problem setting idle

[Joe]

My new 71 280SL won't pass emissions. When I tried to turn the thumbscrew on the back of the IP, I couldn't feel any detent. I turned the thumbscrew all the way CCW, and it jams up and stops turning. I suspect things are not well inside the rear of the pump. Any advice is appreciated.
Incidentally, I managed to get the CO down to about 4% with the Gunson meter by turning the air screw CCW, but the idle speed is about 1700. The emissions place said it has to be below 1350 RPM for them to certify it.
Joe

[ja17]

Hello Joe,
It sounds like the adjustment screw is wound all the way out. When this happens the little hex bolt moves past the leaf springs so no "click" is felt. Search "injection pump tour". You can remove an oval shim from under the warm up device to lean the pump in all speed ranges. After doing this you may need to wind the adjustment screw in enough to get the "click" back. The danger with having the adjustment screw too far out is that it may rub against the adjustment screw when running.
These cars should idle at around 750 rpms. Make sure your venturi stop screw on the intake venturi is allowing the flap to close all the way or you will never get a low idle. The stop screw on the venturi should be set to just barely keep the flap from binding. Actually the venturi adjustment should be done before making any pump adjustments! Keep us up to date.

Joe Alexander
Blacklick, Ohio

[Joe]

Thanks for the info, Joe. Here's an update. Before I started this, I adjusted the venturi plate. I also replaced most of the sockets and balls in the linkage, as well as the two brass bushings in the crossbar that connects the linkage to the injection pump to the other side of the engine. Everything is tight, and opening the injection pump coincides with opening the venturi very well, I think.
I took off the spring-loaded adjuster on the rear of the IP and the plate that holds it on, and found that, just as you suggested, the adjustment screw inside was screwed too far ccw, and was just loose. I screwed it back in (cw) and can again feel the click when I turn it. In your IP tour, you don't mention whether the adjustment screw can simply be screwed back in if it has been turned too far. Is it ok to do this?
The CO level is still too high. I guess my next step is to dismantle the thermostat housing and remove a shim. Is there anything else that can cause too high CO readings? I know my valve guide seals are bad, but I don't want to attack that job right now. I checked the csv and it doesn't leak at all.
Joe2

[ja17]

Hello Joe,
Yes the rear adjusment screw should  wind back in and be fine.
These injection pumps wear rich, that is they run richer as they get more miles on them. Removal of a shim or a rack adjustment will lean the pump in all rpm ranges.

Joe Alexander
Blacklick, Ohio

[Cees Klumper]

A summary from the old yahoo! posts:
"A common cause of a rich mixture across the whole range is a worn barometric compensator diaphragm, which can be corrected by removing some of the thin shims once you screw it off. The compensator (the round cylinder on top of the fuel injection pump) sits on top of the fuel injection pump and is about 7-8cm in diameter. It pushes down on a lever inside which adjusts fuel mixture over the whole range. With age these compensators wear out and the pin does not push in as far as required. That is why you have to remove some shims to get things back to normal. Be careful to check that you don't go too lean, e.g. through your CO meter or by "reading" the plugs after a few test runs. Just use a thin 19mm wrench to grab the hex flange at the lower side of the barometric compensator, to gain access to the shims."

Also have you made sure that the thermostat closes all the way when the engine is warm? Hold a piece of fuel or similar hose to the inlet of the air filter on the side of the FI pump (that other, smaller, round canister) and make sure there is no air being sucked in.

Cees Klumper in Amsterdam
'69 white 280 SL automatic

[Joe]

Update. Today, I removed the filter on the thermostat tower and started the car. There was lots of suction. After the car reached operating temperature, there was zero suction. I checked this by holding a small scrap of paper next to the opening, and it was not pulled at all.
Next step: took off the tower and base and found two shims, each .020 inch. Don't know if these are thin or thick, having no reference. Took off one and restarted the engine. Using the Gunson CO meter, I saw no effect. I then adjusted the idle mixture screw on the IP ccw, and there was still no effect. Things are getting weird.
I then removed the remaining shim, and there was still no effect. The idle mixture screw is now all the way ccw.
Further exploration found that the air cleaner element is closer to dirty than clean. Also, the bellows between the air filter and throttle body has a small tear, which I could see, because it was on the bottom. I'll put some electrical tape around that and restart the car without the air filter element and see if that has an effect. Now I have to show the better half some attention (darn it, perfect weather to work on the car!).
The plugs are new NGK BP6ES, set to .035, and the plug wires are new Bosch copper. I've not looked at the points or timing, but don't think they would affect the CO reading.
Any more advice? Should I take off the throttle body and clean it out? I started to do that, and found it has a water line to it. The 230SL gets by just fine without this addition.
Joe2

[A Dalton]

If you are having problems with your adjustments and shims not having an effect, you may want to be sure the cold start squirter is not leaking by.  This will cause richness and if someone has tried to compensate for this by leaning out the pump idle screw, this would account for the high rpms..
 High idle rpms can also be advanced ignition timing.. One should always set timing BEFORE any pump adjustments...both high rpm timing and retart/idle timing.

