Bericht: Motoruberhohlung E28S M5

It is about time to move forward with installing auxiliary parts. I started with the AC-spanner rail that I sourced from IvoC. It is a bit larger then the original for some reason, but two additional washers did the trick.

[Blockierte Grafik: http://rwoe.150m.com/tmp/20080301/img595.jpg]

Since the alternator pulley was no longer available I had the old part powder-coated.

[Blockierte Grafik: http://rwoe.150m.com/tmp/20080301/img596.jpg]

The paint-finish of the engine-shackle got damaged during the engine installation so I had it powder-coated as well.

[Blockierte Grafik: http://rwoe.150m.com/tmp/20080301/img597.jpg]

After tightening the 24mm nut from the alternator pulley with a power-tool, I reinstalled the alternator back onto its sub-frame on the engine. I then reconnected the ground-return lead with the M6 bolt to the alternator. However, upon tightening the bolt, I noticed it turned through, indicating a damaged thread. To solve this, I relocated this connection to the suppression capacitor-bolt for the time being, about a centimeter away.

[Blockierte Grafik: http://rwoe.150m.com/tmp/20080301/img598.jpg]

Last but not least; the water pipe in front of the engine. I already replaced this part at the end of 2004 so there was no need to have it powder coated.

[Blockierte Grafik: http://rwoe.150m.com/tmp/20080301/img599.jpg]

The next step is the reinstallation of the engine wiring-loom; but lack of time today prevented me to more forward with this.

I last week resumed with the electrical installation for which I wanted to install the idle-control valve prior to the engine-wiring loom. However, upon installing the flange for this part, I discovered it was cracked around the upper mounting hole. A new replacement part still is available in Munich so I quickly ordered one as replacement. This will arrive next week together with the new viscous fan for the cooling system. These are the two issues for this month;

1: Completing the electrical installation
2: Completing the Cooling system

Even though the old flange for the idle control valve was FUBAR, I pre-fitted the idle control valve to check for other issues. One would hardly expect one, but the M88/3 uses a different ICV-design then the M30 and S38 engine families and there is little clearance between the M8 wire-ends for the throttle-bodies and the ICV itself. And indeed, there was no clearance left so it seems as if the M8 wire-ends were too long. A comparison with my spare M88/3 head confirmed this to be the case indeed so I replaced them with the correct parts that I found in a box with excess parts from the engine rebuild. The following pictures clearly show the difference.

[Blockierte Grafik: http://rwoe.150m.com/tmp/20080301/img600.jpg]

Even though I protected the intake ports with tissues and there was little chance for these to enter the intake ports, I replaced the wire-ends one at a time.

[Blockierte Grafik: http://rwoe.150m.com/tmp/20080301/img601.jpg]

The following picture shows the progress of this week. The purple arrows shows the progress of the coolant system. The short summary is that the thermostat, thermostat-house and all the coolant hoses, except those to the radiator have been reinstalled. The existing parts are less then three years old and have been used for no more then 1000km so I did not bother to replace them.

[Blockierte Grafik: http://rwoe.150m.com/tmp/20080301/img602.jpg]

Another big-step is the reinstallation of the engine wiring-loom (yellow arrows). I did not secure the new branch for the temperature senders and sensors yet as I first want to add some protection against mechanical and thermal stress around the heat-shrinkable tubes (red arrow).

Reinstalling all the goodies is not the most challenging work, as it doesn’t challenge my intellect nor can I make a really good write-up from it. But it has to be done anyways as otherwise #231 will not run in the foreseen timeframe. Anyhow, I made a hughe step today by starting to check the tightening torque of the engine subframe to the chassis, the ground thrust arms, the steering box, pitman arms, etc etc.

Last week, I reported that I broke the flange for the idle control valve. BMW still has some replacement parts stored in Munich and for 20 Euro’s one cannot really bother. I gave the ICV a good clean before installation.

[Blockierte Grafik: http://rwoe.150m.com/tmp/20080301/img603.jpg]

I replaced the fan-blades and viscous clutch with new OEM parts. I didn’t really want to reuse the old parts and for the cost, one cannot bother.

[Blockierte Grafik: http://rwoe.150m.com/tmp/20080301/img604.jpg]

I wrapped some wire-protection around the new section for the temperature senders before final installation. Note that I still have to connect the timed switch.

[Blockierte Grafik: http://rwoe.150m.com/tmp/20080301/img605.jpg]

The next step was the installation of the V-belts and adjusting their tensioner bars. Note that the V-belt for the AC-pump has been omitted for the time being (See next pictures)

[Blockierte Grafik: http://rwoe.150m.com/tmp/20080301/img606.jpg]

Whilst placing the V-belt for the AC-pump around the pulleys, I noticed that the V-belt was much too small. Please note that this is exactly the same part that I removed about a year ago, but even with the spanner rail of the AC-pump in its minimum setting, there is way to place it around the pulleys.

