WaGuns.org

My compressor gave up the ghost.

No, not the big one with the DIY check valve, the anemic one. Trying to nail down floor sheeting, and for a while had to drive a nail, and wait a while. Drive another, and wait a while. Would have been faster to just break the nailgun nails apart and nail the floor by hand. LOL Then finally, it just gave up the ghost entirely.

Time to fix the BIG compressor...

Fix an oil leak. Fix a couple air leaks.

Replace a condenser. Now it starts up and runs beautifully!!

Until...

Well, it starts, runs, and fills up. Stops running, because it is at the upper pressure limit. Then, when the pressure reaches the lower limit, it tries to kick back on, but even the new condenser doesn't give it enough oomph to start back up.

Problem? Air pressure at the head.

There is a check valve at the end of the line that comes from the head to the tank. This valve screws into the tank itself. Then the copper line is attached to it. A check valve is essentially a one-way valve.

Just above the point where the air is not allowed to come back out of the tank, is the line that goes to the pressure switch.

The way it is supposed to work is...

The check valve prevents air from returning to the head. That line from the pressure switch also has a 'dump' on it. (Another way to know that your check valve is not working, is that you will hear air constantly escaping from that dump valve.) The check valve closes. The dump valve then dumps the residual pressure in the copper line. Now, when your compressor tries to start again, it is not fighting against all the pressure that is left in the tank. An "at pressure start" is the same as an empty tank start.


What happened is that my check valve actually broke. Physically. Looking at a new one, and comparing it to my old one, makes me wonder if their over engineered part is actually designed to fail... (Which is why I decided to fix it instead of replace it. My 'design' should last until after the compressor head has melted itself down from so much use...)

Here is a pic of a factory part:


When I took my check valve out of the tank, it was missing the entire bottom part. It was in essence just a straight-through pipe. Here is a pic of it straight through. (Sorry about the crummy pics from here on, I have the cheapest possible camera and it just doesn't want to focus.)



My first attempt at fixing it didn't go so well. I tried to simplify what they had done. Used a spring and a circle that I cut from a juice bottle cap. (Thick stuff. Has silicone rubber on one side. That side went toward the valve.) It just blew right off, just like the factory design itself had done.




Finally I gave up on trying to fix it the way they had done it in the first place, and go with what I knew would work.

Here are the parts to my new valve. The original valve. A 1/4-20 round head stainless screw. A spring. And a rubber faucet repair washer.



Down inside the valve body is a lip that the spring rests against. It probably would have worked fine, but I bent the bottom of the spring out just a bit, to be sure it stayed on that lip. The head of the screw was pretty much exactly the same as the OD of the spring. but to make sure that it never worked it's way past the spring, I bent the top of the spring to be an exact fit on the shaft of the screw. The pic of that is so bad that you can't even see what I am talking about, so no pic this time.

I can't add any more pics to this, Please delay responses until I have finished this. (I'll write finished at the end.)

The head of the screw was only slightly smaller than the ID of the valve body. That's no good, not much air is going to get past that. (Bad pic, as most of them are, but you might get the idea.)



So, I carved the head of the screw into an 'X'. Again, bad pic, but you should be able to see that I have cut the head into an X shape. Plenty of air gets past this. I also rounded over all edges to keep the screw head from getting wedged against the inner walls of the valve body.



Finally, it all went together this way.




What I didn't take a pic of was, I put a steel washer after the rubber piece, and a nylock nut after that.


Points...

This is way beefier than the factory part.
The rubber piece has an angled face. I chose that one so it would find the center of the bore, when closed, and would seal better, and for longer.
The only way I could see this failing, was if the rubber piece unthreaded itself from the end of the screw. The steel washer holds the rubber piece against the valve body more firmly. The nylock nut will keep everything in place, regardless of vibrations.


For most people, just replacing the valve with a new one is the way to go. It's cheap enough that I only saved about half the price of a new on, if I ordered it on ebay.

The biggest reason I did this is because it just pisses me off to see things deliberately engineered to fail. They could just as easily have designed it as simple as my own design. And it would have lasted as long as the compressor itself. Instead they engineer this complicated device that is deliberately designed to fail in a specific way. (Every broken check valve that I have seen or read about online, has failed in exactly the way mine did.

Aside from the expense of a new valve, most larger compressors, like mine, require more work than just unscrewing this valve, to replace it. I had to remove the head from it's mount, to be able to remove the copper line. Then I could remove the valve. So I'd rather not do the work again. (Or pay a factory representative to do the work, as a lot of people would do...)

Finished.

Yes, it works. Yes, I have been running it.

There are no leaks. It seals perfectly.

