Could This Be the Root Cause of Rough Idle?

EricSilver · #1 10-15-2003, 01:21 PM

1989 260E
189,000 miles

For some time, I have been plagued by rough idle problems. Electricals (distributor, rotor, plugs, plug wire, ignition coil) are all new or near new. Fuel/Air mixture is correct. Fuel delivery – pumps, accumulator, lines, filters, distributor, injectors, injector seals) are all OK. Yet the problem persists until, hopefully, now.

There has been one common denominator in this mystery: the oil pump and its connected parts.

When the engine stumbles at idle or in any other low RPM state, it is visually indicated by the bouncing oil pressure needle. I have always assumed it was simply a visual symptom of a fuel or air starved engine.

Now, however, I am beginning to suspect that the oil pump or a component of the oil transport system (I believe there is also something called a “Sender Unit”) is actually causing the rough idle.

This morning after the first start, the realization began to hit me as I thought, “I am idling at about 550 – 600 RPM, yet the engine is smooth. Why? Normally it should be stumbling. What is different?” The only thing different was the high oil pressure indication.

We all know that applying a direct or indirect mechanical load to the engine, via the transmission, PS pump (on lock to lock turns) A/C compressor, or alternator measurably impacts engine RPM. It is therefore reasonable to assume that a faulty component of the oil pump/transport system -- or the pump itself -- which is intermittently performing normally then not-normally, is perhaps applying back-pressure to the oil pump, causing an intermittent engine load, and thus the erratic idle.

Because the problem only occurs at low RPM And lower oil pressure, and does not occur at high RPM/High pressure nor at low RPM, high oil pressure, I am tending to believe this is the root cause. (At high rpm, any back-pressure caused by a dirty or faulty component is overcome by the pressure out of the oil pump. At low RPM – assuming warm.thin oil, that is not apparently the case. But low RPM with cold/thick oil masks the problem.)

What do you all think? I don’t know enough about oil systems, but this seems to make sense to me. I am due to get an oil change and new filter in a few days as part of normal maintenance, what else can I check? And what is this oil sender unit? Am I on the right track??

Peyton300TD · #2 10-15-2003, 01:44 PM

your car sounds VERY nice. id love to see a picture of it.

you wouldn't by any chance be 'the eric silver' of advertising would you?

if not , just so you know there is a very noteworthy, highly respected guy in advertising by the name of Eric Silver.

Q · #3 10-15-2003, 01:45 PM

I would say that the inverse is true. The fluctuating oil pressure is a result of the rough idle.

Duke2.6 · #4 10-15-2003, 02:17 PM

Agree. Below the 3 bar level and the point where the pressure relief valve opens, any variation in engine speed is going to cause a variation on oil pressure.

Lack of idle smoothness is somewhat inherent to continuous flow fuel injection systems, and any variation in injector flow rate is going to make a cylinder run a little lean or a little rich, which is also going to affect emissions.

If everything else checks out - ignition, idle stabilization system, and basic mixture setting as indicated by the duty cycle, then the injectors are suspect. They can be cleaned and flow tested and should not have no more than a maximum of 3 percent variance from the highest to lowest.

They can also be replaced, but if there is evidence of loose dirt or other particles in the injectors, the entire fuel system should be flushed to attempt to remove any foreign matter that could wear new injectors and upset their flow calibration.

If the injectors including the shrouds have deposit buildup an Italian tuneup with a big dose of Techron in the fuel might help.

Duke

EricSilver · #5 10-15-2003, 02:39 PM

Peyton,

Nope, that is not me, but I'll keep in in mind; who knows what doors it may open.

I should go ahead and take some good photos of the car, but probably not until after it is properly polished and shined!

EricSilver · #6 10-15-2003, 02:40 PM

Q & Duke, thanks.

So far this is sounding simple. I did the injector seals, so the injectors themselves can be readily done as well.

How, exactly, does one clean an injector?

Duke2.6 · #7 10-15-2003, 03:24 PM

There is specialized equipment that is used to clean injectors, most of which are the solenoid types for EFI systems. The machines have various adapters to mount different injector designs, and I'm sure some can handle the KE type mechanical injectors. Enough system pressure is required to open the check valve.

Combined with an ultrasonic wave generatior, a strong cleaning solvent is run through the injectors and the flow rate is also measured, so once they are clean you can determine if the flow rates are in spec and also visually inspect the spray pattern. Any that are out of spec should be replaced.

Any deposit buildup in the injector or the check valve seat will affect flow rate, and any deposit buildup on the injector skirt can affect spray pattern. Either of these can cause idle stability problems.

Some shops may have their own cleaners and some sent them to outside vendors. You'll have to check around locally to find a shop or vendor that has a machine that can handle the KE injectors. In the meantime, the following link to GB Remanufacturing has some information on injector cleaners. They sell them to service shops, but I'm not sure if their product line will handle the KE injectors.

http://66.70.39.114/

Duke

Kestas · #8 10-15-2003, 04:32 PM

If you search the internet, you'll find a number of outfits where you can mail the injectors for cleaning. These places check the injectors coming in, ultrasonically clean, reverse flush (gets rid of solid contaminants), then check them for flow and balance among the group. All for about $25 each injector. Others can even blueprint injectors.

EricSilver · #9 10-15-2003, 06:08 PM

At $25/injector to clean, it makes sense to instead buy new ones (at about $23/each).

