Actually, the airflow sensor is the device that makes the major adjustments to fuel flow in that as the plate is deflected it displaces the control plunger in the center of the fuel distributor, increasing or decreasing the orifice opening to the injectors. If it is out of position, it places the plunger in the wrong position, thereby providing an incorrect amount of fuel to the injectors. This is especially a problem during start procedures since some major aspects of fuel control, i.e. lambda, are not part of the start cycle, only the cold start items have much input.

Dirt, air leaks, oil, and corrosion are common problems with this beast, as are improperly adjusted stop pins and uncentered sensor plates. As with any system sensitive to small adjustments, patience and diligence are essential.

M.B. DOC:

There is always a wait for appointments, so mine is scheduled for 10/3. But, the dealer mechanic who did a quick check on the start and idle problems thinks that my start problem is caused by a vacuum leak. That's a new opinion, yet I also remember the previous owner of the car had complained about a vacuum leak on this car -- though he said it was taken care of by his dealer service. Is this something difficult to locate and fix? Could they be right about a vacuum leak? How does a car develope a vacuum leak? How does one prevent it in the future? Thanks.

YF Yin

Vacuum leaks are reasonably easy to identify, and often easy to fix. Common leaks include injector seals, hoses, and the idle control circuit. Often found with a squirt bottle of water with the engine running. If your injector seals have never been replaced, they're good candidates for leaks as they age harden and crack.

If nothing is known about a particular vehicle, there is a certain sequence to evaluating problems; a logic diagram, if you wish. Once vehicles become better understood, the logic diagram has a different starting point: this is what you pay for with expert help. If your tech knows that 70% of MB 190 16V engines with hard starting problems have vacuum leaks, that's where the diagnosis will start. If that is not the cause of your problems, then a tech willl likely go to the next most common problem before undertaking a rigorous, expensive diagnostic series.

This opens a potentially heated discussion about the proper way to diagnose and treat vehicles, but I always preferred to establish the most likely issue first ( air, fuel, mechanical, electrical) and go from there. This is not intended as a superficial "fix the part and get it out" approach. If there is a blown fuse, what caused the failure? corrosion or circuit overload? short to ground? bare wire? bad motor? A good tech soon learns, especially when concentrating on one brand, which problems usually cause which symptoms, but good techs are also willing to look efficiently for other issues when the common solution is inadequate.

So, if it's a vacuum leak, great. If not, we've already given you several possibilities that may be the actual problem. If the vehicle has not been well maintained, you may have a plethora of problems amid your dearth of diagnosis.

Richard:

Thank you. Your comments and suggestions are sensible and reasonable. I, too, prefer to solve the problem at its root whenever possible instead of a quick fix. A good mechanic, as you described it, is also a good engineer. They are hard to find. Unfortunately, in my area, it also means that sending a car for dealer service does not guarantee that a good tech will be working on the car.

Luckily, the previous owner of my car loved it and babied it. I really like my car and have taken very good care of it. The car is very clean and I don't let any problem linger.

After reading you message, I feel better now about getting the vacuum leak checked out first.

Cheers,
YF Yin
'87 190E, 2.3-16

I suggest you move to Florida and let Steve work on your vehicle, or maybe Georgia, and let the DOC do it. Why not ask the forum, under a new topic, for recommendations in your area, even specific people within a given dealership? You may have noted the credits some people here place after their names. Those indications of mastery are often good places to start, but at a dealership, some of the names listed may be no longer involved in actual daily repairs. Be willing to ask, and if you don't like the answer, reevaluate your decision to be there. If you're going to pay the rates, you'd better receive the service!

P.S. Mastery is the start of a journey, not the end. One who attains mastery and stops seeking is no longer a master.

Gentlemen,
I have a similar problem with my 16V, 1986, 155k miles.
Starting is OK, but when cold outside (almost winter here in Colorado) the idle speed is very low, 500-700 so the engine stalls easily, especially when engaging in Drive (which will reduce the already low idle.) In the summer, the idle was fine, around 1100.
All fuel related parts are new as is the O2 sensor.
I am suspecting, as RCMKTG and MBDoc suggested, that the Air Flow sensor potentiometer is bad.

Question: Anyone knows a source for a new one? I have checked around and the best price was $420 (ouch.)

