I think that before modifying the system further, you might begin from a better position if you fixed the IP relief valve, and found the root cause of the air purging problem. It sounds to me like mod after mod is being tried to cure a straightforward issue.

I remember reading a thread not too long ago that involves installation of an auxiliary, electric fuel pump back near the tank to push the fuel forward. Any thoughts on this? Seems like if you raised the suctin pressure of the lift pump even a little bit it would help it's operation.

__________________
1983 M-B 240D-Gone too.
1976 M-B 300D-Departed.

"Good" is the worst enemy of "Great".

Delivery end pressure of om615,616,617 lift pumps FP/K 22 M 13..14...101; FP/K 22MW 3...7
-with the fuel return line pinched:
-at idle speed, 1.1bar minimum
-at 3000 rpm, 1.3 bar minimum
Vacuum,measured before lift pump,idle speed-0.1bar min.


om621 FP/K22 M7(with spring 2 444 617 010)
Delivery measured between the IP and fuel main filter-0.8...1.5bar and 2.2 @ 3000 rpm.
Discharge end pressure of lift pump -min 2.5bar at 3000rpm.
Vacuum prior to lift pump inlet,idle speed-0.3-0.4bar min.
Pes4M, 5mm. plunger diameter,by-pass valve same as on w115,w123`s,1mm throttle hole in fuel return union bolt.


In another thread Mr.Whunter posted some info about lift pump rep.kits:

http://www.peachparts.com/shopforum/mercedes-parts-reference-library/183451-om617-951-952-lift-pump-repair-kit-w126-120-300sd-w123-133-300d.html#post1461282
OM617.912 Lift pump Repair kit
Spring
0000910619=Bosch 2 444 616 010
This spring is also used in w115 and early w123 `s.Also in FP/K22MW3(w115 om617.910)

OM617.951 and OM617.952 Lift pump Repair kit
spring
0009934801=Bosch 2 444 617 010(WSF2156S8x)
This one is used in FP/K 22M7 in w110 621`s and some very early w115`s.

FP/K22 MW7(0 440 017 006) used "kolbenfeder fur 1 bar"
This was used from 07.76 to 08.97 on w123 300D and w123 240D`s
http://www.automotive-tradition.de/download/ausruestungslisten/mf1355.pdf

Here is a list of MB lift pumps with info as to what spring was used:

http://www.automotive-tradition.de/download/kennlisten/einspritzausruestung/foerderpumpen/0440017.pdf

0 440 017 023=FP/K22MW21 used 1bar spring

300D turbo;
http://www.automotive-tradition.de/download/ausruestungslisten/mf1357.pdf
FP/K22MW22=0 440 017 022 used 2.5 bar spring

In short,there is a great chance all om616 LP`s used the weaker spring.



FP/K22M5----M13 (w110,w115)
http://www.automotive-tradition.de/download/ersatzteillisten/einspritzausruestung/ep-78_01-75/0440007.pdf
http://www.automotive-tradition.de/download/ersatzteillisten/einspritzausruestung/sonstige/fpk22m.pdf


FP/K22M13
http://www.automotive-tradition.de/download/kennlisten/einspritzausruestung/foerderpumpen/0440007.pdf

Info concerning w115:
..the lift pumps FP/K 22M 13=0 440 007 014 is said to be equipped with weaker springs(2 444 616 010)-"kolbenfeder schwaecher", as opposed to the "standard" 2 444 617 010 one( .pdf`s above).

Anyway,I assume 2 444 616 010 spring is the weak one and 2 444 617 010 is the stronger one...but there might have been more than these two installed,who knows...

To see what Bosch equipment your car came with,to compare numbers etc. see:
http://www.automotive-tradition.de/en/wissen/recherche.htm
...can be really tiresome,the quality of .pdf`s is far from good...

Greasers use it to overcome the thickness of their improperly prepared and poorly heated "fuel". With the lift pump functioning properly there shouldn't be any need for a booster pump.

Glycerine filled 0-30 pound mechanical gauge used at idle on base fuel in injection pump. Fairly steady but needle vibrates usually. It is doing a good job at averaging things. The high pressure spikes are so fast an event in duration it can almost totally ignore them.

Mechanical 0-30 pound non dampened gauge at idle.
it does a better job of tracking actual activity in the base of the injector pump. It will try to destroy itself. It of course is still trying just to average activity as well. You cannot usually read it as it tries to destroy the needle against the stops. If nothing else this proves there is not one constant pressure in there. Instead there are quite dynamic events happening. I was getting a little concerned the common viewpoint was that of a simple fixed pressure.

Lift pump is a constant pressure pump. Pressure it supplies to the mix still falls away through the injection pumps cycle until it restores itself again after all the elements have been serviced sequentially. We cannot measure this event. The higher the indicated base pressure the higher the flow/ steeper decline profile as injection pump overflow is greater.

