Sorry guys it was my brother in laws old Jetta.
He had just installed a used pump.
But the principal remains the same.

Decided to set for average hottest burn.

So read all glow plug voltages combined by reading the general harness.

Was 15mv he had the pump set by ear.

Loosened and tapped pump until hit maximum point of 36 mv.

Did not expect that large a voltage swing and it surprised me a little.

Also read three other Volkswagens on the property, two were reading 15-20 Milli volts and one was 66 Milli volts.

Do not know if the wide variance was newer glow plugs or older plugs.
Was not leakage from the relay as with key on and engine not running there was almost nothing read on harness of the high reading car.

These three were not adjusted but just was curious about what would be read from them.
It is obvious there will be no standard values and it does not really matter.
Went a little past with voltage tapering off on the one I was adjusting.. Reset to 36 mv.

Told him that since his injectors had never been checked there was a possibly the average highest readout would probably be the best.

Better than setting to one unknown injector at least.
No noticeable pre-ignition was present after or before the adjustment.

He changed the pump out because of fuel leakage not general performance issues.
Just told him to see if mileage was normal as well as performance as he like myself has been driving old diesel Volkswagens for years.

What I would like to have done is check the timing by the conventional method to see how it stacked up.
Just decided to see how it works out instead.

Mileage and smoking should confirm if it really is in the ballpark in a few weeks.

Garage cost to set pump timing is about 240.00 average here as you require an adapter to screw into the pump as well as a dial gauge to get the readout.

If he has no problems with it's setting might do it the conventional way to establish if to design accuracy.

Also a possibility the Milli volt is a better setting that the factory recommended now exists as it does deal with aging symptoms of many forms where the factory method does not.
It will take time to prove that though.

Have to wonder how close it is though.
Again more has to be known but my first thought is how quick and painless it is to do.
If a scribe mark was put on any pumps present position and it moved to the highest voltage reading nothing would be lost by trying it that I can think of.
You could always return your pump to where it was.

Now hoping someone out there will time with the drip method and just do a simple comparison by marking the highest voltage point read before and scribing a mark.

If you do the drip method right after that and they turn out exactly the same then for practical purposes this will be the better, quicker, and perhaps more accurate way to time the 123 engines.

Again a little more feedback will and should be generated.
Two noticeable items were the quick response or voltage change when tap adjusting.
Almost instantaneous for practical purposes and how such a little tap was changing the voltage.

My two cents for what it is worth.
Not meant to be a suggestion of validity by any means just a description of what transpired.

Gathered a crowd of guys and the first common observation was the speed of the process if valid.
Five minutes can reset a Volkswagen pump from start to finish with the accuracy of the procedure still unknown.

.

Barry I encourage you to continue to share your 'adventures of the glow plugs'. I find this subject more interesting than a hungry dog at a chicken bar-b-que.

Hopefully in time I can add my experiences when I get a good DVM.

Mike

__________________
Diesel Discussion. Where learning is FUN!

1983 300D 296k work in progress
1985 300D (sandwiched) parts
1980 240D engine issue

Just got home from a meeting about 75 miles away this evening.
On way home I decided that I might do the car I am currently driving daily, another Volkswagen unfortunately.

This is based on my knowing the results faster than my brother in laws car.
Have suspected my car of being off a little pump-wise for quite awhile now.

Anyways will be able to post in about a week the changes in mileage if any plus driveability as it is a little sluggish compared to most I have owned.

This Milli volt method is used by the factories to set some pumps to engines for quite awhile, so am starting to think it has general application potential and if it proves so over the long-haul, it will mean a lot more pumps get set as the procedure is so fast and simple.

We will see if there is any downside soon.

Needless to say I will scribe my pump to flange just in case I have to find my way back to the original position in a week or so.

A true case of nothing ventured, nothing gained.

Also the fuel mileage on this example is about 5 miles per gallon substandard.

Plus I seem to have my foot a little more into it for highway cruise.
So there is justification in trying it out in my opinion.

