Rough idle when cold

[raueda1]

In my travels in fiddling with the mixture adjustment screw if it starts to bog when you very slightly depress or lift the air flow plate then you are in the sweet spot.

My experience too.  Robert, I'm basing on Koan's comments this old thread, though I've read the same elsewhere.  But the whole thing is a bit mysterious to me.  I've fiddled with the enrichment screw and never got the engine to speed up by depressing the plate regardless of mixture screw setting.  Despite all my CIS adventures and misadventures I've never fiddled with the plate, so I can't help but wonder if my plate gap is right.  It goes back to when I got the car and discovered that somebody "fixed" a faulty WUR  by setting the system pressure = control pressure.    So, if plate is set wrong would that be one of those things that could mask or create other maladjustments per my "nothing's right until everything is right" dictum?  For example, if mixture is set according to a plate that's wrong, would mixture then be wrong as well?  Time will tell and perhaps another mystery of K-jet will be revealed.....          Cheers,

[raueda1]

Maybe some progress.  I couldn't find much on the air sensor plate setup on W116 K-jet systems.  However, I did find Bosch's Porsche CIS manual.  The system is virtually identical, especially the 928.  Air plate setup is described per the picture below. Anybody have experience or comments?  Also, FWIW, this manual seems more detailed and easier to follow than the MB version.  Porsche always gets the good stuff.      Cheers,
 

[daantjie]

Dave take this as you wish, but I always thought that the mixture adjustment is the very last step in the process?  Meaning that one should make sure ALL aspects of the CIS has been set and optimized fully before trying to get the mixture adjusted?  Maybe you have already done all this and thus it would make sense to finesse the last bit, in which case apologies

I have just seen so many folks over the years trying to "fix" their running issues by fiddling with the mixture screw and related when not all of the CIS components, like the WUR and FD and such have not been fully optimized

[ramiro]

There is a documentation in the WIS how to center the plate i would say it's more detailed then the porsche version    , i even found it on this site in english :
https://handbook.w116.org/cd/Engine/107/M117_45/073-245.pdf , it's for the m117 but all the injection parts are the same for the m100.
If you really want to monitor your cis injection i suggest you get rid of the gunson tester and get the cheapest wideband kit then you can always see what is going on and don't have to guess if the engine runs lean ar rich or if the ignition is the problem.

[raueda1]

Dave take this as you wish, but I always thought that the mixture adjustment is the very last step in the process?  Meaning that one should make sure ALL aspects of the CIS has been set and optimized fully before trying to get the mixture adjusted?

Yes, absolutely! No apologies.

Maybe you have already done all this and thus it would make sense to finesse the last bit, in which case apologies

At this point I can indeed say that all the other stuff has been done.  And I haven't spent4 any time analysing the ignition, though I shall.  Could I have unwittingly botched something?  Yes, almost certainly.  But I've been over the CIS over and over again so I'm confident that it's all by the book.

I have just seen so many folks over the years trying to "fix" their running issues by fiddling with the mixture screw and related when not all of the CIS components, like the WUR and FD and such have not been fully optimized

Absolutely!  That's why I pushed to get the CIS manual in  the tech section.  Interestingly though, the MB CIS manual doesn't even refer to the sensor plate!  It hadn't

[raueda1]

There is a documentation in the WIS how to center the plate i would say it's more detailed then the porsche version    , i even found it on this site in english :
https://handbook.w116.org/cd/Engine/107/M117_45/073-245.pdf , it's for the m117 but all the injection parts are the same for the m100.
If you really want to monitor your cis injection i suggest you get rid of the gunson tester and get the cheapest wideband kit then you can always see what is going on and don't have to guess if the engine runs lean ar rich or if the ignition is the problem.

Perfect!  Thanks, just what I was looking for.  Just from the pix I'm almost certain that  my plate is too low.  l will try it and see where it goes.  Cheers,

[raueda1]

There is a documentation in the WIS how to center the plate i would say it's more detailed then the porsche version    , i even found it on this site in english :
https://handbook.w116.org/cd/Engine/107/M117_45/073-245.pdf , it's for the m117 but all the injection parts are the same for the m100.
If you really want to monitor your cis injection i suggest you get rid of the gunson tester and get the cheapest wideband kit then you can always see what is going on and don't have to guess if the engine runs lean ar rich or if the ignition is the problem.

Perfect!  Thanks, just what I was looking for.  Just from the pix I'm almost certain that  my plate is too low.  l will try it and see where it goes.  Cheers,

The good news:  plate is perfectly in spec.  Not too low.  In hindsight I guess I'm used to looking at with engine running and hence drawing air.  The bad news: must look elsewhere.  OK, next items to be checked in order of my enthusiasm for the tasks:

• Ignition?  A plausible culprit, maybe, though I'm baffled that there could be temperature-dependent low speed missing without other ignition problems.  And car has new plugs, wires, coil, cap and rotor.  The car ran a little smoother and zippier afterwards but no effect on the rough idle.
• Cold injector?  Seems improbable as well but easy enough to confirm.
• Aux air valve?  Per the CIS manual can cause rough and/or high idle, no surprise.  Seems most likely to me but I'm no expert and pulling it doesn't excite me at all.
• Weird vacuum leak?  And wouldn't it have to be a temperature dependent vac leak?  And what could cause that??
• What else?