If this is true the injection pump will not remain properly calibrated with a fixed maintained source pressure replacement for the lift pump. It must be calibrated to existing conditions.

When this injection pump is calibrated at a service shop I would think either its own lift pump or a substitute of the same item is probably used.

If the lift pump is substandard. It to me is reasonable to assume the injection pump calibration is no longer on its original lift pump profile. Or that portion missing of the actual real senario in the base of the injection pump is changing things.

Up to 19 pounds lift pump pressure all is well or close enough. Injection pump is still close enough to its probable originaly intended calabrated profile. Higher pressures may change that to some extent.

There is either no more or less smoke usually reported from people elevating their lift pump pressure to 19 pounds and verifying feed system has no abnormal restrictions.

This is a solid indication we are just not dumping more fuel in. The smoke would be greater. We are in fact of course getting more fuel in though. My suspicion is we are getting it in more equal quantities per injector as well.

Now we come to the real issues. Done properly fuel milage is increased over previous much lower feed pressures. Same or less smoke is produced, power is up. So it is a given that efficiency is really improved. . This of course has to mean the apparent timing between elements combined with/or overall timing has been improved.

When trying to understand what is really occuring in the base of the injection pump there are lots of events. Intense pressure spikes. Integrated with the lift pump supply that is not in itself of constant pressure. Combine this with almost instantainious reflected pulse waves etc.

So if all this is so it is perhaps better just to get the lift pump operation and any fuel supply restrictions dealt with by installing a pressure gauge and call it a day.

I feel and have felt that what is occuring in the base of the injection pump may be too complex to easily understand totally. The need to know is not absolute to our needs either.

If you cannot sustain a proper base pressure at cruise. Either the lift pump is tired or restrictions exist.
if these are dealt with properly the car should produce less smoke. Get the optimum fuel milage possible. Allowing for the other conditions that may be present and typical of a very old mechanical piece of equipment. The same logic applies to equalising power strokes as much as practical in a given situation.

My viewpoint the number one rod bearing is worked harder at substandard injection pump base pressures is still a healthy suspicion of mine and will be proven one way or another by testing eventually I believe.

These same engines may have been pushed harder in certain european countries and the intense heat of the middle east. The 616 and 617 engines where in production too long for mercedes to have not improved the oiling supply to the number one bearings if it was indicated. Back then their engineers where pretty conciencious. They suggested their engines could be run wide open without harm.

Their high quality oiling system was not modified to my knowledge from the introduction of these 616 and 617 engines from start to finish of production.

The 616 seems to experiece a greater percentagewise number one rod failures than 617s. Yet the oiling path to their number one bearing is shorter. Now if we can believe that apparent element timing/or and/fuel quantity in relationship to base pressure is involved. Combined with fewer power strokes/ Less frequent events. It is starting to become interesting.

The design stress is almost certainly less with no turbo and normal loading on the 616 per cylinder. The rear end ratio is higher to compensate somewhat for a simular weight car as the 617.

The 617 has a turbo usually, longer oil feed path to the number one bearing. Higher average power loading per cylinder still with the lower rear end ratios. Yet the number one rod events seem infrequent compared to the 240d on site and the 616 certainly is found in much lesser numbers.

More of my speculations prove nothing really on the rod issue. There are more. I just preffer to make sure any 616 I own is going to have proper functioning injector pump base pressure. It costs next to nothing to have and has no downside I am aware of.

Hi Barry, The problem with a non-damped gauge is that the gauge's own inertia will cause it to over swing, and amplify the pressure changes beyond their true extent. I wouldn't pay too much heed to it - all it is telling you is that there are dynamic pressure changes, you cannot use such a gauge to infer anything about their magnitude. As far as I can tell, we have good reason to think that if the IP relief valve is removed, and the system works at no pressure, then the engine will run poorly. However, I havent seen anything that suggests any measurable difference in operation between running at, say, 15psi and running at 20psi.

I think I remember one member posting that his non dampened gauge was banging against the top pin. I think his lift pump pressure was also low when he switched to a glycerin gauge. Not near zero just low.

The non dampened gauge should have tended to hit the lower pin even easier as well. Maybe it did. I cannot remember or find the post. If it was not in my opinion it would indicate those pulses are there.

The only easily obtained proof I can think of that they were the reverse pulse charges. From from the elements fill ports as they were closing off. Is testing.

Most if not all piston/cylinder type injection pumps have the reverse discharges I think. Just the last bit before the port closes. Intense high pressure spikes are produced. The pressure has to be quite a lot higher than the lift pump provides. Or the element ports would not discharge back into the base.

As the lift pump pressure is incrementally increased that may also be a way we are getting a little better element loading profile as far as quantity goes.
The return pulse is defeated a small amount.

