The air reading is now right, like you said the air leaks will show lower reading (less air going past the sensor)
The two pin plug will be an oil sensor if it plugs by the alternator, if it plugs on the alternator it the charge active wire.
The acs (ASC?) is related to the throttle control as it backs of the engine output in low traction situations.
I think you’re on the right track with airflow/leak.
If it was me , I would see if you can introduce an air leak after the airflow meter( maybe a bit of vacuum pipe 2mm nothing to big) into one of the inlet manifold pipes maybe and it gets significantly worse I’d take a guess at the airflow meter, if it’s worse but not drastically I’d say you still have an air leak.
You’re welcome for the help, I’m actually on a BMW new technology course this week so the response time may be a bit slow.
The loose connector was a single wire from oil-pressure-switch next to the generator to instrument cluster idiot light. Now I remember unplugging it so I could get my hand in there.
Re-plugged the fuel tank pressure sensor. (accessible by removing boot around gas filler) That cured the p0446 code I caused following a dead end.
No codes!! No pending codes either!!!
By the time I drove 10 miles to the mechanic, the P0121 code and ACS warning light returned!!! The shop read and reset all codes I cannot access with OBD-ii, and said “leave it alone until symptom recurs. (writing on a blank slate); then return to shop to have codes read out again.”
(Before resetting, we tried unplugging the MAF; the car continued to run. So probably no serious air leaks any more. The pending P1188 and P1189 fuel trim codes are gone.)
Speculating: perhaps an intermittent wheel speed sensor causes the ASC and P0121 code to pop at random times.
With those indicators, light, subtle symptoms of surging came back at 1200 RPM or so.
The experts ALSO said the exhaust ‘sounds somewhat restricted’ at high RPM.
Muffler or Cat? Who can tell.
Thanks for the update, that throttle codes a bit weird returning,
are there any ASC codes or is it just the light?
If you want to give me the last 7 digits of your chassis I’ll see if I can find an engine wiring diagram as I don’t want to miss something being a US spec car as I’m not familiar with emissions equipment.
A blocked cat can cause it but there’s usually a really drop in all performance.
Seeing the ASC idiot-light in the instrument cluster, and feeling a little hesitation, prompted me to look for OBD codes, and there was the P0121 (and no others).
I can’t read ASC module specific codes with my OBD-ii.
Vin is W…AV22499
I found a schematic PDF file PN 001 00 1 470 215. 386 pages including M44 and M52.
I wonder if bad grounds could contribute to my symptoms. I haven’t chased down every one of them to inspect and retighten.
Unplugging/replugging the right connector can sometimes cure a problem by breaking through oxidation. (one of my careers was servicing 2-way mobile radios, where I learned most electronics system problems originate from something mechanical)
I wonder if prolonged operation with a vacuum leak can cause the cat to load up with carbon, and whether a few hours driving at speed can clear it out (somewhat).
I count 5 inputs to the underside of the intake manifold:
(a) CCV port; (b) the IAC port; © the vacuum circuit I fixed (3mm nipple); (d) the vapor canister purge (~7/16" stub);
(e) one other 3mm nipple I haven’t traced out.
Could that last one just be plugged, or is there another vacuum circuit?
Is the instrument-cluster miles-per-gallon gauge just connected to manifold vacuum via a 3mm tube?
Hi, I had a look this morning but I’ve got no diagnostic wiring schematics ( probably being a US only set up) your correct electronic faults are Generally mechanical In nature, ie connections corrosion etc so you’re doing the right thing checking earth and power supply, preferably with a decent meter. The throttle sensor should be a three wire set up an earth , power and sensor, it’s best to check at the DME Engine ecu as you’re taking into account the wiring as well, check you have a good off idle then smooth voltage increase then a resistance ( power too sensor) as it’s opened by someone on the pedal ( not by hand on the motor) The E 36 did suffer from corrosion around the lower screen and subsequent leaks into the DME ( it’s behind the cover held by a few 8 mm screws)
In one very rare case years ago I had one building static electricity in the front wheel bearing due to corrosion in the abs ring, this then discharged via the wheel seed sensor but that caused a electrical shut down of the engine, but that was a once in a lifetime fault, saying that when cars get old they can have some very peculiar things go on. An air leak can’t cause a cat to block because it’s running lean an engine running rich can. The mpg gauge is electronic via the DME from memory
Thanks for answering that blast of questions.
The pending P0121 TPS code eventually set as a hard check-engine code… again.
No other faults were read out by the shop’s very smart reader, just the incorrect TPS value code. In fact, his reader showed the TPS reading as sometimes “idle” when it should have been “partly on.”
The +5V end of the TPS shares connector X6000-pin-44 with the +5V side of the fuel tank pressure sensor.
Page 97 of the schematic has a hint at what the problem could be. That page depicts a resistor between pin 44 and the +5V supply in the DME. Since that resistor sources current to both TPS R6252-pin-3 and the fuel tank pressure sensor B72-pin-1, fluctuations in the current drawn by the pressure sensor “modulate” the throttle position signal. B72 could be faulty intermittently drawing excessive current, or the wire that supply current to B72-pin-1 could be intermittently shorting to ground where pinched or where insulation is worn, or the connector could be corroded due to water damage.
