N54 water to air intercooler street build

Rob@RBTurbo

Lieutenant
Dec 7, 2016
626
401
0
St. Louis, MO USA
www.rbturbo.com
Ride
'08 335i, '14 M6, '15 Tundra
This is an amazing thread, massive props to the pioneers involved. Really hope this can be integrated/controlled effectively, as it is a very nice game changer for the platform for sure. This could simplify everything from inlets, to outlets, to IC'ing- all for the better assuming the bolt-on provisioning and clearances truly are there for all the other hardware. If it were to really materialize let me know- I wouldn't mind making some simple samples to send out to have tooling made for formed silicone to connect N54 based turbos to the S55 based In/Out plumbing where/if needed.

Rob
 
Last edited:

bantam

Corporal
Nov 20, 2017
127
102
0
Ride
2009 335i
Oh man is it beautiful to see IATs stay at a steady 30° the entire time lol. I really hope it's as effective when you aren't running surprise WMI :tonguewink: you dont happen to have any traditional pull logs with that setup yet do you? My apologies if I just missed it somewhere in the thread, and I understand if you wanna wait on that until you get a leak-free setup!

My worthless 2c is to go with an OEM replacement over a Chinese clone. Since you're testing something that so few people have done it's probably best to try to keep it bulletproof and like you said, I could see a Chinese knockoff being likely to leak. But that's also a bit of selfish of me because I want to see the 100% complete end result and it's not my money going towards the project, so take it FWIW I suppose!

Not sure that I have posted many, here are the latest pulls:

 

NoQuarter

Major
Nov 24, 2017
1,662
1,066
0
Indiana, USA
Ride
Z4 35is, 535xi, X5 35i
I removed the fan's PWM wire and provided a ground to the lead on the FAN controller producing a 100% neg duty cycle.

It didn't come on as expected.
 
  • Informative
Reactions: typedRew

RSL

Lieutenant
Aug 11, 2017
937
501
0
Sucks about that leak, ouch. The temp delta is outstanding compared to A2A, I hope it's not the leak. I don't let that WOT sway me, I think the exchanger is way undersized ATM. Maybe @Asbjorn could help get a sizeable (area) core made.

I don't think the fan has a check, dealer has left mine completely unplugged before and never got a light, even to the point of near overheating months later in traffic when it finally got warm out. I don't remember if the WP has a start check, but the thermostat does.

@Hydra Performance pointed out a tiny Arduino controller for standalone water pump with feedback controls for OE electric WPs, but not sure if all of those pumps are needlessly large in size or volume. I imagine it would probably operate any PWM electric pump at a basic level though.

This project is awesome. I've been reticent to switch to hot side inlets, but if enough pieces come together (outlets/exchanger would probably be the largest hurdles for most), I'd be more inclined to start ripping things apart.

@RSL good find! :) One note on that *C temperature delta: it is (IAT - ambient).
That table was buried in there like a tick, geez. Never would've found it without help. Correct, IAT diff to ambient, which means some dialing in to find that sweet spot for the first break point between increased IAT when stopped and dropped IAT when moving. It will probably need some adjustment in seasons and doesn't have the light switch response like VS would (takes time for IATs to move up/down between stop/go), but does produce the intended effect.

@NoQuarter not sure you would get the fan to run 100% without supplying consistent 12v to the PWM channel on the fan, but I've never tried. It sounds like the ground test confirms that, but there are manual switch/fixed DC settings in the rom somewhere too.
 
  • Informative
Reactions: typedRew

Rob@RBTurbo

Lieutenant
Dec 7, 2016
626
401
0
St. Louis, MO USA
www.rbturbo.com
Ride
'08 335i, '14 M6, '15 Tundra
This project is awesome. I've been reticent to switch to hot side inlets, but if enough pieces come together (outlets/exchanger would probably be the largest hurdles for most), I'd be more inclined to start ripping things apart.

Wonder what the odds are of the S55 inlets/outlets fitting (or rather being adaptable)? Does anyone pull their OE plumbing/charge coolers (when swapping for aftermarket) and sell them off for reasonable cost?

Rob
 

NoQuarter

Major
Nov 24, 2017
1,662
1,066
0
Indiana, USA
Ride
Z4 35is, 535xi, X5 35i
@NoQuarter not sure you would get the fan to run 100% without supplying consistent 12v to the PWM channel on the fan, but I've never tried. It sounds like the ground test confirms that, but there are manual switch/fixed DC settings in the rom somewhere too.

