2003 Bluewater Metallic 525iT

[BimmerBreaker]

I can only work off the clues you give me, and you do have several things going on.
Taking it to an A/C guy means you have no A/C charging equipment. And no gauge to put on now to see if the system has lost enough charge to lock out the compressor.
The A/C system will work with a bad pusher fan, just not well after a few minutes. It also won't come on when the A/C pump is not on.
Back to charge level.
The fact that the system worked when charged and then shut off, points to the expansion valve.
Now you could be chasing other problems. Your electrical circuits probably didn't fail between then and now.

Ok I mean that is fair and all, but doesn't explain why the aux fan won't turn on when ISTA commands it to turn on

[BimmerBreaker]

Tried to read the signal from the car at the aux fan harness with my multimeter, one seems to show a 12v, the other I can not get any reading from on my multimeter either in volts or %freq/duty cycle
I'm curious if anyone knows the proper way to test the harness (assuming I'm doing it wrong, with the negative on the ground and the positive on the positive I get 12v, with the positive on the other/PWM wire, with ISTA test activated, I get no reading in volts or frequency/duty cycle)

Talked to someone much more knowledgeable than I and he confirmed my testing method was correct. I should see something like 0v then 5v then 0v... on the PWM signal (because its, you know, pulse modulating). I believe how fast they occur (0v...........5v...........0v vs. 0v 5v 0v 5v 0v) determines how strong the signal is - faster pulse, faster fan.

I am getting nothing, even with the ISTA test, on my PWM wire
Going to dig a bit deeper and trace the PWM wire today with a continuity tester

[BimmerBreaker]

Yeah, sure. But you must jumper the pressure switch to complete the circuit, unless you have confirmed that the system is up to pressure.

No, I don't. That is not how it works. ISTA and INPA are BMW's diagnostic equipment. When they tell the aux fan to come on, it ignores all other checks in the system and forces the fan to run at a % duty cycle as indicated by ISTA. 24% then 76% then up to max speed for 30 seconds. One of the fans we plugged in functioned properly with this test then stopped functioning after driving it around. If your "theory" was correct then the fan wouldn't have worked under any circumstance.

When you turn the A/C switch on in the IHKA panel, it sends a message to the DME saying it wants the A/C on. The DME then checks for stored aux fan codes (which mine has - which is why my A/C is not working) and looks at the pressure switch. My pressure switch is fine, as confirmed by INPA, and also confirmed by the fact that it operated completely properly for extended periods of time on at least two occasions, and by the fact that the compressor momentarily comes on when the fan code is cleared until the fan code reappears and it immediately stops.

To clarify again, when ISTA commands the fan on, it ignores everything else in the system and just tries to run the fan. My aux fan is not coming on but it has run before with this test. When the aux fan works, the a/c works perfectly. My problem is with the aux fan. I observed it as an A/C fault because the A/C shuts off when there is a fan code stored in the DME

I will test the PWM signal with the harness connected before I go too crazy tracing it, read in another thread it needs to be connected to give a value. If the PWM is giving a signal, then my car is killing fans (or maybe just used 540i fans don't play nicely in the 525i)

[geargrinder]

I believe how fast they occur (0v...........5v...........0v vs. 0v 5v 0v 5v 0v) determines how strong the signal is - faster pulse, faster fan.

Yep you had it right all along.
Not really 'faster' pulse, but 'longer' pulse. Hence the pulse width modulation. The % of time that the signal is on per full cycle = "duty cycle %" in other words 5ns-on 5ns-off = 50% duty cycle, 7ns-on 3ns-off = 70%, etc. One of those things where a picture is worth 1000x thread posts.


If you drive a PWM signal into some capacitance, it will smooth the pulses out and that begins to become voltage modulation, or, it can just be used directly to pulse a circuit on/off.

read in another thread it needs to be connected to give a value.

Hmmm. Yeah maybe. Maybe the DME does some diagnostic test on the output to be sure there's a fan there. If so you could also try to wire up a couple leads to a spare fan (firmly clamped to something so it doesn't fly away...)

But give it a try first, I wouldn't be surprised if the INPA/ISTA tests bypass the diagnosis thing and push a signal out the wire.

Inside those PWM fans is also a driver circuit - basically a big transistor - so the actual power isn't coming from the DME directly, the PWM signal is just a control signal. The PWM switches that transistor (prob some kind of MOSFET) on/off and its that fan on-board 'driver' that then supplies power to the motor (its that line that the monster fuse is on, not the PWM side). So... have you checked the connector to be sure there's good power and ground on it? The DME could be supplying PWM but if there's intermittent power being supplied, then of course it can't turn the fan on.

