HOW-TO: M104 / W124 Top-End Rebuild [Cylinder Head / Gasket]

[gsxr]

Take your cylinder head into a machine shop that knows MBs, and have them look at it. They can give you an honest assessment as to whether the head is salvageable or not. Often times they can be repaired/welded with good success, depending on the nature of the damage.

A failed head gasket doesn't automatically mean that the cylinder head is bad. Get a knowledgeable professional to look at it, and then make decisions from there.

Any used head you pull from a junkyard engine has an equal chance of requiring all the same work your existing head probably needs. Unless your head is warped or otherwise damaged beyond repair, pulling a used head may just cost you more time & money with minimal benefit.

 

[jftu105]

gsxr and Gerry,

Thank you for the advice. In fact, I talked to a machine shop locally and they quoted a very affordable price to repair the head. I will bring the head in this Friday to get a more accurate estimate and proceed from there. Pulling a head with all my tools at hands is already a tough job. Doing it at a junkyard is even tougher, plus the coldness, the snow, and the mud. Yes, I will go with the repair route first. Thanks again. jftu105

 

[jftu105]

Had the cylinder head machined in a local machine shop for $50 plus tax. Before maching, the head was warped with the middle 0.007" lower. They took off about 0.010". The machined head is now flat. My straight edge (within 0.001") and thickness gage (0.0015" thinest) cannot detect any flatness error. Also measured the engine block. It is also flat within 0.0015". Before machining, I had to remove both camshafts. The machine shop told me that I must bleed the hydraulic lifter, which I did. Bleeding the hydraulic lifter allows the valve gap to be re-adjusted for an optimal value. After machining, the corrosion spot is now behind the metal ring of the gasket. Will start the assembly in a few days. Hopefully, things can go smoothly.

jftu105

 

[gerryvz]

Sounds like taking the head to the machine shop was the right thing to do. Hopefully they looked over everything else and took care of it all. Did they check the valve guides and the sealing at the valve seats?

Good luck in your reassembly. It's the most satisfying part. Take extra care on the procedure to torque the head bolts. You arms will be quite fatigued once that process is done !!

 

[jftu105]

Had a second thought on installing the old head after it was machined. The corrosion spot, although behind the metal seal of the head gasket, is still a concern. Did not want to invest further into this head. Went to the local yard and pulled another head, which is much cleaner, with only 0.002" flatness error and zero signs of corrosion. Took this "new" head to the machine shop for machining and a valve job. The work will be performed by a fellow from South Africa originally, with very high level skills in rebuilding and machining engines. The cost is $12 per valve. Will also install new seals. The total cost is $12x24 + $50 (machining) + $10 (cleaning) + tax (6%) for $367. The old head will be a spare, for study, or returned to the yard for $15 credit while keeping the hydraulic lifters and camshafts. The "new" head should be back in a week. With this much effort into this project, it does not make sense to save $367 for a potential problems in near future.

jftu105

 

[gerryvz]

Corroded spots can often be welded by a machine shop, as long as they aren't too far along, or in a critical area. I hope your gamble with the wrecking yard cylinder head works out!

 

[jftu105]

In order to weld, all the valves have to come out because of the heat from welding. It is a $400 job, not just $100 (welding and machining). With my old engine head being over-heated and corroded badly, the bet with a head from junk yard is not that much a gamble because it is straight,and free of corrosion. I would not have bought this head from junk yard if it is not in such good condition. I am willing to invest $360 to machine it and do the valve job. I did not want to weld and do the valve job for the old head because the corrosion was pretty bad, not just one spot in particular. Almost 0.010" was taken off to make it flat. It was badly over-heated once (I did that when coolant was lost through the heater hose but the reservoir was full due to blackage). Yes, it costs more but not that much more.

jftu105

 

[jftu105]

The engine head is still in the machine shop. Went there and spent two hours watching the master machinist at work, who was fixing a Porsche head, four-cylinder with 8 valves. Saw my own engine head all cleaned up with valves all removed. They scheduled the repair to my engine head next Monday, hopefully finishing it by Tuesday.

This master machinist learned the trade from his father in South Africa and is so skillful. He explained the work he was doing to me. I know the theories but to see how the theories are implemented practically is quite an experience. He already ground the valves. The intake valves were provided by the custemer, made by AE, while the exhaust valves are Porsche original. They make a differnece. He explained to me the roundness of the AE valves is not as true, resulting the sealing problem with the valve seat.

I then watched him rim the valve guide and then hone them for proper fit and surface finish. I got to feel the clearance of a valve and its guide. He explained that he typically goes for 0.001". There will be thermal expansions of the valve, so it cannot be too tight.

