Another Idle Surge Thread

[Province]

Bought an OBD1 '95 s50us M3, personally been tackling issues left and right but the only thing that is left is a weird idle surge that occurs within open loop. It will idle surge under two conditions, open loop & cold engine temperature. As soon as the temp gauge reaches edge of the blue region, car doesn't surge as hard, slight misfires.

Currently 1/31/2023; 1286 code was stored, no idea when it got stored. I cleared ecu and after driving cycles it popped up again. Planning to get a different DME and test.

Update.
3/20/23; It's been a month or two replacing DME, kept the Turner tune, car drives great. Ignition timing is still being pulled randomly, probably nature of the DME.

This is what I've done so far.
Complete Vanos overhaul - No change in idle, but regained some lost power
Valve cover gasket
spark plugs - less misfires
spark plug boots - less misfires
Retimed engine - No change in idle, engine was still in time even before after vanos. checked timing 2k miles after vanos, still in time.
Checked ICV, cleaned - ICV was in perfect operation, found that a lot of the hoses were replaced or are in very good condition, this was replaced with topend rebuild
Smoke test - no leaks, x2
Unplugged maf during cold start - Car runs even more rough, conclude no vacuum leak
Unplugged iac - car dies, conclude no vacuum leak
Replaced intake air temp sensor - No change, the old one all be it oily, still retained the same resistance resolution as the new one
Bled cooling system - no change, was thinking that coolant temp sensor was reading air (hot), causing it to richen up to cool an already cold engine.
Replaced crank position sensor - no change.
Bypassed evap system - no change.


car does seem to emit smog on the higher end of the spectrum. Car passes 25mph with flying colors. 15mph car does fail only @ HC

This first test was done ONLY with new spark plugs and full vanos replacement <--- Before starting 15mph test, smog tech stalled the car then started rolling test. also car overheated after smog test. it was like a 100* day too.
106PPM HC 0.27% CO and 291PPM NO

This test was done after previous smog but with new spark plug boots, motul engine oil, and new fuel cap. <--- no issues from the tech, hot day but not 100* hot
84PPM HC 0.11% CO and 211PPM NO

[Province]

Going back to this wanted to add some thoughts. Open loop, car shouldn't be reading from sensors unless there's an external input, that's common sense / elementary knowledge. Maybe if maf is sending incorrect air flow, then the car might switch to closed loop and richen up the mixture. Either way, maf/o2/sensors are not my issue as I can confirm I am staying in open loop throughout the duration of the surge. The only realistic sensor that would be my issue is my intake air temperature sensor but with that being replaced and ohm'ed out to be in good condition as per Bentley manual, the only thing I can think of is mechanical operation of the engine and not electronics.

The IAT actually changes ignition timing per intake temperature, obviously. What I am getting at here is that since electronically the car is good, mechanically timing must likely be off causing ignition to be off time relation to cam timing, but its off marginally to where the IAT can compensate once air temperature reaches a certain amount. Obviously once the car reaches closed loop, ignition timing can be dialed in.

I'll be tackling timing for the 3rd time tomorrow, but I am very confident that it has to be timing. It is a shame though because I didn't expect myself to be off - twice in a row to be exact. I guess maybe the engine being tilted by several hours just exacerbates parallax error. Timing these things ain't the same as Japanese cars haha but stay tuned, I'm sure this'll be the end of my woes.

[Province]

Alright, I've actually retimed the engine twice this time around. The first time I put it back together and the car fired but would not maintain an idle. I was thinking maybe reset adaptions but I didn't decide to do that and just opened up the valve cover once again and removed vanos. This time I took pictures to document. I first took the engine to tdc verified via both crank pulley and also sticking an extension to see cylinder 1 is top of its stroke. As you can see, the camshafts are both pointed 45* ish relative to head surface so we are obviously on the compression stroke.
IMG20220925175858.jpg

Took out the vanos and then double realigned camshafts to top dead. The square end with the dots are facing upwards. I do not have the camshaft blocks nor do I need them. I verified their timing by placing a straight edge across the two square ends and used feeler gauges
IMG20220925182640.jpg

After that, I brought the vanos sprocket all the way counter-clockwise to see my vanos range. Then I proceeded to rotate all the way clockwise to prepare reinstallation. You can see the timing marks on the vanos gear move. I do not know if their position is correct but based on other videos I've seen, the mark is typically cantered off center so my gears check out for sure.

