I've been trying to troubleshoot the annoying idle lope in the '80 300TD 617NA. Valves adjusted... all the easy stuff done.
Fearing I might have a weak cylinder I put a compression test off 'til yesterday. The numbers revealed decent compression which tells me to look elsewhere. Here are the numbers:

1 345lbs
2 340
3 348
4 355
5 340

I thought I'd have a look at the injectors.
Dismantling the "115 bar" injectors to clean and inspect the nozzles I found teflon tape. Someone has been here before...
Nozzles all looked good, shims measured:

1 .050"
2 .048"
3 .047"
4 .045"
5 .041"

Mesured twice as I questioned the .001" diff in thickness which seems like an odd standard. Maybe "Mr. Teflon" honed them. Where's a pop tester when you need one. Parts to DIY one are ordered.

All cleaned and re-assembled, without teflon, the lope is still present but not as bad as it was. I'm hoping by balancing the injector pop pressures the idle will be smoother. It's hard to tell which cylinder is the culprit when cracking the lines open. Each has about the same RPM drop.

Out of interest I looked at the millivolt output of each glowplug and here's what I found:

1 9.1mv
2 6.8
3 6.9
4 7.8
5 9.6

Clearly #1 and #5 holes are either running a little hotter or the GPs are not consistant. These measurements were taken with the engine at operating temp, under a mild load (AC compressor).
After shutting it off and sit for a few minutes to allow the head temp to balance out I measured around 4.8-5mv from each GP telling me they appear to be faily consistant as thermacouples.

It will be interesting to compare the GP readings to the injector test.

After all this the IP vacuum valve leaks, wont shut off. I thought I knocked a line off but nope... mity vac says the valve is a leaker. Great, a shaker I can't stop without opening the hood. I just need a few backfire noises and smoke to really draw attention.

edit: shut off problem fixed. Someone must have leaned on a vac line cracking the rubber piece while takin' measurements.

There is alot of discussion of element adjusting using the GP to test delivery of each valve. Since you have it all apart why not rotate the GP around for sng and see if you get the same readings. Also how many readings did you take.

Search for the millivolt method.

__________________
http://superturbodiesel.com/images/sig.04.10.jpg
1995 E420 Schwarz
1995 E300 Weiss
#1987 300D Sturmmachine
#1991 300D Nearly Perfect
#1994 E320 Cabriolet
#1995 E320 Touring
#1985 300D Sedan
OBK #42

This is interesting -- thanks for bringing up the subject. I had not previously read anything about this [I must have slept through that class ]. Since the glow plugs are not true thermocouples (no dissimilar metals to "couple"), I would want to see a calibration curve for several new glow plugs made in an identical environment. If repeatability could be assured, it certainly would be a way of comparing performance between cylinders. I'll have to try it myself.

Jeremy

__________________

"Buster" in the '95

Our all-Diesel family
1996 E300D (W210) . .338,000 miles Wife's car
2005 E320 CDI . . 113,000 miles My car
Santa Rosa population 176,762 (2022)
Total. . . . . . . . . . . . 627,762
"Oh lord won't you buy me a Mercedes Benz."
-- Janis Joplin, October 1, 1970

Since they poke into the combustion chamber itself, there's probably a big difference between the temp the GPs see while the engine is running vs. the temp they cool to after the engine has been off for a while. If there's as much temp difference as I think there is, you probably can't reliably say much with only one calibration point at a temp so far away from your desired measurement temp.

You could do the whole calibration thing.. or just swap the highs and lows around and take new measurements

__________________
'83 300DTurbo http://badges.fuelly.com/images/smallsig-us/318559.png

Broadband: more lies faster.

Since somebody has been messing with the injectors before its very possible the glowplugs have been changed at least once as well.

The different mv readings could be due to having different brands and/or age of glowplugs in your engine. You should take them out, measure their cold resistance, ream the holes and preferably use all new glowplugs before you put any trust in the mv method or make any adjustments based on their readings.

