I have a '91 300TE that idles around 1000 when you start it, then increases to around 14-1800 once warm.

Background (sorry it's long, but it's all inter-related):

I had a bad idler buffer and when I fixed this and put a new belt on it. About a month later, all the dash lights came on. It wasn't charging, only 12v at the battery. The belt was too loose. I tightened it, and it worked for about 2 days. But I guess it strained the alternater or something because voltage at the battery dropped to 12.5v. A new alternator tested fine when I first put it in (14v), but after a few days all my bulbs started burning out. I rechecked the voltage and it was 18v at the battery! Luckily while trying to diagnose the earlier problem I bought a spare regulator. I slapped this on and now the charging system works fine (and I was able to get my money back in the regulator instead of returning the alternator). However, I'm not sure how much damage was caused by driving around for a few days with 18v through the system.

What I've done:

I had a spare OVP relay I bought because it would die immediately after started the car when cold (except, strangely, sometimes when it was well below freezing out). This didn't help my cold start problem, so I kept it as a spare. Swapping it out doesn't have any effect. When I disconnect the idle air control valve, it seems to run fine. It idles normally. It doesn't start quite as easily, but other then that seems to run fine (I haven't taken it out of the driveway like this yet). If I plug it back in after starting, the idle jumps. After reading the engine diagnosis article on this site, I tried to test the dwell setting, but I couldn't get a reading. Not sure if I hooked it up wrong or if my meter is fried. Also, I was getting very little current draw (.13ma), that didn't seem to change.

So two questions: Does this sound like a bad idle air control valve, or a bad controler? Is there any harm in driving around with my idle air control valve disconnected?

Thanks,

Dave Sawyer
'91 300TE
'83 300D backup

Did you check the throttle valve switch and linkage microswitch?

Duke

No, I didn't check those. I'll try searching the board for a test procedure.

I know I need the manual CD and I've had one on order since Dec, hopefully that will make this stuff easier to diagnose

Dave

Dave:

Welcome to the board! Still making beer?

Sounds like the idle control valve is going wide open instead of closing if you get faster idle by plugging it in -- it should fail open unless really dirty.

You can clean it by spraying carb cleaner in it and shaking with the hose nipples covered, repeating until the fluid comes out clean.

Check the hoses on the idle valve too -- the ones on the TE were rock hard and leaking. Pure b,...h to get the one on the mainfold back on with the clamp, but the car runs much better.

Yes, 18V would probably fry the electronics, but if the OVP is working, all the sensitive stuff should have been switched out at high voltage -- that's what it's there for!

If you have not done so, replace ALL the rubber vac line connectors -- chances are they are all shot and leak like crazy. I did them all in the TE, cured the hard shift, throttle lag, and most of the rough idle.

To check the duty cycle, you want to put the red lead of you frequency meter (set to read duty cycle) in pin #3 of the diagnostic socket on the EZL computer (driver's side fender) -- socket has a round screw-on cover. Black lead in #2 or ground, or free like for my cheapo Sears meter that won't read at all if the black lead is grounded. With the O2 sensor hot, duty cycle should be around 50% and changing. If it's always 50%, the O2 sensor isn't working, if it's high or low and changing, mixture is out, if not 50% but constant, some electronic problem. Make sure you're not set to frequency, as that is constant. You can cheat with a digital VOM and read voltage instead, but it's not as accurate -- you'll have to check with another car to see what the meter actually reads at 50% duty cycle -- should logically be 6V, but I don't think it actually works that way.

If you have a direct reading mA meter (not an indirect one, you must complete the circuit) you can check the current to the idle control valve. If the current is high and the speed is high, either the speed signal is bad or the control circuit is toast. If should be very low above idle, then swing up and then steady when the idle speed is being controlled by the valve.

Current to the EHA (back side of the fuel distributor) should be near zero at 50% duty cycle. 20mA key on, engine off, should go positive when the throttle is opened, then rapidly return to aroun 50%.

Fix all the vac leaks before attempting to set the mixture.

You probably need a new belt tensioner, too -- the "spring" is the rubber bushing in the tensioner (like the rear suspension links) and they usually go bad in place. Work OK until you loosen them, then refuse to properly tension the belt when you crank the screw down.

Peter

__________________
1972 220D ?? miles
1988 300E 200,012
1987 300D Turbo killed 9/25/07, 275,000 miles
1985 Volvo 740 GLE Turobodiesel 218,000
1972 280 SE 4.5 165, 000 - It runs!

Peter, good to hear from you! Have no fear, the beer brewing is going strong.

Unfortunately, the 300TE isn't. I was too busy to play with it today. What doesn't make sense is that unplugging the idle control valve, idle drops to 500-600 rpm. Plugging it in raises the idle rpm to 1300-1800. So it's not failing wide open, which would indicate it's stuck. Instead, it's opening wide open when plugged in and closing upon being unplugged.

I haven't had a chance to replace all the rubber yet, but will. When I sprayed WD-40 around the rubber, I didn't detect any engine speed increase (of course, that was before I disconnected the idle valve, I don't think the hoses went on as tight as they came off).

It's running fine without the idle control valve. It idles at 500-600 rpm. Is there anything wrong with running without the idle valve? I will keep trying to track it down, but I want to know if it's better to run around at 500-600 with the valve unplugged, or at 1300-1800 rpm with the valve plugged in until I get it figured out.

Dave

I also forgot to mention, I did replace the tensioner while I had everything apart.

Dave

Dave:

Current should hold it open. Probably need to hook up the ammeter and see what kind of current you have. I think I remember that 600 mA is wide open.

