Parts Catalog Accessories Catalog How To Articles Tech Forums
Call Pelican Parts at 888-280-7799
Shopping Cart Cart | Project List | Order Status | Help

Go Back   PeachParts Mercedes-Benz Forum > Mercedes-Benz Tech Information and Support > Diesel Discussion

LinkBack Thread Tools Display Modes
Old 05-15-2006, 03:33 AM
Join Date: Dec 2005
Location: CA... No. of S.F.
Posts: 890
OK I'm a late comer, but...

OK guys I'm a late comer to this intriguing little project Barry has hooked us on... but what can I do to help this experiment along some?

I do have access to my 1980 model 240D[na], my son’s 1980 300D[na], and my father-n-law’s 1985 300SD [turbo]… and propose that I get my “banana plug” connector test rig put together and initially collect precise readings for each of the glow plugs on these vintage M-B(s).

From that point on I will be need to know what additional steps I might take to help this project along.

I have recently talked my local independent M-B shop owner into investing in a new diesel pulse sensor [piezo technology] accessory that enables him to more readily check the timing on these cars using his timing light… so I guess I could get some data on what the present timing shows up as using this electronic equipment... what do ya think??

The thought just occurred to me that after a car has been peek tuned using this simplified "thermocouple" method... wouldn't it be interesting to have a local diesel injection shop [one with the top-of-the-line "dynamic" bench test equipment]... wouldn't be interesting to see what they might conclude from examining the CAR and/or the IP?

Sam Ross
Reply With Quote
Old 05-15-2006, 12:08 PM
Registered User
Join Date: Nov 2004
Location: Nova Scotia, Canada.
Posts: 6,510
Sam, sounds like an excellent ideal.
Your car with the poorest overall mileage and performance might be selected for the test.

I would set the timing by the piezo method.
Mark the pump and see if it coincides with the highest leading edge of the Milli volt method.
I would not be shocked if it was not a few degrees advanced over the piezo method either with the Milli volt method.

Allowing for aging characteristics or modern fuel perhaps.
This is exactly the kind of information we need.
Can see no reason the two methods cannot be used simultaneously when doing the timing to compare.
As there is no interference between the two systems during application.


Last edited by whunter; 03-19-2013 at 01:35 AM. Reason: spelling and readability
Reply With Quote
Old 05-15-2006, 07:31 PM
Join Date: Dec 2005
Location: CA... No. of S.F.
Posts: 890
Barry... see Brian's post at

Barry - Thanks for getting back to me so promptly.
See Brian Carlton's post at:
and you will see that the pulse timing light method has to have a fudge factor that luckily Brian has already worked on. He wrote: "With the SD [his 1985 300SD] set perfectly at 15ATDC via the M/B RIV timing, the pulse timer registered 13.5° BTDC. So, I would presume that the n/a 300D [my 1980 300D(na)] should be close to this value."
Which I interpret to mean:
(1) The basic primary timing spec is 24 DegBTDC
(2) which=(s) ~15 DegATDC on the IP using M-B's "RIV" method
(3) which=(s) ~13.5 DegBTDC by timing light equipped with piezo pulse accessory.

Personally I get befuddeled [<-sp?] when I think about the several factors here that have potential to keep these three from being more absolute relative to one another, at least in my muddled brain.

(a) age of vehicle [ e.g. stretched timing chain],
(b) condition of glow plugs,
(c) condition of the injector nozzles,
(d) condition of the heat shields protecting the nozzles,
(e) how well the valves are adjusted, and
(f) how well the individual IP "elements" are set-up/timed relative to one another...

doesn't one or more of these [if any] keep us from establishing a precise comparison between the three numbered timing methods above?

If you guys can get my mind straight about these factors and timing methods I think I would be of some help going forward... what affects or changes what aspect of IP timing?

Reply With Quote
Old 05-15-2006, 08:37 PM
Registered User
Join Date: Nov 2004
Location: Nova Scotia, Canada.
Posts: 6,510
Sam, you have just hit the highlight of the Milli volt timing function on the head.

It does automatically compensate for many factors present as you list them.
All other methods but the ear just basically refer to the factory recommendations from years ago.

Basically #1 injection point in reference to crank marking.
That I expect may be crude in comparison to reading the flame front.
Todays fuels alone are of less quality as well.

It may turn out perhaps on Mercedes that the peak Milli volt reading is perhaps a little too far advanced.
Even if so we will learn to set them a little before the peak.
What gives me hope that the factory recommendations are wrong for a large percentage of cars is the few cars that give substantially better fuel mileage than the majority.