The pattern of the stumble, because it has been going on so long, would seem to support your link to the injectors because it seems the same cylinder is misfiring. The feel is consistent every time.

I thought about using more Techron, or that BG44K (?), but it am doubtful of how effective it would be. It would also be great if I could figure out which injector was the really bad one.

So, another question:

Why does the injector, if that is the culprit, behave normally
on cold starts? What would make the fuel flow consistent at
start, and then inconsistent after the car has been running
(assuming also the 5 year-old O2 sensor is working properly)?

Duke2.6 · #10 10-15-2003, 07:40 PM

During cold start and warm-up the system operates in open loop mode and the mixture is on the rich side. Once the system goes into closed loop the idle mixture is maintained at stoichiometric, so if one or more injectors have a flow or spray pattern problem, the resulting leanness causes combustion instablity or intermittent misfires that cause idle roughness.

I notice on my 190E 2.6 that the idle is butter smooth - as should be the case for an inline six - until the temperature reaches about 60C. Then it picks up a slight roughness/uneveness until the next cold start. Sometimes the system will also "hunt" - slight cyclic variation in idle revs - until it finds stability. As I said before, a slight idle roughness seems to be in the genes of the KE system, and mine isn't bad enough to tear into the system to try and fix it, but I try to give it Italian tuneups as often as possible.

Pre emission cars with carburetors idled at an average A/F ratio of about 12.5:1 to maintain best idle quality. With port injection and well matched injectors and air flow, a leaner idle will usually be satisfactory, but running at stoichiometric - about 14.7:1 - will cause roughness unless the injectors are matched within a couple of percent and a long duration high energy inductive spark is required to prevent misfires.

Also, high overlap camshafts cause more exhaust gas residual that can increase roughness due to combustion instability without a very rich mixture. These 103 engines only pull about 15" manifold vacuum at 700, which indicates more than average overlap for an emission controlled engine, but fairly aggressive valve timing is required on a two valve engine to make the high specific output that is characteristic of this engine family.

Back in the fifties and sixties the nervous, lumpy idle of a special high performance Corvette engine was highly sought after and a mark of performance. The '64 - '65 versions of these engines with mechanical lifter camshafts only pulled 10" at 900-100 RPM idle speed, and the carbureted versions generally had better idle quality than those with the Rochester continuous flow mechanical fuel injection system that is very similar to the KE system in operating principle.

Duke

haasman · #11 10-16-2003, 12:15 PM

I don't know if this would apply or even make any difference, but when we first checked out the 190e for purchase, it had very similar symptoms to Eric's car. Now this car, I should also point out, needed a lot of work: water pump, cap, rotors etc but an MBZ tech recommended bleeding the injectors at the fuel distributor.

Frankly, I was dubious, but I gave it a try. With the engine running, multiple rags to catch the fuel close by, I systematically bled each line at the FD. I was amazed at the amount of crap that came out and how the engine (2.3) settled down and ran much more smoothly. I followed this up with two bottles of Techron Fuel Injector cleaner on less than ½ tank. The car now runs great! I continue to be very pleased with the results. The car idles well hot or cold and accelerates well.

My 2 cents,

Haasman

ED C. · #12 10-16-2003, 02:28 PM

Hassman,

Can you go into a little more detail on the procedure for bleeding the injectors? Did you just disconnect the fuel line at each injector?

Thanks,

Ed C.

haasman · #13 10-16-2003, 03:07 PM

It is actually very simple, with one big caution: You are dealing with raw fuel, so be sure and take all appropriate safety precautions-

Remove the air cleaner to gain access to the fuel distributor.

Start engine and crack open the fuel injection line, one at a time going to the fuel injector, AT the fuel distributor, not at the injector.

The engine idle will fall, when closed and no longer "bleeding" the idle will rise again.

Do this several times for each line at the fuel distributor.

Have rags around, below etc to catch the raw fuel. It doesn't take a lot of bleeding, just a purging if you will. Often more or less the amount of a thimble or so.

You should see some nasty gunk coming out from just cracking the line and letting bleed, if there is any accumulation there.

On the 190e it kind of looked like a little bit of liquid mud. The engine instantly responded after doing this.

Hope this helps,

Haasman

PS - Have you checked your idle control valve and especially the rubber connections to it? These often get brittle and unpliable allowing air in.

The idle control valve often gets very gunked up and needs to be cleaned out with carb cleaner and/or replaced.

EricSilver · #14 10-16-2003, 05:23 PM

Hi Haasman,

I might try that.

When I replaced the injector seals, there was leakage from the distributor when I removed teh lines, but all looked pretty clear.

I hear a lot about the idle control valve, but I have no idea where it is or what it looks like. Since all the other rubber I have seen is rock hard, it seems reasonable that the ICV would be too. (My old injectors were leaking because they were brittle too.)

#15 10-16-2003, 07:51 PM

380sl races at drive and neutral to 1500 rpm initial idle is about 700rpm at cold start runs smooth at 700 but is rough if shift at stop to neutral. could this be the idle control valve a tech looked at it recently and said it was probably rthat since everything else is almost new.
directions on removing ang cleaning idle control valve iknow on the 380sl they are about $250 can they be cleaned?
can parts be replaced in them?
anything else cause racing idle at neutral and park?

Thread Tools
Show Printable Version Email this Page
Display Modes
Linear Mode Switch to Hybrid Mode Switch to Threaded Mode