IMG20220925182928.jpg

IMG20220925183129.jpg

Then reassembly but that's straight forward. Before asking "did i put in the vanos actuator properly, yes I did.
I also had the battery out for about 3 hours in the midst of it all.
Everything back together and it fires this time around and holds an idle. Doesn't surge like crazy but does search here and there. Now I start to drive it, felt fine until I start doing drastic throttle inputs / large throttle inputs. If I ease into the throttle the car moves, but as soon as I open throttle say 30% the car pulls, reaches about 2k or 2.5k then immediately gets cut off while emitting a very loud wooshing sound. Too much valve overlap.

It's getting late, the next diagnostic thing I'd probably do is unplug the vanos and see maybe if its vanos related. If that fails once again, then back to rechecking timing again....

[aeronaut]

"I do not have the camshaft blocks nor do I need them. I verified their timing by placing a straight edge across the two square ends and used feeler gauges"

This is your problem. The timing faces at the end of the cam are not square. They are around 5deg off from square, and I'm not even sure the intake is the same as the exhaust.

[Province]

Perfect. If then basing the top square face, aka the horizontal, (dotted face) will yield me an incorrect angle, then squaring off vertically should yield me a more correct angle. I mean, that is what the camshaft block does after all: is square the vertical portion of the end to the head. Appreciate the input.

[Province]

Well, just redid timing, this time I used a calibration block to square the cam sides with the head, turns out the the top face with the dots are not squared just as aeronaut said. Well, everything is back together - I found that the exhaust cam went slipped out of time as well as the intake cam being off. Now that is rectified and then vanos installation is just as it is. Got late into the night but this morning just started it up and its back to uh I suppose the original state with the open loop idle surge. Clearly, this is a sign to open the valve cover once again for another several hours of fun.

Edit: The surging isn't as bad as before which is good, but I know for a fact I need to advance timing on the intake cam to dial in the car. I'll enjoy the extra top end power though.

Double Edit Update: Surging just got worse / car returned to how it was surging before. Car literally sounds like its pulling ignition timing just like before. Again, same scenario where open loop cold ambient air it'll surge. Now I am starting to doubt mechanical timing... Back to the drawing board then. Maybe car finished doing adaptations for the worse.

[Province]

Borrowed cam locks and went upon timing for sure. Now I can surely say timing is out of the equation. Still having idle surge same symptoms as before, literally no change at all.

Besides that, I did only like two diagnostics afterwards; I unplugged vanos - idle doesn't surge as hard. it'll surge, going back and forth between 500 and 800 but with the vanos in, will surge with the same frequency but harder- 300 to 1200. It literally sounds like the car ignition cuts which lets the rpms down that low then after some hundred miliseconds does it decide to fire again, catching up to 1200. Something like this I wish I had an oscilloscope to monitor the injectors/spark but I'm ill equiped.

Unplugged maf with fresh timing, not a single change in idle.

Also should add that during the idle surging, I could rev the car and hold the throttle cracked a little and the car will still surge ie; hold target at 1k rpm and rpms will bounce to like 1300 and surge back and forth.

Well now theres several things left that I can bet on (somewhat).
-Tuned ecu for unmodified engine / wrong ecu?
-Leaky injectors
-ICV erratic signal? <- bs cause closed loop is good....
-Coilpacks
-Vanos seals? They only have like <1k miles on them....

[bmwstephen]

does the DME look tamper with? Make sure the vin matches that of the car. check the tabs around the enclosure. If they are broken off our bent then good time to open it up and see if the chip is original or not.

[Province]

I come back with a big discovery.
IMG20221101165117.jpg
Time to ID this, but I honestly can finally be at ease knowing the real culprit and that timing not the issue from the getgo but I actually dialed it down per aeronaut's insight and the car feels wonderful on the low end.
Edit: if my OPSEC is correct, then the only thing is 21.5lb pinktops and I am running greentops. Everything is finally lining up now.

[Province]

Back once again for more disappointment and nonsense.
IMG20221104155433.jpg
Remanufactured metal 21.5 pinktop boschs.