[Blockierte Grafik: http://rwoe.150m.com/tmp/20080301/img608.jpg]

The belt itself says 13x810. This contradicts to REALOEM that specifies 12,5x800!!! According to the bill of the main dealer who replaced them as part of the last inspection II service, the correct part (12,5x800) has been installed. I checked my notes from the disassembly process now about a year ago; this belt could only be removed by unbolting the AC-pump from its sub frame.

[Blockierte Grafik: http://rwoe.150m.com/tmp/20080301/img607.jpg]

Anyhow, since the AC-system needs to be checked and refilled with the old and trusted R12 coolant in 2010, I decided to omit this V-belt for the time being. I do not really need the AC system before the summer. I will buy a longer V-belt; I think that 830mm or 840mm will suffice.
The next step is the connection of the reverse-gear switch. Whilst removing the gearbox last year I disconnected the switch and removed the rubber boot when I noticed that the wiring has been patched with faston wire-wire crimps in the past.

[Blockierte Grafik: http://rwoe.150m.com/tmp/20080301/img609.jpg]

I replaced the wire-wire crimps with a soldered connection and heat-shrinkable tubing.

[Blockierte Grafik: http://rwoe.150m.com/tmp/20080301/img610.jpg]

After pulling the wiring through the shifter boot, I placed a new heat-shrinkable tube over bother wires to improve their protection. Last but not least, the both faston-crimp contact were isolated as well before they were connected to the gearbox its reverse-gear switch.

[Blockierte Grafik: http://rwoe.150m.com/tmp/20080301/img611.jpg]

I did contemplate about replacing the radiator with a new OEM part, but it an original Behr part and visually in excellent condition. The metal still shines and the plastic water-galleries including the hose-connections is in excellent shape. This was confirmed by a pressure test that gave it the all clear.

[Blockierte Grafik: http://rwoe.150m.com/tmp/20080301/img612.jpg]

Work is moving fast forward at the moment and I expect to reach one of my objectives for January; the reinstallation of the butterfly valves already this week.

I installed the exhaust system yesterday evening. This is a relatively new part from spring 2008 with little use. I only need to buy new chrome-ends with a diameter of 50mm. (Unlike the *** that mentiones 45mm)

[Blockierte Grafik: http://rwoe.150m.com/tmp/20080301/img613.jpg]

With the reinstallation of the throttle bodies, the 'engine-installation' is nearing its completion. The lead-time is now determined by the injectors that will be send to von Prickartz consult in Germany for thourough testing and an overhaul. They will provide a full test-report with the results of the relevant test-items prior and after the overhaul. This will give me an insight of the previous overhaul of these injectors that was carried out by a local Bosch-specialist in 2005.

[Blockierte Grafik: http://rwoe.150m.com/tmp/20080301/img614.jpg]

Later more.

Thomas, my youngest son is completely fascinated with technical stuff and likes to check my progress on the E28S from time to time. When he spots new parts on the car and finds the same part in my trash-bin, he tries to claim them for his own use. :hihi:

[Blockierte Grafik: http://rwoe.150m.com/tmp/20080301/img615.jpg]

BTW; I did find some time to install the intake plenum though.

It has been a while since I last wrote an update. The reason for that is that I am awaiting the injectors that I have send to Germany for further investigation just in the period that Prickartz consult is closed for three weeks for holidays and resumes operations as of next Monday. This was a known fact before I shipped the injectors, but the added value of Prickartz justifies the delay.

In the mean time, I have ordered the last batch of parts, including new spark plugs and a new adapter-spacer for the distributor-rotor. These arrived yesterday.

[Blockierte Grafik: http://rwoe.150m.com/tmp/20080301/img620.jpg]

Since the Bosch X5DC require an 18mm nut for fastening and the internal clearances are restricted for normal tools, I simply use the appropriate BMW tool that is part of the tool-box. When properly used, it will apply the required 23Nm pretty accurately.

[Blockierte Grafik: http://rwoe.150m.com/tmp/20080301/img625.jpg]

With ~46 Euro’s ex VAT, the adapter-spacer for the distributor-rotor is not cheap, but I didn’t want to reuse the old-part. The new part fits much better compared to the old part. Just make sure, you lock the camshafts somehow before tightening its bolt!!

[Blockierte Grafik: http://rwoe.150m.com/tmp/20080301/img621.jpg]

I removed the valve-cover to add some lubricant over the cam lobes. TMR already added a healthy supply before assembly, but I wanted to add some 15W40 mineral just in case. Furthermore, I re-checked the TDC alignment of both cams with the crankshaft. Whilst assembling the clutch, I manually retarded the timing with 30degrees to increase the safety margin on the intake valves of cylinder #6.

[Blockierte Grafik: http://rwoe.150m.com/tmp/20080301/img622.jpg]

I also installed the air-filter cabinet, the AFM and some other small parts. The AFM has been replaced just a few months before the engine with a new OE part.

[Blockierte Grafik: http://rwoe.150m.com/tmp/20080301/img623.jpg]

The fitting of the ignition related parts are simple and straightforward. Not much to tell about them.

