This is rather an odd one. It concerns the K100 fuel pump.
I have an early K100 which is fitted with a Mercedes Smart Diesel engine.
The original in tank pump circulates the diesel fuel as it did in the bike and feeds the high-pressure injection pump.
 
It is running at about one bar.
 
It has done about 800 miles or so with this setup without a problem but recently, after running for just a short period of time (it first manifested itself on a run out) the bike started to slow and eventually cut out.
 
I discovered that switching the fuel pump on and off (the kill switch does that) it’s okay again for a mile or so and then it starts slowing down.
 
As the problem seems to resolve after the fuel pump gets a bit of an electrical kick, I am wondering whether there is a problem with the fuel pump. I know it wasn’t meant to run on diesel fuel and nor was the filter (new) and I can always make other arrangements but the reason for posting is to ask if there are any known problems with these pumps. I can’t imagine that diesel would cause problems where petrol will not although of course, the fuel pump is having to work harder because the fuel is thicker.
What is surprising me is that the fuel pressure is not dropping.
 
The other thing to ask is whether there are different sizes of pump or whether they are all the same.
 

Thanks for the answers guys.

You say you are using the original fuel filter?  If I am not mistaken, diesel fuel uses a different filter media and construction due to it's higher viscosity.  I suspect the fuel pressure is displacing the media in the filter and blocking the flow.  shutting off the pump allows the media to rebound before the next time the pump starts.

When you say running at one bar, the in tank pump should run at about 60psi [for Brexiteers] and the K fuel pressure regulator [FPR] runs it at about 36psi if you still have it fitted. If you are running the in tank pump without the FPR it should be producing more than 1 bar, more like about 4 bar or have you fitted an FPR with the Smart engine?

Second is what pressure does the Mercedes high pressure pump need the incoming fuel to be at to correctly run?

I am also wondering does the K pump arrangement cause frothing of the diesel that might affect the running of the engine.

Pumps in Ks as far as I know are same for all 8 valve Ks and run about 36psi at the injector rail.

Mercedes plumbing between the diesel fuel tank pumps and their high pressure diesel pumps was a well known issue for causing air locks. Been there on that one.

That's a smart looking conversion.

Aha, the collapsing of the media in the filter may have some merit. Although I did cut one apart once and they are very substantial inside.
It’s the fact that the pressure doesn’t drop (I have a constant pressure gauge on it) which is bafflingly somewhat.
 
Without the pressure regulator it is virtually open circuit. It runs at about 0.1 bar.
However I have a restrictor in the line and a pressure gauge and I can make it run at anything I like.
 
The engine will run quite happily and did for many miles) with low-pressure fuel at about 0.25 bar although ideally it needs to be about one bar.
 
I thought of frothing in the pump myself but I don’t know how it would generate frothing if the pump was totally submerged.
 
Remember also that it ran for several hundred miles without a problem.
 
If you want to look at the complete bike building thread for interest, it is here:
 
http://www.suckindiesel.com/thingy/viewtopic.php?f=31&t=3216&sid=f5305209373bd3ebeff9c9fe5e5c7824
 
and a video
https://www.youtube.com/watch?v=f_TBvZLK5oY&t=120s
 
I have a C220CDI Merc and if ever it runs out of fuel, it is a Demon to bleed.
 
What is the bike in the picture? It looks really neat.
 
It seems that there are 2 different diameters of pump.
 
http://www.ebay.co.uk/itm/Bmw-K-Series-Fuel-Pump-K100-K75-K750-1000-1100-43mm-Direct-Fit-1983-1999-New-/231424766266?fits=UKM_Make%3ABMW&hash=item35e1fdb13a:g:3q8AAOxy039TNDB4
 
Thanks for thoughts guys

The first place I would be looking is the filter as it may have taken some time for the diesel to get at the filter medium and the next place would be the rubber seal where the alloy body is crimped to the top of the pump, the diesel may have destroyed that seal and you are losing a heap of fuel out there.

What’s confusing me about any potential loss of pressure here is that if I squeeze the return pipe, I can set the fuel pressure at virtually anything I like.

I wonder if any materials (pump internals, fuel filter internals)deteriorate more when they are immersed in diesel fuel than in petrol. I would have thought that Petrol would be more problematical.