IMG20221104193603.jpg
Yucky greentops, one of the injectors was blocked with debris.

IMG20221104193611.jpg
Return line looks funky and cracked but I was able to manipulate the clamp to get it to seal for now. I'll have to replace it eventually.
Upon starting up the car ran super shoddily, heavy misfires and it seemed like the injector pulses didn't line up with the spark. I unplugged battery to reset adaptations.

Starting it up once again, mind you the previous start up was cold too and I only had the car ran for 10 seconds, and it started up nicer. As soon as it went to open loop, the car immediately ran horrible and surged just like before on greentops. I let it warm all the way up, holding the throttle open just for some assistance. I went out to drive a bit and car felt smoother - a lot smoother but I felt a "dead spot" 1st gear around 2k. I went out on the highway and cruised 65 to find that the mpg gauge on the cluster was reading really high, like around 35 (?) compared to before when I cruised 65 on greentops and was getting 26 27, more-so lower side. I obviously don't think the 35 is accurate but the car seems to like it. I'll be going out on a 180 mile commute tomorrow so I'll be seeing if the cold start surge changes at all.

Oh yeah also, during the cold start after battery unplugged - car ran extremely rich, faint white smoke emitting from tailpipe. After closed loop triggered, still had white smoke. I went on a drive as I mentioned before, came back to find no smoke or extremely less smoke than before. So at this point, what I am probably going to do again is unplug maf on the next cold start to test for vacuum leak.

Does anyone know if the turner chips are reprogrammable? Maybe the previous previous owner sent it in to reprogram another addon say a 3.5 maf which could be another factor thats screwing me over.

[bmwstephen]

That chip number was tuned for 21lb injectors when stock s50 were 17lbs. You need to get a proper tune.

what year production was your car because the ews could get in the way depending on the year

[Province]

That chip number was tuned for 21lb injectors when stock s50 were 17lbs. You need to get a proper tune.

what year production was your car because the ews could get in the way depending on the year

413 ecu, so it has EWS and I referenced the chip to be 21.5 pinktops and EWS delete. Yes, I bought the correct injectors per the tune requirements, seen in my previous post [BOSCH 0280150440]. It's to early to reach a definite conclusion but I can eliminate fuel from potential issue. These injectors are audibly louder, its extremely probable theyre getting the right signal.

[catalyst.]

I have this exact same problem with my s50 swapped 325is. I have the TRM chip for my setup, injectors, everything. I have never gotten it to stop but tbh I gave up because I had other projects and the car has been sitting under a cover for 3 years. I did everything you have done in this thread, I had the injectors cleaned and rebuilt, replaced the maf, sprayed about a gallon of carb cleaner on the intake trying to find some kind of vaccuum leak.

I was going to try to fix it over the next few weeks as I really want to drive the car, its a real beast even though it does not idle.

Please for the love of god I hope someone has some other ideas that can fix this because it is such a royal pain in the azz.

If I remember correctly I also replaced the coolant temp sensor because I thought maybe it was causing the car to fluctuate idle on coolant temp, not even sure if that is possible off the top of my head, but I did try it. Also I replaced the fuel pump and fuel lines I think. I really tried everything short of installing a new cylinder head. I did leak down test with good leak down on all cylinders, perfect leak down actually. 80/80 psi

[catalyst.]

IDK maybe this will help, I think i tried this though... been a few years... maybe consider idle control valve issue and also throttle position sensor issue. Maybe I will throw some more parts at this motor hoping it will idle LOL

Hello guys I won 2 championships in GTS-2 with the TRM tune, and for sure it was not the tuning!

I thought I might share some light on this subject/issue, I want to make my story short and to the point.

I had this 1215 code after swapping my original M50 for a M52 in my GTS-2 race car, the old M50 had 180psi compression across all cylinders but #4, #4 came in at 160psi, so needles to say after 10 years of track abuse and winning 2 southeast championships it was time to retire it.

Nothing change when I did the swap, the TRM tune was dialed in for the ODB2 pink injectors and that was it. At the dyno we saw no issues but soon during the first race of this year I found out the check engine light flashing at high RMP, and mostly on 5th gear.