[Blockierte Grafik: http://rwoe.150m.com/tmp/20080301/img624.jpg]

The only thing that has to be done now is the completion of the simple issue list and the fitting of the injectors. I hope that these arrive in two weeks from now. Then a full inspection II service shall be carried out before completing the installation phase and start with the recommisioning.

As I already wrote, I had the injectors ultrasonically cleaned and tested according to the ASNU method by a local Bosch service agent in the winter of 2005. Even though less them 2000km have been driven since then, the risk of reusing them as-is simply is too high considering that the total cost of this project has exceeded the five figures in European currency by far.

Since another main engine-problem with another car in our fleet has raised questions about the quality and capabilities of the very same Bosch service, I wanted to have them tested by another company. My requirements were rather simple; (1) Inspection and testing the injectors as received and log the results in a test-report; (2) Carry out the I injector service and retest the serviced injectors and log the results in a test report; (3) compare the test results prior and after the injector service. This test-report is very important, as this is a main part in this dossier before starting the re-commissioning of #231.

Problem is to find a company who works that way; it is not difficult o find a company who offers an injector service, but not according to my requirements. Then in the summer of 2009, I read an article in the German ‘Oldtimer Markt’ about problems with fuel injection systems. In that article, the author used a 1970’s Mercedes Bens R107 350SL as an example to prove the effectiveness of the ASNU method that is widely used. The procedures and methods used for that example very much complied with my requirements so it did not take long before I contacted Prickartz consult in Germany.

When I removed the injectors a few days after Christmas 2009 and prepared them for shipping, I documented each injector and marked its position. This allows me to compare the test results with the engine failure and hopefully draw some conclusions for the Failure Mode Analysis that simply is not complete without analyzing the injectors.

[Blockierte Grafik: http://rwoe.150m.com/tmp/20080301/img629.jpg]

On last Monday, I received an Email from Bernd Prickartz with the inspection and test results of my injectors. Upon the visual inspection, he already found some abnormalities.

1: The internal filters do not comply with the type specified for the Bosch 0 280 150 201 injector.
2: The O-rings were not of the correct size
3: The injector enclosures were unpainted
4: The spray nozzles were not of the correct type.
5: The spray-image of all six injectors did not match the criteria.

After the required corrective measures, Bernd Prickartz carried out the ASNU cleaning service and retested the injectors. The result was that five of six injectors passed on all criteria (dynamic-and static test and spray-image). Injector number four (#4) was rejected though and required replacement. I agreed to Bernd’s proposal after which he ordered a new injector, tested it and updated the rest-report with the results. That was last Wednesday.

Already today, UPS delivered a parcel that contained seven injectors (the defective #4) and the six tested injectors (including the replacement for #4). The injectors were shipped with protection caps on both sides and packed in an encapsulating sealed bag.

[Blockierte Grafik: http://rwoe.150m.com/tmp/20080301/img630.jpg]

The injectors themselves were also individually packed in a sealed bag.

[Blockierte Grafik: http://rwoe.150m.com/tmp/20080301/img631.jpg]

It is nice to compare the injectors prior and after the injector service. The right injector is the rejected #4 with the incorrect spray nozzle. The left injector is the overhauled #1 with the correct spray-nozzle. I have to admit that the spray-nozzles is not the fault from the company who performed the injector service in 2005. They just cleaned the injectors and did not replace the nozzle, but they should have spotted that it was incorrect.

[Blockierte Grafik: http://rwoe.150m.com/tmp/20080301/img632.jpg]

The rest is simple; placement of the injectors according to the imposed placement restriction based on the deviation of the average dynamic test.

[Blockierte Grafik: http://rwoe.150m.com/tmp/20080301/img633.jpg]

All in all, I am very satisfied with the injector service from Prickartz consult. With a price-tag of 30 Euro’s per injector, he certainly is not the cheapest, but the added value including a full test-report justifies the extra investment.

Prior to the ASNU injector service, the static flow test results for all six injectors are within 250 and 254 with an average of 252,67. This is a good result, hence why there is no reason to assume a relation to the burnt exhaust valves of cylinder #4 and the lean running condition on cylinder #3.

Prior to the ASNU injector service, the dynamic test results are within 87 and 92, whereas the minimum criterion for an injector to pass the test is 90. This means that injectors #3, #4 and #6 are rejected. With other words, there is a direct connection between the too low dynamic test results of cylinder #3 (87) and the white color-tint of its exhaust valves that indicate a lean running condition for this cylinder. Although the injector for cylinder #4 test slightly better (88,50), its spray-pattern was very poor. The direct result is that the injectors leak which causes poor combustion with increased carbon build-up on the affected parts. This is confirmed by the emissions test that was carried out six weeks before the engine failure last recorded emissions test (20 April 2008) before the engine failure in which the values for unburned hydrocarbons and CO2 exceeded the limits.