 PS..it is common for these to have idle CO of 3.5%/4.5%

[ja17]

Hello Joe2,
Good tip by A Dalton. Joe2 how does the engine react to adjustments of the air idle screw on the intake manifold? Yes it does seem like you are getting extra fuel and/or air somewhere. You should have seen a noticable difference from removing the shims.



Joe Alexander
Blacklick, Ohio

[Joe]

The csv does not leak at all. The CO goes down as I turn the air screw ccw, but the rpms increase. I showed the emissions guy yesterday how to turn the air screw, and then watched the effect on his meter as he did so in a vain attempt to get the car to pass. By law, a car being checked for emissions must be checked at idle, and the idle speed cannot exceed 1350 rpm. At 1342 rpm, the CO was about 5.7, which is too high; the max allowable is 4.5%. Today, after removing the shims, I set the idle about 800 rpm using the idle air screw, but the CO was in the 7% range. When I turned the air screw cw till the idle got close to 1350, the CO was still above 6%.
This guy seemed to think it was a vacuum leak that was causing the high CO. However, he could not find any place where a leak was evident. (I suspect that is a common reason for this sort of problem, and that's why he mentioned it.) There's a small brass tube sticking out of the throttle body, opposite from the one that sends vacuum to the distributor, but the cap on it appears to be in good shape. The line to the distributor looks ok, as does the large line to the brake booster.
I'll check the timing tomorrow (what should it be?). And, is there a suspect place to check for a vacuum leak? Is there a vacuum line associated with the automatic tranny? I probably should read the plugs, too, right?
Thanks for sticking with me on this!
Joe2

[Ben]

To consolidate an earlier point, the ignition timing and dwell should always be set before any adjustments are done for idle and CO otherwise its futile, as the emissions will change with any variation in timing !

 

quote:When I turned the air screw cw till the idle got close to 1350, the CO was still above 6%.

.......this is strange as it should be the opposite, unless you havent turned it IN enough. If you turn the idle screw all the way CW, effectively blocking the air it should stall ! If not the this would confirm a vacuum leak !

Regards,
Ben in Ireland.
'64 230SL 4sp.
'03 CLK Kompressor

[ja17]

Hello Joe2,
As mentioned, check to make sure the cold start valve on the intake is not leaking. Also the linkage lever on the injection pump should be resting on it's stop at engine idle.

Another test;  disconnect the linkage between the injection pump and linkage crossover rod. Next move the linkage to the venturi open with the engine running. This will increase air without increasing fuel. If the engine speeds up more than a couple hundred rpms, then the mixture is still too rich.
An internal rack adjustment can be made if all else checks out ok.
Stick with it we'll get you there!


Joe Alexander
Blacklick, Ohio

[TheEngineer]

Ignition adjustment affects idle speed big time. Check dwell angle and following that set the timing. If all fails, get "collector vehicle" plates. Here in WA, with those plates, we don't have go to emissions. The shim adjustment under the barometric pot is very sensitive: A few .001 inch will make a considerable difference over the entire speed range. My injection pump gives me too much fuel in the first three cylinders, the others are fine. I suspect, I wouldn't pass.

1969 280SL,Signal Red, A/T, P/S, A/C, Both tops.    Ret. engr. 1967 230SL - Sold

[Joe]

Today's update on getting the 1971 280 SL's CO to within limits:
Checked the dwell, and it was spot on at 30 degrees. When I revved the engine, it didn't vary, so I know conditions inside the distributor are probably ok.
Now to the timing. After researching under "Ignition" in the site data base, I decided the timing should be set to 30 degrees ATDC at 3000 rpm. Because I'm at 6000 feet above sea level, I'd want to retard the timing, so subtract about 6 degrees. That makes 24 degrees ATDC. I thought something was terribly amiss when I found the timing was 50 degrees ATDC at 3000 rpm! Double checked, and this was indeed right. Oddly, there is a spot of white on the crank at 50 degrees, indicating someone before me thought that was correct.
I loosened the clamp on the distributor and moved it ccw and the car barely started. It ran fine before I did that. To keep the car running, I had to turn the air screw out several turns.
Given how far off the timing had been set, I decided to put the two shims back under the thermostatic valve thing. That brought the idle up to the point the car would run. I then connected the CO meter and found it was about 2.6%, telling me I had fixed the problem of too-high CO readings.
I tweaked the timing to about 25% ATDC at 3000 rpm, and then fiddled with the injection idle mixture screw and the air screw until it idled at about 800 rpm and CO was around 3.5%. The idle mixture screw is now near the middle of its range, so I don't have to worry about it unscrewing.
Tomorrow, I'll take the car back for a retest; I expect it to pass.
Had I checked the timing and dwell first, as a few guys here have suggested, I'd have been spared this exercise. All in all, though, I think I enjoy working on the car more than driving it, so this was not all futile.
Thanks for the help and support, everyone.
Joe2
1971 280 SL, Red/Parchment, Auto