[BimmerBreaker]

Yep you had it right all along.
Not really 'faster' pulse, but 'longer' pulse. Hence the pulse width modulation. The % of time that the signal is on per full cycle = "duty cycle %" in other words 5ns-on 5ns-off = 50% duty cycle, 7ns-on 3ns-off = 70%, etc. One of those things where a picture is worth 1000x thread posts.


If you drive a PWM signal into some capacitance, it will smooth the pulses out and that begins to become voltage modulation, or, it can just be used directly to pulse a circuit on/off.

That makes more sense, thanks for explaining that. None of my other cars (I've been in Z3/E36 land since selling the E39 M5 years ago...) have this sort of stuff...

Hmmm. Yeah maybe. Maybe the DME does some diagnostic test on the output to be sure there's a fan there. If so you could also try to wire up a couple leads to a spare fan (firmly clamped to something so it doesn't fly away...)

But give it a try first, I wouldn't be surprised if the INPA/ISTA tests bypass the diagnosis thing and push a signal out the wire.

Inside those PWM fans is also a driver circuit - basically a big transistor - so the actual power isn't coming from the DME directly, the PWM signal is just a control signal. The PWM switches that transistor (prob some kind of MOSFET) on/off and its that fan on-board 'driver' that then supplies power to the motor (its that line that the monster fuse is on, not the PWM side). So... have you checked the connector to be sure there's good power and ground on it? The DME could be supplying PWM but if there's intermittent power being supplied, then of course it can't turn the fan on.

Yes, I tested the ground and the main power wire and am seeing 12v. When on the same ground pin and checking the PWM wire in either volts or freq/duty cycle, I get no reading (with the harness unplugged) with the ISTA test trying to turn the fan on

And fans don't need to be strapped anywhere, they aren't powerful enough to blow around and move themselves. Have tested many a fan off the car like that (just resting against the front bumper) and never had an issue

To clarify again (maybe my other posts were too verbose lol) I did check the first replacement fan with the ISTA test and it was working before we fully installed it. When we installed it A/C worked perfectly for about a block then died. When we got back to the shop the fan would not come on with the ISTA test. I believe that was the last time the A/C worked.
That is partly why I am starting to suspect the PWM wire got caught/snagged somewhere, and maybe what was a bad connection/frayed wire before (causing intermittent A/C) totally let go after we removed the bumper and uninstalled the original fan.

Fans are all the same. Maybe you just have another bad fan.
As a test, I ran my A/C with the fan unplugged. Comes on and works perfectly.
Yes the fan has power all the time and the ISTA turns it on with a signal. You should see how much circuitry is under that motor cover. At turn on the fan does a reverse bump as a check at each turn on.
But you still have no A/C. The fan problem is a symptom, not the cause.

So you are telling me if you unplug your fan with the car off, then start the car, then turn the A/C on it, it works?
If it is unplugged at the startup, it will fail the DME startup test and throw the internal DME code. If you unplug it after the car is running the A/C should continue to work because it won't have stored the code during the startup test



I just made an interesting discovery as I was going to purchase an aux fan as I sort of felt it was going that way...
While many sites list the part #'s as interchanging, on RealOEM, it lists a fan 64548380780 for M52/M62 engines only. My car asks for 64546921946 or 64546921395 which lists compatibility with M52/M62/S62
Willing to bet I've got the M52/M62 specific fan on my car that doesn't play nicely with the 525i PWM system. It was a little weird during the low-speed mode on ISTA, I just figured it was an old fan, now I'm suspecting it was because those fans weren't designed the same as the later ones.
The correct part # for my car was a bit more $ too but hey that's life. Hopefully that is available tomorrow and we will see what happens with the correct fan.

Cool fact, my wagon was made in the last month possible for 525i wagons - 6/03

[BimmerBreaker]

Just for you I rechecked. It's easy because my fine ride wagon is missing the inlet plenum for the radiator and my wires are zip tied to the core support.
Unplug, turn on car and hit A/C. Came right on .
Interestingly enough, unplugging the pressure switch prompted the fan to run at max speed, both A/C on and off.

Well maybe things work differently in your neck of the woods. It's also 110 degrees here, so maybe where you live the car doesn't need the aux fan on anytime the a/c is on, but I think for me in this temperature it does - and if the aux fan doesn't come on and the car wants it to the a/c wont run. Maybe your car just isn't asking the aux fan to come on at that moment.

