1.5 from cold starts, then dies and won't restart!

[Mikexx]

This has happened a couple of times, where the engine has started seemingly ok.

 

If I move the throttle back to reduce revs the engine then dies.

 

It won't then restart even with 20secs of glow-plug heat before trying.

 

If I bleed the High Pressure side a few times it then start, with a lot of relief on my part!

 

Any ideas? I'm wondering if air gets in somewhere into the pump after I've turned the fuel tap off as things cool down and contract as a consequence? Long shot I know!

[Tony Dunkley]

This has happened a couple of times, where the engine has started seemingly ok.

 

If I move the throttle back to reduce revs the engine then dies.

 

It won't then restart even with 20secs of glow-plug heat before trying.

 

If I bleed the High Pressure side a few times it then start, with a lot of relief on my part!

 

Any ideas? I'm wondering if air gets in somewhere into the pump after I've turned the fuel tap off as things cool down and contract as a consequence? Long shot I know!

 

Check the spring loaded valve and the gauze underneath the [injector] pump inlet union in the transfer pump endplate.

The fine gauze can clag up with the same gummy deposits that also form on, and gum up, the spring loaded valve plunger.

[Mikexx]

 

Check the spring loaded valve and the gauze underneath the [injector] pump inlet union in the transfer pump endplate.

The fine gauze can clag up with the same gummy deposits that also form on, and gum up, the spring loaded valve plunger.

 

When you say "transfer pump" I presume you mean the cam operated fuel pump?

 

This pump is brand new, though the fuel level in the tank is such that this pump is near redundant.

 

The engine runs well when going so don't have the impression of fuel starvation. The sedimentary filter body and fuel filters have both been cleaned / changed.

 

Many thanks for your reply.

[Tony Dunkley]

 

When you say "transfer pump" I presume you mean the cam operated fuel pump?

 

 

No, . . . the transfer pump is part of the DPA injector pump. It's a rotary vane pump that supplies fuel at intermediate pressure [ more than lift pump pressure - less than injection pressure] to the high pressure plungers in the hydraulic head.

 

The pressure regulating valve and [ nowadays nylon] gauze that get gummed up are under the larger hexagon fitting in the steel end plate which is held on with 4 x small diameter studs and nuts.

 

Be very careful when you open it up, . . . there's a small spring behind the gauze which has a great fondness for the bilge somewhere just out of reach, underneath the engine.

Whilst you've got it opened up, check also that the plunger is free to move, and not sticking in it's bore due to gummy deposits and muck.

Edited July 22, 2016 by Tony Dunkley

[Tony Brooks]

Read the "still no joy" posts for links to the injector pump diagrams Tony is talking about.

[Mikexx]

Many thanks for the help.

 

Two things, first that many of the union nuts have been rounded, such I was considering getting a 1/2" flare nut (hex) spanner.

 

Second, the engine runs fine and I don't get any impression of fuel starvation. Is this really likely to be the culprit?

 

I have now found some manuals for the pump, thanks.

[Tony Brooks]

Well often but not always fuel starvation on these engines causes them to rev and then die but the fact that you need to bleed the high pressure system points to an air leak into the system but still do what Tony advises.

 

If it has rounded union nuts there is a very good chance the olives have been crushed so no longer give a good seal. It also suggest that as you only have 4 posts that the previous owner has been trying to cure the problem by bruit force.

 

Note what Tony said in the concurrent thread about broken pick up pipes in the and and the need to be very careful with the seals on the filters.

[Arthur Brown]

If the nuts are rounded then it's a sign that someone has needed to work there before and done it badly maybe often.

 

Start to look for replacement fittings of the correct size and then replace all the olives and unions. If the nuts are kaput expect the olives to be damaged or worn too, which will cause or allow fuel to leak out or air to leak in, neither of which would be good.

[Mikexx]

I'm waiting for the spanner to turn up before dismantling the transfer pump innards.

 

I can confirm that after standing for 2 weeks the engine started and despite maintaining full throttle died and had to be re-bled.

 

For info:

https://www.dropbox.com/s/7n2d67zhujudvv0/2016-07-25-2051.jpg?dl=0

 

Having little knowledge of these pumps, I wasn't initially aware of which filter input/output should be attached to which end of the pump. Something tells me these have been fitted the wrong way round.