All the components were designed to be used with petrol not diesel and it is quite likely that the diesel would react with them or cause them to swell or break up. Diesel has some amazing penetrating properties it may even be causing the glue that is part of the paper fuel filter to lift away from the body of the filter.
Check the internal surfaces of the lines, you may even need some special lines to work with diesel.

RicK G wrote:All the components were designed to be used with petrol not diesel and it is quite likely that the diesel would react with them or cause them to swell or break up. Diesel has some amazing penetrating properties it may even be causing the glue that is part of the paper fuel filter to lift away from the body of the filter.
Check the internal surfaces of the lines, you may even need some special lines to work with diesel.

Thanks Rick.
I do actually have lines with Fuel printed on them, so fingers crossed, Fuel is multi-substance.
I guess I will just have to replace a bit at a time see which one works.

Have you checked for water in the filter? Does the system have water separation.

Laitch wrote:Have you checked for water in the filter? Does the system have water separation.

I haven’t checked for water in the filter although I guess that’s easy to do when it’s removed. A new filter would resolve  that in any event. I’m not certain how turning the fuel pump off and back on again would relate to a potential water problem.
Bear in mind also that the system is brand-new.
I would have thought that any water would have settled in the seams at the bottom of the fuel tank (where the tanks rot).
There is no separate water separator.

touchwoodsden wrote:I would have thought that any water would have settled in the seams at the bottom of the fuel tank (where the tanks rot).
There is no separate water separator.

If that were the case, diesel engines wouldn't have water separation systems, would they? Some are dedicated; some simply consist of regular filter changes. Intermittent performance is a symptom of many conditions; one of them is water contamination.

You've undertaken a mighty project and had pretty fair results. If this idea doesn't conform to your thinking, I'm not the person to contradict you.

Laitch wrote:

touchwoodsden wrote:I would have thought that any water would have settled in the seams at the bottom of the fuel tank (where the tanks rot).
There is no separate water separator.

If that were the case, diesel engines wouldn't have water separation systems, would they? Some are dedicated; some simply consist of regular filter changes. Intermittent performance is a symptom of many conditions; one of them is water contamination.

You've undertaken a mighty project and had pretty fair results. If this idea doesn't conform to your thinking, I'm not the person to contradict you.

I am not contradicting you Laitch, far from it. I am grateful for any suggestions and water wasn’t something that I had considered.
It’s the intermittent nature of the problem which is confusing me and the fact that if I turn the pump off and back on again, the problem disappears for a short period of time.
I can’t help feeling that the problem is somewhere in the low-pressure fuel system fuel, (all changed), pump, filter, fuel lines. After all, there is no more to it!
If I change all that, one bit at a time, it either eliminates the problem or moves it somewhere else.
As you have said, it’s a mighty project and I suppose that replacing the fuel filter and the pump one at a time is relatively minor compared to everything else I have had to overcome. I just hate grovelling around inside the fuel tank and having to drain it each time !

touchwoodsden wrote:I suppose that replacing the fuel filter and the pump one at a time is relatively minor compared to everything else I have had to overcome. I just hate grovelling around inside the fuel tank and having to drain it each time !

Cure by part replacement isn't too much fun either.   How have you tried to track fuel pressure and restriction?

Laitch wrote:
Cure by part replacement isn't too much fun either.   How have you tried to track fuel pressure and restriction?

Indeed. One thing at a time, otherwise we don’t know what the problem is!
I have a fuel pressure gauge permanently fitted in the return line and I have a restrictor which I can adjust fitted to the return line immediately after the gauge.
The guys who manufacture the ECU have looked at the log and they are convinced that its fuel pressure related because the high-pressure pumps either work or not at all.

touchwoodsden wrote:It’s the intermittent nature of the problem which is confusing me and the fact that if I turn the pump off and back on again, the problem disappears for a short period of time.

I got this type of problem on the K100 with the chair. Once in a while when riding, I would lose power (The bike would run fine on idle, but would choke when opening the throttle).
Stopping for a 5 min break would have the bike running fine again.
I bought a cheap used fuel thank with all the internals (pump and filter) and it solved the problem. The reason for a new used tank was that it was cheaper than buying a new pump .

