we sure the coolant reading from the MHD logs is the charge cooling temp not he regular radiator temp ?

[kern417]

Originally Posted by ivanj308

Are we sure the coolant reading from the MHD logs is the charge cooling temp not he regular radiator temp ?

There are no temp sensors in the IC circuit, so the only actual coolant temp you can measure is the radiator temp.

The only thing you can measure to evaluate IC performance is intake temps before and after the intercooler.

[OzBMR]

Originally Posted by kern417

There are no temp sensors in the IC circuit, so the only actual coolant temp you can measure is the radiator temp.

The only thing you can measure to evaluate IC performance is intake temps before and after the intercooler.

I updated my earlier post to clarify that the Coolant Temps are Engine Coolant, not intercooler coolant temps. I can delete the reference to the Engine Coolant Temps if you think they don't add any value.

[Glemm1970]

https://datazap.me/sites/default/fil...oct_102ron.csv
I have CSF. ATI 22-28C at an outdoor temperature of 10-14C

[OzBMR]

Originally Posted by Glemm1970

https://datazap.me/sites/default/fil...oct_102ron.csv
I have CSF. ATI 22-28C at an outdoor temperature of 10-14C

The IAT increased from 24C before run to 29C at 6428RPM in 6th gear, 17.5 PSI boost (about 20 PSI steady in 5th prior to changing to 6th), so a 5C change

In my higher boost run IAT increased from 38C to 44C at 5092RPM in 6th, 17.2 PSI (also about 20 PSI steady in 5th prior to changing to 6th), a change of 6C

Is this a valid method of determining cooling performance of the CSF cooler, the change in IAT from cruising to 3rd/4th/5th/6th gear WOT run peak?

This example indicates 16% - 17% improvement in IAT compared to the stock system. IAT's still increase, just less than they would with stock system.

The CSF example indicates IAT's don't increase at all, 34C at the beginning of the run and 34C at 5900 RPM. This would be very boost/turbo efficiency dependent.

Stock boost, as in alyabiev example on page 1 the IAT remained the same, or actually decreased from 29C to 27C during a 4th gear run.

Pushing 20PSI at 6000RPM in 5th and 6th gears charge air temps get a lot hotter and IAT's will increase.

[kern417]

Originally Posted by OzBMR

Is this a valid method of determining cooling performance of the CSF cooler, the change in IAT from cruising to 3rd/4th/5th/6th gear WOT run peak?

This example indicates 16% - 17% improvement in IAT compared to the stock system. IAT's still increase, just less than they would with stock system.

The CSF example indicates IAT's don't increase at all, 34C at the beginning of the run and 34C at 5900 RPM. This would be very boost/turbo efficiency dependent.

Stock boost, as in alyabiev example on page 1 the IAT remained the same, or actually decreased from 29C to 27C during a 4th gear run.

Pushing 20PSI at 6000RPM in 5th and 6th gears charge air temps get a lot hotter and IAT's will increase.

This is exactly my thought process. On OTS maps, it's not an issue because the turbo isn't working that hard. I'd even venture out to say that a bigger turbo wouldn't need it until it was being pushed hard.

I was fine until I got my custom tune. 19psi creates exponentially more heat than 16-17. So the only way you can really tell is by doing logs to see if it's worth it in your situation.

I don't think it will stay steady in my tune, but it should definitely help reduce the increase in temps over a multiple-gear pull.

[Solid67]

Kern are you going to make an install video as well or are we asking for too much?

[kern417]

Originally Posted by Solid67

Kern are you going to make an install video as well or are we asking for too much?

Yes I'll be doing an install vid. the BMW repair instructions actually recommend doing it from the back, so my video will be different from the others. i'll be by myself so I don't want to remove the whole bumper/crash bar and worry about disconnecting all the sensors, cameras, lights, etc. I think it'll be faster than removing the whole front end too.

[kern417]

Also do you guys want to see anything specific with the heat exchanger before I install it? Dimensions, side by side comparison, etc. I don't think I'll be able to see the inside of the core since the inlet and outlet are so small and I don't have a scope.

[yokata]

Originally Posted by kern417

This is exactly my thought process. On OTS maps, it's not an issue because the turbo isn't working that hard. I'd even venture out to say that a bigger turbo wouldn't need it until it was being pushed hard.

