Mercedes-Benz Forum banner

1 - 9 of 9 Posts

·
Registered
1988 560SL
Joined
·
60 Posts
Discussion Starter #1
Sorry in advance for the long post. I’m just trying to give as much background and info as I think is relevant. I also hope the vacuum diagnostic test article I linked may be helpful to others… it’s pretty simple tests.

I’ve been trying to resolve a bit of a rough cold idle that my ’88 560SL has had since I bought it about 20 months ago. I’ve done a long list of preventative refreshing… but also with hopes that one or more things would correct the problem… but so far no luck. Things I’ve done include: 1) New plugs, wires, cap and rotor, 2) replaced all engine bay vac lines, injector sleeves/orings/seals and air pipes, checked that EGR, Air Pump etc hold vac., smoke tested (I do have a slight trace coming from somewhere under the throttle/FD, but am not convinced a very slight vac leak is my issue) 3) both fuel pumps (they were humming) and fuel filter along with new rubber fuel hoses back at the pumps and in the engine bay. Most recently I decided to try two new injectors on cylinder 6 and 8 (chose those two cyl because those plugs seemed to get more carbon buildup than others). That seemed to help a slight stumble when accelerating from stop, but not rough idle when cold.

When the engine is cold, the econ gage is about ¼ in off the peg, but once the engine warms up it goes back down to the peg. I know this suggests issue is due to vac leak (perhaps from a seal or gasket that seals up when engine gets hot and expands), but I don’t think smoke test supports this, and I think duty cycle measurements (below) indicate ECU is trying to lean further, which also would not support vac leak as issue.

I went back in search of other diagnostics and things that could cause engine to produce low vacuum. I found this article which I think could be very helpful, but am having trouble making sense of my Vacuum, RPM and Duty Cycle measurements. Mastering the Basics--Reading A Vacuum Gauge | MOTOR

So here’s what I’m measuring for vacuum, RPM and duty cycle at various warmup points.
When engine is cold (i.e. sitting overnight in 40F garage, startup, pull car out to driveway and begin tests with coolant temp still below the 50C mark)
In park RPM fluctuates from approx. 600 – 700 RPM, Vac fluctuates between 14-18 in
In drive, RPM drops to just a bit over 500 RPM (top of tach needle is touching top of 500RPM line). Vac is 14in -15in. Both are steadier than when in park.
I didn’t take any duty cycle measurements since I think the engine was still too cold at this stage to kick to closed loop. Idle is noticeably rough, and has a little surging when in park as noted by the RPM fluctuation.

Drove around block until coolant temp just touched 80C (hot enough where lambda should be in control, but not fully warmed up), and re-took measurements.
In park, RPM is just over 600 and steady (no more surging) and vac is up to 18 – 19 with slight jumping around in that range. I forgot to test vac and RPM in drive as I was focused on the rev testing which was normal and seems to eliminate possibility of clogged exhaust (vac increased and held steady at 22in when holding rev around 2500RPM, jumped to peak of just under 24in when I let off throttle).

Duty cycle at this temp (with carbon canister line to intake plugged up) was 23% - 25% at idle and 29.5% - 32% at 2500RPM (this was with a Fluke meter in duty cycle mode, red on pin 3, black on pin2, which measures correct for the engine off tests 69.8% when ignition is on engine off, 10.0% with air sensor plate deflected, and 19.8% at full throttle). I think from reading HD’s great posts on KE Jet on the 107 and 126 forums, that means the ECU is trying to lean the mixture (but I could have it backwards… as there is a lot of conflicting info on interpretation out there.). Idle at this temp is still a little rough… doesn’t seem like a constant miss, but a little bump once or twice every few seconds. Econ gage in park has settled down to the peg, but in drive still comes off peg about 1/8 in or so.

Next, drove the car on an errand to the store and back to get good and warm. The rough idle pretty much goes away once the engine gets good and hot as indicated by the oil pressure coming down to 2 when car is idling. .
The vacuum measurements in park were the same as at 80C (pretty steady between 18-19in); however, the duty cycle was better. 35% - 37% at idle and 39% - 47% at 2500 RPM.
Just as a check, I hooked the carbon canister line back up and measured duty cycle 45% - 55% with hot engine.

The main questions/thoughts I have are:
  • Am I interpreting duty cycle correctly… specifically that duty cycle below 50% (and mine are quite a bit below) indicates ECU is trying to lean the mixture, which would be opposite of behavior if there was a vac leak. From what I’ve read, vac leak would also manifest as a higher than normal idle RPM, which I don’t have (if anything mine is a hair below normal)
  • The article with the vacuum diagnostic tests indicates that low vacuum can indicate retarded timing. I’m pretty sure I don’t have a cam or ignition timing issue (I replaced my chain guides 6 months ago, and crank was only a couple degrees off when sprocket marks were lined up on both sides… which is why I decided to not replace the chain yet). I triple checked this before closing the valve covers back up. I also double checked the rotor timing lined up at TDC when I replaced the cap and rotor a couple months ago. Is it plausible that poor injector spray could mimic the symptoms of retarded timing (bigger droplets take longer to combust, right?), but this issue becomes less noticeable when the engine gets hot as the heat from the cylinder helps vaporize the bigger droplets before the spark?
  • In reading more about how the O2 sensor measures average of exhaust from all cylinders, and how the measurement will be dominated by the richest cylinder, I think I’ve come to realization that trying to troubleshoot injectors by replacing just a couple is probably futile. Even if I did replace the worst injectors, I may have just pushed the problem to other cylinders.
If anyone has any theories for what might be causing my rough idle when cold, or suggestions for other diagnostics please let me know. I’m thinking my next step at this point is to go ahead and replace the other 6 injectors (they have 160k miles, so probably due anyway)
 

·
Premium Member
560SL 1986 244k miles astral grey / black
Joined
·
399 Posts
Nice post!