If your filter is disintegrating then the intermittent bit makes sense because the pressure will push the loose bits into the flow path and when you stop it will become unblocked then when you restart because of the low flow due to the restriction you have installed then it will take a few minutes to block the path again.
I would not run that type of filter with diesel but a combined filter water trap like a CAV. Any diesel I have owned I have always run 2 filters one with a water trap and a second just a cartridge type like an oil filter.
http://www.ebay.com.au/itm/CAV-Dual-Filter-Housing-Glass-Aluminium-Marine-Boat-Tractor-4x4-Diesel-Car-Truck-/222368057806 Doesn't mention motorcycles but I am sure it will do the job.

I actually have a new filter on the shelf so I’m going to try that first.
If it solves the problem, and then it manifests itself at some stage in the future, then I will have to use an external filter. I’m tight on space so will need to find something which is relatively compact. The CAV filter with the glass bowl I think just allows you to see the water because the ones without the glass bowl still have the bowl at the bottom albeit it is aluminium. The ones I have had also had a small drain screw.
 

If the new filter doesn’t resolve the problem even for a relatively short period of time, the next big is the fuel pump and if that doesn’t work, I think a gallon of petrol and a zippo lighter!

It'll be good to hear how you fare with the new filter...although, I suspect, if it is still one designed for petrol systems, you will be back in your fuel tank again before long having suffered the same issues again.

While you'd think that petrol is more aggressive than diesel, diesel does have some nasty additional issues associated with it, all of which can lead to filter/system blocking, which ultimately leads to fuel starvation. For example, automotive diesel often will contain Fatty Acid Methyl Esters (FAME) which is a bio fuel variant (can be as added as high as 10%). In the marine medium grade distillate fuel world we specifically limit the allowable FAME to 0.5% as it causes all sorts of problems with elastomer seal swelling and also has a habit of cleaning years worth of crap out of the system - which then blocks the filters. We would like 0% but with product pipeline configurations in the refineries, we can never get a guarantee that there won't be some cross contamination. While your Smart car engine has probably run on FAME containing diesel blends for most of its life and therefore the system should be already clean - it is possible that by mix and matching with K100 petrol components has introduced new crap into the system to be dislodged. Another diesel issue is associated with it's wax content - most will recall a few harsh winters where truckers have lit fires under their fuel tanks to try and dissolve the wax that has come out of solution and gunged up the system. Ordinarily, once the wax comes out of solution, it doesn't easily go back in and chooses to build up a surface layer on tank, pipe and other fittings. Again, in the marine world, we test bunker fuel for cloud point - i.e. the temperature that the waxes will start to come out of solution and specify particular limits depending on where a ship will be operating. There is quite a large variation in cloud point for automotive diesels (in the UK) and, although the fuel quality is better than it was a few years ago (when auto breakdown companies noted a much higher incidence of waxing up north), it can still be a concern. Although, I think it is unlikely that waxing is your issue (small volume of fuel and relatively short time before it is burn and replenished on a bike), it could have contributed to some filter blocking if you have experienced a particularly cold snap immediately before your problems started. 

As for pressure appearing steady - I think you stated 0.25 bar (about 4psi); if the pump (with petrol) usually kicks out 36psi (~2bar) (as per Olaf's post), then you are well down and this could indicate that either the filter is blocked and only enough fuel was getting through to sustain your much lower pressure...if you have any further deterioration of the filter, then the fuel rail pressure would drop even lower. When you stop, the pressure would perhaps creep back up slightly on restarting to be just sufficient again (for a short while). Meanwhile the poor old pump will be slogging it's little heart out trying to pump a higher viscosity fuel to give you what you need, but potentially, the volume will be too low - leading to fuel starvation.