BTW, I even put the car on the dyno at ran it thru every gear to try to replicate the issue, but never did it on the dyno……..

As I try numerous things the issue change its behavior, at some point it was only in the top of 4th and 5th gear but right before I fix it, it got really bad and the CEL will flash at any gear but high on the RPM.

Here is a list of things that I tried, in the order I replace them (nothing on this list fixed it):

1. cleaning the Air mass meter (HFM)
2. Tested the intake system for vacuum leaks.
3. Changing the air filter to multiple brands (OE intake box)
4. Installing a CIA w/a ITG filter and a K&N
5. Changing the OBD2 intake manifold to the OBD1 M50 manifold
6. Replacing the Crank sensor
7. replacing the engine temp sensor
8. replacing all coils
9. swapping the O2 sensor
10. Isolating the intake box from chassis harmonics.
11. TRM Tuning burn me a new chip
12. Different gasoline.
13. Injectors got serviced
14. replacing the air mass meter
15. Swapped DME’s
16. replacing the ICV

So Michael at TRM and I tried different things, and scratched our heads on this, and probably replaced sensors that were the originals, so some of them was just a must, and since the car is on solid mounts everything gets extra rattled.

So one day I started looking at my throttle data from one if my races at CMP, and I’ve notice one unusual read at idle, so base on that I took my bag of spare TPS sensors and ICV’s out and started to swap them, I found a newer ICV cleaned it and install it, then I found out that one of my spare TPS sensors actually got the DME to throw the TPS code rather than the 1215 AMM code… I smiled and started to log voltages, that one sensor was reading at idle below 1v somewhere around 0.4v.. so pull another one out the bag and that read 1.1v at idle, went for a test drive and voila the issue was gone….

I was able to do my last race at CMP with no issues, I hope this helps the community and if you want to say thank you, please support my sponsor since without them finding this issue would’ve taking much longer and probably would’ve have to race at the NASA Nationals with it…

WWW.TRMtuning.com
WWW.jacksonmotorsportsgroup.com


Rafa,

https://www.bimmerforums.com/forum/s...-MAF-code-1215

Maybe i will order this:

https://www.ecstuning.com/b-vemo-par...411744713~vmo/

[Province]

After driving it for a while I mentioned before about a dead spot but this dead spot is actually occuring by pedal position and not load. Either way this morning before I went out I started it up and it surged but because of the dead spot I wanted to try unplugging the TPS and voila the car idled perfectly fine.

Appreciate the extra input Catalyst.

[catalyst.]

Are you saying you need to calibrate the tps even after installing a new one?

The more I think about it I think that my issue may be the ICV even though I tested it and visibly checked it for operation about 10x. I did actually install a new tps and it did not work for me but my issue is different than yours.

[Province]

Are you saying you need to calibrate the tps even after installing a new one?

The more I think about it I think that my issue may be the ICV even though I tested it and visibly checked it for operation about 10x. I did actually install a new tps and it did not work for me but my issue is different than yours.

I dont know if there is any adjustment ie slots when it comes to tps but im just going off the assumption that there is from previous cars Ive had. If it is slotted then without a doubt, calibration would be necessary or else you get incorrect voltage feedback per throttle position.

[catalyst.]

Alright thanks yeah I can't really remember at this point, need to get the cover off and take a look at it

[Province]

Another day final diagnostic (probably). So I took out the tps and found that there is no adjustment and so I decided to do sweep through the tps to see if theres any dead spot. Funny enough I found no dead spot. My resistance range on this tps is 1.1k ohm to 4.9k ohm but the thing is, at throttle plate closed, the tps resistance sweep is at 1.8k ohm. I decided to reinstall the tps but max out the sweep so I can pin the tps @ the 1.1k ohm mark and started the car up. (the tps is not screwed in here, just pressed in but rotated to max sweep for 1.1 k ohm closed.) I started it up (cold) and it idles perfectly fine! I then decided to rotate the tps opposite direction getting it to the 1.8k ohm (which is where the tps hole lines up with the throttle body mount) and the surging began. I also sprung a new code from this cold start, 1222. Could it be that I finally fouled my o2? Doubt it. I wanted to voltage test the tps sweep but the pigtails are sealed and I can't see how I would pull it back without stretching or ripping apart the seal.

