Fuel Management

Here in Saint Louis there are a lot of fleet vehicles that run on natural gas. Natural gas is more expensive, and these vehicles don't have the range that gas burning vehicles do, but one nice thing about these engines is that they burn so much cleaner that they last a lot longer.

If your engine is running rich, extra fuel is getting into the pistons, diluting the oil, and allowing premature engine wear of some very expensive parts. It's also possible for the oil to get diluted in the MFIS injection pump, and again allowing the engine to wear prematurely because the oil is not properly lubricating the engine. I had this happen to me on a GM vehicle that I once owned. It was running rich, I didn't worry about it, the next thing you know, it didn't have much compression!

The EFI systems supposedly do a better job of being efficient with the fuel, and Karl is able to set the idle mixture very low, but the EFI system doesn't get better gas mileage than one of Dan's properly tuned cars? So, basically, is the injection pump (in top tune) capable of metering gas properly enough that no gas is wasted? Or, does the EFI do a better job?

Peter in St. Louis
6.3 #1349

Peter, great questions. First, a perfect Mechanical Fuel Injection System (MFIS), is simply a work of art. I've had restored MFIS that were far from perfect, thus, just because a guy has one rebuilt, you never know the outcome.
The only perfect MFIS we have is on the Bullet. Everyone of ours have been rebuilt, some twice, but all have minor flaws. Those flaws, are only noticible if you have other 6.3s to compare.
That said, a perfect system is exact, to within one gallon of fuel, when ran side by side to Karl's EFI system, in a trip of 2,000 miles to California. Exact speeds. Exact amounts of power on starts.
Get that. Karl's system compares to a perfect system. Now that is saying something.
OK, easy to see why we installed one, is it not. I simply was damn tired of rebuilding the MFIS is search of perfection. BUT, it is WAY DIFFERENT, way different. It is smoother. Does a gy bwant smooth on a 6.3 or brute force? That is the question, as Hamlet should have stated. If a guy is close to Dale Cherry, it's a snap. For my old fashioned Benz guys, it was tough.

EFI does a better job. The Bosch inline pump depends upon a moving 3-D cam for proper mixture control across the operating regime of the engine. Compare that with any modern EFI's multi-contour fuel and ignition maps.

It's like a Swiss mechanical watch compared with some cheapy quartz plastic wristwatch. Technology trumps craftsmanship in this case, sorry to say.

quote:

---

Originally posted by Mark in SBA

EFI does a better job. The Bosch inline pump depends upon a moving 3-D cam for proper mixture control across the operating regime of the engine. Compare that with any modern EFI's multi-contour fuel and ignition maps.

It's like a Swiss mechanical watch compared with some cheapy quartz plastic wristwatch. Technology trumps craftsmanship in this case, sorry to say.

---

Regarding the watches,i watched a TV show in the UK called "the secret life of machines" where they explained the $5 swatch was the most accurate watch ever made and yes you can get the EFI to run perfectly ,which is my ultimate aim for my 6.3 .
The only real design issue is the position of the injectors in relation to the intake port,On the 6.3 they are right up in the runner away from the intake valve pocket,as opposed to the 6.9 where they live right above the intake valve itself.I suppose a braver person than me could drill new injector ports into the heads to bring the Electrnic injectors right down to the valve pocket.This would have to release a few more horses.

...There are old cars,and then there are Classics..(Mercedes Benz Ad. 1999)

Dan quoting Hamlet, priceless!

Way to go Dan.

Francis E. Abate
1970 6.3 No.3791
1968 6.3 No.0250 (Parts)
1998 ML320

Ron,

Don't the mechanical EP/DEC injectors take a 90° turn down the intakes and into the head? These extensions look pretty long (#5 in the diagram):



Funny how the more modern 6.9 uses a continuous flow injector while the vintage PES 8 injector pump is fully sequential, similar to the most modern EFIs. You'd think they'd move the CIS injectors way up the intake to promote better mixing while the intake valve is shut.

What's old is new, I guess.

Brutal is a word often associated with this motor, but it does indeed require more fine tuning than a Swiss watch will ever need.