After that, using a mandrel as guide, he machined the valve seats. Then he tested the sealing with vacuum. The exhaust side is perfect, but the intake is quite there. As he turns the valve, the vacuum reading changes, indicating not round or straight fitting. He then lapped the valve and seat individually for each intake valve, spinning a spindle by hand with some lapping compound. After a few tries, all valves seal properly.

He also showed me a VW head with 20 valves (5 valves per cylinder). Next, after he installed the valve springs, he would machine the valve end so that they would be at the same height because the valve seats are not necessarily the same when he machined them. He explained that he could only take out a little bit each time or it might chatter, ruining surface finish. I asked him what if he cuts too deep, he then showed me the valve seat inserts which can be inserted with shrink fit.

It is all about concentricity.

I came away with a full confidence that he would me the engine head so good and my engine will be like new again. He actually measure the clearance of my valve guide and valve stem, and noted it is about 0.0015", a bit on the upper side but should be fine. He went to the computer but he could not find the spec for E320, 1994. He found C280 and diesel. The cost to fix the valve guide is another big ticket item. I should not need it.

Should be able to get the head back next week and then the assembly process will start.

A final thought is that skillful person like this master machinist is a dying trade. This person is still quite young, probably in his early or mid 40s. He is the only one working on the valve jobs at this machine shop and this machine shop is the only one in the triangle area of North Carolina. He told me that no young people are interested because they go for bigger money, mostly computer and internet related stuff (bitcoin, for example), where the big money is. Sad but true.

jftu105

 

[gerryvz]

Yes, having a competent machine shop with specific experience in Benz engines is a “must” when redoing motors. If one doesn’t know where such animal is, ask your local Benz stealer what machine shop they use for any machine work they have to do. They will usually tell you, and it’s a safe bet that shop will know how to work these motors, particularly if they have been around a while.

It is definitely true that “trades” (any type of work where someone uses their hands to craft/make/repair something in a skilled manner) are nearly a lost art these days. The funny thing is that skilled trades folks can make more money than paper pushers like me.

I wish I had gotten more deeply involved in Benzes a good 5-10 years earlier than I did, before I got married and had kids. I had one shot in 2006 to take over my former mechanic’s business (in Portland, he was retiring after 40+ years running the shop) but we were about $200,000 apart on the value of the business — mostly the building& land — and the due diligence just didn’t support paying more. The owner really didn’t want to sell the building/land, just the business. About 3 years later he did just that, sold the business to a guy from Florida but kept the building and land.

It would have been a dream for me to own a well-established Benz repair business, but alas, it was not in the master plans for me.

Now I just live vicariously by watching GSXR diagnose & fix his cars, try to learn a thing or two, and drool when I visit the shops belonging to Jono in Atlanta and my friend Robert Fenton in California.

 

[jftu105]

OK. Got the new engine head back and the valve job was done so nicely. It was also machined flat. The total cost $370 including tax. Started the assembly process. Have a friend to help to lift the engine head with exhaust manifolds over the engine block and sat it down. Put in all the head bolts and started the tightening process. The head bolts are to be tightened as wet, indicated by the head gasket instruction. First tighten to 56 Nm and then two 90 degrees. I took it slowly. Armed with my digital Saltus torque wrench, I was able to torique it very precisely, from 35 Nm, 45 Nm, then 56 Nm.

After that, I did the 90 degrees. I only eye-balled them. Based on the torque wrench starting angle and eyeballed a 90 degree location and turned it until I reached the "targeted" location. According to my torque wrench, after the first 90 degree, the torque went up to 95 to 105 Nm. The second 90 degree, the torque went up to 115 to 120 Nm. Once the turning started, it was not too tight to turn until almost the end. The torque reading did not increase linearly. I assume that the gasket was crushed so that the torque did not increase dramatically. Besides, it is wet torque with oil as lubricant, the torque, which is related to friction, will be less but the tensional force acting on the bolt will be quite high.

I hope the head bolts are all tight enough.

Next, I tried to set the timing. Did everything I could, but the closest, I can get is 2 degree (or 2.5 degree) after TDC. This is when both the intake and exhaust camshafts are at its correct position, the crankshaft is at 2 degrees after TDC. I took some pictures.

I assumed that the chain has been stretched a bit. When I took the valve cover off last, the crankshaft was off about 4 degrees. I assume that the engine will still run fine because, it is variabl adjustable. The timing will be advanced and 2 degrees probably not critical. Let me know if I am wrong and I should get a new chain. I really don't want to get a new chain.