Your vacuum leak under AFM is likely one of the donuts underneath. Quite a job to replace, but the opportunity to clean your intake manifold is very rewarding.

Indeed, as a next step I would as well replace the other 6 injectors + seals around them, so you know for sure that is not it and with 160k miles done it's rather logic.

2616887
 

·
Registered
1988 560SL
Joined
·
60 Posts
Discussion Starter #3
I appreciate the encouragement that injectors would be a good next step and wise to replace based on age/mileage even they aren’t my main issue.
I would like to hear more about why you think donut seal is more likely cause? I can see some of my symptoms would fit a vac leak at a seal/gasket (Econ gage and vac test showing lower vacuum when car is cold that goes away when car gets hot). But other diagnostics seem to point away from vac leak as cause (duty cycle trying to lean, and lack of elevated idle rpm). I know from smoke test there is some very small leak down there(barely enough to even see with flashlight in dark garage), but would like to understand better what points you to these seals as culprit.
 

·
Registered
560SL '88 Suzuki GS1000E '78
Joined
·
60 Posts
Hi Pascal, I need more time to go through all your stuff but a couple of things so far:
Duty-Cycle need to be closed to 50%, focus is the 2500 RPM-point. At Power on you should see 85% if you have a '88 MY California-Model! 70% indicates a system without OBD. Maybe someone adjusted duty-cycle for the system with all the leakage and now with new seals it's just to rich.
During the change of the chain and glides you may have lost the synchronisation between crankshaft and distributor. For the 560SL this won't affect the ignition timing (this is made by the electronic ignition unit), but the distribution may suffer! That was one of the main issues at my engine.
Martin
 

·
Premium Member
560SL 1986 244k miles astral grey / black
Joined
·
399 Posts
I appreciate the encouragement that injectors would be a good next step and wise to replace based on age/mileage even they aren’t my main issue.
I would like to hear more about why you think donut seal is more likely cause? I can see some of my symptoms would fit a vac leak at a seal/gasket (Econ gage and vac test showing lower vacuum when car is cold that goes away when car gets hot). But other diagnostics seem to point away from vac leak as cause (duty cycle trying to lean, and lack of elevated idle rpm). I know from smoke test there is some very small leak down there(barely enough to even see with flashlight in dark garage), but would like to understand better what points you to these seals as culprit.
I don't know if the donut seal is the exact cause of your issue, but it's known KE-jetronic is sensitive to vacuum leaks, so if you know you still have one due to the smoke test you did, in my opinion that should be fixed first. To establish the cause or causes of your issue at a certain point you need to be sure certain components of the KE is working properly. Either by testing it or by replacing it. With the injectors as an example, knowing they did 160k miles, my personal taste is to replace them instead of testing.
 

·
Registered
1988 560SL
Joined
·
60 Posts
Discussion Starter #6
Martin. Thanks. I don’t hav Cali model, so 70% for engine off is my target (which I’m getting and using to confirm my mm leads aren’t flipped). Please explain more about distributor-crank alignment? I put distributor back on when crank and cams were aligned to tdc, and made sure it went in so rotor was pointing to tdc also on cap. Is there something I’m missing?
 

·
Registered
1988 560SL
Joined
·
60 Posts
Discussion Starter #7
Bauk. I understand. Also based on martins input I may be putting too much confidence in what duty cycle is telling me. Assumes prior owner didn’t fiddle with it at reset 50% point. One thing I am starting to appreciate about KE jet is that system seems to assume all cylinders are operating the same. That is certainly not true for my injectors after replacing two with new ones.
 

·
Registered
560SL '88 Suzuki GS1000E '78
Joined
·
60 Posts
Martin. Thanks. I don’t hav Cali model, so 70% for engine off is my target (which I’m getting and using to confirm my mm leads aren’t flipped). Please explain more about distributor-crank alignment? I put distributor back on when crank and cams were aligned to tdc, and made sure it went in so rotor was pointing to tdc also on cap. Is there something I’m missing?
Hi Pascal, attached a pic that may help to understand better. If the crankshaft is in TDC-position, the finger in the distributor needs to aling with the mark on the housing. On the pic you can see the missalignement from my engine. I took the distributor out, turned it one tooth back until I had a good alignement and from this day idle was almost perfect.
Martin

2616954
 

·
Registered
1988 560SL
Joined
·
60 Posts
Discussion Starter #9
Martin, thanks for the pic. I did line that up.,, that's what I meant by lining the rotor up to the TDC mark when the crank (and sprokets) were also on TDC. In re-reading my response I can see that I mistakenly said the mark was on the cap, but hat mark on the distributor housing is really what I meant.
 
1 - 9 of 9 Posts
Top