Hi Guys,
Got my rebuilt cylinder head back on Friday and installed it this week.

Went to start the car and of course there was the difficult first start while I bled the injector lines, but then it fired, smoothed out, but would do nothing but a low idle at maybe 400-500rpm with my foot to the floor. Anything less than half throttle and it would die.

Prior to removing the head I rotated the engine to TDC, pulled the timing chain sprockets out of the head and zip-tied the chain up tight to itself to keep it from rolling into the engine and screwing up the timing of the injection pump and distributor.

So it's running like something's timing is very retarded. I popped the valve cover back off, rotated to TDC on the crank and the cam mark is straight up at 12 o'clock (maybe just a bit on the passenger side of 12) and the distributor rotor is pointing at the mark on the distributor housing. That's how it's always been set up and it's always run fantastic.

I'm going to put a timing light on it in a few minutes, crank it over and see what the ignition timing is doing.

Anyone have any ideas?

__________________

I advanced the cam timing one tooth and that puts the cam's mark just on the driver's side of the pointer that's on that #1 cam tower. Engine started, but basically did the same thing.

The check of the ignition timing "looks" like it's really retarded. If I look at the distributor cam lobes, the heel of the points is midway between 2 lobes and doesn't come up on a lobe until something like 20* after TDC. But, at TDC, to just look at where the rotor is pointing, it's pretty much right there at that mark on the distributor body for #1.

Since the distributor is driven off the front of the gear that drives the injection pump, the only thing I can think is that the chain jumped a tooth which is making both the distributor and the fuel injection pump about 20* retarded.

Guess I'll be tearing into it in the morning. Son of a @#!%& !!!

__________________

Hello,

Yes it sounds like your ignition timing is very retarded. It is an easy fix. Remove the distributor, pull the gear and spring and rotate it in the direction needed to re-align the distributor mark. Be sure to note that the cog on the bottom of the distributor is off-set and must be oriented correctly. You can force it in 180 degrees off.

You should have a small spring between the distributor and this gear. Be sure to press the distributor all the way down during re-installation and tighten the cynch enough or the sprig will push the distributor up enough to dis-engage the cog after start up.

There is also a chance you could have re-set the engine with the injection timing 180 degrees off. You would have to have have noted if the engine was on TDC of compression stroke and was not changed to TDC of exhaust stroke to be sure. These engines actually run decent but not quite right with the IP pump 180 degrees out of time.


Here is a photo Todd;

Attached Thumbnails

 

__________________
Joe Alexander
Blacklick, Ohio
1964 220SE Rally (La Carrera Panamericana someday)
1966 Unimog 404s (Swedish Army)
1969 300SEL 6.3 (sold)
1969 280SL Pagoda
1973 280SEL 4.5
1974 450SLC FIA Rally car (standard trans)
1982 300D turbo (winter driver)
1986 560SEC
1989 Unimog FLU419 (US Army)
1991 300TE (wife)
2002 SLK 32 AMG (350 hp)

Thanks for the photo, it's nice to see the layout of all the pieces.

I know I can re-clock the distributor, but that wouldn't take care of the timing for the injection pump.

I'm still fighting the idea that the chain jumped because:
1) I kept it held under tension while I was zip-tieing
2) I took it apart at TDC for number 1
3) I didn't touch anything until I started re-assembly
4) The guide rails in the block, and proximity of the chain to the block at the injection pump sprocket make it really hard for the chain to have enough movement to lift off the sprocket teeth, become slack, bunch up and set back down again into the sprocket.

This blows!

__________________

Well, things went from bad to worse.

I pulled the complete distributor housing/cover plate off, removed the chain tensioner, and intermediate sprocket and took the sprocket off the cam. Then I put some slack in the sprocket that drives the injection pump (and distributor) and moved the sprocket clockwise (if you're facing the engine) in order to advance the sprocket 1 tooth on the chain.