Put in spark plugs and now when I turn it by hand, I can hear some breathing of the engine. The belt tensioner is also in and when I slightly turn it back, I can see the tensioner gets pushed back. I think gsxr said this is the check to make sure the tensioner is doing its job.

Please advice on the timing issue.

Thanks.

jftu105

 

[gerryvz]

If I was you, I’d replace the chain and top rails. It’s incremental work at this point.

You’re not too far off at 2 degrees, but since you’re doing everything else, why not start fresh with everything?

 

[gsxr]

If it's only 2° off, I'd leave the chain alone. Although the M104 FSM does not provide any details on allowable stretch, the OM 602/603 (diesel 5/6-cyl) manual states that chain requirement is not required unless the chain stretch exceeds 4°CA (crank angle) - see attached screenshot. The M119 (V8) manual says that up to 2°CA variation from specs is acceptable even with a brand new chain. I can't find it at the moment, but one of the 124 engine manuals even provided a spec for new chain vs a chain with >20kkm (12kmi) on the engine, and the "used" chain spec was +2°CA.

If you have the tools and the time, it would not hurt to roll in a new chain... but it's a PITA if you don't have the tools.

 

[jftu105]

I don't have the tool and I really don't want to do the chain. Thanks a lot. I looked all over for the timing tolerance for M104. Knowing that 4 degrees could be acceptable is very comforting. Thanks again. jftu105.

 

[jftu105]

Gerryvz and Gsxr,

Success!!! Put everthing back without incidents. Started up right away. However, the engine was extremely noisy in the beginning because the hydraulic lifters were working their ways to be correct. High pitch clicking sound. All of them were bled. Drove around the block and the engine grew a bit quieter, still noisy. Took a 17 miles drive to the machine shop and brought two bottles of special 22 Oz beer (double IPA) to treat the two master machinists there and also wanted them to check out the engine. After 17 miles drive with speed up to 75 mph and engine speed to 2800 RPM, the engine was a lot quieter but one lifter was still making huge noise. One master brought out two bottles of MOS2 from Liqui Molly. Pour them in. I then drove back for another 17 miles. Half way, I noted the engine was much quieter. Got back home, the engine was purring, quiet and smooth. This machine shop at Raleigh is called T-Hoff. Extremely high quality work and very reasonable pricing. Great people to talk to too. Highly recommended. I still need to flush my cooling system several times to get the oil residues of last six years out. One bottle of Liqui Molly radiator flush, then a few more rounds of dish washer detergent. Almost all hoses were replaced. The car now has a new life and its owner (me) has a new sense of achievement.

Despite the success, I don't wish to do this job ever again. This How-To thread is a god-send and I cannot thank both of you enough.

jftu105

 

[gerryvz]

Glad it all worked out. Always good feedback to hear that the HOW-TO articles are helpful guides.


Sent from my iPhone using Tapatalk

 

[jftu105]

I would like to provide some real data for the engine performance and touch on the timing issue.

Two charts are shown for engine data monitored by Hfmscan. You can read my posts on how to acquire and use Hfmscan on benzworld.org for the 124 forum.

In those two charts (see attaached), the engine was started from cold and then drove around town. I don't drive fast and you will see the engine RPM barely went over 2000 rpm. The curve of O2 sensor control indicates that the control had an average about 4% rich, which is normal. It is better a bit rich than lean. The MAF signal was between 0.75 to 2.25V, correlating to the engine RPM closely.

On the second chart, two more signals were shown together with engine RPM. One is the ignition angle. It shows that the ignition is about 8 degree advance during idling. It then went up as high as 30 degrees in advance depending on the RPM. When the ignition is advanced quite high, the camshaft actuator was activated to advance the camshaft angle (signal goes from high to low). The data show that the camshaft advance is an on/off control. It is either full advanced or not advanced.

In conclusion, I think I understand why my 2 degree off in timing chain is not causing problems. First, there is a crankshaft position sensor. The engine computer will issue ignition command based on the position sensor and the driving condition, also the camshaft position (from camshaft sensor). Second, the camshaft will be open within +- 30 degrees. When my engine is off by 2 degrees (after), it becomes between +28 and -32. As long as the ignition advance is too much, the compression and explosion cycles will be fine. When the ignition advance is too high (nearly 30 degrees), the computer will advance the camshaft to keep things right. The timing accuracy is a lot more important in a non-computer-controlled engine, I believe.