Water-wise - a big problem with diesel fuels which is why (as noted by RickG) that most diesel engines have filters that incorporate water separation capabilities and the ability to drain off excess water. This is particularly important on most diesel applications as the fuel filter is usually remote from the fuel tank. This is not the case with the K100 fuel tank, pump and filter configuation. Given the relative small volume of the fuel tank, it is unlikely that you would see much water separating out in the filter - you might get some separation in the tank, if you don't use the bike regularly and this could be picked up by the fuel pump, given it's location at the bottom of the tank (exactly where the water will gather). I don't believe the standard K100 filter has any means of trapping water as it is expected that any tiny amount of residual water will simply go through with the petrol and will be spat out of the exhaust as steam. Of course, to add to your misery, if you do have a layer of water at the bottom of your tank, it could lead to Microbial Bacterial Contamination (MBC) - where bugs grow in the fuel/water interface and these bugs will definitely block you filter and spoil you day. In the worst case, you can get sulphur reducing bacteria (SRB) which are evil little buggers as they can eat through steel! MBC/SRB are more prevalent in the marine industry with bigger tanks/volumes and sometimes deliberate fuel/water interfaces (sea water compensated tanks) and in the yachtie world where the tight fisted yachties only use their engine a minimum amount of time and end up keeping a tank of increasingly buggy fuel for a long period of time (and then wonder why the engine doesn't run when the really need it). Personally, despite putting the frighteners on you, I don't see MBC/SRB being your problem...but you never know. If you find gloopy gunge on your filter...perhaps we'll think again!

End of science lecture! My tuppence ha'penny worth would be to find a filter designed for use in diesel systems and fit that - you may need to consider a different pump too (for the reasons above with FAME and seal issues), but start with a decent filter.

I look forward to hearing how you get on!

I work on small marine diesels.  These engines virtually all have Racor filters.  Some of the boats with gasoline engines have the same model Racor heads, but the elements ARE NOT INTERCHANGEABLE. 

If you are running a petrol filter GET IT OUT OF THERE AND REPLACE IT WITH A FILTER INTENDED FOR USE WITH DIESEL. 

If you are running a petrol filter you are wasting our time asking us to help solve your problem.

Point-Seven-five wrote:

If you are running a petrol filter you are wasting our time asking us to help solve your problem.

Ouch!!!

touchwoodsden wrote:I have news.
The FrankenBrick rides well once more.

Thanks for the update and photo.

Thanks for the update and the photos. 

The pressure that crushed the filter element was the output pressure of the fuel pump which is about 4.5 bar less the return pressure your gauge was reading which I think you said was about 1 bar.  This means you had a pressure of as much as 3.5 bar working on the surface area of the element.  This could easily result in as much as 100kg of force on the structure of the filter element.

As far as the fibers on the pump's filter screen, I suspect they are from the crushed filter's media which is nothing more than a mass of fiber compressed into a membrane and held in place with bonding agent.  When the element is deformed the fibers and the bonding agent are broken up and will flow downstream and what doesn't go through the injectors into the engine is returned to the fuel tank where it is caught by the pump's screen. 

I hope you didn't replace that element with another petrol element.

Very good point point-seven-five.
You cannot put a single restriction on the pressure line. The excess pressure has to be relieved to the tank. Otherwise the internal fuel pump regulator will not quick in under 66PSI which is 4.5 bar
If you want to play with the fuel pressure, you have to replace the original relief valve with an adjustable one. Adjustable fuel pressure regulator

touchwoodsden wrote:I have news.
The FrankenBrick rides well once more.
I emptied and removed the fuel tank and removed the fuel pump and the filter.
Please bear in mind that this was a completely clean installation with a new filter.
The very fine gauze on the bottom of the fuel pump (which actually needs replacing) was full of some kind of a very fine fluff. I don’t think it was enough to stop the fuel flow completely but I think it was probably contributing to the problem.

Great news! 

As per my previous post, if you are already on the hairy edge of not getting enough fuel through the filter, then it really doesn't take much debris to push you over the edge into fuel starvation. That's one seriously deformed filter!

I have fitted a full-size car filter
http://www.ebay.co.uk/itm/260357383770?_trksid=p2057872.m2749.l2649&ssPageName=STRK%3AMEBIDX%3AIT
immediately behind the fuel injection pump between the frame rails and held with a jubilee clip to a bracket on one of the frame tubes.
You will note that the inlet and outlet go in opposite directions which would have been great if I was using it sort of in-line there was nowhere to mount it and when it’s full of fuel it’s not going to be particularly light.
Because I need one of the fuel lines now to do 180° and I don’t trust the robber pipe not to kink, I put some quarter-inch copper fuel pipe the inside with the 5/16 rubber pipe and that allowed me to put quite a nice bend in it without it kinking. The copper stays in place.
When I come to replace the filter, I will replace it with this one
http://www.ebay.co.uk/itm/351839961251?_trksid=p2057872.m2749.l2649&ssPageName=STRK%3AMEBIDX%3AIT
 
Because it makes the pipework neater