The cold resistance of the glow plug has nothing to do with its function as a thermacouple. Besides, very low resistances can be difficult to measure.

Also, as the head temperature equalized all five glow plug thermacouples were well within 1mv of each other. On the other end of there range I tested each one after a fixed amout of "on" time. Applied a fixed amout of power (IxE, not just +12volts) using a voltage and current regulated power supply to each plug for a specific time period then measured its output. In each case they were around 28mv hot. Consistant at the other end of the temp range. So, I think the GPs are pretty well matched in this case.

Next I may whip out the Tektronix O-scope and see if I can see any piezo effect on the compression stroke. Maybe when it cools off a bit.

Ah... she may have a lopey idle but, she has ice cold R12 AC.

I would swap #s 4 & 5 glow plugs and see if the milli volt spread remains. Temporarily swap the injectors on 5 & 6 next if it does and check the milli volts. Then see if the spread remains. If nothing else it gives you a solid position for the injector testing at a shop later. Or as you mentioned do the injector shop test first. Let us know how things go.

Swapping injectors into the same cylinder one at a time is also a very sensitive dynamic check of each one. That is one specific area that might even be superior to interpretation of a pop test. At basically almost zero frequency an injector might seem okay. Yet dynamically at 1000 pulses per minute be a far different story.

If nothing changes at all the injector pump most likely is at fault. Your milli volt spread is too great in my opinion. Thats if interchanging components verifies it. My fear has always been the lack of absolute verification of an indication by an individual.

Also I feel a person should make sure the valves clearances are reasonably set. If they were done or checked in the last 10k miles or so they probably are acceptable for a general test. If an owner has no knowledge at all when they were last done is another story.

You also brought up an interesting ideal on establishing consistancy of the glow plugs. You could get an answer to your methology by just swapping the last two glow plugs in your engine. Or at least an indication that you are on the right track perhaps with your testing the plugs for consistancy. As a thermocouple they are far from linear I believe. Yet nobody has experimented with one to graph their charactaristics. . Thats just to the best of my limited knowledge.

Since the loping idle has been reduced somewhat by what you have done already that is a good sign you are perhaps on the right track. This might be a side effect of unequal burning cylinders I had never thought of.

It sounds like you are going to contribute some newer valid information to the milli volt method before you are through. My opinion has never changed. At some point for certain problems it has to become the only possible tool to prevent guessing and spending money on speculation. Or even the only tool that can identify a certain problem.

Do you have a feel for the milage your car has been getting? Substandard, average or just unknown? This area has interested me for a long time. Just too much of an apparent spread between examples. I know a lot has to do with the driver. So it is hard to make it a science.

For the cars that exhibit low milage I would consider heavy milli volt variences an injection pump individual element timing problem. Cars getting reasonable milage an individual pump element volume problem. Thats if the initial platform was as good as yours is with pretty equal compression and the other items mentioned all verified. This area needs some exploration as well.

can't do it.
The most a millivolt reading will tell you is that the engine is running.
You would gain more data monitoring each cylinder exhaust port temperature.


Note:
Glow plugs are not thermocouples.
Tip temperature of the glow plug is measured using a K-type thermocouple


Basic uncontrolled variables:
* Electrode stress/torque damage
* Coke load in each prechamber
* Carbon load on each electrode tip
* Injector spray pattern
* Glow plug tip degradation pitting/burn off
* Cylinder compression variation
* Glow plug age degradation
* Glow plug brand/manufacture variation
* Injection pump issues/variable
* Fuel addatives, summer/winter variation


Any of these can cause false data.
A combination of these will cause false data in perfect LAB conditions.

A glow plug instrumented with K-type thermocouple runs roughly $500.00 - $7000.00 each.