In conclusion, Hfmscan provided me lots data in addition to those presented in these two charts. From these data, the engine is running great after the gasket replacement. I am still flushing out the old oil residue but the coolant is getting cleaner and cleaner, indicating no oil leaking into coolant anymore. After driving and after the car parked over night, there is no high pressure gas when I open the coolant reservoir, indicating that no exhaust was leaked into the coolant. These are the symptoms I had before the gasket replacement.

So far so good. Thanks again for all the help.

jftu105

 

[gsxr]

If only we could get data like that from the LH system!

 

[jftu105]

That's the reason why I stay with 1994-1995 E320, nothing older or younger. Hfmscan does not cost much but the data it provides are invaluable, making debugging so much easier.

jftu105

 

[jftu105]

Just want to follow up with some final assessemnt of the head gasket repair project. It has been two weeks since I finished the head gasket replacement. The engine performs gloriously. First time ever, I have an E320, 1994-1995, which does not leak any fluid on the ground. My other three E320s, though no heavy leaks, still leaves some drops. The gas mileage is the typical 21 mpg with mixed driving. I have flushed the coolant at least ten times with dishwasher detergent and Liqui Molly radiator flush. Now the fluid is clear and I put in two bottles of pentofrost, the blue fluid. At 144xxx miles, I want this car to last at least until it hits 200xxx miles. Thanks again for the valuable information provided by this forum.

jftu105

 

[gerryvz]

Just want to follow up with some final assessemnt of the head gasket repair project. It has been two weeks since I finished the head gasket replacement. The engine performs gloriously. First time ever, I have an E320, 1994-1995, which does not leak any fluid on the ground. My other three E320s, though no heavy leaks, still leaves some drops. The gas mileage is the typical 21 mpg with mixed driving. I have flushed the coolant at least ten times with dishwasher detergent and Liqui Molly radiator flush. Now the fluid is clear and I put in two bottles of pentofrost, the blue fluid. At 144xxx miles, I want this car to last at least until it hits 200xxx miles. Thanks again for the valuable information provided by this forum.

jftu105

Thanks for the follow-up. It’s always good to hear of a successful conclusion. Always seems like there are minor teething problems and small things overlooked on the reassembly.

For both of the motors I’ve done, there was a notable improvement in sharpness and crispness after the job was complete.

 

[jftu105]

Sharpness and crispness, very true indeed.

 

[jftu105]

After this head gasket repair, I am now more aware of hydraulic lifter noise. Since the repair, I have been driving the repaired one and leave the other one in the garage. Today, I started the one which has been sitting and immediately heard loud noise and noticed the severe engine shake. Alarmed but not scared, I poured in another bottle MOS2 and drove it around the block for about 10 minutes. When returned, the engine was quiet and smooth again. When it was making noise, I used my stethoscope to trace the source of noise. Water pump, quiet; steering pump, quiet; belt tensioner, quiet; top of the engine, some noise; and then I checked the exhaust manifolds. Wow, it was noisy. It makes sense. It came from some hydraulic lifter/valves from the exhaust side. The combustion probably slammed the exhaust valves if it was not closed properly or something. The exhaust valve is larger, so it would be subjected to larger forces (pressure x area).

MOS2, the solid lubricant, is a godsend. I used to pour in one bottle. Now, I used two bottles.

jftu105

 

[jftu105]

It has been two months since I completed the head gasket replacement with help from this forum. The engine still performs flawlessly. Decided to change the engine oil. It was pretty dark, probably due to MOS2. It was regular engine oil. This time, I go for Mobil 1. Got them from Sam's or Costco when on sale for about $4.5 per qt. It has been my daily driver to work, 35 miles a day. The mileage is consistent at 21 miles/gallon, mixed driving.

Since the head gasket replacement, I also performed a few jobs on this car to keep it running at the top condition.

1. Fan clutch rebuilt. Inject some silicon oil with 10,000 cst. The temperature now stays low during long idling.

2. Brake hoses, all four. The old ones were original and there were surface cracks. Safety item!

3. Lower control arm boots replacements, both sides. The rubber boots were cracked, but the ball joints are still fine, no play.

4. Power antenna replacement. Got a good one from junk yard.

5. A rebuilt radio unit with DIY aux input.

jftu105

 

[John5788]

Here's a likely very controversial subject to broach - I spoke with a retired MB tech who has done 100s of head gasket jobs on this vehicle. He claimed the reason for the consistent failure of the gasket was not the gasket design but the small oil journals in the front and rear corners of the block (about 1/4" in diameter next to the bolt holes) that ultimately go nowhere. My head is at the shop so I can't confirm this on the head. By these seemingly useless journals pumping oil by the gasket, the gasket breakdowns and leaks in these areas. According to him, the fix is to plug these oil journals on the block with a number of options. As a dealer employed tech, he was not allowed to make such a modification. Have you heard of this and what are your thoughts on it? Anyone other members aware of this? Thank you!