Based on the inspection results, it is concluded that local Bosch service did not follow the appropriate repair instructions during their injector service in 2005. Due to the lack of a test-report from that injector service, it is impossible to compare the test results. Based on that, it cannot be proven that teh local Bosch service did not service the injectors incorrectly or used the wrong test-criteria. However, the fact that they have missed (1) the wrong spray-nozzles for this type of injector, (2) used the incorrect filters and (3) used incorrect size of O-rings, clearly is proof that they did not inspect the injectors well enough nor did they follow the appropriate repair instructions.

I completed the last installation work underneath the engine bay today. The injectors are installed and connected to the fuel rail and engine wiring loom. The throttle bowden cable is connected to the throttle lever on the butterfly valves and the last electrical connections have been made.

[Blockierte Grafik: http://rwoe.150m.com/tmp/20080301/img634.jpg]

I also reinstalled the Motronic unit and the connectors for the relays. These will not be connected untill after checking the oil-pressure.

[Blockierte Grafik: http://rwoe.150m.com/tmp/20080301/img635.jpg]

Apart from the usual pre-start checks and filling of the power-steering reservoir, the installation-work is completed. The checks will be carried out next week after which I can crank the engine to build up oil-pressure and hopefully the first start the week after. I won't drive it until March so I am still debating weather it makes sense to start the engine already in the 7th week. An option is to delay that to the 8th or 9th week.

Well Gents,

The all expected milestone was reached yesterday evening. I first crancked engine 42243748 to pump oil through the engine until the oil pressure light turned off. I then connected the DME, fuel pump relais and ignition coil ground lead after which the long awaited moment neared. I felt the tension and asked myself, (1) was this worth the 14k Euro investment and (2) Does all go well?

Then we started the engine; a brief moment of doubt; but then I realized that the fuel had to be pumped around first as well so it took another 10sec or so before 42243748 bursted into life. There was a small dejavu moment about a missing part, but that was corrected immediately. We then checked and topped up all the fluids, crancked the engine to build up oil pressure again and started for the second time. This time, 42243748 bursted into life immediately.

There is a hughe difference compared to the behavior of 42243748 that I knew before. Timing-chain is silent and can hardly be heard; the engine does not vibrate at all; idle-speed is constant and can be adjusted properly; the irregular throttle behavior disappeared and the injectors can be heard ticking as they should do. The idle running characteristics is much smoother then the 190k km S38B36 in my E34S ///M5. I'd rather compare the idle-running behavior with a well-running M30B30. In fact 42243748 idles smoother then the 180k km M30B35 in the E32 735i. Engine 42243748 truly sings, instead of blurking like before the engine failure. We kept 42243748 idling for about halve an hour to warm it up; check for leaks; to allow trapped air to escape and to verify the proper operation of the thermostat.

Now I have an engine that is as close to new as possible, I have more respect for the engineers that designed and manufactured this landmark of engineering. The running behavior before and after the rebuild is a huge step forward, proving that the bulk of the current M88/S38 engines have aged gently, during which they develloped some symptoms that some might describe as character.

All in all, I am satisfied with the results of this milestone. It allows me to close the bulk of the existing issues, but as I found some others (operation of the coolant-temp gauge), #231 is not ready for recommisioning yet.

To be continued.

Gents,

One of the issues that came out from the first-start was the failure of the water-temperature gauge. I investigated this problem and checked all relevant connections, series resistance of both coils from the gauge, replaced the NiCd batteries on the SI-board and the damn thing still didn't work (not by starting, but by shorting the NTC-resistor in the temperature sender). Since the Si board is placed in series with the temperature sender and the water temperature gauge I measured the voltage on pin-4 of the diagnosis connector; 0V => Not OK so it appears the SI-board is partially defective.

To test this theory, I bypassed the SI-board. This connected the temperature sender directly to the temperature gauge; I measured the voltage on the diagnosis connector; result 6080mV => Bingo. As a last test, I shorted the temperature sender to ground to check if the gauge moves to the full-scale => It did. With these two tests, I can proove conclusively that the SI-board is broken.

A new SI-board is 342 Euro ex VAT. I can always decide to do that, but for that money, I'd rather make an attemot to repair it on a component level. But I will first have to analyze the circuit so the spare SI-board (from a 524td) comes in handy. Untill then, the SI-board bypass is a work-around that allows me to drive the car without worrying about missing an engine-overheating condition.

The main focus for last week was to get the car back on the road. For this, I had to pass the Dutch APK, which is an annual safety inspection combined with an environmental test (emissions). The brakes were still an issue that was not addressed so we checked them last week and at the same time replaced the brake-oil. What’s left were some minor issues as the secure fitment of the battery, the operation of the headlight height aim control and some other points that were not addressed during the engine rebuild and engine reinstallation.

We started with the emissions test. It is not required for my E28 M5 to meet other requirements then CO; but by performing an emissions test according to the four-gas method that is mandatory for all cars from 1/1/1993, one can check the health of the cars ignition and fuel-injection systems. The following picture shows the test-setup with a Sun MGA1200 calibrated emissions analyzer.

