Piston Clearance - Air Pump

[Dhutch]

What is an appropriate clearance for the piston bore on an airpump? (Modified/Doubleacting Edwards)

I currently have the air/vacuum pump from emilyannes steam engine pieces on the kitchen table, and while I have found an number of areas of concern and a highly worn valve, I also have some reservations as to if the piston/liner clearance is good enough for it to ever perform as we wish it to.

The piston brass, in a bronze liner, and is around 2inches in diameter, the pump operates at at speed of around 200stokes/min and 3inch stoke. There is and can be no rings, but three labyrinth grooves. The pump pumps condenstate, and is intended to produce a vaccum of around 22-24 inch/hg or .75-.8 bar

Thoughts on a postcard?


Daniel

[jschoenly]

Is the Edwards design similar to the Bailey and Navy air pumps with the inlet port in the middle of the cylinder and valves on each end? Crescent had a Bailey/Navy Style pump and the few I built had basically zero clearance from the piston to the valves. You want that zero clearance to evacuate all the condensate possible and draw the best vacuum. I believe my speed was lower than your 200 strokes/min but maybe not by much. On a 3x5 single, this pump with a flat plate condensor quickly drew a 23-25in vacuum. Great setup!

[Dhutch]

Not familur with the Bailey and Navy air pumps, but ours is a double acting pump with the inlet port in the middle of the cylinder and valves on each end, yes.

I plan to work out the valve clearance (stoke vs liner length) but its the clearance clearance on the bore I am most interested in a present.


Daniel

[gondolier88]

Daniel,

Could you explain why there cannot be any rings fitted to your piston?

Greg

[jschoenly]

I had a pretty close fit, probably about 0.0015 to 0.0025" for my pumps making sure they didn't bind. Those groves do wonders for sealing with a little water and you don't need rings. O-rings would cut instantly on the ports and regular style split rings would be a mess. With a good fit, they do a great job on the condensor.

[Dhutch]

Daniel,
Could you explain why there cannot be any rings fitted to your piston?

Should have mentioned, but as above, they cannot pass over the ports, o-rings get chewed and I am sure metallic rings would too. I was remaking it I expect I would leave one or two of the eight ports out, keeping the spacing the same, and then the rings could pass on the unported section and I expect pass over else where without issue.

I had a pretty close fit, probably about 0.0015 to 0.0025" for my pumps making sure they didn't bind.

Ok, that's the magic figures I was looking for, although 2 thou (50 micron) seems incredibly tight?

What material have you used for the piston and liner? I would also be interested in what you have used to seal the 'upper disk valve' to the piston rod.

Very similar looking pump by the way.


Daniel

[jschoenly]

Truthfully, it could have been a little looser, but I don't have any notes or measurements from when it was made. It was a REALLY nice fit though and if the valve was placed on the cylinder, it took quite a while for the piston to drop out from the vacuum.

The cylinder casting, liner, and piston were all bronze. First being a casting, the later a bronze bushing and stock. The valves were also bronze and as you can see there is a groove which was a snug to tight fit for an o-ring. Once pressed in place in the valve it was held quite well. Not shown are the springs that keep the valves tight to the cylinder face on both ends. The "seal" on the valve to the piston rod was again just a nice, very close sliding fit. It never seemed to hang up on the rod and this pump worked tremendous well. This was fitted on a scotch yoke driving this on one side and a boiler feed pump on the other. The unit was chain driven and could be fitted to various engines. This is/was the Crescent design. I sold the business a number of years ago when I was buying my house and it's still local, although pretty dormant right now.

[steamboatjack]

I think there is some confusion here, the first question was concerning diametrical clearance of the piston in its liner. Also this type of pump is not an “Edwards air pump”. The diametrical clearance should indeed be very small, Really as close as you can get with out seizing up and 0.0015ins to 0.002 ins on diameter is probably about right.
These pumps (and Edwards pumps) are never usually fitted with rings just water grooves.
Clearance volume: The double acting pumps are very good as they have zero clearance volume, an Edwards pump usually has some and this is where many launch engines spoil an otherwise good vacuum, the clearance volumes both below and above the piston (more properly called a bucket) should be AS SMALL AS POSSIBLE as this space will expand any gases in it and so reduce the efficiency of the next stroke.


Regards Jack

[Dhutch]

Ok, well certainly ours it nothing like that at the moment, more like 0.5mm than 0.05mmand while it pulled about 15inHg like this with good disk valves, it sounds like a new piston is in order. The bore looks in suitable shape, although as yet unmeasured, and clearly thats the large and more costly part to remake or rebore so would be avoided if possible.

What would you make the new piston out of, brass again, bronze? What sort of grade, where would you source it from?


Daniel

[steamboatjack]

Daniel, can we see some pictures to get a better idea?
Assuming the liner is gunmetal and truly cylindrical then make the bucket of a similar material, otherwise it may need a re-bore or at least honing. I would have a look at leedsbronze.co.uk they do supply small pieces pro forma. Probably LB2 would be the best bet.

Regards
Jack