I hate to necromance a little bit on this thread, but I stumbled upon this post on page 5 from Googling "m104 oil passage" looking for this exact information. I posted a similar question on another forum and a response I received was that the gasket has a small line that leads oil out onto the chain tensioner. Crosspost here: http://www.peachparts.com/shopforum...f-unused-oil-passage-under-cylinder-head.html

Just wanted to add this little bit of information incase anyone else looks for this in the future.

 

[jftu105]

I hope that this is the last puzzle I have to hunt down. My rebuilt engine (head gasket replacement and valve job) is performing great. However, two things still bother me. First, the check engine light will be thrown if I did not drive slowly to allow engine to warm up when start from cold. Second, the idling, although smooth and steady, has some vibrations. I thought I might need a new transmission mount but it was replaced not that long ago. I want total stillness. My other three E320's are doing just that, stillness at idling. I know that the first problem is related to the upshift delay. I also knew that it does not hold vacuum when I checked last time. This means that the upshift module (on the passenger side of the transmission) and/or the vacuum tube to it may have leaks somewhere. The vacuum is derived from the intake manifolds to a control valve (see picture). The green tube brings in the vacuum. The control valve is activated when the engine is cold so that this vacuum will pass down to the three to one connector right next to it. From there the vacuum goes down to the upshift delay module. I don't know what are the other two connections for.

As a result, I don't have upshift delay. It could be good or bad. The bad thing is really that the check engine light comes up here and there and I have to reset it. I did not think it would affect the engine performance because upshift only occurs when the engine is cold.

Over last weekends, I decided to block the vacuum line from the intake manifold to the upshift control valve, which is the green tube in the picture.

Well, it makes dramatic changes. The idling is now smooth and the acceleration becomes crispy. I had a slight flare from 2nd to 3rd (improved after some transmission valve kit with stiffer springs a few years back) but it is now prompt. At 2000 rpm, it shifts crisply, not flaring to 2300-2500 rpm, anymore. I guess that the vacuum leak was there to cause those issues.

Perhaps, the upshift control valve (see the picture) is leaking; therefore, it affects the engine performance all the time, not just when cold.

By the way, I realized that it is not holding vacuum by attaching a syringe to the green tube to test it. I pull the syringe but no vacuum forms. Therefore, I know there was a leak somewhere. A vacuum gage may not detect this small leak because the gage reading was still in the proper range.

jftu105

 

[gerryvz]

It's not uncommon for that upshift delay vacuum valve to go bad (that the green line/rubber connector are attached to) on the 104 engines. Definitely with a rough idle/vibration, it's going to be a vacuum leak with a 104 if it's not engine/transmission mounts, but that's a different type of roughness that is felt at idle through a rattly dash.

Don't discount the rubber connectors and vacuum lines themselves, either. Rubber connectors often have micro cracks from age and heat that you can't see unless you flex them. It's why I replace all lines and connectors when I do a top-end job with the heads off, M104 or M117.

 

[jftu105]

Decided to confirm where the vacuum leak is. Again, using a syringe, first unplug the green tube to the upshift delay control valve and attach the syringe to the input rubber of the valve. Pull the syringe, and it is not holding vacuum as previous discovered. Next, pull the rubber connector out and test it. It is actually fine, no leak. Then, pull the vacuum tube to the transmission upshift module. This is the white tube connected to the yellow 3 to 1 connector, next to the upshift control valve (see the picture in my last post). To my surprise, it is actually holding the vacuum. This is actually a good news. Because the engine is not running, the upshift control valve is not energized and should be closed. It appears that this upshift control valve is bad. This valve has a tube open to the air. If I use my finger to block the open tube (in front of the valve and underneath the electric plug), it holds vacuum. Therefore, it is bad. Check my other cars, they hold the vacuum.

Will find a replacement. This explains why the idling is not as smooth as total stillness I am after. It is apparently due to the small vacuum leak through this upshift control valve. However, I don't know why it affects the shifting after the engine is warm. Will find out more after I acquire a valve replacement.

jftu105

 

[gerryvz]

These valves are pretty cheap, and also very easily found on any late 124 at a wrecking yard. If you need a spare, let me know. I have 4-5 of them in my parts stock.

https://www.autohausaz.com/pn/0015407097

Cheers,
Gerry

 

[gsxr]

Or, Importec has the same Pierburg valve for ~20% less than Ahaz (but requires a $75 min order for free shipping).

https://importecautoparts.com/parts/part_number/001 540 86 97/3

 

[jftu105]

Hi, Gerry and gsxr,

The link to the new parts shows a three-way valve. The valve on my car is a two-way. The third is a dummy to the air. Two of my cars have bad valves. When I pull the vacuum, the air is drawn in through the dummy one. The good one can hold vacuum. I am going to the local yard to see if I can secure a few. I am also thinking to block the dummy tube and test if the valve will work after that. Please provide update.