Reduce the feed pressure at idle to zero. Hook up a mechanical non dampened gauge. If it is still hammering it's stop or stops we know. As a poster recently mentioned it is time for some assorted tests.

This may also in some fashion answer how the injection pump can function with zero injection base pressure or close to it. What is filling the elements specifically then? There is not enough time for a very good gravity feed. A small portion of what remains of the reflected pulse residues.?

Some pumps apparently provide harder discharge shields to eliminate errosion problems of the injection casting they are so intense.

Cervan pulled the relief valve spring and lost it. Litterally and figaritivly I imagine at that moment . Car still ran. Pressure from the lift pump had to be low as proven by the gentleman again with the valve in the return line. He reads zero with his return port open and no spring. For some reason the various tests to establish things are not being conducted.

Again since the reverse pulses may contain a small amount more fuel in the discharge as basically the loading pressure has been reduced. .Or even less fuel as the elements did not fill properly.. This in combination perhaps could cause other effects.

No hammering gauge no element pulse discharges. The non dampened gauge may be even under reacting if the frequency is high enough. At seven hundred rpm is it 1400 pulses per minute? Or about 23 pulses per second on a 616 injection pump?

The best test might be at cranking rpms other than at the slower speed the pulses may be much smaller in amplitude. I have to wonder if the reflecting pressure waves experience any delay adaquate enough to be helping the elements in any way at very low supply pressure? If the relief valve is closed. These are events of very short duration. Information in this area is very sporatic.

You know this is pretty complicated for a simple guy like me. Simple tests would allow me to understand things better and perhaps others if done and posted.

Then we might move on to methology of increasing milage on these old indirect direction diesels. It of course is basically simple in concept.

A way to increase efficiency that does not involve the injection pump. I have a pretty good ideal I believe with a chance if we all contribute. Need this base pressure thing out of the way first.

What I call the effect for want of a simple term is already present in another indirect diesel engine. I had known of it for years but thought it could not be developed because of certain limitations. I found a way around them if what I believe is true.

It should be looked at for the entertainment value alone. I believe there would be some opinions and a few disscusions. Anyone want to get the injection pumps base pressure tests done and out of the way?

Anybody test the new spring as of yet????

The one I ordered has not come yet. Is anyone else trying a different spring?

Hi Barry, I think the pressure pulses from the closing spill ports are rather interesting - it's effectively a water hammer type source - there's some flow velocity of diesel out of the spill port which is suddenly stopped as the plunger rises. I think these more acoustic (in terms of their frequency content) pressure fluctuations can't last long enough to have any gross effects on element filling, but might have local effects in terms of material errosion. What I'm not sure about is how well the (comparatively!) large flow of fuel out of one spill port is balanced by flow into another - doubtlessly better on a six than a five. I think the damped gauge shows whether or not the lift pump's working, and whether or not the fuel filters are clogged. Beyond that, I'm not sure that mechanical gauges are much use. Fast measurement, probably electronic, would be the best way to see what's really happening.

Glycerin gauge, air chamber after main fuel filter. I installed a check valve before the air chamber ( might not be needed ) as the lift pump has one in it.
I made the air chamber out of 2" schedule 40 pvc pipe about 4" long with 2 end caps. One end tapped for a 1/4" pipe thread by 3/8 hose barb. The gauge is all
most vibration free. Hope this helps.

__________________

1984 300SD 175,000 Miles
1987 GSXR 1385cc
1985 GSXR Limited Edition 750cc
2005 Suzuki Burgman 650cc

Did the spring thing. It did help the idle some. Did the air chamber and it helped some more. New rack screw and it helped some. All together the shaking is almost gone.

__________________

1984 300SD 175,000 Miles
1987 GSXR 1385cc
1985 GSXR Limited Edition 750cc
2005 Suzuki Burgman 650cc

Some inline IPs have steel plugs across from the fill holes in the Elements Barrel to prevent erosion of the IP housing from the Fuel expelled out of the fill hole before begin of injectin.

__________________
84 300D, 82 Volvo 244Gl Diesel

I guess as at least most owners are now aware. Having reasonable pressure in the injection pump base is somewhat desirable over very low pressure.

If that point has become common knowledge it hopefully is a step in the right direction. Restrictions anywhere in the fuel system as well as weak lift pumps and weak return valve springs will be the usual causes.

I will start a new thread on observations. Just to see what people think about them relating to fuel milage.

Dad and I were discussing this last night....

In all the years that we worked on these cars, I do not ever remember checking lift pump pressure, so we are both very interested in this thread. However, there is something that I wanted to post in relation to this.

I have read where guys are changing the spring in the lift pump, but wouldn't the relief valve on the IP be what governs lift pump pressure????...Robert