News:

The ORG - 100% FREE advice!

Main Menu

Rough idle when cold

Started by raueda1, 01 November 2018, 10:03 AM

raueda1

After solving a long crazy WUR/fuel pressure mystery my car is running very well.  However, it's now cold enough here, 35-40°F,  that all the cold start mechanisms should be in play.  The car starts immediately but idle is very rough for a minute of so before it smooths out, almost like it's not firing on all 8.  I know that the WUR and fuel pressures are all ok.  At one point I cleaned the aux air valve on principle.  Any thoughts on what to look at?  In the spring it's my plan to replace injectors and maybe rebuild fuel distributor.  Thanks!

-Dave
Now:  1976 6.9 Euro, 2015 GL550
Before that:  1966 230S, 1964 220SE coupe, 1977 Carrera 3.0

daantjie

You know what they say about things you know for certain, right ;D?  Might be time to recheck the WUR operation and fuel pressures?
AAV could be the culprit yes.  At full cold it must be completely open.  Unlikely that it will cause super rough cold running unless of course it is stuck closed, which sounds like might not be the case, but good to check.  Usually they cause rough hot start/idling as they are stuck open, thus the inverse of what you might have.
Daniel
1977 450 SEL 6.9 - Astralsilber

ptashek

Quote from: raueda1 on 01 November 2018, 10:03 AM
After solving a long crazy WUR/fuel pressure mystery my car is running very well.  However, it's now cold enough here, 35-40°F,  that all the cold start mechanisms should be in play.  The car starts immediately but idle is very rough for a minute of so before it smooths out, almost like it's not firing on all 8.  I know that the WUR and fuel pressures are all ok.  At one point I cleaned the aux air valve on principle.  Any thoughts on what to look at?  In the spring it's my plan to replace injectors and maybe rebuild fuel distributor.  Thanks!

My first port of call would be the cold start valve, a.k.a. ninth injector.
It should inject extra fuel into the intake until the car warms up, which would normally manifest as higher idle speed that settles down to <800rpm over the course of about two minutes.
1993 "Pearl Blue" W124 280TE
1988 "Arctic White" W124 200T
1979 "Icon Gold" W116 450SE

Randys01

The fact it starts readily when cold indicates the cold start injector is working Ok. it's role in start up is like a mayfly..it's all over in a few seconds. But maybe yours is sticking open for too long causing rich and rough idle until it turns off.

Disable the injector wire connector at the unit and introduce a sep 12 v power supply to the injector. You should hear it click on and off as power is applied.
If it's clicking then it suggests the valve is OK..the problem lies in the sensor that determines the fuel turn on/ shut off.
With the injector wire still  disconnected try a cold start. It might be cranky to get it started but if it immediately settles down to  a good smooth fast cold idle then the injector sensor at the rear of the rh cyl head becomes the centre of your attention.

If the injector checks out as Ok, ..then as Daniel suggests,... it might then be prudent to double check your WUR..fuel pressure values.

TJ 450

After checking the above, I would also check that the airflow meter plate is moving smoothly and that the screw is adjusted. This is assuming there are no vacuum leaks.

When it's running roughly, see if you can get it to run smoothly by manually pressing or lifting the plate to see if it is running too lean or rich.

Does the engine hesitate or feel like it's going to cut out when you open the throttle during this condition?

Tim
1976 450SEL 6.9 1432
1969 300SEL 6.3 1394
2003 ML500

raueda1

Quote from: TJ 450 on 03 November 2018, 07:18 PM
After checking the above, I would also check that the airflow meter plate is moving smoothly and that the screw is adjusted. This is assuming there are no vacuum leaks.

When it's running roughly, see if you can get it to run smoothly by manually pressing or lifting the plate to see if it is running too lean or rich.

Does the engine hesitate or feel like it's going to cut out when you open the throttle during this condition?