[Ben]

Glad it all worked out for you......however one thing still concerns me !

The dwell !

 

quote:Checked the dwell, and it was spot on at 30 degrees

.........I thought it should be 38 - 40 degrees ??  It IS on my 230 !!

Regards,
Ben in Ireland.
'64 230SL 4sp.
'03 CLK Kompressor

[A Dalton]

Almost all the pumps that are out of adjustment are caused by different owners playing with the adjustments WITHOUT setting the ignition up FIRST.
 Then ,they complicate the setting even more by doing tune ups on the ignition after the pump..
 Jernold told me tat 95% of the pumps sent out for repair/adjustments
have basically nothing wrong with them...
 ALWAYS , ignition first, then valves, and lastly, fuel..
 Same goes for alignment..the last adjustment is always toe.  All other adjustments change toe.

 Glad you found the problem/s...

[Joe]

Well, The car passed emissions this evening. Now the problem is that it won't idle in gear. It is idling about 800 rpm in park and neutral, but when I shift to reverse or 4, the rpms rapidly decrease and the car dies. I can keep it revved and drive it, but I'd like to fix this problem. Is it a common problem?
Thanks,
Joe2
1971 Auto

[ja17]

Hello Joe2,
Sounds like your learning a lot. One important thing for everybody  to remember (as we have heard so  many times before), always check all tune-up items before begining injection adjustments.

Now the problem at hand; does your car have the vacuum dashpot on the venturi and does it have the slip joint linkage rod between the venturi and linkage cross-over rod?
does it have the constant idle solenoid on the intake? These items both aid idle while in gear. However if the engine is healthy and hitting on all six cylinders at idle with correct fuel mixture, the engine will usually continue to run without stalling even without the dashpot or idle solenoid. The engine rpms may drop, but continue to run. The dashpot and solenioid should keep the idle fairly constant in or out of gear. Of coaurse by doing this they will also lessen the chance of stalling when in gear.

Start the car and let it warm up in park. Disconnect the linkage rod going to the injection pump from the linkage cross-over rod. With the engine running in park push down on the linkage if the engine rpms increase over 200 rpms the car is idleing rich. The engine rpms should raises only slightly and then faulter.
Next leaving the linkage unhooked, open the venturi linkage, if the rpms increase over 200 rpms then this confirms the idle is rich. If the engine faulters right away the idle mixture is lean.

You can turn the engine off and adjust the idle mixture knob on the pump and then retest. As you get closer to the correct mixture the engine should continue to idle under load in gear.

The dashpot and constant idle solenoid can be adjusted so there is very little difference in rpms in gear or out.
 
Here again good tuning, plugs, wires, timing, compression etc. is important for a good smooth idle before mixture adjustments can be successful.

I always do final idle adjustments, (with the aid of a helper), with the car "in gear". Idle "out of gear" is only experienced during start up and just before parking. Idle "in gear" is experienced many times during a drive.

Let us know if you have the dashpot and solenoid.


Joe Alexander
Blacklick, Ohio

[Joe]

Thanks, Joe Alexander.
I'm not sure what you mean by the dashpot on the venturi and the slip joint thing. Connected to the right side of the crossover rod are two rods. One goes directly to the venturi body and the other goes, of course, to the linkage that goes downward through the manifolds and thence to the accelerator pedal. Mounted on the rear of the intake manifold, very close to the tranny dipstick is some sort of solenoid that has two wires to it. One wire is grounded and the other goes to the temperature switch on the venturi body and to the vacuum advance/retard switch, or whatever it is called. Another wire that is tied into this wire goes down under the ac compressor to a point I could not trace in the dark. I did see some places on these wires where the insulation is worn off, and it may be grounding out.
The solenoid obviously functions to increase the engine speed, and it is this item that is probably not working correctly.
Joe, when you say you do your final adjustments with the car in gear, do you mean the IP idle mixture and air screw adjustments?
I assume you get the car through emissions and then tune it to run best, right?
Joe2