Hunter is completely correct that "glow plugs are not thermocouples." You might be able to use the glow plug as a crude thermistor, i.e., a heat-sensitive resistor. All resistors are somewhat heat sensitive, of course. Thermistors are designed to have enhanced sensitivity to heat and to be repeatable, unit to unit. Glow plugs are not, so you would have to calibrate each one. Since their cold resistance is only about half an ohm, you would need a milliohmmeter and very low resistance connections. I have a Fluke 8800 digital VOM with a bottom ohms scale that reads down to 0.001 ohm but I'm concerned about my ability to make a repeatable low-resistance connection to the glow plug.

In any case, I'll do some experiments and see what kind of numbers I get when the engine is running as compared to when it is off but hot and then off and cold.

__________________

"Buster" in the '95

Our all-Diesel family
1996 E300D (W210) . .338,000 miles Wife's car
2005 E320 CDI . . 113,000 miles My car
Santa Rosa population 176,762 (2022)
Total. . . . . . . . . . . . 627,762
"Oh lord won't you buy me a Mercedes Benz."
-- Janis Joplin, October 1, 1970

Thats the trouble with these engines you cannot get a good exhaust stack reading on an individual cylinder. Brian, I have earlier thought about most of the items you mention at length. In fact when first starting to consider this area. I thought there was no way it could ever be of any use because of all the variables.

My primary concern still is a person not taking the effort to verify or at least try to verify what is being indicated.. It is difficult to be sure even when you verify. Yet a lot better than absolutly nothing. I regularily read threads where the milli volt method would at least indicate something. Or conversly eliminate something.

Yet as crude as it is it is about the only method usable at times. The glow plug is not a good thermocouple but does tend to generate voltage with the dissimular metals mix of the element as far as I can see. Lets see how the gentleman does. I suspect he will get indications or elimination of problems if he verifies as he proceeds. Lets watch and see.

For example I feel it is now becoming important to comparison measure the flow from the injectors if the the pump is implicated. The same way volkswagon mechanics at good dealerships do into calibrated containers. This is only after the injectors have been properly checked out of course. Plus glow plug calibration verified by some method.

. If a good thermocouple were installed into the pre chamber it would be subject to the same variables you mentioned. It is not practical and may never be to do anything on the exhaust side of the engine because of the common exhaust manifold.

It is just an attempt to get an indication of the average rather than the instantainious temperatures occuring in the prechambers. I can think of no other tool that can give an indication or at least suspicion about the injection pumps or injectors condition we have at the average home garage.

Time will resolve this whole issue eventually anyways. Your points are well taken and understood . It is important to point them out periodically so people will understand the need to attempt verification always.

Do they sell glowplugs, with thermocouples in them?

__________________
Have you ever noticed that anybody driving slower than you is an idiot, and anyone going faster than you is a maniac?

As long as they would add one additional commandment for you to keep thy religion to thyself.
George Carlin (Wonder where he is now..)

1981 240d (engine donor 1983 240d) recently rebuilt engine hurray! - No more.. fought a tree and the tree won.

pearl black 1983 240d 4speed (Converted!@$$%) atleast the tranny was rebuilt.

As a side note...pertaining to your lope. Is it occuring only at below 800 or so rpm? I had a similiar problem in a 617. I got so fed up chasing it that when I pulled the IP to change the gasket I put a different IP in. Problem solved. I havent sent the offending ip out yet. But I do know the car sat along time, without running.

Not that I am aware of. I am using the stock GPs as crude devices to see a relative indication of the offending cylinder.

It is an easy test, just unplug at the relay and measure. The connector pin numbers are consistant with the cylinder number with exception to cylinder #4 which is pin 7 at the connector.

This car has 57K mikes and has been sitting for the last 10-12 yrs.
My mileage is around 22 mpg with the AC running. Top speed at 97f with the AC on is 86 mph GPS verified.

The answer is yes.

Post #8. last line.

The cost is high because these are Automotive Research and Development tools = custom built for each application.

Note:
Durability life is generally three days - one year.
Calibration must be constantly checked, and there is NO warranty.