Also FWIW the times the A/C worked best have been the long drive back from San Diego (at night, cooler temps) and it only started to die when we got near Phoenix (90F+ even at night) and only under higher engine load (when accelerating to pass). It then worked the next day, which was cooler than usual here, until the car got up to full operating temp and it died. It then worked the next day, in the cold morning, on a long highway drive, until I got to the offramp to pull off, A/C died when I was waiting at the red light.
So you may be right that the A/C will run with the aux fan not working. My aux fan may have died at some point (it was working in San Diego, at least to some extent) and now the A/C is only working when its cool enough to not need the aux fan on...

[BimmerBreaker]

I went back and reread your thread, starting with the A/C. recharge.
Your problem went from being cold to warm, and now no fan.
I've asked, but not heard if your system is fully charged. (Still charged, checked by you)
I stand by my diagnosis of expansion valve at the first symptom of being cold and cutting out.
Then the problem was cold at idle, but warm moving. (Classic low charge)
Then only briefly in the morning (No charge)
Now the fan problem and testing.
I'm old, and a curmudgeon, and go with old school testing procedure's. But these cars (love and hate) with the electronic interface still have basic systems underneath.
Geargrinder is your man for all the electronic, electric diagram, coding, etc.
But we have to get down to one problem at a time. And I wont be happy till i know that a low pressure isn't interrupting your fan test.
Our here in the West the cooling fan does come on any time the A/C is on.
And the majority of used fans are bad.

I know you're busy and your personal car can't be looked at continually, but by now you have probably discovered that you have a bad compressor.
Check inside the lines for gray material coating the tubing. That is probably what was clogging the expansion valve. If you get lucky, flushing everything will clear it, saving tons of work.
For a circuit test on the fan, use your ISTA test to see if the fan signal comes thru for the high coolant turn on.
I have NO idea how to do a ISTA test, use INA, or any of the coding, laptop programming, etc. that these cars use.
And yet all my cars run, and work.
I also know I'm biting off a BIG one calling it, so we'll see what you report.

Interesting theories for sure man, I'm really digging that creativity
I advise you to channel those efforts and write a book, sell a bunch of copies and meet Oprah

[BimmrMeUpSnotty]

Meeting Oprah’s a high thing in your bucket list???

[geargrinder]

Grahamalammadingdong:

Check this out. Sorry for the sh177y Scribd link but I don't have a link to the proper PDF and I hate signing up for their stupid trial and then getting 1000 spam emails a minute for the next 3 years.

Hit pages 14-15, then page 17.

https://www.scribd.com/doc/60087943/...r-conditioning

Some particular excerpts:

The engine control module signals the IHKA (Signal KOREL) when the idle increase hastaken place and the IHKA will then activate the compressor clutch, as long as all othercompressor requirements are satisfied. Compressor cut off conditions include:
•Evaporator temperature below 2C
•Coolant temperature > 117C - causes the compressor to run in a pulsed operation.
•Coolant temperature > 120C - the compressor clutch is deactivated until thetemperature drops below 117C.
•Full acceleration at low speeds - below 10 MPH with a full throttle input will causethe compressor to shut off for 10 seconds.

(Note that idle increase is a requirement as well as the list)

For ‘99 Model Year, E39 IHKA now receives a variable input from the refrigerant pressure“sensor”. The sensor is mounted in the receiver/dryer.Based on the refrigerant pressure, the pressure sensor input allows the IHKA module todetermine the “start up” torque of the A/C compressor. If the pressure is too low or highthe compressor will be deactivated. The IHKA also anticipates the necessary auxiliary fanspeed and passes this information on to the ECM via the K-bus. This change now removes the pressure sensor from the compressor activation circuit.

Taking that information, I would continue to use INPA and observe live data from the IHKA (pressure sensor, evap temp sensor, coolant temp, AC clutch activation) while you drive and it cuts in/out.

HTH... I'm off to watch some Oprah Porn (just when you thought there wasn't a type of weirdo porn you hadn't heard of...)

[BimmerBreaker]

Taking that information, I would continue to use INPA and observe live data from the IHKA (pressure sensor, evap temp sensor, coolant temp, AC clutch activation) while you drive and it cuts in/out.

No A/C at all anymore. Worked once with Dannys '00 540i fan for about a block. Suspect the DME sent a PWM signal the fan couldnt respond to and it shut off and shut off A/C with it

Suspecting earlier intermittent failure was the aux fan failing to respond to some speed modes.
Should have replacement fan later today. Fingers crossed

[BimmerBreaker]

Still waiting on the aux fan...