I have suspected that knowledgeable mechanics may have set the pumps a little ahead by experience.
The 240d is a classic example of a possibility as some seem to deliver 15% better mileage than other examples in generally good condition as well.
Plus some of the same owners report much better power than average.
Also we can stand the additional power as well if it is available to us without over stressing the engine.

I am almost certain if the mpg are higher after the adjustment the car is not too far advanced.
If mpg are the same or less than it is probably over advanced.
That is the flaw in setting the pump by ear to some extent.
Too much trial and error.
You might luck into the right spot but chances are you will not.
We will see.

I also would do a fast preliminary check to see that the engine is in pretty good general condition.
Timing adjustments will not mask serious flaws.
But will help compensate for minor ones.
That is okay as these engines are not babies.

A good quick check of your engine is to just read each glow plug voltage separately.
If one has noticeably higher or lower glow plug voltage find out why. Swap the glow plug with it's neighbor first.
If no improvement perhaps swap the injector next.
I assume the valves have been checked at reasonable intervals before this.
Next might be a compression check of that cylinder if fault has not been isolated.
If that is good then you might want to start evaluation of the pump.

I did say evaluation not moving things until you are really sure.
Always remember two glow plugs beside each other that are reading low indicate a possible head gasket problem.
Great verification process of a suspicion.
This method has great potential if we can get past the preliminaries.
Remember it is also the method of choice already for some diesel engine builders to set their pumps to engines during the manufacturing process.


Last edited by whunter; 03-19-2013 at 01:41 AM. Reason: spelling and readability
Reply With Quote
Old 05-19-2006, 08:21 PM
Join Date: Dec 2005
Location: CA... No. of S.F.
Posts: 890
Some basic data... mv glow plug readings:

1980 240D: [ has 3 relatively new Bosch GPs and 1 Monarch GP ]
1980 300D: [GP's are a Non-Bosch mix ]
mv Readings:
"Composite" readings taken with all GP's connected in parallel !
240D... #1=9.9,..#2=10.3,. #3=9.9,. #4=8.6.................Composite=9.5
300D... #1=10.8, #2=11.6, #3=10.3, #4=7.9, #5=6.6... Composite=9.0

*As I understand it, the absolute accuracy of these mv readings is not really important so my cheap digital meter should suffice.... what's important here will be to eventually dynamically adjust these to peak/balanced mv readings, right?

The 1985 300SD data will have to come later.

Again the 240D and 300D are both geting about the same 22-24 mpg. I expect the 300SD is getting the best highway mileage. Things might improve for the 300D soon by virtue of infusion of a new/rebuilt tranny from Peter Schmid [Redwood City, CA] next week.

The 300D is currently the smoothest/quietest running engine and the 240D is the noisyest and roughest running. If nothing else I hope to get the 240D to settle down by fine-tuning the IP elements and IP-to-Engine timing.

Now I need information on how to proceed from this point. The IP-to-Engine timing I understand and I think my independent M-B shop owner is "on-board" with me on the notion of building a heavy, long-bar, tool that will clamp onto the front of the IP to facilitate this dynamic adjustment. So now I need input and/or will have to educate myself on the fine-tuning of element-to-element timing... something I do not want to jump into helter-skelter!!

Note- I picked up sone "Banana" type plugs at Radio Shack that make it a breez to connect to the GP Relay's connector to take the mv readings... individual one as well and the composite. I can supply further info on this if anyone wants.

I'm still working closely with my local M-B shopowner to try and come up with a workable piezo-electric pulse sensor timing light to quickly, dynamically check the engine/IP timing on the many diesels he sees.

Sam Ross
Reply With Quote
Old 05-20-2006, 12:45 AM
Registered User
Join Date: Jul 2004
Location: Miami FL
Posts: 253
I highly suggest you dont mess with the IP elements just yet. Do your primary Timing and then do take the mV readings and let us know what you find.
1983 300SD White with Grey interior
Reply With Quote
Old 05-20-2006, 01:32 AM
Join Date: Dec 2005
Location: CA... No. of S.F.
Posts: 890
JMan300sd - Trust me I read your past posts...

and I fully appreciate the difficulty you had in trying to restore the timing of two of your your IP's elements that had accidently been changed. And I do not intend to make that mistake. I'm proceeding slowly with this matter! Thanks for the appropriate warning.

By the way I failed to say that in taking these readings I did NOT experience any appreciable "floating" or "drifting" readings using my cheap digital VOM and I did NOT use a R-C circuit to dampen the meter readings. The meter only infrequently changed by 0.1mv. Maybe my $40 VOM was designed by a genius!

So as I understand things at this moment, my next step should be to accurately check and if necessary correct the IP-to-Engine timing.? It might be a while before I get around to that.
We are presently trying to obtain and satisfy ourselves with a piezo-electric pulse accessory that will allow us to use a timing light to dynamically check and set the car's IP-to-Engine timing.