Will give an extensive report in about a week or less.
Who knows?
Perhaps I will not have to buy fuel anymore and the ball joints will self rejuvenate as well.

.

you strike me as being firmly based in objectivity and scientific methodology.

Related Sidebar - Right now I'm working with my small local independent M-B shop and have convinced him to invest in one of the pulse sensors that interfaces with an inductive timing light... this to more quickly check IP timing. And now if your method of using mv readings off of the #1 glow plug proves to be valid, this would be an outstanding adjunct to our fast timing check.
Questions:
Q1 - Do you think there is a way to attach a long, heavy, bar to these IP(s)... such that one could more easily/accurately shift the IP body and its timing? If so, this would go well with your method of reading the glow plugs for the highest mv readings!
Q2 - What if you install a brand new glow plug in the #1 pre-combustion chamber just before warming up the engine and taking the mv readings and/or making adjustments? Wouldn't this likely give you more consistent results?
Sam

What generates the voltage? The dissimilar metal of the heater bonded to the case, and the hotter temperature where this bond is at the tip? That would be my first guess, but it's just that.

As of one hour ago it is no longer theoretical.

My brother in law and myself did my own car as well.
There was a complicated peak reading of about 1/2 of a Milli volt greater than our start point perhaps less than an 1/8 of an inch from the original pump setting away.

One immediate change was a small puff of black smoke that was very small when quickly opening pump linkage from idle.
Brother in law said that is an indication of peak possible pump timing point rather than a new problem.

He does work on a lot of Volkswagen diesels.
Took car for a drive after finding it started quite a bit quicker as well.
Much more power than having foot into the pedal.
Not subjective but a absolute highly noticeable difference.

Now the voltages and observations.

We did this one from the value of all plugs still linked together.
Start voltage was a floating thirteen Milli volts.
As we moved the pump the voltage average started upward a little till average of 13.5 Milli volts was reached before going past it and voltage dropping off.

Reset pump to our punch marks and did it again.
Same basic results so locked pump down at that 13.5 average.

I do not need a week to find out as the improvement is major in nature. We will see about the fuel mileage but even if no change I want this setting over any other.

Car kind of floats along now where I had to kind of force it a little to perform.
Also it wants to rev not only easier but with a lot more apparent power.

We will do a few more early next week as brother in law thinks my results where better than the manufactures recommendations could accomplish.
He said he has only seen a few set by the manufactures specifications give that slight puff and he does a lot of those engines.

We were also getting voltage change when there was no visual obvious moving of the pump other than a light tap being applied.

Now the downside although minor.
It appears to minimize the voltage hunting effect over about 0.4 Milli volts, that a condenser will have to be inserted across the test meter leads to establish an automatic average to further speed up the process, plus eliminate the mental gymnastics of establishing the average float values.

If similar results are obtained on the next six cars lets say and any complications noticed and rectified.
It means we have a vastly superior method of timing the pumps to our engines in sight now as really think it is far more accurate than a drip or well up test.

This method automatically seems to compensate for existing deficient problems that are inherent to quite an extent in older engines within sensible limits.
The other systems that I am aware of do not.
In other words if you where to take your car to Mercedes to have the timing checked my guess is other than a new engine the true optimum setting will not be there when you pick up your car and even on a new engine it may not still be there.

But before getting totally carried away we have do do some more cars and get rid of the hunting effect of the meter that we experienced on my car, yet was not apparent on the one we did yesterday or to a much lesser degree.
I do not know what the hunting effects indicate as have not had time to reflect upon it but they might have some significance.

Idle sounded very good and smooth so do not think pump is over advanced either, although with the power one might almost think it is.

So far so good but it is important to realize until we get about a dozen cars done we will not be absolutely sure this is going to replace the older methods.
At this point I think there is a very good chance it might transpire though because it is physically impossible to get such small movements of the pump with hand manipulation, plus you are almost guessing about the actual instance of when the wellup occurs or drip is absolutely right.
At that point of pump adjustment all you have really accomplished is making the injector spray into the cylinder at a given point in relation to the crank.
No compensation whatever has been done to allow for slightly out of value components basically.