Installed these euro ambers while it is sitting. It made it look suuuuper base spec, but I'll be tinting the windows soon and getting my chrome trim wrapped in gloss black vinyl that I think will make it look a bit sportier. Can't help but think this color really needs M bumpers to look aggressive. It needs enough paintwork here and there that I can justify the bumpers... eventually



Having trouble doing the coding though - can't find a lot of solid info. Have NCS Dummy and NCS Expert set up, anyone who has done this mod before have any assistance they can provide? It would be the same process as for those who installed clear euro touring lights
May be able to figure it out once the A/C is fixed, so I can more comfortably sit in the car while scrolling through the NCS options, but maybe someone has done it before and can offer some help...

[dannyzabolotny]

We made some major progress on this today! We got the new aux fan in from FCP Euro (who somehow got it to us faster than the local AutohausAZ!) and immediately plugged it in to test it. Initially, no luck, even when running the aux fan test in ISTA. So off we went, probing around the wiring harness, checking the fuse, checking voltages, etc. With the engine running, we saw good voltage on both sides of the fuse, so that was flowing correctly. Then we probed the aux fan harness again, and after wiggling it around a little bit we started getting voltage there. We then plugged in the new aux fan and it started working correctly. Awesome!

After that, we removed the bumper and replaced the aux fan on the car with the new aux fan. It took us less than an hour, we've gotten real good at

We went on a short test drive and the AC now works consistently. No issues with the compressor, no weird noises or anything, just ice cold AC! We're guessing the wiring at the aux fan connector had either gotten pinched or bent in a way that made it unhappy, which is why it worked so intermittently before. We'll be keeping on eye on the aux fan the next few days, though we might just use one of my parts car aux fan connectors to replace the one on the wagon now, just for peace of mind.

With the AC fixed, we turned our attention to the front right window regulator, which was mostly dead. That needs to be working for the car to get tinted, as tint is an absolute requirement here in Arizona. Graham bought a Üro Premium window regulator, which is supposed to be an improved design (it also has an excellent warranty). Installing it was pretty straightforward, and we really took our time to make sure the cables were adequately lubricated before installing the new regulator. We also spent some time adjusting the window glass to sit perfectly so the window would go up and down smoothly without any hesitation. I took special care to make sure nothing was loose inside the door, because there's nothing worse than closing an E39 door and hearing something clunk around inside. I'm happy to report that the window regulator now works perfectly and the door still has that wonderful bank vault feel to it.

Tomorrow we'll be thoroughly washing the 525it, clay barring it, and then compounding/polishing it. It'll be a long day but I'm really looking forward to making this car look awesome.

[BimmerBreaker]

Last night a few blocks from home the A/C died after running some errands with my girlfriend... once I got parked at home the A/C started working again and fan was working

This morning had A/C at start. Drove about a block, no A/C and car was heating up very quickly, I assumed the fan was not working. Pulled over, pushed the harness together, no change, pushed the wiring in on the car-side harness and eventually heard the compressor click then the aux fan whirred to life

So snipped off the aux fan harness



New one soldered in and wrapped up...



Running solid now

Now to figure out NCSexpert... watched some videos online just can't seem to read the data off my car for the .trc file... gonna detail it this afternoon too

[BimmerBreaker]

Gave it a mild paint correction yesterday on some of the worst panels, gonna buff/polish the entire car when its a bit cooler outside over the winter. Wanted to address the worst parts today, the hood and some scuffing on the sides.



Really loving this car right now

edit:
Just got home from a 400+ mile roadtrip with my gf and dog, car did spendidly. Did eventually throw a check engine light for torque converter but whatever. I'll make sure the donor 530i has a good trans

[BimmerBreaker]

Tint is on... not suuuure I love it but its a decent quality tint job at least. Much darker than I wanted, I also specified ceramic tint and I don't think it is (I didn't really pay ceramic pricing either to be fair). If I'm really unhappy with it next summer I'll get it redone... but it'll probably stay

Wanted to make a little list for myself of what I need done:
A/C fixed
Windows tinted
New tires
Front window regulators
CCV replaced
Rear subframe bushings

A/C and tint done. Subframe bushings ordered. Think the CCV might not be as much an issue as I thought, going to do the valve cover gasket today then monitor oil consumption from there.
Gonna order tires once I finalize what wheels I want. Tempted to run Style 42's but afraid they'll look too small on the touring, may end up just rocking the flower power wheels. I do like them quite a lot and think they look great on the touring

[geargrinder]

Yeah that's a little ghetto dark. Car looks super clean tho. I don't love the flowerpowers but if its a DD then they fine...

