The never, ever, ever ending cold RE-start problem... again

E28 technical advice asked and given! Troubleshooting, modifications and more.
oldblackcar
Posts: 233
Joined: Nov 11, 2009 11:03 AM

Post by oldblackcar »

TSMacNeil wrote:From my post on the Bosch Motronic...

(peeot was on the right track...but you don't need the CSV)

(1) you have never mentioned the IASV (ICV) does it vibrate on ignition on? Have you ever cleaned it?

(2) Since the ECU enrichens the first cold start, then leans a subsequent start (as described above) I'd check to make sure you have good clean injectors with plenty of fuel pressure at the rail to begin with. If your pressure is down, and the ECU leans from that...you may be too lean on your subsequent start attempts.
I've never touched the ICV - never had a reason to suspect it and to the best of my knowledge it has always worked. I guess a good cleaning would be in order.

I like the fuel pressure issues you guys bring up and I'll order a test kit. I tried to buy one from the stores around here but none of them have the right adapters and mounts to attach to our type of fuel rail. I was going to look at one from BAV unless you know of one better.

Thank you.
TSMacNeil
Posts: 7014
Joined: Jan 22, 2009 5:37 PM
Location: Phoenix, Arizona

Post by TSMacNeil »

You can get a generic kit from any parts store, and put it inline...pull the rubber injection hose off the rail, put the gauge inline there, use the rubber hose that comes with the gauge to attach to the rail.
Start car....observe fuel pressure.
oldblackcar
Posts: 233
Joined: Nov 11, 2009 11:03 AM

Post by oldblackcar »

I've looked at the kits at Advacnced, AutoZone, the local store and they're all the same - designed to fit into a specialized, ready fit adapter on the fuel rail. After examining the packages it seems like they are intentionally designed to not be used in the fashion you mentioned here. I don't see the rubber hose any more and I don't see any fittings that would lend to putting it inline. Believe me. I've looked.
TSMacNeil
Posts: 7014
Joined: Jan 22, 2009 5:37 PM
Location: Phoenix, Arizona

Post by TSMacNeil »

Well, I bought one from Checker that had some brass fittings, and two pieces of injector hose.

Or, buy a gauge, and two 12" pieces of injector hose and two hose clamps....less than $25 out the door.

Time to think outside the box here....the fuel line that attaches to your rail is just rubber line with a hose clamp...there is no "fitting"
so you don't need any fittings.
oldblackcar
Posts: 233
Joined: Nov 11, 2009 11:03 AM

Post by oldblackcar »

TSMacNeil wrote:Or buy a gauge, and two 12" pieces of injector hose and two hose clamps....less than $25 out the door.
I've tried this method as well and none of the chain stores have this either. There's one last place to look - then I have to order online. I was thinking of fitting some stuff from Summit Racing on the rail so that it's easier to check.

The real problem has been finding the gauge and hoses - I've asked in every shop in town and they all look at me like I have a third eye. I've even asked at the local HVAC shop. I think they're all afraid of liability or something.
oldblackcar
Posts: 233
Joined: Nov 11, 2009 11:03 AM

Post by oldblackcar »

I was thinking that something like this might do.
http://www.summitracing.com/parts/ANM-CP7802/?rtype=10
TSMacNeil
Posts: 7014
Joined: Jan 22, 2009 5:37 PM
Location: Phoenix, Arizona

Post by TSMacNeil »

oldblackcar
Posts: 233
Joined: Nov 11, 2009 11:03 AM

Post by oldblackcar »

What the hell is that over your TPS?
TSMacNeil
Posts: 7014
Joined: Jan 22, 2009 5:37 PM
Location: Phoenix, Arizona

Post by TSMacNeil »

A condom

It's an automatic tps, but I have a 5 speed. Yay.
oldblackcar
Posts: 233
Joined: Nov 11, 2009 11:03 AM

Post by oldblackcar »

I already have a length of hose coming off the CSV port on the rail. I just need to pop a gauge in and see whats what. I like the permanent job you did on yours. Did the gauge come with all the various elbows and hardware below it?
TSMacNeil
Posts: 7014
Joined: Jan 22, 2009 5:37 PM
Location: Phoenix, Arizona

Post by TSMacNeil »

No the Summit gauge came alone. I bought length of new injection hose, and went with some nice AN fittings from a hydraulic hoses and fittings shop in town.

Unless your CSV hose is newish, I wouldn't use it for a gauge
oldblackcar
Posts: 233
Joined: Nov 11, 2009 11:03 AM

Post by oldblackcar »

The CSV hose was added when I did the engine compartment hoses this past fall. I wasn't thinking of the permanent "BOSS" solution - just a temporary test line.
oldblackcar
Posts: 233
Joined: Nov 11, 2009 11:03 AM

Post by oldblackcar »

I checked the ICV and it's vibrating along happily when idle.
I'm not sure how to clean it.