Put it back together and that did move the heel of the ignition points more toward the distributor lobe when the engine was at TDC.

But, as I was putting it all back together, I dropped one of the washer shapped bushings that goes rearward of the intermediate sprocket.....down into the timing chain are.......clink, clink, clink and gone!

Don't panic, I thought, it's too big to go anywhere, I'll pop the sprocket out and get the bushing with a magnet. And then clink went the other bushing Son of a @#$%&!!!!!!!!

Off came the distributor drive/cover assembly. Couldn't find them. Drained the oil and swept inside the pan with a magnet. Nothing. Removed the oil pan, nothing. Reached up into the engine block, nothing!!!!

I finally located both bushings laying up on the face of the timing chain, at about the 3 o'clock position, wedged between the chain and the block.

Took out the radiator, fished out the bushings, and put the engine back together. Still starts hard and will only idle at a couple hundred RPM if floored. It's still waaaaaaaaay retarded.

So I marked the current position of the distributor and then rotated it in the housing about an inch or so counter-clockwise (looking down at it from in front of the car). Engine fired right up and ran almost normal, but had a weird hunting, loping due to what I'm guessing is the injection pump still being waaay retarded.

Man, I just don't get how this thing got so out of time. I might as well have just let the chain fall down into the engine!

__________________

Today I removed the injection pump. Man, what a job that is.

Sure enough, the timing mark on the pump was a good 1/4" or more "retarded" from where it was supposed to be, with #1 cylinder at 20 degrees ATDC on the firing stroke.

I reset the pump and installed it.

Then I rotated the engine about 1/4 turn backward....and then came forward to TDC for #1. I loosened the cinch clamp for the distributor, and pulled the distributor completely out. Then I used a magnet up pull up the distributor drive gear and re-clock it one gear tooth in the clock-wise direction. I set the distributor back in, gave it a rough adjustment and the engine fired right up.

I did a quick setting of the mixture, idle and timing, and the engine runs great. The headgasket is retorqued and tomorrow I'll take it for a short drive and then come home and dial in the mixture, linkage, timing, and idle.

Too bad I didn't do this right the first time eh?

__________________

Got some miles put on it today, and spent some time dialing in the mixture, timing, idle and linkage. My buddy's shop's exhaust gas analyzer is down right now, so I couldn't take the easy route.

What do you 6 cylinder SE guys like to do for ignition timing? The book says 30 degrees BTDC at 3,000rpm+ and if I go by that, I wind up with something in the 3 degrees ATDC at idle and it won't start hot after sitting for 10-15min unless you crack the throttle during cranking. If I give it 5 or so BTDC at idle, I wind up with 36+ degrees at 3 grand, but it starts more normally.

There are a couple steep residential street around, and they make for good ignition timing testing by hitting them at say 20mph, shifting into 2nd or 3rd, flooring it, and listening for the engine to start pinging. So far, I'm not getting any pinging under load, and I'm thinking about giving it more advance, like say 8-10 degrees BTDC (as long as I can keep the idle in the 750-800rpm range using the mixture and idle air settings).

Any tuning advice welcome.