[Blockierte Grafik: http://rwoe.150m.com/tmp/20080301/img651.jpg]

At first, we measured the exhaust emissions with the idle, CO and basic-TB setting in the same positions as they were prior to the engine failure in June 2008. This allows a comparison with the exhaust-emissions test-results from April 2008 in order to proof that the injectors were the main culprits of the engine-defect.

To obtain sufficient data, I recorded the analyzer results every ten seconds for about five minutes in order to find the minimum and maximum values. The results were as follows.

CO: 2,16 – 2,31 (%)
CO2: 13.11 – 13,15 (%)
Hc: 245-258 (ppm)
Vol O2: 0,42 – 0,44
Lambda score: 0,942-0,948

Although the official emissions test-protocol only uses the CO value for determining a PASS or FAIL for a 1985 car, this parameter is not really that important from a technical point of view. Much more important is the ‘lambda’ score and ‘hc’ readings. If for instance the spray pattern from one or more of the injectors is not optimal or the ignition has even a slight defect, the ignition will be poor with as result that not all hydrocarbons (hc) will ignite upon ignition. This will also result in a very high lambda score; i.e. lean running condition that will eventually destroy your engine as happened to me on that sunny Sunday in June 2008.

The test results prove that the lambda score is typically 0,945 with a tolerance of +/-0,03. This indicates an engine that is running a tad rich. The ‘hc’ reading should not exceed 300ppm, which it does not, even at the raised CO-reading of 2,31%. All in all, these results verify the ASNU test results of the injector overhaul carried out by Prickartz consult in Linnich Germany.

Even though I would be perfectly happy running the engine in above setting at least throughout the break-in phase, it does not pass the official emissions test since these primarily rejects based on the CO reading that for a 1985 car without a catalytic converter should be no more then 1,5%. In order to obtain a healthy safety margin in case of a verification-check by the Dutch RDW, we set the CO to typically 1% and repeated the measurements.

CO: 0,95 – 1,04 (%)
CO2: 13.71 – 13,73 (%)
Hc: 220-221 (ppm)
Vol O2: 0,71 – 0,79
Lambda score: 0,993-1,001

The higher lambda score was expected, but sill is on the safe side from a lean running condition (lambda >> 1). The ‘hc’ reading also is about ten percent lower. To reflect back to June 2008, I compared these results with the emission test results from April 2008 that also was carried out at a CO of typically 1%. Based on these results, I ca safely say that the engine failed due to a lean running condition caused by faulty injectors and an incorrect operating ignition. The faulty injectors are what angers me specifically since these were overhauled by a local Bosch service station (whose name shall remain disclosed) not long before the engine failure occurred. The faulty ignition is something that occurred suddenly on the eight of June 2008.

The replacement of the brake fluid and the checklist for the APK-items went uneventful. Nothing was stuck and it was just a matter of pumping new oil through the system whilst at the same time draining the old oil at all the four calipers.

[Blockierte Grafik: http://rwoe.150m.com/tmp/20080301/img652.jpg]

Since I cannot ask from an APK accredited official to falsify the CO test results, the APK was performed at a CO of typically 1%. It is suffice to say that #231 passed with flying and that once the official clear in the RDW database was given, we raised the CO to typically 2,31%.

Last friday, E28 M5 #231 hit the road for the time in 21months and my first impressions are very good. As a coincidence, I had borrowed both ///M's to my local BMW dealer to represent the older five-series for their dealer introduction of the F10, so the first 50km with the rebuild engine were driven towards the dealership in Enschede, albeit for a different reason then most cars do. As it turned out, the location manager of Smudde in Enschede is a true enthusiast of the older cars as well so I had a nice conversation with him for about an hour on last Sunday.

Anyhow, I almost forgot how good it was to drive #231. Granted, I am restricted in engine speed (2500RPM max) during the first phase of the break-in (500km), but teh difference is so significant compared to before the engine failure. At idle, it almost is smoother for instance an M50 engine, which is astonishing, concidering the racing-pedigree of teh M88 engine. Also at constant speed it is much better then before; the slight surging it suffered from prior to the engine failure has disappeared completely. I picked up #231 again yesterday and added another 100km and experience no problems at all. Everything went smooth and as expected. I will add up another 400km untill the first oil-change, hopefully somewhere next week, but need to schedule my agenda first.

The gearshift overhaul (new shifter-mechanism, bearings etc etc) worked perfectly. It is much more precise then before and the occasional misshift that occured before the engine failure has gone. Off course, I have a few pending issues, but these do not affect the drive-train and can be solved sequentially at a later date. My planning is to solve the open issues (e.g headlight wash system and the refill of the AC system) in the next couple of months. I have to find a way to protect the wiring loom to the temperature senders from the impending heat of the exhaust. IMHO, this has been a design flaw, at least on the DC91 cars that will affect its long-term reliability.

Overall, I am very happy with the results that have been achieved untill so far and am glad that I can share some of the driving-experiences from now on.

Yesterday morning, I used #231 by exception for my commute to Hengelo. It just is a 27km drive, mainly over the autobahn. When I returned home, the trip-meter indicated 261km. I used this opportunity for a last check before completing a larger trip yesterday evening when I picked up my recently acquired BC2 (on board computer) that is intended to enhance #231’s functionality.