Thanks.

The two-way valve is on eBay

https://www.ebay.com/itm/Mercedes-w...e=STRK:MEBIDX:IT&_trksid=p2060353.m1438.l2649

jftu105

 

[gsxr]

jftu105, AFAIK they are all 3-way. The third port has a cap over the vent.

EPC shows a different part number (A0015408697), but that is still a 3-port valve:
https://importecautoparts.com/parts/part_number/001 540 86 97/3

 

[jftu105]

Took off the shift delay control valve and verified its operation. It turns out it is working properly. When the power is off, the port on top is connected to the port underneath the electric plug, while the port at the front is blocked. Once a 12V is supplied, the port on top is now connected to the front port and blocked from the one underneath the electric plug.

The cap to the port underneath the electric plug is supposed to be open with a fine screen. So, the port is working properly.

However, this does not make sense. Because the upshift valve is only activated when the engine is cold; therefore, the top port will be connected to the underneath port when the engine is warm. Therefore, there will be a vacuum leak through this connection. Why?

Now, this brings up another puzzle. Currently three of my cars are in my garage with the fourth driven by my daughter out of town. Two of them are the same, both 1994, with the top port and the underneath port connected, while the third one, a 1995, the top port and the underneath port are blocked when the power is off. This 1995 is very distinct in the shift delay and I could not shift out of 2nd gear until the engine is warm. The two 1994 are not as obvious.

To prevent vacuum leak, I now blocked off the underneath port. It does not make sense to have Mercedes to design it this way.

Finally, when I took off the valve, I heard a vacuum release from the front port connection to the yellow three-to-one connector. Some vacuum is formed there for something. More puzzles, I guess.

jftu105

 

[gsxr]

When the valve is energized, vacuum from the engine should be applied to the Bowden cable vac diaphragm, and the "open" port should be connected to nothing.

When the valve is NOT energized, vacuum from the engine should be applied to nothing, and the Bowden cable vac diaphragm should be connected to the "open" port.

If it's not working like this, something ain't right. Engine vacuum should never be applied to an open port. Any chance the different valves have different internal plumbing / routing, and someone replaced it with the wrong part number? Can you verify the part number on the valve installed on your car?

 

[jftu105]

Went to junkyard this afternoon and acquired three vacuum valves from a 1995 E420 and a C220. They all work as I reported. When not energized, the top port (from vacuum line) is connected to the open port. When energized, the top port is connected to the lower port, which then is connected to a two-one check valve. From there it is connected to the transmission upshift module via a black vacuum tube.

This morning, I returned the setting to the factory setting. When the car was cold, I measured the voltage to the vacuum valve and it was 10V. I then drove to my tennis match for about five miles. Upon arrival, when the engine was fully warm, I measured the voltage again. To my surprise, it was still 10 V. The valve was still energized. This actually makes sense because there will be a serious vacuum leak if the valve is not energized.

However, the car does not drive as well with the factory setting. After my tennis match, when I drove back, the check engine light came on again. I then decided to seal off the vacuum line to the upshift valve. It drives better that way.

Now, I want to see how it works with my 1995 E320 special edition. This car has a strong upshift delay when the car is cold. It won't shift out of 2nd gear when cold. I did the same measurement when it first started from cold. I got 10V. I then drove around until the upshift delay was gone. I got back, and measure the voltage again. Still 10V. So the valve is always energized. This valve is a bit different because the top port is not connected to the open port when power is off. I probably should take it off to verify but I don't want to touch it. An't break, don't fix.

I am very confused on how this upshift delay works. I also got a piece of 2-1 yellow check valve next to the vacuum valve. The two ports on side are connected together. One is connected the vacuum valve low port, while the other is connected to the black tube to the transmission module. The other side of this yellow check valve is connected to a while tube, which is attached to the throttle body. Air can only go one way from two-port side to the one port side. Therefore, if the throttle air pressure is high, it will not affect the vacuum to the transmission module.

Finally, I pull the black tube to the transmission and check if it would hold vacuum. It won't! Therefore, sealing off the vacuum line to the vacuum valve is a right decision to prevent vacuum leak.