edit: Just got a used tps sensor - it measured 1.6k @ closed throttle. I have done a cold start with it yet but I don't have high hopes because thats pretty high resistance. I'm thinking of just slotting the tps.
Double edit: I removed the metal retainers on the tps which gave it enough slop for me to shift the tps position so that it is at 1.35k ohms. Should be within adjustable tolerence. I took apart my original tps just to see if cleaning it and readjusting the metal fins to see if I can get a better reading but I could almost muster out 1.7k ohms at most. What I think I might do is add a conductive grease or lubricant to see if I can add conductivity and further reduce the resistance. This could honestly be alleviated by getting a new tps but I'll hold off on that.

[Province]

More than a week goes by and several hundred more miles of driving and cold starts are getting progressively worse. Logically it'd make sense that the tps could be causing the issue because @ 1.8kohms closed the voltage reading could correlate to throttle open making the engine run rich and mess with timing. But I never accounted for the TPS sensor and with that being verified, leaves me with only one last sensor that would be causing my woes and that would be the crank sensor. I want to include cam sensor as well but I believe that only serves as a minor assist in timing in comparison to the crank sensor which gives a fundamental feedback signal that dictates ignition unlike than the cam sensor. But hey, if replacing the crank sensor does nothing then the cam sensor is last and after that, I've truly exhausted every single thing that could be bad / contributors to the idle surge.

I still have this dead spot / hesitation at low rpm, so with the tps being confirmed in operation, leaves me to think of the crank sensor pulling timing at those rpms. Heres a random picture of a video I took when I was trying to dial the bottom end to top dead when doing the vanos. Quite some gunk there but I might give the sensor a wipe and see if the idle changes. It's not a hall effect sensor from what I've searched so that means the feedback signal is not a discrete value but a continuous value which means the sensor might be weak causing timing drift. The drivability actually has been getting progressively worse at the low rpm and I want to believe that it might be the sensor causing timing drift which ultimately could be the problem in the first place.

Too bad its the weekend so I'll have to wait a while till the new sensor comes in.

Edit:
Crank sensor installed, no change. Cold start idle has been getting worse - the car surges bad enough to where the car stalls out now when it runs for the first time in the day. I've just been starting the car up with TPS sensor unplugged. I feel as if performance around the 1500-2500 range has been diminshed a lot. I was thinking knock sensor, but I looked back at the work order from the head replacement, both knock sensors have been replaced. What was not mentioned was the cam sensor being replaced but before I decide to bite the bullet, I am thinking of doing more diagnostics. Also, when car is warm, the car idles higher now. Before the car would idle 550 and this was when I didn't really feel the low end hesitation. Now the car could idle between 700 to 800. Not surging just, holds a higher idle for some reason. The car would idle back to 550 sometimes but it's a little rare now.

I'm going to check the evap purge valve if its stuck open/operation, then I'll move onto the iacv electronical operation but I'm still doubtful because iacv has been replaced with a brand new unit.

[Province]

Took out evap purge valve and tested it using a vacuum pump, vacuum leaks through the valve when not powered. So I capped off the throttle body port that goes to the purge valve and started it up only to have it still idle surge. I don't have the huge low end hiccup I did before so thats a plus. I ohm'ed out the IAC, 12-12-24 all in ohms was the set of readings so electronically that is good.

Still, I unplug tps to have the idle go back to normal. I tried unplugging maf and the car dies out - again, signifying no vacuum leak.
The only thing I just can't fathom is by unplugging the tps, I'm bypassing something / overwriting something. No BMW text explicitly states what actually happens. I was thinking of unplugging cam sensor to force group ignition via crank sensor instead of bank by bank but the car warmed up by then so it'd be futile. I'll also be uploading two videos one of the open loop idle b4 I "deleted" the purge valve and one after. Both open loop.

[bmwstephen]

All things point back to software. I am curious how your car would behave with a stock tune and injectors

[Floxer]

Summarize this whole thread. Nobody reads walls of text.

[Rugbyfan]

This may be silly or you already addressed it (I skimmed), but what about the set screw holding the OBD-1throttle plate too far open?

[Province]

Fully closed.


If knock sensor's pulling timing, the CEL would indicate so?