Mark, obviously, given that the injection timing is dependent on the cam, are you suggesting that the cam cannot address exact fuel metering needs at all points of the engine's range? I hadn't thought of that....the EFI would be able to continuosly adjust - can it not? Whereas the cam has fixed timing that it would use at idle, pulling a hill, cruising, or accelerating. Can EFI adjust continuously, in the same way that these modern engines have variable cam timing in the heads and variable intake runner lengths? It was my understanding that they have set fuel mapping, or whatever it's called, and while you can update that, it is fixed and not capable of adjusting infinitely.

Dan, your response obviously begs the question 'what is the difference between the systems?' Is it simple mechanical differences in the moving parts? Different rebuilders? Different injection pumps? Or, have you not figured out what the differences are? I know you have swapped around governors and possibly other parts to get an ideal transmission. Is it the case that you can send two of the same pumps to the same rebuilder and they will come back and perform differently? Or, are there other factors such as chain tension, or ignition systems that might be outside factors?

As we have discussed, these motors can idle perfectly smoothly, and simultaneously be capable of tremendous power. I know that a common cause of rough idle is hot air and hot fuel, but no doubt worn injection pump cams, injection pump timing, distributors out of whack, and worn chains also affect the idle too.

Good performance is obviously a benefit to having the injection pump/etc. properly tuned, but having the thing run rich and dumping that extra fuel into the motor is a serious concern. The parts and labor to rebuild these engines is skyrocketing, and it was never cheap. Having them wear prematurely by diluting the oil with fuel is insanity. It would not take long at all to seriously destroy the rings in this manner. I know that it surprised me when this happened to my van, but what was I thinking? The damn thing was getting about 10mpg, when it should get 17 to 20mpg (miles per gallon, the US way of measuring fuel economy). So, I thought, what a bummer that it's getting bad mileage, but I didn't realize that it was washing out the rings. When a friend of mine changed out the oil for me, he told me it was heavily diluted with fuel. I fixed the running rich problem, and it ran much better and had much more power, but not the kind of power it once had. Climbing hills it would downshift, etc. Basically, the engine was shot. Fortunately, it was only a cheap GMC van, and I bought another van in top working order for little more than the price of a set of M100 pistons. I also sold that one for more than I paid for it. I would be hard-pressed to pull off that kind of a deal with an M100. By the way, the purchaser knew the condition of the motor, I didn't try to hide it.

So, I take it from this conversation so far, that while EFI is more efficient, the MFIS is capable of metering fuel properly enough that a clean burn is created?

Dan, does the smoothness affect the power? I know that you said the same thing about electronic ignition, and again I would ask if the smoothness afforded by the electronic ignition robs power, or just gives a different feel?

Thanks,

Peter in St. Louis
6.3 #1349

The MFI dopesn't use the cam in the pump to meter the fuel,it uses the relationship between the throttle position and the pump govenor mechanism to do this.Thats why it is so important to get the throttle plate opening tuned to correspond with the pump.The cam only determines when fuel is pumped.
As the govenor moves the rack,the fuel elements turn in the pump housing opening little ports which in turn allow more fuel as the rack is pushed forwards.The ports are open to the fuel gallery in the side of the pump housing.
The MFI is an extremely accurate setup ,but EFI is easier to tune with a laptop and the oxy sensor tells us exactly what is happening to the fuel charge.This is adjustable on the MFI too,but involves removing the left hand manifold and getting in the back of the pump...not easy.
With regard to the injector positon,on gasoline engines it's far more desirable to get the injector as close to the valve head as possible,this enablews a fully atomised charge to be delived with no worries about droplets etc which cause unburnt fuel.On the 6.3 I am sure that it would have better if the injectors had of been placed as per the 300 Sl engine with the injectors in the head,but i guess we will never know.

