Bleeding engines!

[alpha]

It's like this... Despite having checked, double-checked, and triple-checked all the unions between tanks and filters, I am still having very occasional trouble with engine stoppages caused by air in the fuel filters. No major difficulty as I'm now extremely adept at a quick bleed and re-start, but Mrs Alpha is a little fed up with my shouts of 'Where's the Bleeding towel' in front of the young Alphas (I keep an old towel for the purpose, you see).

So, I got to pondering...

Is it practicable to take a very fine tap off the bleed screw on the top of the filter, back to the tank, probably not by joining it at the HP pump return line (as the pressure is too great and there would be flowback into the top of the filter), making an effectively self-bleeding system?

I'd have to be careful about tank use, as the return fuel flow into the mains would preclude setting off on the auxiliary tanks with the mains full, but that's not a problem.

I've left my Nigel Calder books on board, so would anyone like to comment?

Engines are Ford Thornycroft 2700 series with CAV filter holders.

<hr width=100% size=1>

 

[Freebee]

I dont pretend to know your fuel system but if you run this small pipe with a shut off valve what have you got to lose? If theengine falters with the line open-due to starvation you can close the valve, you have lost nothing and the next time you have to bleed just open the valve. Sounds like a simple solution to me.

<hr width=100% size=1>

 

[claudio]

Wouldn't the air be drawn in through the 'bleed pipe' with the engine running?
Do you currently bleed the fuel system with the engine running ?
You shouldn't need to install this bleed line, have you checked your fuel lines for air leaks?

<hr width=100% size=1>

 

[pvb]

Compromise solution...

What you're suggesting is a compromise and, like most compromises, it won't be satisfactory. Far better to invest your time in tracking the air leaks. Have you replaced all the rubber seals in the CAV filters? Have you re-made the pipe joints with sealant? There can't be that many places where air can be getting in.

<hr width=100% size=1>

 

[alpha]

Re: Compromise solution...

Thanks so far.

Yes, all the joints have been checked and appear, even subject to serious scrutiny, to be sound. There is no evidence of diesel leaking out anywhere.

No, the HP pump won't suck in, because there will be positive pressure in both lines, albeit less mass flow in one than the other.

Compromises... Compromises!! Yes, but when I look at my boat, and all I see is a series of compromises, at the end of which, I've got some machinery to go boating in!! Most of my expertise is in aviation, and all aircraft are the results of compromise. It don't think that's a reason to dismiss the end products as unsatisfactory!

In fact, not only would my 'compromise' solve my present problem, it would guard against similar future problems.

The filters are new, and not in doubt.

<hr width=100% size=1>

 

[claudio]

Re: Compromise solution...

Alpha,
Have you tried pressurising the fuel line with compressed air and using soap based leak detection?
To get air into your filters there 'must' be an air leak somewhere, might only be a pin prick and maybe not enough to leak diesel but definately a leak.



<hr width=100% size=1>

 

[alpha]

Re: Compromise solution...

I haven't subjected it to anything at high pressures, as that shouldn't be necessary. I'd rather avoid going down that route, if I can.

The problem only arises after about thirty or so running hours, so it's not a big deal.

Any more thoughts on the proposed solution...? (I could simply bleed the system every third day or so, but that would be a messy pain, too!).



<hr width=100% size=1>

 

[gus]

Re: Compromise solution...

I would suggest fitting a header/day tank then wrapping tissue around each joint. By positively pressurising the supply using gravity, any small leak will show up on the tissue eventually.

<hr width=100% size=1>

 

[oldharry]

You need to find out where this air is getting in, and not to be mesing about with unauthorised and risky mods which could make the problem worse not better. Spend the time and effort on going through it all yet again. Replace every sealing ring and washer in the system, and check the seatings.

Have you checked out the fuel lift pump? A very common source of leaks. Although this provides positive pressure, the injector pump can induce negative pressure in the feed line each time it sends a charge to the injectors, so that air is drawn IN against the normal pressure. This will then either go through the injectors causing faltering, poor running, and inexplicable stoppages (because with this type of fault, the engine will often start again immediately it has stopped), or will find its way into the filters where it accumulates and causes stoppages.

If you really do want to try a self bleeding system, the normal arrangement on modern cars is to use the Injector pump return line, and replace the bleed screw on the CAV filter housing with a proper banjo union and hollow bolt. This is standard practice on many road engines, is easily fitted, and will cost very little time or money to fit (or to abandon if it does not work)



<hr width=100% size=1>

 

[Dave_Snelson]

Re: Compromise solution...

What about air from the tank itself? Does the diesel froth in the tank and then the risers pick up the small aerated amounts of diesel which accumulate in the filters.

Try a fuel additive to prevent frothing. Take a look at the grade of diesel you are buying, some marinas sell crap diesel, or worse still, heating oil and pass them off as marine diesel. Both of which froth like mad and get very aerated. A good grade of low sulphur diesel doesnt froth, or at least it froths a lot less.

Just a thought

<hr width=100% size=1>Madoc Yacht Club
<A target="_blank" HREF=http://www.madocyachtclub.co.uk>http://www.madocyachtclub.co.uk</A>

 

[summerwind]

Re: Compromise solution...

The thirty hours running time is a bit of a clue to this problem. From this I reckon you do not have a circulating system through the injector pump and that the only bleed off is from the injectors back to the tank?

There is no doubt you have a mini-air leak somewhere in the system up-line of the lift pump. The best way to find that is with a temporary header tank as already suggested. Talcum powder sprayed onto the joints will give a good indication of a leak. If your feed is out of the top of the tank, make sure that the pipe into the tank is long enough. It is also worth removing this pipe to make sure it is clear. I once had a problem with a small pack of cr*p in this pipe partially blocking it. The flow would gradually pull the plug up to the elbow at the top of the tank and give more resistance to fuel flow. Lift pump sucked air from somewhere and engine eventually stopped. Cr*p fell down pipe and bled engine strarted again. Took a bloody long time to discover the problem.

A good permanent solution will be to fit a line from the low pressure gallery of the injection pump back to the tank. You will then have a constant flow of fuel through the system and any small amounts of air sucked in will escape when the fuel goes back to atmosphere in the tank. You will probably find a tapping on the gallery somewhere that you can use. Alternatively, use the bleed screw tapping on the pump. You need to be a little bit careful of the size of return pipe, it shouldn't be too big as you need to make sure that there is enough fuel for injection at full revs.

<hr width=100% size=1>

 

[Guest]

be logical, i know you are pissed off highly but continue the search, dont compromise, you have a leak somewhere.
one thing that always used to fool us when we were apprentices, and no one has suggested it, was the stack pipe in the tank, some of them screwed into the connection from the tankand came loose, some of them got pin holes in them, some of them have ickle gauze filters on them that get blocked and the worst one was when the operator of the machine lost the filler cap and stuffed a rag in, said rag would then migrate to tank and do intermittant blocks, are you sure also that you do have air coming in? and not a partial or intermittant blockage?
stu

<hr width=100% size=1>