 

The injector leak-off pipe is routed back to the sedimentary "filter". A little diesel does leak out between the nut and bolt threads, perhaps expectedly. I have a domed nut than should solve that issue with a coper washer. I don't think ingress of air here would cause the symptoms I'm getting.

 

I've felt around the pump end plate area and after the engine has been running for a while it stays dry which surprises me if air is getting in.

 

Will advise again when I've undone the big "nut"!

[Tony Dunkley]

 

For info:

https://www.dropbox.com/s/7n2d67zhujudvv0/2016-07-25-2051.jpg?dl=0

 

Having little knowledge of these pumps, I wasn't initially aware of which filter input/output should be attached to which end of the pump. Something tells me these have been fitted the wrong way round.

 

The injector leak-off pipe is routed back to the sedimentary "filter". A little diesel does leak out between the nut and bolt threads, perhaps expectedly. I have a domed nut than should solve that issue with a coper washer. I don't think ingress of air here would cause the symptoms I'm getting.

 

 

That photo may well be showing the cause of your problems, . . . the fuel system piping on the engine is all to cock.

 

The pipes between the secondary filter [the one mounted on the cylinder head above the flywheel housing] are the wrong way round, . . . the pipe to the injector pump inlet connection [big hexagon fitting on the pump endplate] should run from outlet connection No.3 on the fuel filter head, and the return pipe, from the engine block end of the injector pump body, should go into inlet connection No.4 via a non-return valve.

 

The injector leak off pipe banjo should not be connected to the tank return pipe banjo with what appears to be an ordinary nut and bolt. They should both be connected, by means of an extended [long] banjo bolt with a small centre drilling [to restrict bleed back volume] into the boss adjacent to the centre bolt on top of the filter head into which someone has fitted a blanking plug.

 

With the piping arranged and routed as described above, as it would have been originally, the low volume bleed off of fuel back to the tank from the top of the filter head constantly purges out any small quantity of air, which may get into the system between the tank and the lift pump, before it find it's way into the injector pump.

 

Before undertaking any dismantling/cleaning of the pressure regulating valve and gauze in the injector pump endplate, you need to re-pipe back to the original/correct layout, and change the filter element [because any muck in there will then be on the wrong 'side' of it].

 

We may possibly have got to the cause of your problems a little sooner if you had posted that photo at the start of this topic.

Edited July 26, 2016 by Tony Dunkley

[Tony Brooks]

It also has no non-return or pressure release valve fitted into the filter head where the return from the injector pump goes.

 

Further to Tony D's comment about photos. I think the boat is new to you and there is a world of difference between something like this starting to happen after years of satisfactory running and someone buying a boat and finding they purchased a problem. If you had told us the boat was new to you we would almost certainly have asked for a photo in post number .

 

I also think that the faults the photo shows are more likely the cause if the problem rather than the damaged union nuts so agree with Ton D - get the pipework and banjo union sorted before anything else.

[Mikexx]

The boat has been used as a live-aboard by a member of the family for a number of years. I haven't been on the boat for a few years. I know they has used a generator for a while rather than using the engine to charge the batteries. They're also shot and the second alternator has bearing issues!

 

One issue is where to get a banjo bolt from. I'm not at the boat for a few days and don't know the likely thread size etc. I assume imperial from the sizes of the nuts. It looks like someone has used a plain bolt on the filter body!

 

Many thanks for all the help. I added the photo to confirm the fuel direction after your posts as I had assumed the connections were correct.

 

I shall be at the boat this w/e and hope to be in a position to at least swap over the pipes.

[Tony Dunkley]

The boat has been used as a live-aboard by a member of the family for a number of years. I haven't been on the boat for a few years. I know they has used a generator for a while rather than using the engine to charge the batteries. They're also shot and the second alternator has bearing issues!

 

One issue is where to get a banjo bolt from. I'm not at the boat for a few days and don't know the likely thread size etc. I assume imperial from the sizes of the nuts. It looks like someone has used a plain bolt on the filter body!

 

Many thanks for all the help. I added the photo to confirm the fuel direction after your posts as I had assumed the connections were correct.

 

I shall be at the boat this w/e and hope to be in a position to at least swap over the pipes.

 

Swapping the feed and return pipes from filter to pump won't do anything to re-instate the fuel/air bleed from the top of the filter head back to the tank, and it is essential that this is done.

