Sweet pea water pump

[newington]

Still struggling with this pump. pump fabricated to the drawing sizes.There are no leaks and it pumps ok when not on loco.When on the loco it pumps but does not seem to have enough pressure to over come the boiler. "O" ring a good seal and as I said no obvious leaks. I am now considering the distance the ram travels may be insufficient. It is only 7/16". If a member would be able to measure the travel of the ram on their sweet pea and let me know I would appreciate it. At least I could rule that avenue out. Thanks Newington

[smallbrother]

I would have thought the problem is elsewhere. Does the bypass valve stop the return flow completely? Have you checked the clack valve is OK? Are the ball valves on the pump itself operating freely?

I have recently made the axle-pump eccentric for my 3.5" Juliet. The offset is 1/4" making the stroke 1/2". 7/16" on a Sweet Pea does seem a bit short in comparison but the pressure would not increase by lengthening the stroke, it would increase the quantity only.

Pete.

[johnthepump]

Looking at the Sweet Pea handbook while having a tea break, the ram would move 5/8 of an inch

[SteveW]

Are you sure there is no air trapped in the thing?
Given the short stroke and a bit of back pressure from the boiler you could just be compressing against an air spring and never getting a 'solid' hydraulic push against and through the clack.

[newington]

Thanks chaps. The clack valve into the boiler has been stripped and it looks fine. If the pipe from the pump is disconnected and the loco pushed along water pumps from the uncoupled pipe.The bypass returns to the tank and water can be seen issuing from the pipe with the valve turned to tank,with it turned to boiler there is no water coming from the return pipe. Think this means the bypass is working ok.
Not sure how to check the balls in the valves but operating the ram with a tube connected to the inlet end one gets a jet of water from the outlet,but of course this is only against atmospheric pressure.Would the stroke being short make a difference?

[smallbrother]

Apr 23, 2016 20:18:44 GMT newington said:

Thanks chaps. The clack valve into the boiler has been stripped and it looks fine. If the pipe from the pump is disconnected and the loco pushed along water pumps from the uncoupled pipe.The bypass returns to the tank and water can be seen issuing from the pipe with the valve turned to tank,with it turned to boiler there is no water coming from the return pipe. Think this means the bypass is working ok.
Not sure how to check the balls in the valves but operating the ram with a tube connected to the inlet end one gets a jet of water from the outlet,but of course this is only against atmospheric pressure.Would the stroke being short make a difference?

Johnthepump's avatar above shows how the ball valves work in the pump and their location. These can be a bit awkward to get at, but I had a problem with muck which stopped one working and the pump was ineffective on my Polly 1.

[jma1009]

Hi Alan,

I dont think the pump throw is the problem. I made an axlepump of 7/16" stroke and it always performed perfectly.

The problem lies elsewhere. It is always difficult with something someone else has built.

Cheers,
Julian

[newington]

I made the pump myself by fabrication so it is new.The previous owner had the pump go through the frame stretcher so when I made the pump I put an 1/8" mounting plate at the rear (which is why the throw is shorter) The balls seem to seat when they should.There is a good head of water and the bypass is ok. I feel I am missing something but I cannot figure out what. An axle pump is a simple thing after all! Alan

[nonort]

Can I suggest gaging the output clack and then turning the wheels slowly by hand. I suspect that air is being drawn in past the ram. This problem is generaly caused by undersize pipework on the feed side of the pump, it will draw in air long before it will draw in water if given the chance. Also it may worth checking that the suction ball isn't covering the inlet on the suction stroke. Again the tendancy is for the pump to draw air past the ram rather than draw water from the pipe work.

[newington]

nonort,
If I gag the output and turn the wheels what would I expect to see? water coming out of the bore or something? I can see where you are coming from.

[newington]

Went to the club this afternoon and a member suggested I fit a gauge to the pipe from the pump to the clack. as with the other suggestion from you chaps it makes sense. I have recently seen a thread on this site regarding this. I think a T was being made. My question if I may ask is why did it need a "T" and what was the outcome of the test with gauge attached. If nonnort wouldn't mind explaining what I should look for I can do that at the same time. Thanks Newington

[johnthepump]

Hi Newington,
The T piece that you refer to I put up on page 42 of what I've down today 04.04.16 . The purpose of this T was to fit it in delivery line from the axle pump, as a test point so that a temporary gauge could be attached to check the delivery pressure of the pump and make sure it exceeded the boiler pressure. I have made a batch of five and they were only finished yesterday and have not yet been fitted. I wanted one to fit my Sweet Pea that I acquired recently and 3 others in the club wanted to do the same for there Sweet Pea's. This was also an experiment with CNC working.
My thoughts on your pump, it is the inlet valve not setting properly. Because the delivery has a clack on the pump and another on the boiler and they would both have to be leaking, not to work.
Regards John.

[jma1009]

Hi Alan,

I dont understand why the throw should be shorter as a result of your mounting plate. The throw is set by the throw on the eccentric.