...There are old cars,and then there are Classics..(Mercedes Benz Ad. 1999)

Peter and Ron,

The "cam" I'm referring to is the LP/RLA governor 3-D space cam on the back of the flywheel governor, item 13 in the diagram below, used for mixture control:



This feature permits precise fuel delivery over the entire load range of the engine, integrating engine speed, load, temperature, and barometric pressure using a complex set of levers, springs, and shafts to dynamically adjust the rack. As engine speed picks up the flywheels open up due to centrifugal force moving the space cam left to right and non-linearly bias the rack position.

You are correct as to the "camshaft" (item 11) providing timing correlated to the engine, and obviously delivering the pump stroke to the individual plunger units (item 7).

Peter, as to why EFI is preferrable, keep in mind this mechanical computer has several wear points and springs, which tend to wander slightly over time. No such change with EFI, particulary when combined with adaptive Lambda control.

Also, EFI can integrate several more parameters into engine control. Instead of one 3-D "cam", figure several equivalent "maps" can be simultaneously integrated using silicon:



Don't get me wrong, I'm a big fan of the Bosch MFI as a historical artifact and representative of the ultimate fuel control at the time. Amazing technology and precision.

Priceless information.

Thanks, all!

Peter in St. Louis
6.3 #1349

The adjustment for the fuel range is picuted there,(#15) ,the idle mixture is in the center but the four out side screws control the mid and upper range.To get at those involves removing the left hand manifold and then unscrewing the big nut on the rear of the pump.it's almost impossible to get at with pump on the car...and unless you are using a gas analyser theres no way to determine you are doing things properly.
all this is why I am beginning to build my EFI ...that, and the prospect of paying $3.00 per liter for fuel...

...There are old cars,and then there are Classics..(Mercedes Benz Ad. 1999)

Peter, enjoy your insightful questions. In this case, you quessed it, I am not sure of the WHY are various MFIP so different, but I have a strong hunch. That hunch keys around the various letters in the part numbers. My best MFIS pump is a W. We have no idea on how it ended up on the Silver Bullet which in all of it's history, it has resided in Oregon. It is the pump we sent down to the MFIS rebuilder which I describe as the slow guy, meaning it took them about three months to get it done. The other rebuilder, also in California I describe as the fast guy. No other 6.3 we have running has a W pump, go figure. But, we did find a very decent rebuild subject on a very nice rust free Euro model and we had it rebuilt and installed on another restoration. There is a big BUT. The BUT is that I think this W MFIS pump only gives you extra punch if you have a 1969 9 to 1 ratio 6.3. But again Peter, all this is a theory developed in messing with six 6.3s. The gutless American 8 to 1? Forget it. It cams and pistons don't seem to match up with the W MFIS.
SO, my theory is that the W pump rules, if combined with a 9 to 1 spec rig. We are in the process of starting a new experiment, awaiting a solution to a rebuilt tranny's problems. First we will need to slowly bring the engine into about 500 miles at various speeds, before the time trials start.
On the other parts. We always use new spockets, chains, and tensioners. The guys from OZ know this and proved it. This is as important as the comp ratio and a perfect MFIS. Without all the key parts running correctly, it's all a waste. All must be replaced with new parts. It might look OK, but how can they be OK if a M100 has even over 100,000 miles. They have that 100,000 miles of wear.

Dan, I'd say that the key factor is the 'W' pump. Fortunately, I have 3 so far. My interest in a 6.3 goes down as soon as I figure out it's an 8:1 car. Harsh, I know, but true.

What you say is exactly right - pump and cams etc must all be in harmony, but the Bullet's cylinder head cams are not OEM spec, yet the engine is still in harmony with the pump? So, I guess the pump has a broader range of adjustment than I suspected. Your modified cams must still have the exact same valve timing, but more duration and lift. I do know that the US cams have retarded timing, which would put them out of time with the W pump.

So much to know.....