[ja17]

Hello Joe2,
Some USA delivery W113 cars with automatics came with a disc-like vacuum pot hanging below the venturi. At low idle in gear, this dashpot would increase the idle by pushing the linkage. the cars equiped with this feature had a slip joint on the linkage rod from the crossover rod to the venturi. This slip joint allowed the dashpot to increase air by moving the venturi open without moving the rest of the linkage going to the injection pump or accelerator pedal. If you have this there is a specific procedure for setting it up.
The bakelite switch on the intake venturi influences the transmission by operating a three position solenoid at the transmission which changes the modulator pressure for downshifting, upshifting, and kick down passing.
The solenoid on the intake increases idle rpms in W113 cars with AUTOMATIC transmissions in pre 1970 cars via pressure switches on the transmission. On W113 cars 1970 and later the same solenoid is only hooked up  on models with AC. Late cars without AC have the solenoid but it is not hooked-up to anything!

The injection pump must not be adjusted with the engine running. The air screw on the intake can be adjusted with the engine running in gear at idle for tuning.

Joe Alexander
Blacklick, Ohio

[Joe]

Thanks a lot, Joe. That answers some questions. After you wrote last night, I went to the car with a flashlight so I could answer you accurately, but then failed to do so. TI now know the CS solenoid is connected to a wire which goes through the firewall to parts unknown. The other end goes to the vacuum switch and another wire ties into this wire and goes (apparently) to the ac compressor. These wires look patched to me, but apparently they are routed correctly.? I don't know why the wire goes to the vacuum switch, though. Should the vacuum switch be wired to another place?
Today, I had someone start the car and put it in gear, and checked for voltage on the CSS, and there was none. I thought I would have to crawl under the car to examine connections on the transmission, but you've told me that is not necessary since the CSS only functions with the ac on and this is a '71. I assume the voltage that gets to the CSS when the ac is on originates at the compressor, right? If so, where does the wire through the firewall go? I realize how foolish this question sounds, by the way!
I'm still unable to get the car to idle in gear. Would you come to my house and help me?
Joe2
1971 280SL, ac, auto
Colorado Springs

[ja17]

Hello Joe,
Are you working with any literature (BBB, data booklet, Chuilton's etc) ? A good wiring diagram can answere many of your wiring questions easily. We can get one in your hands if needed. Also you may want to visit another W113 owner to take notes. With Colorado Springs being the National headquarters of MBCA you should be able to find some real enthusiasts in the area to help.
The intake manifold cold start solenoid on your 280-SL the is controlled by a temperature sensor (two wire sensor on spark plug side of the cylinder head), and a relay. It is activated when the starter is engaged and can squirt fuel from 1 to 12 seconds depending on the temperature. Check your fuses, check the relay and check the valve. Keep in touch!
Remember the engine must be hitting on all six cylinders or it may stall when putting it in drive. If you have low compression  or ignition miss firing, then injection and linkage adjustments may not achieve a smooth even idle at 800 rpms. Too bad your 1500 miles away!

Joe Alexander
Blacklick, Ohio

[Joe]

Here's an update on this situation. We drove the 280SL to Aspen on Saturday, and returned last night. The entire trip was about 460 miles, including tooling around (20 MPG!). Prior to leaving, I put a can of BG44K in the gas, and after a quarter tank of gas was used, the idle was up to about 1600 from 1100, and the car idled in gear about 600 RPM. I guess the injectors were dirty or something else in the fuel system needed cleaning. Now that I'm home, I'll readjust the idle settings.
As for getting the car worked on in Colorado Springs, there are two independent shops frequented by MB owners. Once I bought a 300D Turbo, and the turbo didn't work. I took it to both these shops, and they both said the turbo needed replacing. This was on the order of $1600 plus labor, and being cheap, I looked further. A man who called himself "Diesel Dan" on the internet helped me discover an overboost shutoff valve, or something like that, was bad. I replaced it for about 25 bucks. The turbo then worked fine.
One of these shops has been "the place" to take your car, and is especially popular with real estate agents and wives of physicians. They always send out birthday cards and reminders of service required as part of their fawning and obsequious facade. In December last year, the state raided the place and found over 100 stolen MBs there. They were shut down, but are now reopened. I don't know any details, but do suspect their moral lapses are not just recent.
The other shop is a one-man operation, and this guy, a purported ex-German, exhibits the requisite crustiness to be a 113 expert. However, I think he was the person who set the timing on my car 20 degrees off, because the person who sold me the car had this guy do the emissions for it. The fact that the CO was over 10% only 300 miles after the car passed his emissions test indicates possible moral lapses on his part, too. Don't get me wrong. These sorts can be very useful when faced with onerous state laws.
Anyway, like so many fellow list members, I've discovered if I want the car worked on right, I'll have to learn to do it myself.
The car looked wonderful in Aspen, town of designer water, food, women, cars, tans and dogs. The designer car of the moment is the Hummer, though, indicating the owner is serious about chic off-the-road activities. Special cachet is achieved via the proper exhibition of mud on the vehicle.
Returning, we were on an 8-mile downward slope in the Rockies, and I put the car into neutral to save gas. I noticed the RPMs stayed about 2100 while coasting. Does this indicate something in the automatic transmission needs adjusting? Also, is coasting potentially damaging to the transmission, like towing the car with the driveline intact?