Speed and reference sensor are spot on.
CTS ohms out at 205 while the car is at full operating temp. If that is the high number as per Bentley - then it is out of spec - but I don't have a digital thermometer to be able to really test the actual heat.

I'll test the CTS again when the car is cold and try to get an idea of it's cold reading. It's merely guess work without the right equipment.

Engine ground points have been thoroughly ground and cleaned for good measure.
oldblackcar
Posts: 233
Joined: Nov 11, 2009 11:03 AM

Post by oldblackcar »

I measured the CTS this morning - I suspect the car is around 70/75 degrees after sitting all night - and the CTS ohms out at 2512 cold. It would seem that it's within normal range - but hardly a scientific measurement.

I guess fuel pressure is next.
oldblackcar
Posts: 233
Joined: Nov 11, 2009 11:03 AM

Post by oldblackcar »

As it's always the last thing you check - I replaced the fuel pump relay yesterday and so far I've made 1 successful trip to the store without a re-start issue. Which is one more than I've had in a long time. I'll post back about how it's going...
oldblackcar
Posts: 233
Joined: Nov 11, 2009 11:03 AM

Post by oldblackcar »

Just got stranded again the other day. Same problem - so the fuel relay didn't fix it. On a hunch I checked the *NEW* coil I put on last year sometime and the resistance from + to - was .50 - which is fine - but the resistance from the center post to either terminal was 7,020 - which APB is out of spec.

I put an old coil on that hadn't quite given up the ghost and the car started and I made it home. So far I've been able to start and stop without an issue - and I've been able to go to the store 1 time and have it re-start without a problem - so the saga continues.

I don't know if the coil was the solution - or just another stone along the path.. M
gidgaf
Posts: 391
Joined: Feb 22, 2011 10:23 PM
Location: LAX, Kalifornia

Post by gidgaf »

I have an 86, with the same kinda problems. I never noticed it because all my driving is stop-and-go, at least a half hour each time. But it got harder and longer to start, and the short trip to the nearby market (three blocks) meant I had to shop for at least thirty minutes for me to restart the car.
So, after a hundred different other things I changed out the IAC valve. Even tho the other one worked ... this one worked better. Always starts, no short trip problems, and my gas mileage went up @ a mile. My buddy's car still has my old one, and it seems to give him a little better idle.
Win-win.
oldblackcar
Posts: 233
Joined: Nov 11, 2009 11:03 AM

Post by oldblackcar »

So the journey continues. After noticing a really bad fuel pressure bleed down overnight and reading around here - I ordered and installed the BMW fuel line check valve. As of yesterday the car started a lot easier after jumping the FPR and building up initial system pressure.

I went for a ride and shut the car off, only to be stranded again. No start after a short trip. I cleaned the ICV earlier in the week as some have suggested.

Now I'm off to try the next thing - seeing as how there hasn't been any change in behavior no matter what I try. I came across this thread and figured it's at least worth a look.
http://www.mye28.com/viewtopic.php?p=604813

I've tried to duplicate the behavior in the driveway and it hasn't happened. The car only pukes out on me after I've gone for a ride. I've even gone around the block and shut it down in the driveway. Only the have it re-start and fool me into thinking it's been fixed.
oldblackcar
Posts: 233
Joined: Nov 11, 2009 11:03 AM

Post by oldblackcar »

I was able to finally duplicate the scenario in the driveway. I've now confirmed that it is a no-spark situation.

When it happened I pulled a plug, set it on the valve cover and cranked the car - no spark to be had. Wait 10 minutes and the car starts to spark again. this is after 2 computers, 3 coils, 2 relays - same shit.

At least now I know that it's a no-spark which is causing it. Still doesn't explain why.
bm0p700f
Posts: 406
Joined: Oct 23, 2007 3:00 PM
Location: UK

Post by bm0p700f »

Duncan_uk wrote this on forum5 for no start issues. It does not deal with O2 sensors because we don't have them here.

I will echo not spraying the car with new parts and hope for the best. When I get a problem I consult my Bentley and begin testing as per instructions. Sometimes the problem has not been identified by the Bentley but further examination and testing has always revelled the problem. Sometimes I have had to draft help in as my testing skills are not the best.

Now here's the trouble shooting guide.

"Following the number of Motronic no starts of late relating to E28 535i and 525e first gen Motronic cars i though it about time i pulled a few common tests and problems together in one post. Please flee free to add comments below!

The following information has been gleaned from my experience, trial and error and others but mostly from the Bentley Manual, Haynes Engine management manual and the Haynes manuals for the e28 520 &525e and the 3 & 5 series manual. I cannot stress how useful these books can be especially the Bentley manual!

The advice below is offered in good faith and is implemented at your own risk. The Neither the author nor BMW5 can take any responsibility for damage caused accidentally or otherwise from following the procedures below.