What do you say Barry about my next step

In the mean time, I'll work on getting more mv data on the '85 300SD [W126] M-B.

Reply With Quote
Old 05-20-2006, 02:53 PM
phidauex's Avatar
BioDiesel Hopeful
Join Date: Oct 2004
Location: St. Louis, MO
Posts: 806
I just wanted to note that I'm following this thread eagerly... I'm facinated by 'test in place' methods, and this seems like a great example. For those who asked, or are confused still, I wanted to note the reason for the voltage...

It was noted at one point by an Estonian scientist that any metal, subjected to a heat gradient, would show an electrical potential across the piece of metal. The amount of the potential depended on the type of metal, and the difference in heat between the two parts. Of course, to read the voltage, you have to connect another piece of metal, which is, itself, now subject to a heat gradient, and is now producing a voltage of its own. The difference between these two potentials is what we can measure.

So any time you have two dissimilar metals together (as in the internals of glowplug), you have a thermocouple. The difference in voltage can be measured, and will increase with temperature.

If you were crazy, you could fashion a fake glowplug with a calibrated thermocouple on the end, like a K or J type thermocouple. Then the mV reading could be compared to a chart, and correlated directly to a temperature, and you could measure the cylinder temperature directly.

Instead of doing that, we just use the glowplug. They aren't calibrated though, so different brands and ages of glowplugs will show different voltages, even at the same temperature. If you had the right test rig, you could 'calibrate' a scale to your glowplug, create your own conversion scale, and read an actual temperature of the cylinder. 14.5mV = 1200F or whatever.

But since the actual temperature isn't crucial at the moment, thats not worth the time and energy. It makes more sense to just look for the max temperature, or in our case, the max voltage.

The 'hunting' described is probably a combination effect... The cylinder isn't one temperature, its heating up, and cooling down, through the cycles. This pulsing up and down heats and cools the glowplug irregularly, causing a variation in the voltage, corresponding to a laggy variation in the cylinder temp.

One thing to note is that in the 'millivolt world', everything creates potential. Try wiggling the test leads around while they aren't connected to anything. You'll see a little voltage from their inductive interactions. When making a reading under the hood, make sure your test leads are as still as possible, to eliminate that fluctuation.

Anyway, thanks for working on this method, folks! I'm very curious to see how it all turns out, and will be trying it once there is more data (I've never timed a pump before, so I want to make sure I know enough to do it right the first time!). I'll see about taking some mV data on my 300TD soon, just to add to the pool of data we are building.


PS more about thermocouples:
"That f***in' biodiesel is makin' me hungry."

1982 300TD Astral Silver w/ 250k (BIO BNZ)
2001 Aprilia SR50 Corsa Red w/ 5.5k (>100 MPG)

Reply With Quote
Old 05-21-2006, 02:24 PM
Registered User
Join Date: Sep 2004
Location: Canada
Posts: 495
Adjusting Rod

Gentlemen- I think I've got it. A facsimile of the special tool required to move the IP (see pic).

It's part of a door closer assembly. The left end would clamp to the fuel line bolts on top of the IP, while the right end will slip over the engine cam-cover stud and fasten down securely with the nut.

To finish it, I need to do three things-
- drive that pin out that goes across the left end so the piece that goes in there can be extracted
- cut the aluminum tube shorter so it will fit in between the IP and cam-cover stud
- re-insert the removed piece, but instead of staking it with the pin, I've got to somehow weld a retainer on the end of the tube where I cut it, to hold the piece in and allow it to rotate within the tube, thus pushing and pulling the threaded part while turning the tube with a vice grip or wrench.

Problem- I always get stymied on the physical execution of things. First, I tried driving that little pin out with a punch and hammer, but couldn't get it to budge. It does look like it would come out though. Second, I cannot weld- especially aluminum, er- especially nothing- I just can't weld.

Barry- if I can bring this over some time this week, and if you can bring this to completion- it's yours man! I know the idea will work.

Attached Thumbnails
Setting pump primary timing by milli volt method.-tool-01-vga.jpg  
Reply With Quote
Old 05-21-2006, 03:28 PM
Join Date: Dec 2005
Location: CA... No. of S.F.
Posts: 890
I hope your guys N. of the border have something, but...

this aluminum door closer cylinder idea seems a bit "flimsy" to me and I canNOT really follow Dave's explanation of how it would work!?

My independent M-B mechanic has an idea that appeals to me for hit would literally clamp onto the front of the IP just forward of the individual elements where the IP's body is clear/clean and the top and bottom surfaces of the IP body are square. As a quick-N-easy way I could see a "giant" crescent wrench [ metric of course ]... or even vicegrip-like tool [how big do they make them?] connected to this squared off front the IP might give us the leverage needed to dynamically adjust the IP-to-Engine timing.