Perhaps one could say the old method is crude by comparison.
Will stop droning on now until we do a few more cars.

Still looking for any downside indications by keeping an open mind.
Am almost certain now they are not there.
Just thought the meter instability to some extent that appears to be hunting might be the meter itself, will have to check that out.

As for his car that he changed the pump on that had a large leak in it.
The one we timed yesterday in fact.
He thought the basic problem was caused by the old pump.
Turns out the engine block was not in good enough shape for really any meaningful interpretations.
So I am discounting it from the test other than that apparent massive peak of 36 Milli volts it topped out at.
Also do not know the reason why that peak occurred at that magnitude.
Well away to the beach for the weekend hopefully but since I may have to come back to town for parts as the well malfunctioned last week I will check this site.

This is no longer a joke as it really appears to work.
This is also a dynamic test that is being conducted under operational conditions, Not a static test.

.

Guess I will try to answer a couple of questions after some thought.

As I mentioned the adjustment of the pump seems to be so fine that a long leverage bar may or may not allow the incremental movements required by this system.

As for going off of the first glow plug that is OK if you know your injector is popping off at factory recommended values.
By using the reading from all plugs simultaneously it will give an average value of all injectors and cylinders combined.
In other words a more accurate result on older engines.
Also do not forget you are doing a dynamic test with the engine running.
This has to be better than the static mechanical test.
You are also tunning for the absolute power peak.

My next concern is if that is also the best fuel economy point with this procedure, if not perhaps the timing may have to be set a little before the absolute peak on the retard side.

Since my accumulated mileage is only 75 miles on my old Jetta I have no way of estimating fuel consumption yet.
The thought is that most 123 owners especially 240d owners will want the peak setting versus the highest possible mileage setting.
If the two settings coincide it would be great but life being what it is may not be so.
Forget engine sound at idle as well.
This system is so fine you will not notice any difference with the idle after you get close to the final point.

As for the gentleman that asked the function of the dissimilar metals in the plug that generates higher voltage with more heat.
I really do not know exactly how the glow plug is acting a little like a thermocouple exactly.
Also still have some questions why some read higher that others.
But at this early stage do not feel it is of any recognizable importance.

Just too early yet to make any absolute conclusions about anything.
Interest will only build once it is proven totally to be of benefit to 123 diesel engines.

There is not too much doubt today in my mind that it is a good and workable procedure for the older 1.6 Volkswagen diesels.
All brands of indirect type diesel engines with glow plugs will probably respond in the same fashion.

This is also a simple procedure that is well within most site members ability to preform.
That was one thing that I was really thinking about before getting into it awhile ago.
Plus as long as you take a pin punch and mark the pump flange and the pump mount flange so you can always go back to those marks I felt it almost foolproof.

As I stated if the procedures to use the glow plugs in many ways as a service tool providing all the initial responses are positive.

It will still take a year or so for acceptance and I personally know that many questions and issues will arise in many peoples minds as this is explored.

If you want your car to perform to it's maximum ability do you really have any other choice other than trying to set the pump and fine tune it many times by driving and adjusting many times and still you might miss the sweet spot.

The factory drip or well up position will in 9 out of 10 cases not be the best pump timing position on older engines in my opinion.
The factory position will be okay probably but just not the optimum possible position.

Just keep an open mind as there may still be a downside that arises yet.
Well back to the beach as have gathered up the last of what I need for the repair out there.

.

Barry,

I have been reading your threads for a while now and have tinkered with this operation and now have some questions.

If you dont remember I was the second individual who had fooled with the fuel elements (because the rubber o-r ring was leaking). I had moved the first and second elements on a five cylinder.