I decided to use #231 for the trip but started a bit late in the evening to avoid rush hour; this allowed me to vary engine speed freely without disturbing traffic. Given the
late departure, the 150km single distance and the restriction in engine speed, this will be a late hour trip, but what the heck.

In Zwolle we entered highway A28 towards Heerenveen where we stopped at an Esso station to fill the petrol tank one last time before proceeding to Leeuwarden. However, Esso stopped distributing the 98RON Super Plus and replaced it with Esso Supreme.

With Euro 1,63 per liter, its price is rather hefty and a full ten cents per liter more then the garden variety Euro95. Since there was no reference to an octane rating, I went inside and asked the cashier. She pointed towards a leaflet that contains a lot of words and marketing blah blah, but no relevant information. This left me to only one conclusion; that Esso doesn’t want to fully inform their customers, hence why I politely replied that I didn’t trust the information and said I would go elsewhere. The next opportunity was a Shell station shortly before Leeuwarden. Although V-Power in Holland is marketed as 95RON, its technical datasheet specifies an actual octane rating of 97RON, which is good enough for an M88/3.

The rest of the trip went rather uneventful and we arrived at Ivo’s place shortly before half past ten. Being a good host, Ivo offered us a cup of coffee before doing business. After he proved the system worked and we closed the deal, he offered us something else to drink, but as it was already close too midnight, I kindly declined and said it was time to leave since we still had a two hour trip left.

Shortly before Meppel, I passed the 500km mark, which allowed me to increase the engine speed to 3000RPM max. This allows me more freedom to vary engine speed whilst driving between 2000 and 3000RPM. I varied more often and slightly increased the load before arriving home at half past one.

Summarized; I have completed the first 556km and have completed the first phase of the break-in. This means that an oil change is due for next week before continuing with the second phase of the break in (500km-1000km). I am now allowed to rev the engine 500RPM higher. Although this isn’t really that much, it allows me more freedom to vary engine speed. I have to admit that from what I have experienced in the last 56km, the engine feels strong and healthy. Although I still am far from being able to use the full rev-band, driving #231 already brings a lot of joy.

This morning, I checked the fluid levels in cold condition and found no inconsistencies. Coolant, power steering fluid (ATF) and engine oil are as on at the same level as the last time they were checked at 206km.

“Me is very happy.”

On Friday the 9th of April, I traveled to the North-west of Friesland to pick up an OBC2 for my E28 M5. There has been some debate about whether or not an E28 M5 has the OBC2 fitted as standard, but a fact is that #231 does not have any. Also the wiring loom was not present and has not been messed-up with. The entire wiring of #231 underneath the dashboard is original. As individual OEM parts, the OBC2 is hideously expensive, but from time to time, complete sets are offered on the bay of e for reasonable money. I did not source my set on the bay of e, but directly from fellow E28 enthusiast Ivo who has a nice parts-bin of E28 parts lying around.

The OBC2 set consists of (see picture below):

1: The OBC2 unit with the coding plug.
2: The outside temperature sensor.
3: A relay box for the alarm horn (PIN code protection)
4: Primary wiring loom.
5: Secondary wiring loom for relay box and option-box
6: The alarm horn
7: Turn signal switch with OBC dim-dip control
8: Gong speaker



[Blockierte Grafik: http://rwoe.150m.com/tmp/20100416/img001.jpg]

Removal of parts

To install the OBC2, the instrument cluster and the existing digital clock from the dashboard have to be removed. When that is done, the knee-cover, steering wheel and the bottom-part of the steering column cover have to be removed. When that is done, the driver’s side looks as follows.

[Blockierte Grafik: http://rwoe.150m.com/tmp/20100416/img002.jpg]

For cars that have the DWA alarm, the relay control box and the secondary wiring loom are not needed. The DWA alarm does have a feature connector for interfacing to the OBC2. The code function does require this connection to activate the alarm horn if the car is being tampered with.

[Blockierte Grafik: http://rwoe.150m.com/tmp/20100416/img005.jpg]

The control-module for the DWA module is located under the knee-cover on the driver’s side (LHD model). Even though it is not required to remove the DWA box, it limits the available working space.

[Blockierte Grafik: http://rwoe.150m.com/tmp/20100416/img003.jpg]

The dim-dip function of the turn-signal switch allows to a sequential selection of the OBC functions. The dim-dip function is not included on non-OBC2 cars and thus has to be replaced unless one wants to push the buttons on the OBC itself.

[Blockierte Grafik: http://rwoe.150m.com/tmp/20100416/img004.jpg]

The installation of the primary wiring loom is rather straightforward and will be integrated in the existing wiring bunch. The green multi-pin connector is for the OBC2 unit; the yellow multi-pin connector for the instrument cluster and the rest of the connector is routed towards the driver’s footwall.