Thanks to all for lively discussion and suggestions.  Will report back with what happens.  Cheers,

[ramiro]

1.Would be really weird, usually m103 have this problem because of condesation in the head mounted distributor.
2.Could be if its slowly leaking but why would it stop when the engine is warm
3.Already wrote that it only changes idle speed not mixture.
4.Would be even worse when the change is warm
5.I think that you WUR enriches to much so the engine runs very rich when it's cold , best way to see that would be a wideband , if se the cold pressure to the lowest factory spec the engine will run very rich when it's cold i got AFR's below 10 with a cold engine.

[rumb]

Delving into the inside of the FD plenum show what is going on there.  The big pin moves the entire arm which moves both the plate and the FD plunger.  The FD mix screw only moves the end of the arm that pushes on the FD plunger, where the air stays the same and only the amount of fuel is changed. This shows why the plate should be set correctly before twidling with the mix screw.

Bosch engineers in the day must have had fun designing the shape of the plenum to have the right amount of air and also coincide with the amount of fuel.  Imagine designing that in the 70's before all the computers we have now!  The plenum bowl is almost a logarithmic increases in air due to the bowl shape as it open up.

[raueda1]

1.Would be really weird, usually m103 have this problem because of condesation in the head mounted distributor.
2.Could be if its slowly leaking but why would it stop when the engine is warm
3.Already wrote that it only changes idle speed not mixture.
4.Would be even worse when the change is warm
5.I think that you WUR enriches to much so the engine runs very rich when it's cold , best way to see that would be a wideband , if se the cold pressure to the lowest factory spec the engine will run very rich when it's cold i got AFR's below 10 with a cold engine.

Yes, fully agreed on 1 - 4.  Still, can't hurt to eliminate all that stuff.  As for #5 (WUR), it's so odd.  Cold pressure is spot on, and checked over a range of temperatures well within Bosch's temperature/pressure maps (Example below for a Porsche).  Still, it could be too rich anyway.  After all, the maps cover a pretty big area, they aren't a straight line.  Maybe I can find the map for my WUR (ending in xxxx10) and bump cold pressure to the top of the range on map.  And then maybe I can find decent exhaust analyser that doesn't cost +US$1500?  Suggestions welcome.  The project goes on.....    Cheers,

[daantjie]

If you think about it, running a bit less rich at start is not so bad unless you really live in super cold ambient and you want to drive your car in winter which most of us don't.

I found that mine was running stinking rich even at mild ambient so I knocked back the WUR somewhat. I think it's a bit of a compromise honestly and you need to find that sweet spot where both cold and hot fuel pressures are "perfect"

[raueda1]

If you think about it, running a bit less rich at start is not so bad unless you really live in super cold ambient and you want to drive your car in winter which most of us don't.

I found that mine was running stinking rich even at mild ambient so I knocked back the WUR somewhat. I think it's a bit of a compromise honestly and you need to find that sweet spot where both cold and hot fuel pressure is "perfect"

Yes, I think you're exactly right.  Stay tuned....

[raueda1]

Bosch engineers in the day must have had fun designing the shape of the plenum to have the right amount of air and also coincide with the amount of fuel.  Imagine designing that in the 70's before all the computers we have now!  The plenum bowl is almost a logarithmic increases in air due to the bowl shape as it open up.

Yeah, that's why I thought that mixture might depend on position being right.  But I never knew about how to set the plate per your nice pic.  And yeah, how they lived without computers is a wonder.  Yet our ancestors created monumental ICE's used in WW2, not to mention The Bomb.  And the B-17 to the B-58 in about 15 years!  We've become soft and weak.   Cheers,

ps - hope that plenum isn't going under the hood of your otherwise perfect project!  LOL

[ramiro]

I can find decent exhaust analyser that doesn't cost +US$1500?  Suggestions welcome.

Just buy a kit like this one https://www.amazon.com/AEM-30-0300-Wideband-Sensor-Controller/dp/B0184TSI84/, you don't have to permanently mount it.
I added a 12v plug to it and welded a thread insert into the exhaust with a plug so when i want to measure the exhaust gases i can put in the AFR meter and remove it after that , although i am planning to mount it permanenly with a small screen where the colum automatic gear indicator normaly goes in the instrument cluster.

I also reduced to cold pressure on mine , the only problem is that if it's not super rich it likes to bog down in the first ~ 20 seconds when you push the throttle at idle an its below 10 C outside.

[rumb]

I really like that AFR meter. Looks like a better way to dial in old cars.