The previous owner of my 1994 E320 rebuilt the transmission with a shady transmission shop. Most likely, they did not put the transmission back in proper setting for the upshift and shift point module. My decision to seal off the vacuum line helps to reduce the vacuum leak; thus, it runs better.

I am very confused but will just go with whatever that is working. Three E320 now have three different settings for upshift delay but they all run good by their own ways.

jftu105

 

[jftu105]

OK, could not resist the suspense. Took out the upshift vacuum valve of my 1995 E320 and confirmed that it is exactly like other cars and functions the same.

Also decided to seal off entirely the tube to the transmission upshift module to eliminate any possibility of vacuum leak.

The only way I can make sense of this whole thing is that ECM control the upshift vacuum valve based on the gear position. Because I always measure it when it is idling, the valve must be energized; otherwise, there will be serious vacuum leak. During acceleration, it is possible that ECM could cut off the power to the vacuum valve so that there is not upshift delay. After that, the valve could be energized again at the 4th gear.

This is the only way it makes sense to me unless a better explanation is offered. All but one of my four E320 are running at the factory setting. The one I sealed off had a shady transmission rebuilt, but it runs better with the vacuum sealed off. No check engine light either, at least not yet.

ftu105

 

[gerryvz]

OK, could not resist the suspense. Took out the upshift vacuum valve of my 1995 E320 and confirmed that it is exactly like other cars and functions the same.

Also decided to seal off entirely the tube to the transmission upshift module to eliminate any possibility of vacuum leak.

The only way I can make sense of this whole thing is that ECM control the upshift vacuum valve based on the gear position. Because I always measure it when it is idling, the valve must be energized; otherwise, there will be serious vacuum leak. During acceleration, it is possible that ECM could cut off the power to the vacuum valve so that there is not upshift delay. After that, the valve could be energized again at the 4th gear.

This is the only way it makes sense to me unless a better explanation is offered. All but one of my four E320 are running at the factory setting. The one I sealed off had a shady transmission rebuilt, but it runs better with the vacuum sealed off. No check engine light either, at least not yet.

ftu105

Is the CEL working on the car with the vacuum sealed off (when you turn the key to position 1)?

 

[jftu105]

CEL does light up at key position 1. CEL is working properly as it threw CEL often when I ran the factory setting. It was code 26. I reset it when I get tired of seeing the red CEL. Since I sealed the vacuum off to the vacuum valve, it has not thrown the CEL yet.

 

[jftu105]

It turns out that gsxr is right (any surprise?). I made a mistake in interpreting the vacuum tube connection. The green tube, which connects to the top of the vacuum valve is actually going all the way to the transmission upshift module. The vacuum input is connected from the yellow check valve to the front of vacuum valve. The vacuum is supplied from a gray tube from the intake manifolds through the yellow check valve. When the vacuum valve is energized, the vacuum is applied to the green tube (to the transmission) and when the valve is not energized, the green tube is connected to the outside air. Therefore, it all makes sense now.

The only puzzle is why the vacuum valve is still energized when the engine is warm, during idling.

jftu105

 

[jftu105]

Anybody familiar with knock angles? I was doing some engine performance tests with hfmscan. It can display the knock angle of each cylinder. I notice that cylinders #2 and #5 tend to have a bit higher knock angles than the rest. When driving at constant speeds, knock angles mostly are zero. When accelerating, depending how hard I press down the gas pedal, some knock angles start to show up. They range from 1 to 6 degrees. Cylinders #2 and #5 are consistently higher. They could be at 3.8 degrees while the rest of them are essentially zero, in my normal acceleration (not fast). I figured that they could be related to the coil shared by these two cylinders. Replace it with a different one. Hardly see any difference. The engine runs fine. Just curious. The knock control was not activated most of the time.

jftu105

 

[jftu105]

I contacted Mike K, the developer of hfmscan software, and he replied with nice answers. Mike told me that knock angle 3 degrees means that the ECM retards the ignition back by 3 degrees. For example, if the timing advance is 30 degrees (typically from 5 to 45 degrees), a knock angle of 3 degrees means that the timing is now at 27 degrees instead of 30.

Therefore, when the knock sensor detects any knocking, the ECM will retard the timing by some angles to suppress the knocking.

Because cylinder #2 and #5 are moving up and down together; therefore, they might show similar knocking behavior. Coil is not a factor because knocking is combustion not due to the spark ignited flame front.