The turn signal switch has a relay integrated in its wiring loom. However, the relay-socket of the OBC2 set did not fit the steering column of #231 so I decided to exchange the wiring looms. Although rather straightforward, the pin-layout of the turn signal switch for the OBC2 variant is different then the part without the dim-dip switch, hence why one has to be careful not in interchange the wiring without noting the connections first.

[Blockierte Grafik: http://rwoe.150m.com/tmp/20100416/img007.jpg]

Although I have a degree in engineering and exchanging wiring looms and the required soldering should be a piece of cake, the primary functions (turn lights, emergency lights and high beam) were tested prior to final installation. The dim-dip switch is tested in combination with the OBC2 unit that should be connected in this stage.

Please note that the coding plug for the OBC2 is E28 M5 specific. Coding plugs for the German version are still being stocked in Munich and are reasonably priced but take a few days to deliver. After testing the OBC2 controls and connection of the gong-speaker, the DWA alarm control box and the routing of the wiring for the ambient temperature sender, the steering column cover and the steering wheel were placed back.

[Blockierte Grafik: http://rwoe.150m.com/tmp/20100416/img008.jpg]

In the mean time, the correct coding plug has arrived and was picked up this morning. When I have replaced that part, I can test the temperature function for which I still have to install and connect its sensor.

As part of a 1500km check, I measured the valve clearances and checked the allignment of the camshafts. The latter may seem superfluous for a 1500km engine but an allignment-check is the only method to check prematurely failing pars due too a production-fault. The allignment can be checked by placing the camshafts at TDC and measure the distance between the TDC marks on the harmonic balancer and the chain-cover.

Caution: Although this appears easy, be carefull when doing this yourself. Rotating the engine in the wrong direction increases the risk of valves hitting the pistons, so make sure you follow the correct procedure.

[Blockierte Grafik: http://rwoe.150m.com/tmp/20080301/img654.jpg]

The TDC mark on the harmonic balancer matches the TDC mark on the chain-cover. The difference is negligent => Perfect

[Blockierte Grafik: http://rwoe.150m.com/tmp/20080301/img655.jpg]

Upon assembly of the engine, I specified the use of a single size shim (3,80mm). The correct clearance has been adjusted by grinding the valve-stem. This method offers the highest accuracy within 0,01mm, allowing a nominal clearance of 0,30mm for all valves. However, with the valve seats being new, they need to settle which can cause a slight change of the clearance compared to the initial value.

[Blockierte Grafik: http://rwoe.150m.com/tmp/20080301/img653.jpg]

The intake valves all measured 0,27mm to 0,28mm; Compared to the 0,30mm construction size, this a small change, but bear in mind that during the assembly of the head, no oil film was present underneath and above the shims. This explains the 0,02mm. With other words, the intake valves are all correct and stable within 0,01mm.

The exhaust valves are similar. Eight of the twelve exhaust valves are still at 0,30mm (with oil film). Three others are at 0,28mm whilst there is one at 0,33mm (cylinder #4). Coincidentaly, the same exhaust valve caused the engine failure in 2008. I don't have an explanation for that so I see this as a coincidence. Although reshimming three of the twelve exhaust valves is the correct corrective measure, the results are notjing to worry about so I have decided to measure the valve clearance again at 3000km in order to check their stability.

Letzter Samstag hatte ich besuch von Stevie, der ich den Schlussel in den handen gab fur eine kurze test-fahrt. Diese wurde auf HD-video festgelegt.

Click mich

Viel spass.

Gents,

It has been a bit silent from my part in the recent two months, but #231 is going well. During the last month, I worked on a friends E34 M5 that needed some attention as well. Please check this thread to find out more.

As for #231: The trip meter is now standing at 2046km since the rebuild, 120km of which were driven on last Friday under high temperature conditions that exceeded the 35°C mark. These were driven on the autobahn as well as urban use. At some point, the temp gauge of the BC indicated a 40,5°C; but this was in a living area with a maximum speed of 30km/h were it is my habit not to exceed 20km/h.

In these conditions, the coolant temperature gauge indicated a slighlty higher then normal operating temperature; the AC fan trips off course, but that was expected. Once the driving speed exceeded the 30km/h mark, the needle returned to the normal level just before the center of the gauge. The viscous fan was needed all the time though.

I have revved #231 up to 5000RPM, but before I go higher, I want to replace the current oil that has been in the sump for the last 1250km first with a synthetic product. I have concidered Mobil 1 15W50, but as this is not a full-synthetic, I think I will rely on the good old Castrol TWS 10W60 even though I had some doubts for S38/M88 applications in the past. I will report about more specific checks later this week.

It has been a while since the last update, but since then, I have mainly been busy with running in and finding issues. Shortly after measuring the valve-lash in May 2010, I also removed the spark plugs to check their color and deposits. This also is called reading the plugs, a procedure that can yield tremendous amount of information about the running condition of the individual cylinders.