I finally got a good enough mulitester that can accuratley read mV to a .1 level, so I attempted the operation today. After the car was fully warmed up i measured all glow plugs (disconnected from the wiring of course) and the readings were about 12.3 mV for the # 3,4, & 5 injectors. #1 was about 10.8 and #2 was about 11.6. The problem I had was that in taking these readings the numbers were jumping around everywhere and the measurements I took were (what you would have called) "floating". I noticed that by twisting the elements clockwise the the voltage for the respective glowplug would go up. But here is the problem.....I noticed that if I would raise the volatage for the #1 cylinder, the voltage for #3 would go down. I believe it started throwing off other cylinders as well.

My thinking is that this is the result of the Inj. Pump. is not being properly timed to the engine. Should I do this first before i retry the mV method? I got horribly frustrated today in my attempt. I dont have a drip tube, but I thought I might use this method: http://diymbrepair.com/easley/iptiming.htm

Also I should be recieving a new rack damp. pin tomorrow or the day after. should Install this before I attempt all of this or will it affect it at all.

Thanks for all your help barry

__________________
1983 300SD White with Grey interior

I would think that to be normal governor behavior. If you increase the fuel delivery of a cylinder to make it hotter, the governor will keep idle speed
constant by backing off the rack, lowering fuel delivery to all cylinders.

I really don't recommend adjusting the individual delivery valves without putting the pump on a test stand.

__________________
'79 300SD W116
'86 190E 2.3-16V W201
'92 300D 2.5 W124
'99 E320 4MATIC S210

Mitch is right of course.
This method is to be used only when pump needs a shop sequential timing because a person has accidentally modified the element positions.
The pump is only going to the shop anyways and what you do if not successful does not increase their charge either.
There is no sense in setting the pump primary timing right now.

For one thing the drip and welling up are in reference to number one element and you know it is off.
Only way you could do it is to isolate the harness to 3,4,and five cylinders and use millivolt method.
But not required anyways right now.

It does take two sets of hands to really manage what you are trying or two meters for one person to try.
Every time you change the setting of one element of course your reference voltage is going to change.
Since cylinders 3,4,& 5 are the same any one can be used as a reference by adjusting the element # 1 or 2 a little higher in voltage, then go right back to see what it has done to your reference cylinder voltage.

You are trying to get them to read the identical same output voltage.

Now for the hunting that is hard to work around.
I will go experiment and deal with the problem as a time constant kind of thing today as I do not think it is an inductive problem.
Suspect we will have to just put a capacitor across the meter basically to average our Milli volt swing.
Will try to do that today and get back.

Do not get particularity frustrated.
Look at the bright side you are already getting a solid indication that one and two are off pretty bad.

Also my old Volkswagen did get back the missing 5-6 mpg as I drove it 400 miles in the last couple of days to the Willie Nelson concert, and on many trips back and forth to town from the cottage to get every last part to get the water well up and running.

Will drop a message as soon as possible.
Also look at the upside.
As we jointly work through the thing your knowledge is going to increase.
That way If I am swamped with alligators at my end guess who I can refer people to.

.

Okay.
I built a little time constant circuit from a 3300 microfarad capacitor rated at 16v, and a resistor of about 700 ohms value.
Just stuff laying around.
Worked fine at stabilizing the readings.
The way it is wired is the resistor is in line with the probe and the capacitor is across the negative and positive terminals of the meter anywhere past the resistor towards the meter.

The condenser is still subject to the hunting but acts like a battery to automatically average the readings.
Last digit of my meter stayed still at a constant reading unless the primary reading average value changed of course.
Then the meter just went to the next number higher or lower in a rock stable fashion.
Values for those components are not critical or expensive either.

Clear as mud?
If you have any questions just ask, that is the only way we will ever succeed.
I will try to support anyone that tries needless to say.

Now I know you can get those two out of calibration injectors back in line.
I could do the mental gymnastics easier with the hunting meter perhaps.

The other day I also realized the hunting was also a major distraction.
Distractions tend to affect accuracy.
Keep going and at any point if you feel frustration drop me a line.

You are going to succeed at this.
Also remember as one person mentioned your actual movement of the element is going to be very small.
Do get a friend to move the voltage probe to what you want to read while you are adjusting the first two elements.
Really almost a two person job.
There will be a lot of back and forth with the probe until you are really close.