I was fine until I got my custom tune. 19psi creates exponentially more heat than 16-17. So the only way you can really tell is by doing logs to see if it's worth it in your situation.

I don't think it will stay steady in my tune, but it should definitely help reduce the increase in temps over a multiple-gear pull.

My 5th gear logs show 82F 3000rpm and 100F 6600rpm IAT ,do u think csf can make big difference, here summer is hot with temps around 90-100F

[jsnm]

Originally Posted by kern417

Also do you guys want to see anything specific with the heat exchanger before I install it? Dimensions, side by side comparison, etc. I don't think I'll be able to see the inside of the core since the inlet and outlet are so small and I don't have a scope.

I would definitely like to see some side by side comparisons from the factory unit. Very interesting log btw,following along and thanks for this.

[NISFAN]

Originally Posted by kern417

Also do you guys want to see anything specific with the heat exchanger before I install it? Dimensions, side by side comparison, etc. I don't think I'll be able to see the inside of the core since the inlet and outlet are so small and I don't have a scope.

Yes, fill them with water and put them on a scale. As I'm a circuit rat, I am reluctant to add weight to the nose unless it really changes the game.

[kern417]

Originally Posted by NISFAN

Yes, fill them with water and put them on a scale. As I'm a circuit rat, I am reluctant to add weight to the nose unless it really changes the game.

Ugh. Doesn't sound like fun, but I'll see what I can do.

Maybe just measure the difference in radiator weight, then extrapolate total increase based on the additional fluid volume?

[NISFAN]

Originally Posted by kern417

Ugh. Doesn't sound like fun, but I'll see what I can do.

Maybe just measure the difference in radiator weight, then extrapolate total increase based on the additional fluid volume?

Yeah, that will work if you measure fluid out when you drain vs fluid in.

[kern417]

Installed this weekend. I'm thinking about just uploading the DIY now and then capturing the data and adding a review later. Some people have been asking me for more info on how I did it without removing the bumper. But the process is pretty simple.

[mike@x-ph.com]

Originally Posted by kern417

Installed this weekend. I'm thinking about just uploading the DIY now and then capturing the data and adding a review later. Some people have been asking me for more info on how I did it without removing the bumper. But the process is pretty simple.

Looks great, congrats!

[weehe126]

Was their a volume difference?

[G.Newt]

Originally Posted by kern417

Installed this weekend. I'm thinking about just uploading the DIY now and then capturing the data and adding a review later. Some people have been asking me for more info on how I did it without removing the bumper. But the process is pretty simple.

Curious from looking at your OEM radiator. Do you have the rock guard installed? It looks like the bottom (especially) and a bit up top where the radiator is exposed got pelted by debris. I'm worried about a punctured radiator.

[kern417]

Originally Posted by G.Newt

Curious from looking at your OEM radiator. Do you have the rock guard installed? It looks like the bottom (especially) and a bit up top where the radiator is exposed got pelted by debris. I'm worried about a punctured radiator.

No, my car was built in July 2017 so before they implemented the rock guard. It's funny because i noticed it was a bit beat up (even the engine radiator behind it had dust and bugs in the fins), but it wasn't until someone else mentioned that I noticed HOW beat up it was on the bottom 1/3. No doubt that could cause OEM efficiency to degrade.

The CSF was installed with the rock guard. But since it's black, I'm still kind of worried that it will be peppered from stuff hitting it over time. That would look pretty bad but we'll see.

[kern417]

Originally Posted by NISFAN

Yes, fill them with water and put them on a scale. As I'm a circuit rat, I am reluctant to add weight to the nose unless it really changes the game.

Originally Posted by weehe126

Was their a volume difference?

Stock System = 3.9L/1.0Gal
With CSF Exchanger = 6.2L/1.6Gal

Stock Heat Exchanger = 6.6lbs
CSF Heat Exchanger = 10.6lbs

I also diluted the coolant, so now it's closer to ~35/65 coolant to water ratio. With fluid you'd be around a total of 9-10lbs increase in weight.

[weehe126]

Wow, that seems like a lot more coolant for something that actually look smaller. The 60% increase in fluid volume alone should help. Hopefully the core actually cools better too.