 

You should be able to get a suitable banjo bolt and washers, and an NRV/check valve for the pump to filter return from a Fuel Injection engineers, or you could try ASAP Supplies or Calcutt Boats. It also looks as though you may need a new throttle cable, and the throttle control lever on the injector pump needs re-positioning a couple of holes clockwise on the throttle spindle. The max. rpm and idling rpm adjustment stop screws are missing and should be replaced and the maximum governed speed re-set to no more than 3600 rpm.

With the max. rpm stop screw missing it will be possible to rev the engine to destruction once the fuel system has been sorted out.

Edited July 27, 2016 by Tony Dunkley

[Mikexx]

I wasn't aware those holes around the throttle spindle were designed for stops. I'll measure the thread size and get some bolts/nuts. Also rotate the lever as you've intimated. The cable might then also fit better.

 

I thought idle was set by the bolt/locknut on top?

 

Same for measuring the banjo bolt etc.

 

There is no return to the tank and doesn't look as if there has ever been one. The bleed-off currently goes to the "sedimentary" filter which feeds the "paper" filter. I'm uncertain of the benefit, part from being currently wrong, of recirculating the fuel in just the paper fuel filter?

 

I am also wondering if the current arrangement of feed/return points are at levels within the filter housing that encourage air to enter.

 

Just read the "Still no joy" thread. Thanks for your invaluable help.

[Tony Brooks]

From experience when trying to save drilling a return into an ex Broads hire boat - the engine starts and runs both on idle and at revs, then it dies and needs bleeding again. Sounds familiar? Take the return back to the tank.

[Mikexx]

From experience when trying to save drilling a return into an ex Broads hire boat - the engine starts and runs both on idle and at revs, then it dies and needs bleeding again. Sounds familiar? Take the return back to the tank.

 

Hmm - yes it does have a ring to it!

 

Trouble is I have 100 ltrs of fuel in the tank after "part" filling it up!

[Tony Brooks]

The return can go into the top or very near the top of the tank. As the tank is probably of unknown thickness so you may only be able to tap one or two threads in it to accept the fitting close to the top is probably the bets idea.

[Mikexx]

One bit at a time!

 

Put the fuel pipes the right way round on the filter housing and changed the filter. After a week of being unused it started first time.

 

Added throttle stops for tick-over and max revs. Slackened the anti-stall device as per book, two flats from when the tick-over is affected.

 

However I don't have a means of measuring engine speed and I don't trust the tachometer as it reads very low.

 

The engine now has a solid tick-over, doesn't race on idle and doesn't die when you put it into gear. In fact, apart from the gearbox, bearing and hint of water flowing over the propeller, you hardly hear the engine note drop . It feels so different!

 

I haven't touched the leak-off and return pipework - yet. However in my exuberance I nipped up the banjo bolt (that should have the leak-off pipes etc) and stripped the thread. It was already very loose and I should have taken more care in anticipation.

 

I've been looking at thread sizes, is it possible this thread in the filter body is a 7/16" - 20 UNF, which might be sufficient to catch the "4" threads of a M10 banjo bolt? The Banjo bolt is an M10 x 1.5mm with just 4 threads. I measured it's major diameter and got 9.8mm. The filter body is a CAV 5836B020 but apart from having 1/2" - 20 UNF unions I can't see any info in the banjo thread.

 

However the (female) banjos in the leak-off and return pipes have an 8mm internal diameter whereas the banjo bolt on the CAV filter is larger! I presume I have the wrong filter housing?

 

Many thanks for all your help.

[Tony Brooks]

Great progress and it sounds as if that banjo bolt had had "attention" before.

 

I have no idea what size the banjo bolt is but I do know that on BMC 1.5s the filter banjo bolt is much larger than the ones on the injectors.

 

I think the later BMC 2.2 s with DPA pumps had a smaller banjo on the filter so maybe it would be worth talking to your local diesel equipment specialist to see if they can supply a new filter head with a suitably sized thread. If you get one with a the larger hole you may have to change banjos a swell and this may involve silver soldering (not soft solder - BSS) the new banjo to the leak off pipe or buying a new leak off pipe with the larger banjo on the end.

[Tony Dunkley]

 

 

I haven't touched the leak-off and return pipework - yet. However in my exuberance I nipped up the banjo bolt (that should have the leak-off pipes etc) and stripped the thread. It was already very loose and I should have taken more care in anticipation.