I really believe you can pull this off then perhaps it will be time to do the pump re-time.
The same voltage on cylinders 3,4 & 5 give hope the engine generally is in reasonable condition or better.

Sure it might be a little problem to locate a condenser either new or used and a resistor.
Any friend with an interest in electronics could help or even any electronic service shop.
These are cheap standard components.
Thinking a few dollars or free used parts.

Will mention although I think you already know not to move cylinders 3,4, & 5 elements position during this procedure.

.

Interesting stuff. I'm wondering if there is any merit / advantage to using an ohm-meter across the plugs during the timing process instead of a voltmeter? The plug resistance definately does change with heat, as that is what they're designed to do. However, as this thermocouple voltage effect seems to be an unsolved mystery and varies widely from one car to another (if I read all the above correctly), would resistance readings give more consistant results?

BTW, when I was in my teens, a long time ago, I worked summers at my uncle's Texaco. He had a contract with the city for maintenance on some of their Diesel powered machinery. I recall when he was setting the timing, he would watch the stack for the exhaust to make a black puff, then just go back from there a bit. I think they were mostly Detroit engines at that time.

Dave
Sackville, NB Canada
1976 300D

Dave, sure we should reset your pump since you are using vegetable oil exclusively.

Since the flame front is slower in my opinion a slight correction will help a little.

The amp meter use is an excellent thought but I kind of feel although we are generating voltage it may not be backed up by any ability to generate current to any extent.

Measuring resistance is not a good ideal as the source is generating voltage at the same time.

All ideas deserve looking at though and they should be raised.
Also as time goes on it will probably prove out that even if different cars generate different baselines it does not matter.
We are only seeking the peaks usually or trying to equal one voltage with another by direct comparison.
The amplitude just increases the percentage accuracy if it is an unexplained higher value.

One never knows until they ascertain the facts by trying.
Again as I have repeatedly mentioned pin punch mark the pump in it's present position before moving it means you can always go back or forward to your present timing if you fail somehow.

What I am still hoping is the hottest burn also coincides with the highest mileage point as well, if so when we work through the really small things like the floating voltage or hunting problem (solved this morning I believe). We are getting close to a tool that will replace what is presently done, and thought now and a cheap one to boot with enormous accuracy.

Time will tell now as the gentleman with the current problem is going to totally solve it in my opinion.
As long as he keeps an open mind and avoids frustration.
This is still new enough to cause application misunderstandings.
I would like to see every question raised and answered as we go along.

Still looking for the fatal flaw but the possibility of it existing is diminishing rapidly.
I imagine the definitive manual will be written in the next six months at the rate things are going.
Seems almost impossible for it to fall flat now.
We have to wait and see though.

The bulk price of vegetable oil in our area will take a little time to find out.
The price I think will be right but the quantity in those containers again may be a problem.
My son in law is dealing with the issue.

.

JMAN300Sd, I will try to address specifics to your inquiries.

You are a sort of test horse here as well.

Yes the voltage will change on numbers 3,4,&5 as you adjust the elements on 1&2.
That is to be expected and is the reason you want an extra set of hands there to monitor one voltage and then quickly check another as you move the element.
With the hunting stabilized to zero with the small circuit I mentioned it will be a lot easier for you.

What you refer as the floating will remain to some extent and is normal, it becomes much less of an issue with the hunting eliminated.

As you get closer and closer to the final position the floating as you describe will lessen.
The floating and hunting are separate things.
The floating is being caused by engine parameters being changed in some way by your application and is normal in my opinion.

.

Alright Barry...I think I understand you. I have still a few questions though as far as my procedure this afternoon:

Should I set the #2 element to the reference (#3, #4 or 5) first and before I do the #1 element?

then...

Should I tweak the #1 element in an effort to raise the volatages of the remaining elements as much as possible, even if it means reducing the voltage of the #1?

I assume after this point I would then go about timing the pump to the engine..Correct?

__________________
1983 300SD White with Grey interior