The “ground side” of both sensors return to the DME through separate pins.
Likewise the +5V wire to R6252-3 could be shorting or intermittently open.
However it’s a straight shot from X6000 to X6052, so a fault is less probable,
and easier to find.
The Tech suggested a parallel set of wires from the DME to the TSP.
I’m betting on the pressure sensor wire.
Principles of designing for minimum noise coupling say to avoid common impedances,
like that resistor from pin 44 to +5V. Shared ground paths too. Single point grounds are also preferred; there are many connections to the chassis in this car.
My other career with physics apparatus provided lessons in noise suppression.
I have some stories that are apropos.
If the tps is showing incorrect idle that could well be a voltage drop and like you’re guy said easiest is to run another 5 v supply. Most of the sensors have a 5 v supply from the DME so utilise one of those. I see according to the factory build sheet it has cruise control and ASC+T (T traction control). If that’s correct that may have a second throttle cable to the throttle body as it uses an electric motor to override the driver input, if that’s correct and it has them you might be worth disconnecting the second cable and give that a go. I still can’t get any wiring information on any our systems although I can see most of the components through the parts system.
I did some exploring of the DMC today with an ohm-meter.
X6000-Pin-44, the 5V power to the sensors, connects to the open collector output (pin 1) of an LM293 dual comparator. The comparator (when commanded by the CPU) probably pulls pin 44 to ground as part of self-testing of both the TPS and tank pressure sensors.
So separating the power leads to the two sensors will probably give me another problem.
So I either find the fault with the wire to the pressure sensor, or I clip the wire at both ends and run a separate wire from pin 44 all the way back to the pressure sensor.
Pin 44 of the DME connector only has one wire crimped onto the terminal. Since the wiring harness schematic shows a junction, I was expecting two. I opened the wire tray above and behind the fuel rail cover. I think I see wire splices in it. So I have a chance of finding that wire splice in there tomorrow.
I found a little documentation on the (Siemens) MS41.1, but I’d like to have a schematic to confirm my finding.
Yes. this car has Cruise control and ASC.
ASC has a separate butterfly (with aux. throttle position sensor) which bolts to the throttle body. (The ASC butterfly self-tests when turning the key)
The cruise control cable connects to the throttle body. CC works normally, so I don’t think it’s part of the problem. I could just unplug the CC motor and take a drive to check.
BMW schematics of any control unit internals are notoriously in accurate. There drawn to depict a generic unit not necessarily your one.
The’re usually a very robust system electrically wise, there’s a lot of filtering of signals by the software, I’ve seen some pretty poor signals on cars with a scope, interference etc and they’ve worked fine. Unless you’re seeing 25% off on the readings I would say you’re in spec. If you do suspect maybe an earth fault try a jumper cable clipped to the engine (lifting eye usually ) and a good body earth .
There maybe only one wire from the DME and it probably splits via a solder joint in the loom ( shown on the diagram by a solid dot on the line as opposed the the usual connector symbol.
Like you said try without the cruise at least you’re ruling it out.
Disconnecting the cruise control motor has no effect on symptoms.
It wasn’t too hard to find the (pin 44) sensor power circuit splice in the wire tray. (The colors differed from the schematic, though) The three wires were crimp in a ferrule with a plastic insulator over it. Disconnecting fuel tank sensor did not cure the throttle stutter & P0121 problem.
(and as expected it gave me a P0446 code)
Damn it. It was such a plausible theory.
So, next, I’ll put a meter on pin 11 of the DME connector.
Pin 11 connects, through an RC filter, to an ADC input on the CPU chip in the DME.
I’ve repaired other appliances like internet access points, digital TV receivers, and even a coffee grinder. What they had in common was age and electrolytic capacitors of relatively poor quality, usually in the power supply. Limited life consumer electronics.
The DME doesn’t have electrolytics. The tantalum capacitors have a much lower failure rate. DME components meet a -40º to + 105ºC spec. The design is mechanically robust. No radial-lead components. So I’m inclined to look for other mechanical problems.
That’s a bummer when you think you’re onto something and it’s a dead end! This is problem that’s is not uncommon in auto repair these days, with all the electronic equipment it’s easy to go past that all there doing is try to burn and regulate an air/fuel mixture. If I get issues like this and don’t get anywhere after a while I try and “mechanically bypass to fault “ after all that’s all we used to do on carbureted engine’s, when there adjusted your just compensating wear. That’s all the engine management is doing- add an air leak and it weakened- restrict a fuel return and its enriched. At least then you know where you’re heading with the fault. As far as engine management ecu share concerned they are usually well built due to the environment there operated in, we tend to forget the unit is a computer and that’s it- they have a logic path for faults A+B=C, we get drawn into thinking they see what we are seeing. If it see’s a richer mixture it can only work out by extending and lessening the injection period what the cause can be and in turn its reading that information of being over rich from the oxygen sensors , if the o2 sensor is wrong off we go on the fault tracing path for an over fuel issue. I had an X3 last year that had a fuel issue, on live data all the o2 sensor were reading spot on, in reality when the second bank post cat sensor was unplugged it was still reading fine! The ecu had an internal fault in the processor.
