Question about air compressors and unloaders

[J Tiers]

I have a small lab-type Gast oilless (diaphragm) compressor. It's decent quality, US made, non-consumer. I have used it for direct pumping air for chip blowing when milling, etc. But it is lowish pressure, although the compressor should reach 100PSI when connected in series mode. (there are two sections that can be either series or parallel).

It's quiet, and does enough for the inside shop. I don't really want anything more than that, as I don't use a lot of air.

I had wanted to hook it up to pump up a tank. Looking around, I got a combo check valve and unloader with the smallest CFM rating I could, which was 1 CFM minimum. Apparently this is even less than that, although I did not think so, and by rough measuring it seems to be more.

The issue is that the unloader leaks so much air that the compressor cannot build up pressure, and does not start charging the tank. I can get it to do so if I block the leak hole, but of course that does not do what is intended. It also will not close the valve if the tank is empty, it just keeps leaking. Without the unloader, it will not start against pressure, of course.

The leak hole for the unloader appears to be somewhere around 0.015" , if that.

I am pretty sure that these compressors are used with tank systems, so there must be an unloader somewhere that will operate with this. I have no idea how this one works, so I do not know if I can modify it (weaker spring?) to do what is needed. I'll maybe take it apart to see.

I found no smaller versions. Anyone know of such?

[eKretz]

Hmm. What brand is it? I used one of these on my small A/C compressor build for my basement electronics workshop (wanted ultra quiet so as not to bother anyone sleeping - I am up at odd hours sometimes). That might be lucky to get near 2cfm but I don't think it's even that high. 2 gallon tank. These are combo pressure switch/unloader. I use a separate check valve.

https://www.condor-usa.com/air-compr...ches-mdr11-usa

[J Tiers]

Those will not work. The problem is that those are too high pressure. I only want 60 psi or so. I have a pressure switch in any case.

The compressor is a Gast, MAA series, not sure which, could not see the model number. One of the possibles is 1 CFM, the other 0.4. They do not have any accessories for that series.

https://www.gastmfg.com/products/com...chnology%3A876

I looked at the combo valve. It has a rough spot, and probably a too-stiff spring. Spring is 0.015 wire, I have some 0.012 wire which would make a softer spring. Apparently the pressure at whatever air flow cannot move the valve shuttle to cover the vent. I will try polishing the bore to get rid of the rough spot that catches the piston, and also make a softer spring. We'll see.

The pump can also have the flow doubled, but the pressure will be reduced, by combining the two in parallel. That still will produce 50 PSI, which may be enough.

[J Tiers]

The slightly weaker spring would not bring the shuttle piston back, so the valve did not work reliably with it. It worked with a length of hose as the "tank", but not with an actual tank, which seems weird.

In neither case did the compressor operate the valve, however. Only if I closed the leak hole would it bring up pressure, which obviously is not helpful

[tom_d]

It's been my experience that small, lower pressure compressors don't need an unloader valve. The motor has enough starting torque to get things moving even if the compressor is dead headed at its maximum psi.

[J Tiers]

So, I reconnected it for parallel operation, doubling the throughput of air. But that did not fix the issue. It would sit there and blow a lot of air from the unloader port, but would not put pressure in the tank. If I close the bleed off, then it will "take up pressure".

Running it open, with NOTHING as a load, it takes a while.... the unloader (mounted with input below, outlet up) bobbles up and down before generally letting the air flow through. But it will not do that with any significant back pressure on the unloader.

It seems that there is not enough flow to pop the unloader, it can't build pressure on the input side sufficient to open the through port and close the leak port. The unloader is not sensitive enough.

It's been my experience that small, lower pressure compressors don't need an unloader valve. The motor has enough starting torque to get things moving even if the compressor is dead headed at its maximum psi.

Nope... was the first thing I tried. Stalled it right away. In both connections.

[gld]

This is just my observation. It appears your compressor does not have enough volume to close the unloader valve while trying to fill an empty tank. Possibly putting the a one-way check valve at the tank will allow the compressor to build enough pressure in the line to close your unloader valve.

[rdfeil]

I like gld 's thought. Try it with a spring check between the unloader and the tank. The spring in the check will force a higher pressure and volume into the unloader and when the unloader closes the pressure will overcome the check and fill the tank. Something to try...

[J Tiers]

The unloader IS a check.

There is a 1/4" "spool" inside. The tank side has a rubber sealed check valve (a rubber disk against a flat surface), which is closed when fully set back toward the compressor. There is a much larger piston on the spool on the compressor side.