Regarding the AirFlow sensor, I did check check the potentiometer (while disconnected from harness) and slowly moved the plunger up/down. I found the readings to be erratic, espacially in certain positions. My mechanic (MB independent whom I trust) told me that his test showed that the sensor is OK.
While I respect his expertise and based on my own measurement,
I tend to think that the part is worn out (and why not, after 155k miles).

Any thoughts greatly appreciated. Pat, if you are reading, you mentioned having a similar problem. Maybe we can get a few people together for a group buy?


THanks All, this is a wonderful forum and resource!!!

Tony Ander
aander@newsignal.com

Tony,

I don`t think you can tell much by unhooking it (air flow meter) and checking it. Best to remove one side of the connector and plug it back in. This allows you to monitor the voltage between pins one and two with either the key on or while driving. You are looking for open circuits. It should be in the .8 to almost 5v range depending on throttle position.

Sounds like you don`t have confidence in your mechanic it he stated it was ok?

I am assuming you have investigated all the other components, ie decel switch, throttle position sensor, auxillary air vavle, valve adjustment, ect....?????

Tinker

The potentiometer senses change in airflow, and modifies the signal sent to the EHA during post start and warm up phases of the running engine. First thing to check on my list is the signal at the eha ( electro-hydraulic aAcurator) the controlled leak that modifies the control side of the fuel distributor, and thusly the amount of fuel available to the injectors. If it floats and varies according to the normal startup sequence, and brief throttle blipping causes an approx 10ma variation, the potentiometer is not the likely cause.

The idle speeds you spoke of are the inverse of normal, making at least two other items suspect: the thermal sensor and air leaks in the intake system or injector seals. Check those first.

This fuel system will not tolerate random checks and scattered approaches to diagnosis. You and your service technician must be willing to follow the proper sequence, from battery voltage available to the computer through every logic step necessary to the particular event you are pursuing. Only if those tests fail to bring a solution should other areas be examined. Of course, if there is an obvious problem, like a broken hose or wire, fix that first.

Otherwise, the bossa nova is just a dance.

Richard

Tony,

I am convinced that no independent mechanic in my area (CT) knows the 16V's enough to service the idle and start/stall problems. Don't change your air flow potentiometer unless you and everyone is convinced it is causing your current problems. I made the mistake of believing my independent mechanic and paid for a new air flow sensor potentiometer assembly and I am back to square one. My start/stall and idle problems persist. It was a costly, and ultimately useless, repair -- not to mention wasted time and money.

Since my last posting, the "vacuum leak" problem has been taken care of (hoses and clamps) as suggested by MBDoc and Richard through the forum. My own dealer said they don't have a mechanic who's trained on this 16V engine so they won't touch the start/stall and idle problem. Another independent mechanic (Porche specialist) suggested the onboard computer has to be replaced (that's over $1K); so for now, I am just living with the nuisance of start/stall and high idle (1400rpm).

Y.F. Yin
1987 190E 2.3-16

Y. F.,

Sounds like were all in the same boat regarding the high idle and stalling with our beloved (ha) 16v`s.

If you go back over some of my prior posts, you will see my unending journey. I also had the mystery vacuum leak that I couldn`t find. It really caused a problem after I cleaned the throttle body with carb. cleaner. The idle shot up to 1500 and would not come down! I ended up removing the whole injection manifold and resealing it with new gaskets and hoses. The result......no big difference! AURGHHH!!!!

But...... I finally got my idle down to a normal level after playing with it for a few WEEKS. But now I have a stalling issue

Here is what I want you to do. When it starts to idle high, check the voltage or duty cycle at the X11 plug #3 pin on the drivers side inner fender. Tell me if you see a fixed or fluctuating voltage.

I think this might steer both of us in the right direction. I want to see if you readings are the same as mine.

Tinker

I was having a simular problem with my 300TE. It would die as if the fuel just shut off. In my particular case, it was the idle speed control valve located just under the air cleaner housing. There is a valve inside that controls air mixture (I believe). The valve rotates with a pivot point located on the outside of the housing. I tested it by taking it off and applying current where the plug connects to. The valve should open when current is applied and close immediately when current is removed. Mine didn't close right a way when the current was removed and I noticed the pivot point had some rust on it. I put some WD40 on the pivot point and let it set overnight and the next day activated the valve with current several times and the valve inside reacted as it should. I put the idle control valve back in the car and it hasn't stalled since and idles beautifully. It's been 3 months. Hope this helps!