[BimmerBreaker]

Yeah that's a little ghetto dark. Car looks super clean tho. I don't love the flowerpowers but if its a DD then they fine...

That picture was taken on an overcast day, the tint isn't quite as dark as it looks in the picture there but it is a little darker than I wanted. It's 18% all around. The main reason I didn't want dark tint was to not sacrifice night visibility. I took it out last night just to test it and I had great visibility still, so I'm at least happy about that.

Yeah its a daily so think the flower power wheels might work. They are comfortable and look good on the touring so why not.

Next up, shadowlining the chrome trim. That should really set it off. Then I'll decide to keep the ambers or go euro clears... I'm a pretty diehard amber guy, but I think on this color the amber just looks "off" somehow. Need to see it with the shadowline trim before making final judgement.
I have a set of euro clear style headlights here though... hmm...

[computiNATEor]

I think Plasti Dip on the shiny window trim would clean up the car. The chrome look out of place, IMO.

[BimmrMeUpSnotty]

Eeeeek! Plaplaplastidip?!? Screw the plastidrip, why that kiddycrap on the car, how about some commitment, scuff the trim, then use SEM high gloss trim paint, enjoy the car and not worry about the time the Plastidip starts to peel....

[dannyzabolotny]

I dig the tint. Makes me wish mine was darker... I'll probably redo it next year to be darker all around. The black vinyl wrap on the chrome trims should look real nice and should last a while, even in the heat. We did plastidip on the body-colored trims to see how it would hold up and it looks decent.

[computiNATEor]

Eeeeek! Plaplaplastidip?!? Screw the plastidrip, why that kiddycrap on the car, how about some commitment, scuff the trim, then use SEM high gloss trim paint, enjoy the car and not worry about the time the Plastidip starts to peel....

I know, I know. The door trims are already plasti-dipped, IIRC.

[geargrinder]

I dig the tint.

Naw. Looks totally ghetto. Now I know you guys have proper actual logical reasons for tint out there for sure. But super dark like that - not only do you look like a jackass gangsta wannabe from the outside, but riding INSIDE those cars BLOWS with tint super dark.

I've never plastidipped trim (wait... maybe once on some little pieces... I seem to recall doing a quick touchup on something factory matte black on an Audi...) but I have wrapped... and wrap can come out great and work well if you do it well. I did have to redo the shadowline wrap conversion on the non-M hardtop for my 'vert... Came out OK from the start, but by the time I got to the 2nd side I was indeed doing it so much better that I later peeled both off and did them again and they came out even better.

But in future - as long as I was never gonna want to revert it - I'd try Snotto's suggestion.

[BimmerBreaker]

I'm not painting it, or plasti-dipping the trim. My buddy does vinyl wrap so I'm having him wrap them in gloss black vinyl. It will give the same look as paint, but if I scratch one or need to re-do one in the future, it's a lot easier to match vinyl then to re-paint and get it looking exactly the same. Plus I just don't like the idea of painting anything on the car like that. That seems pretty ghetto to me.

The tint is darker than I want but definitely not "ghetto dark" - it's nowhere close to limo. But it is slightly reflective, so my camera makes it appear darker than it is since it picks up the reflections in the tint, as opposed to the camera being able to "see through" the tint like the human eye

If you get up close to it it's not that dark but you can sure tell how reflective it is... (which is good, if it reflects light its reflecting some heat too)



Inside looking out has plenty of visibility as well (both images below are looking through the tint)





Car got a little dirty as it rained yesterday... time to give it a good wash and get her looking clean again. Or I could decide to stop caring about how clean it is because its a daily driver. Hmm

[geargrinder]

Ahh ok yeah that doesn't look too bad. Actually "light tint but somewhat reflective" is a good look (maybe it IS the ceramic stuff?)

[BimmerBreaker]

Ahh ok yeah that doesn't look too bad. Actually "light tint but somewhat reflective" is a good look (maybe it IS the ceramic stuff?)

It's not 3M Crystalline but it may very well be a mid-grade ceramic tint. It's definitely not as bad as it looks in the first photo I showed - it is just really hard to photograph right. When I look at the car, I can see through the windows. The camera can't

Forgot to mention as well, I fixed the front window regulator (duh, had to be done before tint) and installed an autodimming rearview mirror - before I knew they had autodimming mirrors WITH COMPASSES IN THEM.

I'm so excited for my compass mirror. Looking for one like this in particular, no homelink buttons and the slimmer style alarm button.