Peter in St. Louis
6.3 #1349

Hi,
Can somebody please help me,how do i lean the mixture, my car is running rich.
Thanks
Stefan

Stefan,before you strat attempting to lean out the mixture,do a couple of things fist;
1/ on the rear of the injection pump is a 13mm hex bolt with slot in it.Thats oil dip stick for the governor.Undo it a check that the oil is not too full.Yu can drain it a bit by sticking a plastic hose into it and sucking the excess oil up the hose...slow but effective.
2/ensure the aircleaner is clear and not clogged up.
3/ the ball joints on the acelerator linkage are in good nick and not loose.
4/ Ensure the two cold start valves on the rear of thec intake manifolds are sealing properly and not dripping fulinto the intake .
5/join up and gain acess to the members area where the process of tuning the engine is clearly explaned in detail.

Use a CO meter to determine if the richness is at idle or throughout the rev range.
Only idle mixture can be altered,by switching OFF the engine and turning the thumb wheel one click at a time until it leans out.
If none of the above fail to cure rich running,the pump governor needs an overhaul as mentioned earlier in this topic.

...There are old cars,and then there are Classics..(Mercedes Benz Ad. 1999)

Dan, do you have any rebuilt "Z" pumps?

Chris Johnson
If you aren't constantly IMPRESSED with your car, then it needs fixing.
100.012-12-000790
109.018-12-000019
109.018-12-003834
www.300SE.org

My fuel mixture was very rich when I bought my 600; I replaced all the filters plugs oil etc, and have decreased the mixture at the back of the fuel pump. The manual says the you should only move it 6 notches, but my mixture still seems rich. Can you move it more than 6 notches, what is other options?

Stefan,
The knob on the back of the MFI only adjusts the idle mixture. If you have checked all the other things that Ron listed back in this thread, them the MFI pump should be sent to a reputable shop for checking and possible rebuild. If the car sat for any length of time, components in the pump may be siezed causing the problem. It is not something that a home mechanic can do. If fact it is not something that most diesel MFI shops can do either unless they have the specs for the petrol MFI pumps and appropriate test gear. The other range adjustment screws that Ron mentioned are easier to get to on a 600 than a 6.3, but whoever adjusts them on the car has to know exactly what he is doing or the outcome will be a disaster. It is also essential that none of these things are adjusted with the motor running, because the outcome of that is a guaranteed requirement to rebuild the pump.
Art

I want to add a theoreticly and practicly extra adjustment for the complete range of the mixture without opening the MFI. Those MFI's are the top of german engeneering at that time. The EFI should not be better, as Dan's experiences proove. Also is well known, that those MFI's could live longer than 2 complete Motorlives, so good were they made in that time. There are several owners that take the barometic Cell off for cleaning, while it looks better and then put it back in a near hight as was, measured im mm, BUT different hight on the MFI. So it could be some-one took it off before and put it back as near as the could measure in mm, but way off from the soll-wert-settings.

If you "re"-adjust the barometric Cell, the mixture will be automaticly re-adjusted. Turn it out and the mixture must be leaner. You should measure now the CO2-values at different RPM's so it will / should be imho function bearly only with a dynotest.

But it is clear you can lean out you mixture through this simple and effective way.
What you could do is "mark" the position of that barometric Cell and turn it every time some Degrees out. Test the car with an extensive drive and CO2-measurments. But do that only when all other itens are adjusted like it should first !!!
One turn has an immens impact !!! so everytime with some Degrees.

quote:

---

Originally posted by Ron B

The adjustment for the fuel range is picuted there,(#15) ,the idle mixture is in the center but the four out side screws control the mid and upper range.To get at those involves removing the left hand manifold and then unscrewing the big nut on the rear of the pump.it's almost impossible to get at with pump on the car...and unless you are using a gas analyser theres no way to determine you are doing things properly.
all this is why I am beginning to build my EFI ...that, and the prospect of paying $3.00 per liter for fuel...

...There are old cars,and then there are Classics..(Mercedes Benz Ad. 1999)

---

That is theoreticly a way of adjustment, but the upper range versus the lower range have to be exactly as it should like the soll-wert.
I would more prefer the way of changing the complete range be re-adjust the barometric cell, so the mixture will be unchanged between the ranges for low and high. And that is of course also much simpler to do on a mounted FIP.
Greetings from Holland

Picking up a very old thread here --

In 2008 Paul wrote: "I would more prefer the way of changing the complete range be re-adjust the barometric cell, so the mixture will be unchanged between the ranges for low and high. And that is of course also much simpler to do on a mounted FIP."