Knocking always happens and I guess 6 degrees retard is not a big concern.

jftu105

 

[gerryvz]

I contacted Mike K, the developer of hfmscan software, and he replied with nice answers. Mike told me that knock angle 3 degrees means that the ECM retards the ignition back by 3 degrees. For example, if the timing advance is 30 degrees (typically from 5 to 45 degrees), a knock angle of 3 degrees means that the timing is now at 27 degrees instead of 30.

Therefore, when the knock sensor detects any knocking, the ECM will retard the timing by some angles to suppress the knocking.

Because cylinder #2 and #5 are moving up and down together; therefore, they might show similar knocking behavior. Coil is not a factor because knocking is combustion not due to the spark ignited flame front.

Knocking always happens and I guess 6 degrees retard is not a big concern.

jftu105

That reminds me, I finally got both of the needed cables for the HFMscan software, need to get things connected on the G-wagen and make it happen !!

 

[jftu105]

Harmonic balancer of M104 shredded!

I thought the in-line six engine of M104 was bullet proof with the head gasket the only weakness. Wrong!!!! Today, I was surprised by the failure of the harmonic balancer, which is the vibration damper between the crankshaft pulley and the crankshaft hub. This happened to one of my E320's with about 142,000 miles. I just did a cooling system overhaul of this car with new water pumps, new hoses, and new belt tensioner, fan clutch, etc. My daughter drives this car and I kept it for a month to go through all possible problems. And it was doing great until my daughter got it for two days. Boom, the harmonic balancer shredded.

At first, the symptoms were some flickering of all the gauges lights at the instrument cluster. Before that, the car runs beautifully. Then, the AC quit. She pulled over to a gas station (being alert! Great). Then all lights came on. The steering became heavy. After she parked the car, she called me. Being 8 hours away, I asked her to take some pictures and some videos.

The serpentine belt is tight but it is not moving at all.

As a result, no AC, no water pump, no power steering, no charging.

I looked up the harmonic balancer from MB: nearly $700 for a piece of rubber!!!

Any suggestions? The car has been towed to a repair shop in Atlanta and the head mechanic happened to be at the gas station and saw my daughter was stranded. He came to the same diagnosis.

The replacement procedure seems to be straight forward if you can remove the pulley bolt (with a pulley holder or a impact wrench (no room)). But the cost of the harmonic balancer could be a killer.

I checked a few places, but none of them carry the harmonic balancer. There are some used ones on eBay but it is a big risk of using a used piece of rubber.

I will get a quote from the mechanic tomorrow. What is a reasonable cost for this job, parts and labor. Being 8 hours away, I could not do this job.

Thanks.

jftu105

 

[captruff]

I have a used M104 at the office. If you decide to go with used just hit me up with an offer.

Jeff

 

[quicksilver500]

Call Jonathan at Blue Ridge MB, he will have a used one. I wouldn't hesitate at all to put a used one on. Jonathan is just outside Atlanta and has told me the harbor freight stubby impact will get them out. The bolt is torqued to 400nm....MAKE SURE the mechanic doesn't knock the key into the oil pan putting the new balancer in.

Never seen or heard of a balancer letting loose. But with age rubber decays.

http://www.blueridgemb.com/child/contact.htm

 

[quicksilver500]

Also, you know MB can get parts next day. I have never paid any extra cost for it.

The job should be a couple hours at most, and I would think the guy would put a new front crank seal in as well.

 

[gerryvz]

Also, you know MB can get parts next day. I have never paid any extra cost for it.

The job should be a couple hours at most, and I would think the guy would put a new front crank seal in as well.

If I was you, I would have the car towed to Blue Ridge MB (Joni’s shop in Lilburn, GA) and evaluated and repaired there.

 

[sheward]

+3 Johnathan should have a HB for you and be able to get things repaired correctly.

drew

 

[jftu105]

Thanks a lot, guys. I will look into used parts based on the information provided. The car is in a shop now, not the suggested but the young mechanic appears to be nice. I will work with him from there and how it can be done.

jftu105

 

[gsxr]

I'm not sure what the balancer is like on the M104, but it appears similar in design to the early (pre-1995) M119, so there is no need to remove the center bolt to the crank... just the smaller bolts that attach the balancer + pulley to the hub. See FSM job 03-3420 for details.

That said, if there is ANY sign of leakage from the front crank seal, this would be the time to replace the $10 seal while you already have 90% of the stuff out of the way.

 

[gerryvz]

Seriously, no matter now "nice" the mechanic is -- unless he works at a Mercedes-Benz specialist/indy shop, you will be MUCH better off shipping the car over to Jono to have the job diagnosed and done correctly. If the car is in Atlanta area, this is a complete no-brainer. Do not mess around - hire the right person and get the job done right. You won't regret it, for a minute.