Warnings by no means exhaustive!!!

The shock from HT leads can cause nerve damage ALWAYS use necessary precautions when dealing with the HT circuit.

Always have the ignition off when making or breaking electrical connections

Always take care to avoid ‘shorts’

Only use a test meter with a high impedance – other types may cause incorrect results or damage.

Switch meter ranges only when disconnected

Do not connect instruments with a 12V supply to terminal 15 of the coil – this can damage the ECU.

So to the diagnosis!


The system is not desperately complex but one must be methodical and check everything to ensure the obvious is not over looked. Do not discount new parts either!

Test equipment needed is a digital multi-meter and some leads with crocodile clips – nothing too fancy but we need to be able to measure volts (both AC and DC), continuity and Ohms.

So your symptom is that the engine turns over but will not fire? We are not looking here at running issues – for instance a duff AFM should allow it top fire if not run above idle.

Do you have fuel? Obvious but often over looked!

Was the car running fine before? If so could it be flooded? Try cranking over with a wide-open throttle for a good few seconds to see if it wants to catch. If so continue trying to start it and it should cough into life but you need a good battery for this.

Is the battery good? Poor batteries have lead to no start scenarios however good you think they are!

So to actual faults or failings:

Do you have a spark at any of the spark plugs?

Test by removing a plug and holding the plug against the block with insulated pliers and have an assistant turn the engine over whilst you watch for a spark. You are looking for a strong blue spark.

If there is no spark (or very weak yellow) then connect a ‘plug straight to the lead (king lead) coming from the coil and repeat the test.

If you now have a strong spark then a new rotor arm/ distributor cap or plug leads are required.

If there is still no spark, swap over the king lead (from coil to distributor) with another plug lead and repeat the test.

If there is still no spark we can assume for now that the plug leads, distributor cap and rotor arm and in serviceable condition.

If none of the above has yielded success we need to test the coil. Break out the multi-meter and set to DC 0-20V scale. With the ignition switched on we should have 12V between the +ve (15) terminal of the coil and ground (eg. The engine block or –ve terminal of the battery.

If not the coil is not receiving its power supply from the ecu. This leads us to suspect either the ECU or the continuity of the supply wire from the ECU and you should refer to the ECU tests described later.

If 12V are seen at the +ve terminal (15) then we need to test the coil.

With the ignition switched of and the multi-meter set to Ohms we can test the coil:

Between 0 & 15 (-ve and +ve) we should have 0.5Ω
Between 15 and the tower (where the king lead connects) we should have 6000Ω

Anything wildly different could suggest the coil is at fault and a substitution should be tried.

If the coil checks out ok we need to look for causes as to why the ECU is not sending the spark signal.

We must now confirm that the DME relay (Digital Motor Electronics or main engine relay) is functioning.

This is the 5 pin relay located on the left hand side of the fuse box, externally at the front of the fuse box. (BMW part Nº. 61361729004)

To test remove this relay and switch on the ignition then gently replace relay – it should be hear to click. If so we can eliminate the OBC security code, ignition switch and supply to the relay.

To confirm this we should check for the following readings at the DME socket:

With ignition on we should see:

+12v @ tab 30 (battery) red wire
0v @ tab 85 (earth) brown wire
+12v @ tab 86 (switched +ve) green wire
+12v @ tabs 87 and 87 (injector & ECU output) red/white and red/black wires

With ignition off:

0v @ 87 & 87

This relay feeds the fuel pump relay (located next to the DME relay) so it makes sense to substitute this relay next and test the fuel pump. The fuel pump relay is a standard type and a matching relay can be pulled from the fuse box for substitution.

The fuel pump should briefly hum when the ignition is switched on if not we can confirm the pump is ok by getting under the car (stands please!) and applying 12V direct to the pump terminals.

So if still no joy let us eliminate the crank position sensors there are two located in the gearbox bell housing and their plugs are located side by side at the back of the engine in a metal channel.

We can test these as follows:

Pull each plug in turn at the bulkhead and check for the following:

Each plug has 3 terminals we shall call them 1,2&3 from left to right. Looking at the plug we assume that the single locating key/ tang is upwards and the tow are down wards.

Between 1&2 we should see 960Ω +/- 96Ω
Between 1&3 and 2 & 3 we should see 100,000Ω minimum.

Switching to our AC volts scale on the multi-meter we should see between terminals 1& 2 nominally 0.7 and 1.4v AC being generated on cranking if not substitution of these components is required if still no figures given we need to consider that thetrigger pin has dropped off the fly wheel…..

Best follow the other tests to ensure that there can be no other possibilities before assuming the worst!

Make sure you put the leads back on correctly – with the leads swapped the car may run and misfire or it may not run at all! Lets not add to our problems!