My M-B mechanic friend is inclined to make a tool from scratch that would have a strong clamp, heavier and longer arm thus providing greater leverage and greater accuracy. However this will have to wait until we get our diesel IP timing light working. The timing light will allow us to readily confirm what timing results from using the dynamic IP [ miliVolt fine-tuning ] method that we think has potential.

Reply With Quote
Old 05-21-2006, 03:48 PM
Registered User
Join Date: May 2004
Location: Flowood Mississippi
Posts: 438
dynamic adjustment control rig...

I like the door opener concept as it allows very small adj. during the dynamic a turnbuckle type adjuster....I may make one out of heavier gauge fine-threaded rod and large fine thread spacer nuts...also more force and finer adjustment available with the screw mechanisim.....I see the square area on the IP you are talking about.....this is good info...maybe combine these two clever designs....A nice rig could even be left in place like the adjusters on the belt driven stuff....keep on thinking guys!....kevin
Reply With Quote
Old 05-21-2006, 04:57 PM
Join Date: Dec 2005
Location: CA... No. of S.F.
Posts: 890
" Dynamic " is the operative word here...

and " DNAMIC " means we are looking to change the IP's timing while the engine is running which means that the heavy fuel lines snaking their way between the IP and injectors act like one big strong spring resisting our effort to twist the IP's body and change it's timing. In a way I like the "turnbuckle" idea especially if a secure enough place on the engine can be found to anchor the other end. Early stage testing for such a tool would most likely tell us whether or not we need such a precise "hold-fast" feature as the turnbuckle. I still think that the inherent strength gained from as much long-arm leverage as practicable will be important in such a tool.

From our prospective here, keep in mind we are thinking of a tool for a profit making mechanic/repair garage trying to check and then easily change the IP's timing, not something that would be left on the engine. Also such "fine-tuning" of IP timing is not likely to be needed that often... should it?

Reply With Quote
Old 05-21-2006, 05:53 PM
Registered User
Join Date: May 2004
Location: Flowood Mississippi
Posts: 438
turnbuckle and lever

I was thinking of a tool left in place just for our own test as to quickly change the pump timing for optimum starting and running...the lever you speak of could be used to help rotate the pump/turnbuckle fixture if a slot or tube was attached to the turnbuckle device that you could slide the bar or arm into....kind of like the old bumper jacks have.... so as to be removable for driving test so you close the hood....the injector lines are very strong and this extra leverage will probably be needed to help move the contraption....of course several designs are possible....I just want to move my pump to see how these voltages track performance....I hope this glow plug thing works out...I hate the drip tube...this is a great site!....kevin
Reply With Quote
Old 05-21-2006, 08:30 PM
Join Date: Dec 2005
Location: CA... No. of S.F.
Posts: 890
Customer-make IP timing some have spoken of…?

I was rummaging around under the hood of my 240D and notices for the first time some marks that have been made on both the engine and IP sides of the IP mounting flange. See photo below [ I hope was able to upload it].

If this is what others have spoken of, what do you make of these marks that obviously indicate that the IP is no longer set the same as the marks… but in this case I think I know why this might be. This M-B dealer imported rebuilt engine only has ~20,000 mi on and no doubt when installed it was matched up with a used/rebuilt IP from another source.

So if I wanted to have MY reverence, I would have to put a new set of marks using a small chisel across this engine/IPP flange line? Your thoughts please so I'll know I'm on track?

Last edited by Samuel M. Ross; 07-11-2006 at 02:28 PM.
Reply With Quote
Old 05-21-2006, 08:35 PM
Join Date: May 2002
Location: Blue Point, NY
Posts: 25,396
Originally Posted by Samuel M. Ross
So if I wanted to have MY reverence, I would have to put a new set of marks using a small chisel across this engine/IPP flange line? Your thoughts please so I'll know I'm on track?
Nothing wrong with putting a new set of marks so that you have a starting reference for your adjustments.

However, I'm curious about your original marks. If correct, the timing is quite late. Have you checked the timing to see if it's near the specification?
Reply With Quote


Thread Tools
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is On
Trackbacks are On
Pingbacks are On
Refbacks are On

All times are GMT -4. The time now is 10:14 PM.

Powered by vBulletin® Version 3.8.7
Copyright ©2000 - 2019, vBulletin Solutions, Inc.
Search Engine Optimization by vBSEO 3.6.0
Copyright 2018 Pelican Parts, LLC - Posts may be archived for display on the Peach Parts or Pelican Parts Website -    DMCA Registered Agent Contact Page