There is a hole through the large piston down the middle of the spool, which terminates in side holes on the spool, open on the tank side when it is fully set forward. (I am not sure where they are when set back, but obviously they communicate with the leak, since the large piston blocks all airflow other than through the center hole.) At some point in the movement from "set back" to "set forward", the movement slides a plastic valve sleeve over the leak hole.

So, when pressure is in tank and line, the spool is set back, and the pressure inside leaks off. Then compressor pressure should push the spool forward opening the check, and then also closing the leak.

I do not have dimensions on everything, but that is how it works. So, there is a check. If there was another one further along, I don't quite "get" what would change. wouldn't the line still have pressure in it and everything would be about the same as direct into the tank?

The valve on the tank side is about 0.3", the actual plunger is about 0.25". The larger piston is between 1/2" and 5/8" diameter. The leak hole is around 0.015 or so, the hole through the spool is perhaps 1/8".

It seems that the leak is so much vs air flow that not enough pressure builds up on the larger piston. It's Ok against the spring alone, but does not seem to work against the spring plus back pressure.

Edit:

I looked at it again at zero backpressure. It does not get pushed all the way to the open position by the compressor. I could pull it another 50 thou or so without "forcing" anything.

Before I did that, it was still leaking a bit from the "leak port". When pulled up fully, it no longer leaked from the port, but it would not stay there..

[no704]

Maybe a two way solinoid valve would be more effective?

[J Tiers]

Maybe a two way solinoid valve would be more effective?

Hard to say. The basic issue is to let the compressor (which can in fact build up pressure just fine) build up enough pressure to operate the unloader. It has to work that way, because the unloader must be open to begin with, to let the compressor start. The only other way that appears reasonably direct, is to have a time delay.

But then something has to tell the unloader to open when the compressor shuts off. The shutoff switch is a logical candidate. But the switches I saw, (and the links above) were for much too high a pressure. The alternative is the check valve, which will close when air flow is cut.

For closing, the compressor has to build up enough pressure to close the valve, which it cannot do if the air leaks away too fast. As for drilling a hole as small as the one in the unloader, I got nothing. Certainly nothing smaller. It's already less than 0.020. Seems like it should be smaller to cut the leak rate and let the compressor build pressure.

[eKretz]

You could run a solenoid valve on the line from pump to pressure switch/check valve in parallel to the compressor motor that closes with the motor turning on. When the pressure switch cuts out the motor voltage the solenoid valve can dump the residual to atmo when it opens to unload.

I wonder if you could find a photo somewhere of the Gast pump factory mounted on a tank and see if you can get any clue as to what they used.

[J Tiers]

I like gld 's thought. Try it with a spring check between the unloader and the tank. The spring in the check will force a higher pressure and volume into the unloader and when the unloader closes the pressure will overcome the check and fill the tank. Something to try...

I didn't do that exactly, but with some hose as a "small tank", or even with just the local piping, and the outlet connector not connected (which shuts off the flow) it is the same.

Also, it works the best with no tank or pipe, just the outlet of the unloader open to the air. It sometimes works right in that condition. But NOT with back pressure.

I was wondering if there should be more volume before the unloader? Not sure why that seems like a possibility.... maybe because the check/unloader spool was bobbling up and down in time with the pulses of the pump. If there was more volume before the check it might be steadier....



The description in words wasn't so great. The check/unloader is pretty much like the following illustrations, one in closed check open unloader, the other in closed unloader, open check position. There is a spring around the stem of the "spool" that I am not showing. It pushes the spool back so the check is closed.

Hopefully that explains what the thing is a bit better

Open check


closed check

[J Tiers]

You could run a solenoid valve on the line from pump to pressure switch/check valve in parallel to the compressor motor that closes with the motor turning on. When the pressure switch cuts out the motor voltage the solenoid valve can dump the residual to atmo when it opens to unload.

I wonder if you could find a photo somewhere of the Gast pump factory mounted on a tank and see if you can get any clue as to what they used.

It was sold OEM to folks like medical manufacturers. Probably at your dentist there is one of them supplying air to the little blowguns they use to dry off your tooth to see it or work on it.

Gast has nothing that I saw. I don;t think they ever put together systems, they make compressors, vacuum pumps, air motors, etc.

[J Tiers]

I'm thinking that I drew the sketch wrong, and that the through holes need to be mostly blocked when the check is closed. I'm not sure if the dimensions guarantee that or not, on the actual item. Otherwise there cannot be pressure built up on the big piston. If there was a good direct path to the vent hole from the compressor side, there could not be a pressure differential across the piston, and it would never move.

There does need to be enough leakage to empty the area on the input side of the big piston. But it seems wrong that there should be a LOT of leakage out the vent hole. It seems like it has full compressor pressure behind it. That may be part of the issue.