Assuming the CPS sensors check out ok we need to look at the ECU we can eliminate this by substituting for a known good one but first we must ensure the ECU is seeing the inputs that it needs.

To do this we need to release the ECU from the glove box to allow access to the multi-plug to allow us to test the inputs the ECU sees. Please carry out all these checks regardless of where you think the fault lies – often it’s the combination of faults that gives us the clues!

The ECU multi plug is numbered as follows:

(1-18 is 18 pins across the top and 19-36 17 pins below)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35

1. Voltage supply to ECU: With the ignition on we should see 12v (battery voltage) between terminals 18 & ground and 35 & ground.

2. Main Grounds/ earths: With the ignition off we should see continuity between terminals 5 & ground, 16 & ground, 17 & ground and 19 & ground.

3. Fuel pump relay control: With the ignition on grounding terminal 20 should cause the fuel pump to run.

4. Starter input: On actuating the starter we should see 8v DC minimum between terminals 4 & ground.

5. Throttle switch (manual): Ignition off we should see continuity between terminal 2 & ground, on opening throttle no continuity and at full throttle continuity between terminal 3 and ground.

6. Idle position switch (auto): Ignition off we should see continuity between terminal 2 & ground, on opening throttle no continuity.

7. Idle stabilizer valve (535i only): 40Ω between terminals 33 & 34.

8. Engine speed sensor (CPS): With the ignition off we should see 960Ω +/- 96Ω between terminals 8 & 27

9. Reference Sensor (CPS): With the ignition off we should see 960Ω +/- 96Ω between terminals 25 & 26.

10. Coolant temp sensor: With the ignition off the resistance should vary depending on temperature (notionally 2000Ω cold and 300Ω hot)

11. Fuel injectors 1-3: With the ignition on when terminal 14 is grounded the injectors should click.

12. Fuel injectors 4-6: With the ignition on when terminal 15 is grounded the injectors should click.

13. AFM: resistance read between terminals 7 & 9 must fluctuate when AFM flap is moved by hand.

14. Coil: continuity between terminal 1 & coil –ve terminal

These tests should either identify you a suspect component or a combination of faults could lead you to a common connector plug or earth point.

In simple terms check:

Fuel
Good battery – perhaps try jump starting the car
Rotor arm/ distributor cap is intact
DME relay is working
Fuel pump is working

The above tend to point to the most common no starts!

Obviously the above does not take into account dodgy alarms/ immobilisers, which are a whole new headache!"
TSMacNeil
Posts: 7014
Joined: Jan 22, 2009 5:37 PM
Location: Phoenix, Arizona

Post by TSMacNeil »

oldblackcar wrote:I was able to finally duplicate the scenario in the driveway. I've now confirmed that it is a no-spark situation.

When it happened I pulled a plug, set it on the valve cover and cranked the car - no spark to be had. Wait 10 minutes and the car starts to spark again. this is after 2 computers, 3 coils, 2 relays - same shit.

At least now I know that it's a no-spark which is causing it. Still doesn't explain why.
Damn dude...
At least you've established it's a spark deal, and not fuel.

This is not my specialty(if I have one) and I know you've searched for "no spark"...but I'd start again with the basics of ground points, etc...hell, you may have a bad wire in a harness somewhere...but you'll find it. Since the coil is triggered with ground, I'd start there.
oldblackcar
Posts: 233
Joined: Nov 11, 2009 11:03 AM

Post by oldblackcar »

TSMacNeil wrote:..hell, you may have a bad wire in a harness somewhere...but you'll find it. Since the coil is triggered with ground, I'd start there.
Thanks for the words of encouragement. It's really starting to suck.
oldblackcar
Posts: 233
Joined: Nov 11, 2009 11:03 AM

Post by oldblackcar »

Yesterday I found some really dodgy wiring going to the CTS - I replaced the old splice work that was done between 7 and 9 years ago. The car ran all day without an issue. I still have to try the ol' "go to the store and get stranded" trick though.. Will report back.

I did confirm voltages across the main relay and coil so far - but I have yet to test the ECU connector.
shagrath
Posts: 14980
Joined: Jun 15, 2009 12:21 PM
Location: pLace

Post by shagrath »

I had an issue with my first e28 where the TPS full throttle wire was chaffed and shorting to the coil ground. Result? Intermittent black smoke rich condition running terrible then no start condition. Thought I would throw that out there.
oldblackcar
Posts: 233
Joined: Nov 11, 2009 11:03 AM

Post by oldblackcar »

After fixing some crap wiring on the CTS, adding a new rotor because the old one was pretty badly burned up, replacing the coil, adding a fuel line check valve, cleaning the ICV and adjusting the valve lash I was able to go to the store like a normal human being 2 times and not get stuck.

I'll keep my eyes on it and see if I still get stranded at some point. This morning was much cooler. I'm not sure if that made a difference.
Post Reply