 

I've been looking at thread sizes, is it possible this thread in the filter body is a 7/16" - 20 UNF, which might be sufficient to catch the "4" threads of a M10 banjo bolt? The Banjo bolt is an M10 x 1.5mm with just 4 threads. I measured it's major diameter and got 9.8mm. The filter body is a CAV 5836B020 but apart from having 1/2" - 20 UNF unions I can't see any info in the banjo thread.

 

However the (female) banjos in the leak-off and return pipes have an 8mm internal diameter whereas the banjo bolt on the CAV filter is larger! I presume I have the wrong filter housing?

 

 

There isn't a banjo bolt shown in that photo, . . . do you mean the bleed screw / blanking plug in the boss on the top of the filter head ?

 

The filter head you've got is the standard, correct CAV type for that engine.

[Mikexx]

 

There isn't a banjo bolt shown in that photo, . . . do you mean the bleed screw / blanking plug in the boss on the top of the filter head ?

 

The filter head you've got is the standard, correct CAV type for that engine.

 

It is a banjo bolt without any "banjos".

 

https://www.dropbox.com/s/y3k3l1o485b1stu/2016-08-08-2080.jpg?dl=0

 

Sorry its so out of focus.

 

Where do the leak-off pipes normally go? In an earlier post you said they should both be connected, by means of an extended [long] banjo bolt with a small centre drilling [to restrict bleed back volume] into the boss adjacent to the centre bolt on top of the filter head into which someone has fitted a blanking plug.

 

The current bolt is M10 x 1.5mm, the thread in the CAV housing seems larger and possibly has a finer thread, and the leak-off banjos etc have an 8mm internal diameter.

 

Sorry, but getting confused!

[Tony Dunkley]

 

It is a banjo bolt without any "banjos".

 

https://www.dropbox.com/s/y3k3l1o485b1stu/2016-08-08-2080.jpg?dl=0

 

Sorry its so out of focus.

 

Where do the leak-off pipes normally go? In an earlier post you said they should both be connected, by means of an extended [long] banjo bolt with a small centre drilling [to restrict bleed back volume] into the boss adjacent to the centre bolt on top of the filter head into which someone has fitted a blanking plug.

 

The current bolt is M10 x 1.5mm, the thread in the CAV housing seems larger and possibly has a finer thread, and the leak-off banjos etc have an 8mm internal diameter.

 

Sorry, but getting confused!

 

That is a bleed screw/blanking plug shown in the photo, but the threads on it are too flat on the crests and look not to have been properly formed, . . . . which won't have done the thread in the filter head any good, and is probably why it now appears to be a bigger and finer.

 

The brass banjo on the end of the leak-off rail goes onto that boss [with the now knackered thread] on top of the filter head, underneath [and hence commonned with] the banjo on the engine end of the tank return pipe.

 

The ID of the banjos that connect the leak-off rail to the four x injector cap-nuts isn't relevant to any of this.

Edited August 9, 2016 by Tony Dunkley

[Mikexx]

 

That is a bleed screw/blanking plug shown in the photo, but the threads on it are too flat on the crests and look not to have been properly formed, . . . . which won't have done the thread in the filter head any good, and is probably why it now appears to be a bigger and finer.

 

The brass banjo on the end of the leak-off rail goes onto that boss [with the now knackered thread] on top of the filter head, underneath [and hence commonned with] the banjo on the engine end of the tank return pipe.

 

The ID of the banjos that connect the leak-off rail to the four x injector cap-nuts isn't relevant to any of this.

 

My apologies I should have considered that bleed screws would be hollow and cross-drilled.

 

The problem I have is that the banjos of 8mm internal diameter won't fit over the thread of a M10, or 7/16" UNF thread?

[Tony Dunkley]

. . . . . . . . . . . . . . . . . .

 

The problem I have is that the banjos of 8mm internal diameter won't fit over the thread of a M10, or 7/16" UNF thread?

 

This is not a problem, . . . . the 4 x smaller [8 mm banjos] connect the leak-off rail to the top of each of the 4 x injector cap-nuts, but the problem you have got is that the larger [brass] banjo on the end of the leak-off rail isn't commonned into the top of the filter head with the banjo on the end of the pipe that should return to the top of the tank, but which at present goes [incorrectly] back to a primary [sediment trap] filter.