__________________

That chassis uses vac retard, not vac advance.
So, you set the 30 BTDC @ 3 K FIRST and then if you want to set the idle timing , you change that timing by the adjustable rod on the dist vac. servo element. It is threaded for length changes. The point is, once you get your 30 set, you lock down the dist. and leave it there. The low end adjustments are not made by turning the dist. [ Your 3ATDC is actually in spec for that car..2-4 ATDC @ idle]
At idle , if you take the vac line off dist, the rpm should speed right up b/c it is then advanced [ no more vac retard]
If it is a stick trans a good setting is 30 @3k and zero[tdc] at idle...
Make sure the throttle plate is all the way closed at idle. A trick for that is to take the intake air filter hose off and place a flat piece of cardboard in front of the throttle intake, blocking off the intake air. There should be very litte, if any, difference in idle rpm ,when coolant temp is normal, with/without blocking intake b/c the engine gets all its intake air from the air bleed screw. If all is correct, You should be able to stall the engine with the bleed screw by turning it in ..if NOT, then you have other intake air, which can be caused by a not fully closed throttle plate OR the WRD is not shutting off intake air when coolant temp is reached to normal spec.
It is the closed throttle plate that causes the highest vac to reach the vac element at the distributor and when it does , it RETARDS the ignition timing.
This is a great system b/c as soon as one opens the throttle , the vac drops off instantly , allowing for a sprk advance ....from that point , it is the weights in the dist [ mechanical advance] that effectively control the advance curve...the trick is to set the 30@ 3k FIRST b/c that is a constant advance you want at 3K, regardless of where you decide to do any tweaking on the dist. fine tuning of the rod length.
The reason for Factory spec of ATDC @ idle was mainly early emmisions
and proper idle, specially w/auto trans. Earlier version distributors had vac advance...I like the retard system better.
To answer your timing preference question, I personally like to set idle timing to Zero [TDC] w/vac on my 250sl, which is basically the same set-up and MFI pump/engine as your 250se....and I have a manual trans, so I run 800rpm and slighly rich on the bleed screw for best results. [ meaning, you are looking for a higher than needed rpm with bleed screw, and then back the idle down with less air from bleed screw, resulting in a slighly rich idle setting...this assures you are not lean at idle and on that R18 pump, the transition to the midrange is much better with the low setting on the rich side b/c those pumps were slightly lean on the midrange [ that was corrected on 18z pump, so check your pump#] ..a vac gauge works well here. Peak the vac @ idle and then richen just a little by turning the bleed IN w/o loosing vac reading..you can hear the exhaust notation drop. That is the sweet spot. If the idle rpm is correct only when the bleed is at its highest obtainable rpm, then you ate actually too lean..you want that final tweak to bring the idle to correct rpm by slighly enrichening via the bleed screw. A 50-100 rpm drop from highest bleed setting is usually about right .
You mention access to a gas/A , so if you use that , you want the idle settings on the high side ..like around 4.5% CO.. You know how to adjust the pump thumb screw, so after that, get the bleed as the final tweak.
And any/all of those adjustments aredone AFTER ign specs are set. or you are wasting your time. I always do both ign and valve adjustment before any a/f mix settings. And those start w/linkages.................
Be aware that most of these pumps wear towards rich anyway. That can all be adjusted and I am sure you have seen that info.

As far as your pinging test, shifting into a lower gear is not a good test for ping...you want to stay in highest gear and go up a slight hill ...if there is pinging, it will appear as you attempt to add throttle while the engine has the load of BOTH high gear and hill inclination combined . That is extreame Load condition on engine.
I own a 121 chassis [ 1956/190sl] that actually has a cable running from the dist to the dash , where the driver can change the ign timing from the cabin in situ...the owners manual gives specific hill incline/top gear instructions on how to turn the control knob back one click for proper timing after pinging is detected,,,,,,,. pretty cool.....They did that back then b/c octane ratings were all over the map from one station to another ..... Gas was probably $5 a gal. back then in Euro.....

__________________
A Dalton

Excellent info (as always from you) Arthur! Thanks for taking the time to type that all out.

My car is a manual trans (4 on the floor in fact).

Sounds like I'm not far off then:
1) I do know that it's a vacuum retard, but I didn't know the idle timing was adjustable.

2) Valves are freshly adjusted and I'll run through them again in 500mi since the head is freshly rebuilt.

3) Just prior to all this head work I replaced the balls and sockets and one transverse linkage rod bushing and then set the linkage so it was really nice and both the throttle plate and pump lever were moving in sync. Of course I had to run through that again today since it's all been apart.