Tim
doesn't hesitate at all, just lumpy for a minute or so, maybe less.  Thanks for everybody's comments, but I'm putting this to bed for the time being.  I'm traveling for the next 5 weeks and it will be full winter when I get back, so will take it up in the spring.  Cheers,
-Dave
Now:  1976 6.9 Euro, 2015 GL550
Before that:  1966 230S, 1964 220SE coupe, 1977 Carrera 3.0

raueda1

So, after 5 years I'm back to this question - rough idle when cold. After just a few minutes it smoothes out and is silky.  Otherwise car runs great.  All CIS and WUR  pressures are spot on. I also checked cold pressure over a range of WUR temps, from 10C to 21C and it's perfect.    Mixture has been set so warm idle is at its best, which is very good indeed. Aux air valve was fine last time I checked.  Additionally, since OP the following has been done:

  • New plugs, wires, cap and rotor 2 weeks ago.  Old plugs were perfectly and evenly colored right (light beige)
  • Rebuilt fuel distributor
  • New injectors
  • Timing good
  • etc, etc, too much to mention

All that's left is cold start injector, and I haven't messed with that.  Can anybody elaborate on the test process?  Or other ideas?  Thanks and cheers,
-Dave
Now:  1976 6.9 Euro, 2015 GL550
Before that:  1966 230S, 1964 220SE coupe, 1977 Carrera 3.0

UTn_boy

Well, contrary to one of the comments above, the cold start injector does not operate until the engine warms up.  It ONLY operates between 2-12 seconds max and only when the key is in the start position.  The colder the engine is, the longer the cold start valve will spray, but again, no longer than 12 seconds.  Conversely, with a hot engine, it does not spray.  The thermo-time switch is what determines how long to operate the injector. You mention that the car starts up instantly when it's cold.  I'd venture to say that your cold start injector system is ok.  The most common failure is the valve/injector itself.  They'll either leak fuel externally, or the pintle valve will get stuck open and will constantly spray or dribble fuel into the intake.  Sometimes the wiring will flake out, and rarely will the thermo-time switch go bad. You'll usually know if the injector is leaking or stuck open.  It'll belch out black smoke from the exhaust, fuel economy will decrease, or there will be a slight miss that can't be rectified through conventional methods. 

As far as testing the system, there is little to do.  With a cold engine, make sure you're getting voltage at the injector when the key is in the start position.  In order to check the thermo-time switch, you'll have to measure the resistance of the switch at it's terminals and compare that reading to the temperature.  There is a table in the shop manual that shows the correlations between the temperature and the resistance of the switch itself.  And of course, check the wiring, and I believe there is also a relay for the system.  But again, reading that the car starts instantly when cold tells me the cold starting system is probably working.....but who knows.  Just my two cents worth. 
1966 250se coupe`,black/dark green leather
1970 600 midnight blue/parchment leather
1971 300sel 6.3,papyrus white/dark red leather
1975 450se, pine green metallic/green leather
1973 300sel 4.5,silver blue metallic/blue leather
1979 450sel 516 red/bamboo

raueda1

Quote from: UTn_boy on 29 October 2023, 05:20 PMWell, contrary to one of the comments above, the cold start injector does not operate until the engine warms up.  It ONLY operates between 2-12 seconds max and only when the key is in the start position.  The colder the engine is, the longer the cold start valve will spray, but again, no longer than 12 seconds.  Conversely, with a hot engine, it does not spray.  The thermo-time switch is what determines how long to operate the injector. You mention that the car starts up instantly when it's cold.  I'd venture to say that your cold start injector system is ok.  The most common failure is the valve/injector itself.  They'll either leak fuel externally, or the pintle valve will get stuck open and will constantly spray or dribble fuel into the intake.  Sometimes the wiring will flake out, and rarely will the thermo-time switch go bad. You'll usually know if the injector is leaking or stuck open.  It'll belch out black smoke from the exhaust, fuel economy will decrease, or there will be a slight miss that can't be rectified through conventional methods. 