[Blockierte Grafik: http://rwoe.150m.com/tmp/20080301/img660.jpg]

During combustion, it is essential that the gas-temperature does not exceed 870 degrees Celcius as otherwise pre-ignition is possible. This can cause detonation with disastrous effect. Since we don’t have temperature sensors build into the spark-plugs, the only way to determine is a checking the tint of the plug insulators. When new, the insulators are perfectly whit, but when operating their tips change color depending on the quality of the combustion. For a perfect combustion (low amount of hydrocarbons left and good air to fuel ratio), the tip should color light-brown. The picture below shows the spark-plug of cylinder #1. The inspection of the other five plugs revealed that there is little variance in coloration so based on the evidence, the air to fuel ratio (AFR) is correct for the use that the engine has seen thus far.

[Blockierte Grafik: http://rwoe.150m.com/tmp/20080301/img661.jpg]

Another important parameter to check is the timing of the ignition. This can be checked by locating the position of the change in color on the ground-strap of the plug. For a good timing, the position of the timing-mark should be located around the middle of the ground strap. The following picture proves that cylinder #1 operates properly. The other spark-plugs show little variance in the position of their timing-marks as well

[Blockierte Grafik: http://rwoe.150m.com/tmp/20080301/img662.jpg]

Deposits on the ground strap and the metal enclosure indicate the AFR at which the engine idles; with other words, the verification of the CO adjustment is done not only by measuring the lambda score, but also by reading the plug. Carbon deposits on the ground strap are fine for the M88/3 engine as the carbon monoxide (CO) at which the engine operates is 1,5% for which it is theoretically impossible to achieve a stoichiometric AFR where all the carbons react with two oxygen atoms (O2) to form green-house gas (CO2).

[Blockierte Grafik: http://rwoe.150m.com/tmp/20080301/img663.jpg]

The AFR of the engine measured between 14,39 and 14,52. These results indicate a slightly rich AFR, which is confirmed by the idling patterns on the spark plugs. Although the AFR at idle is within specification, lack of a closed loop lambda regulation can result in lean operating conditions under load. As long as the operating AFR hasn’t been measured under load in a laboratory environment were worst-case circumstances can be simulated, it is wise to increase the safety margin by lowering the idling AFR.

Unlike the S38 that operates in closed-loop, this can be achieved by adjusting the CO-screw on the airflow meter (AFM). One word of warning though: Never ever try this to do on your driveway without an accurate emissions tester as you risk a lean running condition.

I have to admit that I adjusted the AFR well into the rich region (14,0). This is low enough to violate the emissions legislation, but not low enough to reduce engine performance caused by an incomplete combustion. Some may argue about my motivations, but as long as I haven’t proved that the engine is fully run-in, I won’t be able to test the engine in worst-case conditions under load, which is essential to achieve a safety margin that is needed to (1) meet the emissions limits and (2) still have enough margin from the lean operating range (AFR>14,7)

Throughout June and early July, I didn’t do much work on the E28 as I was occupied with a friend’s E34 M5 that needed major work. This project was originally planned for two weeks, but whilst inspecting the car, new issues were added so in the end, this project took five weeks to complete. Nevertheless, I treated #231 with a new windscreen-fluid reservoir. The old one just didn’t look so nice anymore and some of the Gaskets that seal the pumps were leaking.

[Blockierte Grafik: http://rwoe.150m.com/tmp/20080301/img664.jpg]

Installing a new reservoir just takes half an hour, but it is so much worth the effort. It changes the appearance of the engine compartment to a higher level.

[Blockierte Grafik: http://rwoe.150m.com/tmp/20080301/img665.jpg]

Between half June and half July, the ambient temperatures reached values between 28°C and 35°C. This allowed me to carry out a test of the cooling system. On two days, the average ambient temperature raised to above 35°C with the BC indicating 40,9°C in urban area. The engine performed flawlessly, but when idling for about five minutes, the coolant temperature gauge reached roughly 75% of the full scale, indicating roughly 110°C. Since the operating pressure of the cooling system is 1,4bar there still is a margin of 30°C left before the coolant reaches a gaseous state. The viscous fan was operating correctly, but with an AC-system, there also is an electric fan and that didn’t engage. I enabled the A/C (which is non-functional at the moment) and this engaged the auxiliary fan into the low speed mode. I then bridged the low-temp and high-temp thermostatic switches, both of which engaged their respective operating modes of the fan. The high-temp switch should engage at 99°C. With some tolerance and aging, I can imagine that this part didn’t trip, but the low-temp switch (91°C) should have worked which didn’t. As corrective measure, I bought both high-temp and low-temp switches new at the dealer, but have yet to find the time to replace them.

I had some discussion offline with a friend about the need for the auxiliary fan. He told me that even without the auxiliary fan, the cooling system should stay in a safe operating area. My defense was that this my be the case for cars without the A/C, but cars with the A/C have the condenser in front of the radiator increasing the resistance to airflow. I also added the argument that the auxiliary fan is a standard option in tropical areas and this was exact this condition under which this test was performed. I carried out another test at 25°C a few days later. The coolant temperature indicator reached a tad higher then the center, proving the correctness of my thesis that the high ambient temperature caused the temperature rise in the cooling system.