4) The throttle plate stop is adjusted. I like to back off the screw, remove the linkage rod and spring and then gently rotated the plate until I feel it just touch inside. Then I bring the screw in until it just touches, and then maybe a 1/4 turn more to just lift the plate off the housing.

5) I've also got the power steering boost and the idle return damper and both of those are set to allow the pump and t-plate to return to their respective stops.

6) How you describe leaving the idle and mixture is how I wound up leaving it today. I was going up and down one click at a time on the mixture screw, then going for max smooth idle using my digital VOM which has a tachometer function. With a max idle in the 830-850 range and starting to get that "I'm too lean shudder" I was turning the idle air screw in until the idle smoothed.....as you say, leaving it slightly rich. But I was still getting a wavey idle. Sort of a rhythic up and down of about 75+rpm. No matter where I put the mixture screw it was still up and down with the idle. So I loosened the distributor clamp and advanced the distributor maybe 1/8" and it smoothed out and idle the best it's ever idled. I'll check again in the morning, but I think that put me around 0TDC, or maybe as much as 3TDC at idle. But still, after sitting for maybe 10-30min, it won't start unless I crack the throttle open. Then it fires right away, but has a lopy idle for a minute or so unless I blip the throttle, and then it clears right up and goes back to that nice 6 cylinder rumble.

7) Prior to all this work, I'd set the car up on the exhaust gas analyzer and had it at about 3.5% and 795 rpm. It was pretty rock solid. Still a little subtle miss at idle, but low speed cruise was fine, as was acceleration and sustained highs speed cruise. Based on how the exhaust smells right now, I'm probably closer to 5%.....but that's just a guess.

8) Sorry about the mis-info on the gear usage and pulling a hill under load. I meant to say that I'd be in 1st or 2nd at a slow speed and then go into 3rd and stand on it.

9) I'd like to try your vacuum gauge idle adjustment method. I've never done that.

10) So I guess the plan tomorrow will be to set the timing at 30BTDC and then I may fatten up the mixture a click or 2 more, adjust the air screw to drop back to 800rpm from the 850-875ish I get with more fuel, and then see how it's starting hot. I guess one thing I could do is disco the pump rod and see if the hot start wants the throttle plate open = the engine is too rich. And then disco the t-plate rod and put the pump rod back and see if the engine is wanting more fuel when hot.

__________________

Well..I can see you have a good understanding of these MFI systems .

That being the case, here are a few more finer points I can mention:

When you redo the valves next time, retorque the head.