It should be a grounded PWM where -90% duty cycle actually turns on the fan. This is with the fan at max speed.
26727
 
  • Like
  • Informative
Reactions: typedRew and RSL

RSL

Lieutenant
Aug 11, 2017
937
501
0
Wonder what the odds are of the S55 inlets/outlets fitting (or rather being adaptable)? Does anyone pull their OE plumbing/charge coolers (when swapping for aftermarket) and sell them off for reasonable cost?

Rob
The stock S55 outlets don't seem to be a direct swap and would probably be connection problem even if they were close. They may also be somewhat prone to failing, haven't paid that much attention. IIRC, aftermarket S55 A2W outlet to the TB is what's being used here though.

S55 people upgrade exchangers (99% sure too wide for E chassis) and A2W cores though and can be found for sale used. My main take away is that people upgrade stock A2W, but few seem to ever switch to A2A.

It should be a grounded PWM where -90% duty cycle actually turns on the fan. This is with the fan at max speed.
View attachment 26727
Wow, that's completely backwards of what I expected. If a ground didn't activate it though, maybe it is? I probably wouldn't throw +12V just to find out after seeing that, fans aren't cheap. Thanks for posting that.
 

Rob@RBTurbo

Lieutenant
Dec 7, 2016
626
401
0
St. Louis, MO USA
www.rbturbo.com
Ride
'08 335i, '14 M6, '15 Tundra
The stock S55 outlets don't seem to be a direct swap and would probably be connection problem even if they were close. They may also be somewhat prone to failing, haven't paid that much attention. IIRC, aftermarket S55 A2W outlet to the TB is what's being used here though.

S55 people upgrade exchangers (99% sure too wide for E chassis) and A2W cores though and can be found for sale used. My main take away is that people upgrade stock A2W, but few seem to ever switch to A2A.

Good deal. I will take a look at some of the for sale forums for OE S55 A2W core, but yeah the intention would be to use aftermarket (i.e. BMS) plumbing for everything else then adapt it in accordingly. There is no doubt that the fitment would not work without some adapting to the N54 environment, but it would certainly simplify life to use/modify the aftermarket S55 plumbing hardware that already exists today if at all possible.
 
  • Like
Reactions: RSL

RSL

Lieutenant
Aug 11, 2017
937
501
0
My guess here is anything below the maximums - less than 5% or greater than 95% DC and the fan is off. could be <10 and >90 but my measurements don't match up with that. If this is the case, a simple switch/relay type solution won't do it.
They are unexpectedly confusing (to me anyway lol), but your assessment seems on track, except for the fan being off on the high DC side. Output below 7% is definitely fan off, but 91%+ DC should be on at max speed (conversion table below). I'm not sure what the first/last 7-9% padding is actually there for, but there are factors that act on the output DC after this conversion, so may just be headroom. Either way, <=7% is off, >=91% is max on.

dc_conversion.png
 

NoQuarter

Major
Nov 24, 2017
1,662
1,066
0
Indiana, USA
Ride
Z4 35is, 535xi, X5 35i
They are unexpectedly confusing (to me anyway lol), but your assessment seems on track, except for the fan being off on the high DC side. Output below 7% is definitely fan off, but 91%+ DC should be on at max speed (conversion table below). I'm not sure what the first/last 7-9% padding is actually there for, but there are factors that act on the output DC after this conversion, so may just be headroom. Either way, <=7% is off, >=91% is max on.

The signal looks backwards to you because the DME is providing a grounded signal and not a +12V signal.

Now, I wonder if you were to change the 91 to 100, then provide a ground to the PWM pin on the fan, if you will get max speed? Seems like it should do that already but it doesn't.

Maybe another table that handles the error conditions? Essentially no or lost signal=0% (error) and if the DME sends 100% another error that is handled differently?
 
Last edited:
  • Agree
Reactions: RSL

Jake@MHD

Major
Platinum Vendor
Nov 7, 2016
1,593
2,058
0
Philly
@NoQuarter, the typical dme pwm you are thinking of only has 2 wires, +12V and a pwm'd ground. The fan is different in that it always has a complete power circuit, and a separate wire for speed control. I assume it actually needs to see a pulse width to operate, why your constant ground of the pwm wire test failed.
 
  • Like
Reactions: RSL

typedRew

Sergeant
Feb 25, 2019
412
221
0
Ride
2009 335i xDrive
Thank you to everyone in here, really great to see this is getting cracked.