As far as testing the system, there is little to do.  With a cold engine, make sure you're getting voltage at the injector when the key is in the start position.  In order to check the thermo-time switch, you'll have to measure the resistance of the switch at it's terminals and compare that reading to the temperature.  There is a table in the shop manual that shows the correlations between the temperature and the resistance of the switch itself.  And of course, check the wiring, and I believe there is also a relay for the system.  But again, reading that the car starts instantly when cold tells me the cold starting system is probably working.....but who knows.  Just my two cents worth. 
Thanks Aaron, very clarifying indeed.  I'm going to just disconnect it and see what happens.  But you confirmed my suspicions that it is not the cold injector.  So then what is it?  About all I can think of is aux air valve.  It seems like everything else is pretty much eliminated except some obscure vacuum problem.  Whatever it is, it can't be too bad.  After getting a bit warm (which takes about 5 minutes at most) it runs beautifully.  Cheers,
-Dave
Now:  1976 6.9 Euro, 2015 GL550
Before that:  1966 230S, 1964 220SE coupe, 1977 Carrera 3.0

ramiro

The aux valve is distributing the air to all cylinders evenly through the air meter so in theory it can only affect the idle speed.
Did you ever measure the air fuel ratio, that way you could be sure that the is not running slightly rich or lean during warmup ?
Actually my car does the same think but i think it was because of the cam timings and now it is because off the failed fuel filter the messed my fuel distributor up at part load and idle.

Randys01

Hard to tell being 5000 miles away but I'll wager it is an intermittrnt cylinder failing to fire. Without electrocuting yourself, start the engine from COLD and isolate each plug one in turn until you determine which one makes no difference. I'd say it is the same cylinder.
Once identified work thru the whole electrical  and fuel system  for that pot.

UTn_boy

I'm with Ramiro.  The idle air valve would only affect idle speed since all it does is meter air going into the intake at idle/when throttle plate is closed.  Is there any chance that it's running a tiny bit too rich when cold? A slightly too rich mixture can cause the problem you have.  I know you've checked pressures, etc., but sometimes you have to deviate from what the book says and tune the part to that particular engine. And definitely take Randy's good advice, too!
1966 250se coupe`,black/dark green leather
1970 600 midnight blue/parchment leather
1971 300sel 6.3,papyrus white/dark red leather
1975 450se, pine green metallic/green leather
1973 300sel 4.5,silver blue metallic/blue leather
1979 450sel 516 red/bamboo

raueda1

#12
Thank you all, some good meat to chew on here. 

I've played with the idle mixture screw (not the big white one).  It's set to deliver best idle when hot.  I never played with it when engine was still cold. The engine bogs when the air plate is depressed, suggesting that it's not too rich.  Maybe it's worth setting it towards the lean end when hot and seeing what that does for cold idle.  I've got the Gunnison CO tester but not found it all that useful. 

I'll also revisit the air plate gap.  I set it at some point but maybe it's wrong.  If so that might be a cause, no?
[edit:  Someplace I saw how to set the plate when engine is at rest or idle.  Can't recall.  Anyway, I can't find it now.  Anybody seen this?]

As for ignition, the lumpiness actually does feel like it could be intermittent missing.  Stupidly perhaps, my own ignorance and inexperience biases me against thinking that the ignition system would be the problem.  The issue is  under at low speed and cold conditions (vs high revs and everything hot).  But I guess anything's possible.  A few years ago Mike Morris told me to look at ignition first before anything else, almost regardless of the problem.  Anyway, by a happy coincidence I recently got an ignition oscilloscope probe to fool around with.  Surprisingly cheap! My thought is to start with the lead from coil to dist cap.  If that's OK then move on through the respective cylinders.  Seems like this would confirm/deny ignition problems for once and for all.

I'll report back with findings.  Thanks again and cheers,
-Dave
Now:  1976 6.9 Euro, 2015 GL550
Before that:  1966 230S, 1964 220SE coupe, 1977 Carrera 3.0

rumb

The engine bogs when the air plate is depressed, suggesting that it's not too rich.


My experience tuning SU carbs on MG suggests the opposite. That procedure you slightly lift to carb piston and if bogs means rich, rpms up means lean and slight up and then back just right.

When you press plate you are increasing fuel <and air>.
'68 250S
'77 6.9 Euro
'91 300SE,
'98 SL500
'14 CLS550,
'16 AMG GTS
'21 E450 Cabrio

daantjie

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.
Daniel
1977 450 SEL 6.9 - Astralsilber