As far as your bad start and stumble after some off time with up to temp coolant, I want to mention a couple of things.
First suspect on that condition with that model pump [ R18] is they had a common fault with the check valves in between the injectors and the pump . as they were CONE sytle valves. The problem was they were prone to leaking back [ would not hold the line pressure] , so when you go for a hot start , the lines have drained. [ stumble complaint] . Once you get her going , it clears right up b/c the pump has re-pressured the drain-off. Benzes remedy for that problem was the modification of Ball Valves replacing the Cone ones. This was on the later pumps and those valve can be put in your pump. [ Benz part # 000-074-17-15]
There was also a Benz wiring modification to the CSV . They added a 1 sec relay that would squirt some gas into the intake regardless of coolant temp , so you had that little prime even when the coolant was hot. That helped
remedy hot start problems, [ which these cars were notorious for] and you may even have that addition mod , as it was a Factory Dealer changeout. It is a round relay.
When the car is cold , you do not have that complaint to contend with as CSV operation is automatically operational with the Time/Temp switch sending a signal to the CSV circuit, but when the coolant is hot , it does not...so they added that relay.
Those of us who know these systems simply add a momentary push button sw in the cabin and we operate the CSV when we want/know it is needed...kinda like an old manual choke, but electric activated. I compare it to the little start primer on your lawn mower.....
The advantage here is you still can have the stock TSS and CSV system, where the cabin sw simply Over-Rides those systems...if you do not push the sw, the cars stock start system still works as factory. But if a condition needs it, you have it on demand, manually. Best/ Both Worlds.
The last thing I want to mention on the hot start is the solinoid on the back of the pump....this device is NOT temp sensed, so it should move the rack [ supplying pump fuel to engine w/o opening the throttle plate] only when the starter circuit is activated. This was Benzes way of getting some starting fuel aid to the engine at all crank times , regardless of coolant temp and w/o accelerator pedal operation. But the MAIN starting aid on any/all these MFI systems is the CSV squirter.
Once Benz changed the pump cone valves to ball valve, they actually eliminated the rack solonoid b/c they found that the newer/efficient check valves remedy no longer needed a rack solinoid. [ you will notice the later pumps do not have that sol. b/c of that fact and all pumps without the sol do have the newer ball style checks]
So, addition of a cabin sw is a good idea and most guys I know have one , even if they have all the other mods ..
There are two ways to wire the cabin sw...the first is to get power so that the CSV sw will only operate when the starter circuit is energized...[ being like the TTS circuit] ..or , my preference, is to wire it Hot with Key ON ,so that you can give it a little squirt with key ON, BEFORE you activate the starter. The trick here is you squirt a little into the manifold and wait for about 2 secs before hitting the key..when you do , the gas has already mixed with the plenum air and that car will start instantly.
Once you get the FEEL of the switch for your car, you will be a happy camper.
Those simple mods and an upgrade to a later. higher torque starter w/ballast by-pass, along with point conversion, and these old MFIs will start and run like a brandy new one.......even better b/c of the tech advance of electronic ignition giving the MFI pump a rock steady ign timing, which MFI really like .
My 113 starts on a dime..........
And a final thought... on the throttle plate..you actually want that little stick feel when the plate is completely closed..opening up the plate a little off that is not recommended..that is why I told you about the cardboard in front of the air intake...you want all air for a/f mixture attainment to have to go thru the bleed screw...no air from the plate [ or as little as possible] That is what my cardboard trick verifies..if you can change the rpm and a/f mix at idle by blocking off the front of the intake snoot, then you know that the throttle plate is NOT closing as it should..and that will effect idle conditions and vac steadiness.
I think your main problem now is leaking check valves and it is quite common and a cabin sw usually does the trick. This is also common with a pump that has sat for a long period and the valves are dirty, thus not sealing well.
Another trick we MFI guys use is to allways turn the key ON for a few sec before starter engagement..this simply lets the electric FP pressurize the entire system [ including the line to the CSV] and gets some loop flow back to the tank in case there is any vapor in the lines from engine heat before turning the engine over.. try that when hot and see if that helps your hot start condition..you could have a possible vapor lock condition,,which I see sometimes at the main fuel filter due to its location and heat transfer from beinging hard-mounted to the engine.
...all these little tricks have come from working on these since they were new........and now you can pass it on.

__________________
A Dalton

Again, thanks for the big post and sharing so much of your knowledge and time.

So here's where I stand today.

1) Great cold start with all kinds of suction noise coming through the little airfilter on the injection pump. But the high idle was very shakey indicating to me that it was on the lean side.

2) Once it was hot and idled down, I let it sit for 15min or so and tried a hot start with the injection pump disco'd from the linkage. Wouldn't start hot even while cracking the throttle open, so that told me that it was in fact, too lean.

3) Disco'd the t-plate linkage, and hooked the injection pump linkage back up and it did start hot with depressing the accelerator pedal (again proving a lean condition).

4) Checked the t-plate stop and as you suggest, I did leave it sitting so there was that slight drag on it......so that was already correct.

5) Checked for an air leak from the brake vacuum booster and it's hose. No leak. Engine idles the same with or without the booster connected.

6) Forgot to mention this in my previous long post. While checking the vacuum retard function and line, I discovered that if I tugged a little on the "new" cloth braided hose that connects the plastic vac line to the t-body, the idle would raise! Turns out, this "new" hose was somehow pourous! I replaced it with some new diesel rated hose and curred that little problem. Weird.