-- but I could not see any feedback on this proposal. Did anyone try this, and did it require adding more shims ?

My MFI pump was restored several years ago by an expert company in Germany and they provided a bunch of papers pertaining to the calibration, but those papers are not easy for me to interpret. The barometric cell (or altitude meter) has sealing paint at the threads so I am a bit hesitant to try adjusting it.

What I do know is that my 6.3 engine is running rich, even with perfect ignition timing and dwell, new ball joints in the throttle linkage, carefully adjusted rods etc.

please don't touch the altitude box. Those are NOT self-(mis)adjusting .... so why should this box suddenly not any more correct been adjusted .....

CHECK first all the other possible failures, especially LEAKING coldstartvalves ...

OK, Thanks Paul!

The cold start valves are (I think) the only parts I have not yet checked. Will do this as the first thing before the upcoming driving season.

In case they are leaking, is there a repair possible or do I need to find new valves ?

as far as I know rubber rings could leak petrol. switch those rubber rings ....

Today I checked the cold start valves, not a drop of fuel came out so I can say they are not contributing to the somewhat rich fuel mixture. Guess I will turn the idle mixture screw a couple of klics out.

I do have a CO exhaust tester so I can make sure the idle mixture is not too lean.

Stefan, have you checked the other possible causes of being rich that where listed previously?

The fuel injection system is the equivalent of a carburator, and the same rules apply: verify everything else is perfect FIRST before adjusting the pump because EVERYTHING effects carburation.

How is the oil level in the pump? Are the cold start valves 100% leak free? Is the linkage properly adjusted (very critical)? Are the balls and sockets in the linkages free of worn parts? Is the ignition timing operating and advancing properly across the rpm range?, etc., etc.

I currently have somewhere around 40 MFI Benzes, and I would guess at least 70% had problems effecting carburation that was NOT related to pump adjustment.

It is also true that most less-than-knowledgeable mechanics do all sorts of silly stuff to these MFI cars when trying to balance out fuel mixture, idle speed, and addressing of a dead or weak spot when accelerating away from a stop. These silly things include tweaking the idle screw on the back of the pump, and it is indeed then wrong. A lot of aggrevation comes from trying to adjust the mixture first and then eventually going through the other problems one at a time, only to have the mixture wrong again because EVERYTHING effects mixture (carburation).

Perhaps we, as a group, should put together a very detailed and specific document on everything related to carburation, what wears, how to adjust, how to troubleshoot, and when to rebuild an M-100 MFI pump. It should also point out where the discrepancies in the factory documentation are, etc., etc.

To your specific question, yes you can turn the idle screw more than six clicks, but be SURE to keep an accurate record of how far (and which way) the screw has been turned. The warning in the manual ASSUMES that the pump was very close to right to begin with. If some monkey turned the screw twenty clicks to the rich side (as an example), it is perfectly acceptable to turn it back. But it is still necessary and appropriate to check everything else FIRST so that you have a valid reason to believe the pump is actually out of adjustment.



Chris Johnson
If you aren't constantly impressed with your car, then it needs fixing.
100.012-12-000790
100.012-12-000867
109.018-12-000010
109.018-12-000019
109.018-12-003834
www.300SE.org

Chris, only 40? Have you been cutting back? -CTH

Well, certainly the rate of escalation has tapered off! The only thing I remember for sure is that there are precisely two 600s. I think it is five 6.3s plus a spare drive-train, and I know it is something over thirty 300SEs, and then there is a bastard child or two. I need to count again on the next trip to the hangar.

It's gotten to the point where even I will admit that things have gone to the extreme. I actually passed on a 300SE sedan in CA a couple weeks ago.

Chris Johnson
If you aren't constantly impressed with your car, then it needs fixing.
100.012-12-000790
100.012-12-000867
109.018-12-000010
109.018-12-000019
109.018-12-003834
www.300SE.org