Hi All,
Regarding my cold idle problem, I have watched it some more, but had not time to innvestigate further. What I found is:
1. If I crank and am ready to press the pedal right after the
engine starts (which is 1-2secs) as well as I keep cranking for another second or two, then I can keep the engine running.
After just a few seconds, if I release the accel. pedal the idle speed seems to be OK!

2. If now I put it in drive I may stall it, but wainting maybe a minute will allow the engine to sufficiently warm up (?) so that
the extra load when in Drive will not stall it (it will reduce the idle by a few hundred RPMs).

So I am thinking of to look for (1) a component that only affects the idle right after start for a short time and (2) the switch that
tells the ISC that drive is engaged -- does anyone know where this switch may be located?

Tony.
PS: Thanks for the suggestion for the air-flow meter. Indeed, I
did not check it in-circuit, I only measured what I thought would be the potentiometer wiper's output. My mechanic did check it
while the sensor was "live", so he must be right!

Guys, your idle problems may be idle problems, in which case do the cold start checks in the trouble shooting guides, such as EHA start mA and decay rates, Coolant sensor resistance at various temps, cold start valve leakage, system residual pressure, buildup time, and limits. Also, verify the operation of the idle speed air bypass valve, or indle speed control valve ( you know, pinch the hose off both cold and hot to see how it affects idle speed). If all those things are in limits, you probably don't have an idle speed or start-up problem per se, but another problem, which when compensated for, becomes an idle speed problem.

For instance, adjusting mixture while ignoring a vacuum leak or a malfunctioning injector, weak cylinder, bad plug, etc, is a sure way to ruin the systems's ability to compensate for the non-standard needs of start-up and warm up conditions. I've not seen anybody talk (in this string) about airflow sensor rest and zero positions, centering in the bore, and so on. Everything, and I mean everything, is at its worst ability to compensate under cold start conditions.

You need to have the system pressure gauges and the ability to use digital multimeters, as well as skills at analyzing spark plug coloration to get through this. I had to rig up a way to test my injectors by building a fuel distributor tied in to the vehicle fuel system but still allowing me to observe the individual patterns and volumes of fuel. Once I did that, and all other checks were satisfactory, I had to completely readjust the mixture settings at operating temperature idle, because before that time the lambda system was trying to balance three good cylinders against one bad.

How about cold compression vs hot? How about first compression stroke in each cylinder vs maximum value (5 strokes)? I've used the "1 vs several" with good success to find valvetrain problems.

Sorry for the soapbox oration, but you seem to be running in circles by looking for solutions to symptoms instead of finding and repairing root causes.

Hi All,

Richard is probaby right, but I am not mechanically inclined so I will need to rely on my mechanic to check everything and try to get to the root of the problems.

Until then, I have resorted to Tony's methods for circumventing the start/stall problem. On my car, usually it runs smoothly after I get it going and prevents it from stalling. My car stalls out with cold and hot engine.

I forgot to mention in my last message that my mechanic replaced the idle control valve already and it didn't help at all. After that he replaced the air flow potentiometer assembly. I now believe what Richard is saying: probably neither part needed replacing.

I now detect "vacuum" problem only in 5th gear at 3500 rpm when I try to punch the accelerator. There seems to be no problem above 4000 rpm.

Since I don't work on the car myself, I tend to find a way to live with the problems instead of taking it to the shop every time.

Y.F. Yin

Richard,

I have done everything you have mentioned in your post. Gone through the factory CD trouble shooting matrix a few times. Checked the compression and leakdown numbers, yes the first stroke is important and telling. Checked fuel pressures, upper and lower chamers, and residual pressure. Replaced anything questionable, including a few good parts. Adjusted the valves, resealed the whole intake system and reset the resting height of the air flow meter. Lamba is in spec and fluctuates at temps.

It starts on the first turn of the key when cold......and stalls or almost stalls when hot. Air flow meter is in spec with no open circuit noted while driving around the neighborhhod.

My neighbor, who owns his own shop and has worked on MB`s, BMW`s and Porsches for over 40 years is also stumped.

I am searching for some new ideas......

Tinker