Cheers,
Julian

[nonort]

Hi again sorry if my suggestion was a bit vacant. It is always easier to do these things with the bits in your hand than trying to explain them. The point of gaggig the outlet is to try and see if the ram is leaking if it leaks water then it will certainly leak air into the system. You may be able to prove this by seeing if the pump will self prime and lift water, it won't lift it very far but it may indicate either a leaky ram or a suction seat problem. If the ball lift on the suction side is to much then the water from the pump strke will go back to the tank momentarily before the ball settles. Moving the pump in the frame will make no diference to its' displacment. Although moving it away from the eccentric will increase the dead space at the end of the stroke and may increase the chance of air being trapped on the pump stroke. Having moved the pump body perhaps you should have made either a longer eccentric rod our longer ram to compensate. You could disconnect the eccentric and see how much the ram is not bottoming out by, any more than about a sixteenth of an inch is in my view to much. ps the only otherthing is does the pump work ok when driving slowly ie when the balls have time to seat properly and then not pump enough when traveling faster when the balls have less time to seat this is more problematic on the suction side the the outlet side. I hope this can be of help.

[newington]

Hi all. Now then, I have spent most of the week working on the valve seats. Had really good seats and nothing passing by. Fitted pump back on loco with outlet capped off. On a short length of track and some water in the tank pushed loco along. After a few revolutions of the wheels the wheels locked up and would not move in either direction. There was no sign of any drips. Next I coupled up the outlet pipe to the clack pipe and after ensuring water came from the pipe attached a gauge.
After a couple of revolutions the pressure registered was up to the 100psi max on the gauge. (gauge has not recovered and is now stuck on 60). I then steamed the loco blocked up with wheels clear of the track. With 60 on the gauge and the wheels rotating a good steady stream was pumped back to the tank with the bypass open. With it closed the water level in the glass did not rise,but was maintained. Boiler Pressure was increased 90psi regulator opened and the water in the glass lowered.The bypass was set to open (tank) nothing,no sign of water coming from the return at all. Taped the pump in case balls were dislodged to no avail. Fire withdrawn and a bigger mantelpiece designed to put the darn thing on! Thought I had it sussed prior to steaming the indications were good. Any ideas anyone Alan

[jma1009]

Hi Alan,

For further advice we really need a lot more details.

What bore pipe are you using? What bore ram? What size balls are fitted onto what size seats? If the eccentric straps are disconnected is there a gap on the extremity of the stroke that might be causing an air lock? What is the lift on all balls including boiler check valve?

Did your 'some water in the tank' get emptied so no water left to pump?

Some pics and a drawing would be of help!

A common fault is the water inlet ball getting stuck on it's seat. On a tender loco this is easy to check. Not so easy on a tank loco.

(My own solution would be to fit a decent second injector and dispense with the axle pump entirely!)

Cheers,
Julian

[runner42]

Based on your last post if I assume the clack valve wasn't in circuit when the pressure gauge was connected then it must be the clack valve that is the problem. it appears to be unable to open sufficiently to allow a volume of water to be put into the boiler thus indicating no or little change at the water gauge. Suggest you check the clack valve.

These esoteric problems are frustrating and time consuming I would not give up just yet. Worse case scenario is you may have to remake these parts, this is often quicker than chasing a fault that appears to defy logic.

Brian

[johnthepump]

Hi Alan,
I was thought about your pump problem this morning when I caught site of my injector test boiler, which has an injector test unit on it. I came to mind that this could be used to test pumps as well, this would have the advantage of testing against boiler pressure while delivering the pump water to an overflow which can measured.
I haven't yet fitted the test tee piece into my axle pump delivery line, but it will get done this week, this will allow a temporary gauge to be fitted so as to see that pump pressure is working, l will post some photos when I have. 
Is your pump as the drawing? there is mention of the little grooves which allow the water to reach the valves when the ram is forward of the holes. It is desirable that the ram fills as much of the chamber as possible, as this effects the compression ratio, This is important if air gets in the system for any reason, as a low compression pump won't get the air out against boiler pressure.
Regards John.

[newington]

Julian,John,
There are no injectors on my sweet pea just the axle and hand pump. That is why the axle pump is so essential. I am going to fit an injector. Now the pump is to the dimensions of the drawing but fabricated not a casting. The balls are 5/32" The outside pipe dia is 3/16" (pipes are all coupled up at moment) Think ID is an 1/8" the ram is 7/16" dia . The tank did empty during the test so you and John may have a point there with an airlock. If that is the case how does one get rid of it when running.
I have removed the ram from the bore, I had a hell of a job getting it out against the suction as I had filled the tank back up.I removed the ram as I spotted a bead of water at the open end of the bore. I am going to check the "o" ring size the spec says C11 o ring but I cannot find one of that spec What does it translate to for ordering? I will then check that travel from the eccentric just in case there is a dead space at the end. What is most frustrating is that it did pump and then didn't! I wont give up but I need to get a loco running so may move on to another rebuild out of service loco. Sweet pea was supposed to be a quick fix to alleviate the club loco shortage. No such thing as a quick fix in this game then!

[johnthepump]

Hi Alan,
Firstly are you going to the Sweet Pea Rally at the Fareham & District SME on the 18th & 19th June? there are 4 Sweet Peas coming from the Island on the Sunday.
I think the O ring you refer to is BS No. 011 that has an outer Dia. of 7/16" inner Dia. 5/16" and the cord Dia. 0.070" nominal 1/16". The point I made Yesterday about the compression ratio of the pump is to do with forcing air out of the system the compression has no significance when dealing with water as the pump is a displacement pump. Don't get down hearted about this problem you will get over it, maybe you just need a little help. ( no pun intended Julian and everyone else who knows me) I did leave you a PM if you want to chat about this PM me and I'll give my Number.
Regards John.