Could just be a bad idle control valve, but they are expensive enough that I wouldn't buy one just to check.

Try tapping the OVP, too -- if it's bad, that may correct the idle temporarily, it can switch the idle circuit funny if the relay is bad and only making partial contact.

The only problem you might have is stalling if the AC compressor comes on at idle.

I just did the water pump on the TE this fall -- took me 6 hours and with new visco clutch and tensioner, ran about $600 in parts.

I've got three 6-gallon batches of beer going, will probably do another one next weekend since I can only make lagers this time of year.

Peter

__________________
1972 220D ?? miles
1988 300E 200,012
1987 300D Turbo killed 9/25/07, 275,000 miles
1985 Volvo 740 GLE Turobodiesel 218,000
1972 280 SE 4.5 165, 000 - It runs!

Let me know if you need somebody to sample it.
I don't charge anything for it.

__________________
2007 C 230 Sport.

I ordered some new rubber hoses going to/from the idle valve. I finally found a test for the OVC relay, I don't think that's the problem, but it would be an easy fix if it was. This test (from the BenzBin web site) claims that if you have voltage at the 87 pins, then the relay is ok. Since these are the pins attached to the fuses, it's easy to test with the fuse in place. My relay tested ok. One thing I may have left out before, the voltage at the idle control valve is only 2.5 volts, both with ignition on and not running and running.

Thanks for the great advice so far, I'm sure I'll get it tracked down eventually. It really is running fine without the valve plugged in, so it's good to be able to take my time while diagnosing this.

One more possibly tied in problem, I'm not sure if my 4matic is working. Neither the warning light or the engagement light ever come on. We've had a little bit of snow up here, and I have yet to see the light come on or been in a driving situation where I could tell the front wheels were engaged. I've purposely caused the car to start sliding in a turn, and it won't come on. My tires were almost bald, but I don't think that should have had any effect on the 4 matic system (except making it engage more). I now have new tires, but haven't had any snow yet to test them in.

Dave

Ruminations:

Note that the lambda sensor is out of the loop until the engine warms up. If it then is indicating rich at idle, the ECU will attempt to correct this by adjusting the EHA. If still out of range, I believe it will attempt to add more air via the idle control valve. Note, though, that if you get a hunting behavior in the idle speed, you may have a bad airflow meter pot. This would indicate falsely a need for enrichment (bad spot on pot causing a sudden increase in output voltage), followed by a response from the EHA, and then an increase in idle speed to compensate for the rich mixture that results. Then the pot is no longer in the 'bad' spot, and the idle settles back down.

Steve

__________________
'91 MB 190E 2.3
'08 RAV4 Ltd 3.5
'83 Lazy Daze m'home 5.7

Dave:

I talked to Hans yesterday, and he suspects the OVP relay. It has several circuits, so some things can work and others not (like the 4-Matic).

If yours has a single fuse, it's old -- the new style has two fuses on top.

I'd replace it, they aren't too expensive.

Peter

__________________
1972 220D ?? miles
1988 300E 200,012
1987 300D Turbo killed 9/25/07, 275,000 miles
1985 Volvo 740 GLE Turobodiesel 218,000
1972 280 SE 4.5 165, 000 - It runs!

My new hoses into and out of the idle control valve arrived tonight, so I'll throw those on and do some more diagnostic work tomorrow. The symptoms seem like it's the OVP relay, I just want to make sure before buying another one. I rigged up my ammeter to measure the idle control valve or EHA valve while driving around, so I'll try it out tomorrow (so I don't wake the rug rats who are already in bed). With ignition on, I'm definately not getting almost no current at the idle control valve (.14 ma), which seems to point to a OVP relay; but I'll verify that and also check the EHA tomorrow.

Dave

Ok, I rigged up a harness last night and drove around today testing the current to the idle control valve and EHA. I think the EHA is working fine. The current was fluctuating between 1 and 20 ma. On the idle control valve, I only got minor leakage current (0.14 ma), and it never fluctuates from turning on the ignition through driving around town.
If I didn't have current at the EHA, I'd think it's the OVP relay. However, since I have current at the EHA, but not at the idle control valve, what does that sound like? Is there another relay or fuse somewhere that could be blown?

Dave

Two additional clues, the EHA valve has a slight leak in it. I noticed it looked wet, and wiped it dry. I drove around town, and it's wet again. Also, there is a switch triggered when the linkage returns to idle. I noticed that this wasn't being triggered by returned to idle. I cleaned the linkage, but this didn't help. I then added two more turns to the linkage cross piece, and now it's clinking when the linkage returns to idle.
This didn't change the current to the idle control valve with the ignition on and the engine not running (can't test fire this late at night, will trial it tomorrow), so I still think there's still something wrong.

Dave

Dave:

If the EHA is leaking, it's toast. Get a new one -- may cure the idle valve problem too. It's leaking out the weep hole from an internal leak.

EHA current should be 20mA on starting, drop to near zero and flucuating when the O2 sensor warms up, the stay near zero except when the throttle is opened or closed -- should then go up temporarily when throttle is opened ("accelerator pump"), and drop to -10 or -20 on closed throttle coast for CO reduction under high vac closed throttle (fuel cutoff).

Peter

__________________
1972 220D ?? miles
1988 300E 200,012
1987 300D Turbo killed 9/25/07, 275,000 miles
1985 Volvo 740 GLE Turobodiesel 218,000
1972 280 SE 4.5 165, 000 - It runs!