You certainly sound like you’re on the right track in thinking it’s mechanical and the DME is reacting to something, I can’t remember have you tested the throttle sensor off the car ( wired off the DME)?
The TPS seems smooth when rotated by hand. Looking at the resistance while rotating, it seemed to behave smoothly.
The stuttering doesn’t seem tightly correlated with a particular (narrow) throttle position or range.
I looked at the pins in the plug. They seemed remarkably wide. I’m gonna make them squeeze the TPS pins tighter. Going to monitor the TPS signals at the sensor too.
I hate the thought of soldering wires to the DME in order to monitor the values there too.
If the problem is the DME, how important is the “Index number” of the unit?
I know the trottle-load vs timing and fuel injection tables are programmed in for a particular engine. Mine is “5WK9 0353index 07”
Hi mate, I’ve got a Jaguar xjs with the same issue, the original tps is no longer available so I’ve had to get the later version which requires an adapter mount - now I know what I’m doing , but some where there is movement that come into the system somewhere and puts it periodically out of adjustment. When I check at the ecu it’s out,as soon as I realise the first screw to check it comes back in- so they are a very accurate thing especially as the bm one has the idle signal.
If you trying wires into the DME it’s worth getting new connectors or second hand for that mater and re wire totally separately. The correct pins are fairly cheap, if you can’t find any try a junk European car, most yards in my experience are happy as the don’t usually sell the wiring parts. The pains are coded by the little ears an the blades so be careful you don’t damage your plastic carrier.
Your index should be all good but I can check on Monday at work if you want to be 100% sure.
I poked some wires from a length of cat-5 network cable into the back of the TPS connector pins. I routed the cable into the passenger compartment where I could watch the 5V (with a Digital Volt Meter with resolution of 1mV) for many minutes. No glitches. No dropouts. (The DVM has a max/min/avg feature) Variation of 1millivolt over many minutes. The ASC light came on anyway.
The signal from the wiper never dropped out either. The value stayed around 0.75V at idle; it would drop to zero if there was an intermittent open-circuit. Noise is harder to judge.
I found the CPU chip in the DME. It’s the rectangular 100pin Siemens part.
Pin 50 is the TSP digitizer input. There was residue of solder flux along that row of 20 pins.
Leakage? Plating migration between pins? So with some pure alcohol and a toothbrush I got all the solder flux washed out and took another drive after resetting the check engine codes. The pending P0121 is still with me.
Pin 12 on the DME connector (TPS wiper) has a 98KΩ resistor to ground, and a 4K resistor between the DME connector pin and pin 50 of the CPU chip. (Pin 50 is a 10 bit Analog To Digital converter input). The very last thing I can do is tack a wire onto pin 50 (with a 10K series resistor to protect it) and monitor that node when the car is misbehaving. (Soldering to the DME is not a job for the average mechanic. I have the electronics technician skills and tools to work on surface mount PC boards)
A good, reliable signal at pin 50 will verify the integrity of the TPS signal at the ADC input pin.
The only remaining component is the ADC in the DME.
A Jaguar, alas. Old enough to have Lucas electricals? Is it a Tata?
Did you try to measure the original sensor? Is it truly bad?
Apply 5V between the ends of the sensor, and a few KΩ resistor from the wiper to the (minus) side of the 5V supply. connect a capacitor between the wiper of the sensor and the tip of an earphone. Connect the ring of the earphone to connector to the (minus) side of the 5V supply. Rotate the wiper up and down a few times while listening for noise. You shouldn’t here any popping or sizzling at any position or while rotating the sensor.
Looks like it’s all pretty sound, you’ve tested as thoroughly as is possible and you can’t do more than that . We don’t have much to do within the units themselves as we would trace as far as them then replace, but it’s interesting to understand what’s inside. I’ll see if I can get some information on the DME interchangeability on Monday if you like.
The jag is an older one and a V12 at that, there a bit of a known thing with the tps- you have to love Lucas electrics😃
Yes, please check interchangeability.
The CPU chip can be found on eBay for $10us; it’s a 100 pin package with gull-wing leads.
I have a Norton Command so am acquainted with Lucas electrics.
I’d also like to read a troubleshooting guide for the ASC-T to see if a TPS problem causes the light to go on. I think it does.
No problem, I’ll check that Monday. The only trouble shooting is via the old diagnostic system DIS and only then with the car connected, to be honest and from memory you’ve most if not all- the only thing you can’t replicate is it gave target and actual values. I’d say it’s an algorithm as the ASC is a safety system, that means it has to be satisfied it can perform correctly ( usually if a parameter isn’t met it will give the driver the warning but won’t store a code as it’s an other system related fault)
Hey, had a look on our system today and as far as I can see they should be compatible on any index for that part number, it will need alignment electronically to the EWS system if it has it