7) Attached a vacuum gauge so I could watch vacuum while I made my next adjustments.

8) Turned the mixture on the pump 2 clicks richer.

9) Hooked up my dial timing light, set the dial to 30 and held the throttle at 3 grand. Rotated the distributor to "0" indicating I had 30 degrees of advance.

10) This gave about 2 degrees After, and I left it that way for testing.

11) Back to the idle air screw, I set for fastest idle and then turned the screw back in for about a 50rpm drop and that put vacuum at 13-14" with a very slight up and down idle.........which improves once the car gets driven.

12) Went for a test drive and the engine just screams! Acceleration in first gear to redline is so fast, it's like being on a motorcycle! All kinds of power across the board. Also, the loping that it was doing when driving around in a parking lot at very low speed with the throttle slightly cracked, is gone. That would get so bad the car would actually start jumping/learching and I'd have to put the clutch in, rev the throttle, and start over.

The hot starting is still not good, but I'm going to drive it for a few days and see if it goes away. This injection pump does have the 2 wire solendoid, and not hot start kit (but I've seen those on later 280se's).

Thanks for the suggestion to retorque the head bolts. I almost asked about that yesterday. Seems like most non-stretch bolt headbolts want to be retorqued on the first warm-up, and then again at 1,000km/620mi.

Final issue, and I think this will dial it in for real. The cam timing mark is just on the passenger side of the raised mark on the cam tower. Jumping the chain puts the mark the same amount on the driver's side. I'm thinking it would be smart to buy a couple of offset woodruff keys and dial the cam in so the marks line up when the crank is at TDC. Right now, I've left the mark on that passenger side, ie, retarded cam timing........probably why it starts to really scream as it approaches redline!

__________________

Just borrrowed another shop's exhaust gas analyzer. I had it set at 5.5%, so not bad for tuning by ear. One click leaner on the pump got me to about 4.9% and that's where I've left it for now.

Also, remembered to bleed the power steering assist line, so that's working......doh!

__________________

>

OK
What you want to do now is to verify that the solinoid for the rack is in fact working when the starter is cranking..if NO, then check the relay and power for that circuit. That relay is right next to the CSV relay..they are the same type relay. One for rack sol, one for CSV.
[ you will actually hear the rack/solinoid clicking as the rack moves]
If Yes, then the next time you try a hot start , do nothing different EXCEPT bring a one second ,12 V pos jumper from the battery to the CSV as you crank and see if it starts ..if YES , then the One second relay kit or the manual cabin switch will be the remedy.

I would also do the key ON/Pre start delay for electric fuel pump circulation I mentioned, in the event that you have a vapor condition...
Do that test before the jumper/CSV test......
That will isolate the condition for proper diagnosis.

__________________
A Dalton

Will do!

It's improving. It starts fine within a few minutes of being shut off, but much longer than say 15min and I have to crack the throttle to get it to light. BUT...it's no longer doing the stumble on first start, as if it were building fuel pressure (as you were talking about). It's just lighting right off and going straight to idle.

Other than that, it's really nice.

What say Yee about the cam timing? Should I make some effort to get the mark on the washer/camshaft woodruff key centered on the raised mark that's on the tower?

By the way, this thread has some meat on it's bones! Combine this with the linkage adjustment info that was posted/linked a few weeks ago from the w113 website and folks can really get their SE's dialed in.

__________________

You do not check valve timing by seeing how far left/right the cam tower index marks are....you turn the engine over by hand at the crank [ in normal rotation direction, of course] until the cam index marks are perfectly aligned.
Then you go down and read the degrees on the crank timing index scale. That tells you how far off you are in degrees ..you then 1/2 that for offset spec....but first see how far off you are ..if it is not many, it is not worth the bother.

__________________
A Dalton