Joe2

[n/a]

This is very fascinating.  Can someone explain how changing the timing can affect the fuel richness mixture??  I sure would like to understand that concept!
Thanks,
Don
71 280sl 4 speed
Cleveland

 "I decided the timing should be set to 30 degrees ATDC at 3000 rpm. Because I'm at 6000 feet above sea level, I'd want to retard the timing, so subtract about 6 degrees. That makes 24 degrees ATDC. I thought something was terribly amiss when I found the timing was 50 degrees ATDC at 3000 rpm! Double checked, and this was indeed right. Oddly, there is a spot of white on the crank at 50 degrees, indicating someone before me thought that was correct.
I loosened the clamp on the distributor and moved it ccw and the car barely started. It ran fine before I did that. To keep the car running, I had to turn the air screw out several turns.
Given how far off the timing had been set, I decided to put the two shims back under the thermostatic valve thing. That brought the idle up to the point the car would run. I then connected the CO meter and found it was about 2.6%, telling me I had fixed the problem of too-high CO readings.
I tweaked the timing to about 25% ATDC at 3000 rpm, and then fiddled with the injection idle mixture screw and the air screw until it idled at about 800 rpm and CO was around 3.5%. The idle mixture screw is now near the middle of its range, so I don't have to worry about it unscrewing."

[n/a]

Hello Ben,
The 71 280sl has the transistorized ignition.  For this system the dwell is 30 degrees, as opposed to other cars which the dwell is higher.
Best regards,
Don

quote:Originally posted by Ben

Glad it all worked out for you......however one thing still concerns me !

The dwell !

 

quote:Checked the dwell, and it was spot on at 30 degrees

.........I thought it should be 38 - 40 degrees ??  It IS on my 230 !!

Regards,
Ben in Ireland.
'64 230SL 4sp.
'03 CLK Kompressor

[n/a]

Joe,
The BBB says that the timing should be 25-30 BTDC at 3,000 rpms, not ATDC.  If you stand in front of the car, graduations to the right of 0/0 are BTDC, and graduations to the left of TDC are ATDC.  Can you elaborate on how you set the timing?  
Best regards,
Don
71 280sl 4-speed

quote:Originally posted by Joe

Today's update on getting the 1971 280 SL's CO to within limits:
Checked the dwell, and it was spot on at 30 degrees. When I revved the engine, it didn't vary, so I know conditions inside the distributor are probably ok.
Now to the timing. After researching under "Ignition" in the site data base, I decided the timing should be set to 30 degrees ATDC at 3000 rpm. Because I'm at 6000 feet above sea level, I'd want to retard the timing, so subtract about 6 degrees. That makes 24 degrees ATDC. I thought something was terribly amiss when I found the timing was 50 degrees ATDC at 3000 rpm! Double checked, and this was indeed right. Oddly, there is a spot of white on the crank at 50 degrees, indicating someone before me thought that was correct.
I loosened the clamp on the distributor and moved it ccw and the car barely started. It ran fine before I did that. To keep the car running, I had to turn the air screw out several turns.
Given how far off the timing had been set, I decided to put the two shims back under the thermostatic valve thing. That brought the idle up to the point the car would run. I then connected the CO meter and found it was about 2.6%, telling me I had fixed the problem of too-high CO readings.
I tweaked the timing to about 25% ATDC at 3000 rpm, and then fiddled with the injection idle mixture screw and the air screw until it idled at about 800 rpm and CO was around 3.5%. The idle mixture screw is now near the middle of its range, so I don't have to worry about it unscrewing.
Tomorrow, I'll take the car back for a retest; I expect it to pass.
Had I checked the timing and dwell first, as a few guys here have suggested, I'd have been spared this exercise. All in all, though, I think I enjoy working on the car more than driving it, so this was not all futile.
Thanks for the help and support, everyone.
Joe2
1971 280 SL, Red/Parchment, Auto