Opposed Piston Twin, D.Kelley

[fumopuc]

Excellent  Your experience in making pipework for the Snow certainly paid off  Will a centrifugal pump work well at those speeds or would a gear pump be better 

Thanks Roger.
Concerning the pump type, in my understanding a gear pump will be used if pressure is the main need and not flow, like oil pumps. My idea is to make a similar centrifugal pump like the one for the Show engine, because of it is easy to adjust the flow with the ratio of the belt drive.
I know already that 900 revs of this centrifugal pump is given a good flow, so by measuring the idle speed of the oposed piston engine it should be easy  to define the right ratio.

[MJM460]

Hi Achim, thanks for that tip.  I will look it up.  I guess these days nothing is further away than the computer screen and delivery truck, though postage can be the killer.

The big advantage of a centrifugal pump is that it is not positive displacement.  Once you have one working at some known speed, you can alter the flow by a simple throttle valve in the discharge water line, or if you use plastic tubing even a simple squeeze clamp.

To increase or decrease the flow once you have one working, you can use the simple formula flow is proportional to rpm, and head is proportional to (rpm) squared.

If you just need a larger pump all over, you can scale it up by using impeller diameter, instead of rpm in those formula.

With a gear pump, flow is proportional to rpm, and the pressure determined by the system, not the pump so long as you have enough power to drive it and the pressure doesn't break anything.  So not so flexible.

MJM460

[fumopuc]

Hi MJM460,
thank you very much for the additional information of centrifugal vs gear pump, very useful.


Due to some hard workload in the other life during the last months shop time was nearly not available.
But now I have started a batch of 3 water pumps.
Basically the same procedure as used 3 years ago.
http://www.modelenginemaker.com/index.php/topic,4060.msg84181.html#msg84181
The housings and impellers are finished now, the covers with the water inlet will be made soon.
One pump will be used for this OPT engine, here the direction of rotation is clear, so the water outlet is drilled already.
The two other pumps are for future projects, where the overall details of set up and pump integration are not clear yet, so no drilling of the water outlet done until now.

[Vixen]

I've been quietly watching your progress with this very interesting engine. It was good to see it running. 

Nice work on the water pumps. You should get good flow at low speeds with those big impeller blades. How will you plan to couple it to the engine?

Mike

[fumopuc]

I've been quietly watching your progress with this very interesting engine. It was good to see it running. 

Nice work on the water pumps. You should get good flow at low speeds with those big impeller blades. How will you plan to couple it to the engine?

Mike

Thanks Mike.
Just drawn this morning, a pulley at one crankshaft end.
One crankshaft was made already 10 mm longer at this end.
The pump will be driven by a 5x220x1 mm T150 Neoprene flat belt.
If it will work as created in my head.

[Roger B]

Those are some interesting looking pumps    I will be interested in how they perform. I don't know it the tolerances will be less critical than a gear pump 

[MJM460]

Hi Achim, you are very welcome.  Those simple rules are known as the affinity laws, and they work very well on turbo machinery, whether it be pumps or compressors.  They are used to scale performance up or down from a tested prototype.  Use of these rules means the manufacturers only have to test one of a range of geometrically similar machines, rather than test every size separately, to predict performance over a wide range of sizes.  As you have a pump with satisfactory performance, you can use these rules to scale up or down for a different model, and know what the resulting performance will be.

If you modify the geometry, a new test is necessary, and there are some simple improvements to the basic impeller shape that would increase the performance of the “base model” before it is scaled.  I wait with interest to see what the extra pumps are to be used for.

Roger, the centrifugal pump is not dependent on the clearances around the impeller for sealing as is necessary in a positive displacement machine such as a reciprocating pump, gear pump or a vane pump.  The pressure comes from the dynamic energy imparted to the fluid by the rotating blades.  Needless to say this means they must run at reasonable rpm, as the pressure and flow generated will be very low at low rpm.

The clearances are important of course.  Clearance means that some of the flow can return down on the cover side of the impeller, and recirculate, instead of increasing the flow at the discharge, rather like loose piston rings, so some close clearances are introduced in a more sophisticated pump design to restrict this recirculating without introducing unnecessary friction.  But the rotating blades still produce the pressure.  In a compressor, labyrinths are used for this purpose.

MJM460

[fumopuc]

Hi everybody, the pump is finished and working like hell.


<a href="https://www.youtube.com/watch?v=kw6NcMXv1WQ" target="_blank">http://www.youtube.com/watch?v=kw6NcMXv1WQ</a>



MJM460, thank you for all your information.
My chosen ratio, in accordance to the D.Kelley flow chart of this pump, seems to be successfully.

Next to do some fine tuning of the carb and a fuel cell.

[Vixen]

Hi Achim.

That pump looks very nice and you appear to have plenty of coolant flow. 

What fluid are you using for the coolant?

Mike

[Roger B]

That seems to work well    and the engine sounds good 

Looking at the drawings the impeller is just over 30mm diameter. What size pipes are you using?

[Admiral_dk]

Very good engine running (constant rpm and sound) and now a nice efficient pump too 

[fumopuc]

Hi Achim.

That pump looks very nice and you appear to have plenty of coolant flow. 

What fluid are you using for the coolant?

Mike

Hi Mike,
some month ago I have bought a lot of 5 bottles from this stuff at ebay.
The main reason for my choise was the color, only.
I don´t like the "poison" green or "strawberry" read fluids in my model engines.


 

[fumopuc]

That seems to work well  and the engine sounds good 

Looking at the drawings the impeller is just over 30mm diameter. What size pipes are you using?

Hi Roger,
rotor diameter is 31,75 mm.
The inlet pipe is a 8x0,5 mm brass tube, outlet is 6x0,5 mm.

[fumopuc]

Very good engine running (constant rpm and sound) and now a nice efficient pump too 

Hi Per, thanks.
Now it is time for some long runs.

[MJM460]

Hi Achim, those pumps turned out very well.  I like the casing design, simple and effective, like the impellers.  Nice when the little auxiliaries work well to compliment a great engine build.

A fascinating engine and wonderful to see it running.  A real testament to your skills.

MJM460