By the way, in case it wasn't clear, the air goes through the valve body in such a way that it pushes the rubber away from the bottom of the valve, and the air goes through.

Then the spring pulls it back up, against the bottom of the valve body. And pressure in the tank, pushes it even more firmly against the valve body.

Great job! Very informative.
I had that happen on my IR 7.5 HP 2 stage 80 gallon upright.
Valve was leaking back up into the compressor head.
Just replaced the check valve for about $20.
Don't recall it being that hard.

This is EXACTLY the problem my compressor had on the job site two days ago!

I tore it apart but am not smart enough to figure why the "dump" valve is constantly bleeding air.
I tore it apart to see what parts I could replace or rig, as you did, but discovered that the Thomas compressor folks had cross threaded the thing on in the initial assembly so I'm not even going to attempt a repair as the housing thread are toast and screwing it back on is not possible without retapping the port and resizing the fitting... and still having to find repair parts.

Very cool thread Mr. McWhiteypants.

Pure genius, Jefe.


FYI I think what you are calling a condenser may actually be a capacitor, a starting capacitor to be specific. This data and $7 will buy you a cup of joe at the Bucks.

Huh, I'm kinda slapping myself upside the head now. I have an old IR portable compressor that won't start up again like you described. I figured it was the motor going bad, but didn't think about the check valve. Off to see if I still have all the parts for the dumb thing...

Selador wrote:Yes, it works. Yes, I have been running it.
There are no leaks. It seals perfectly.
By the way, in case it wasn't clear, the air goes through the valve body in such a way that it pushes the rubber away from the bottom of the valve, and the air goes through.
Then the spring pulls it back up, against the bottom of the valve body. And pressure in the tank, pushes it even more firmly against the valve body.

Love these projects.

That was a nice touch on the X-Bolt head.
Probably kind of difficult to find out now, but I wonder if the new check valve passes more/less air than the original. I've seen check valves that look as if they were designed as flow restriction devices.

Why would anyone do that?
Reduce air flow and efficiency, increase heat and load on the motor?

PMB wrote:I've seen check valves that look as if they were designed as flow restriction devices.

Guntrader wrote:Why would anyone do that?
Reduce air flow and efficiency, increase heat and load on the motor?

PMB wrote:I've seen check valves that look as if they were designed as flow restriction devices.

LOOKED as if it was designed that way... meaning poor design.

PMB wrote:

Selador wrote:Yes, it works. Yes, I have been running it.
There are no leaks. It seals perfectly.
By the way, in case it wasn't clear, the air goes through the valve body in such a way that it pushes the rubber away from the bottom of the valve, and the air goes through.
Then the spring pulls it back up, against the bottom of the valve body. And pressure in the tank, pushes it even more firmly against the valve body.

Love these projects.

That was a nice touch on the X-Bolt head.
Probably kind of difficult to find out now, but I wonder if the new check valve passes more/less air than the original. I've seen check valves that look as if they were designed as flow restriction devices.

No doubt more restriction from mine.

If I had to guess from what I saw, the factory valve has maybe 20 to 25% restriction.

Mine is probably more like a 40 to 45% restriction.

I did pay attention to how long it takes to fill the tank from empty.

With the old valve, which was zero restriction, because it was just a straight through pipe... About 3 and a half minutes.

With my valve installed... About 3 minutes 45 seconds...

Bottom line, there is so much flow available, that restricting it more with my valve, made negligible difference.

It's like having a ten inch pipe available, when you only need 3 or 4 inch. Restrict that 10 incher to 6 inches, and you pretty much don't care. Reduce it to 3 and 3/4 inches, and you may see a difference, but not so much as to make any real difference.

I should add that I tripled my tank capacity.

The compressor has a 40 gallon tank.

I converted an old 80 gallon propane tank into an air pig.

Hook the two together, you have 120 gallons.

It takes slightly over twice as long to fill.

Why only twice as long when it is 3 times the capacity... I have no idea.

Less load on the motor and pump while delivering more air with less back pressure.
Pump spends more % of time at a lower pressure.

Selador wrote:
It takes slightly over twice as long to fill.

Why only twice as long when it is 3 times the capacity... I have no idea.

Ok, last but not least.

The way to test your compressor to see if it is a motor problem, or the check valve...

Turn it on, and run it up to pressure. Let it turn off.

First, listen for constant air leakage where the little black nylon air line goes into the back of the pressure switch. If the leak is constant and doesn't stop... Most likely a bad check valve.

If you hear no leaks... Use a blowgun or something similar, to let air pressure out, until it goes below the lower pressure point on the pressure switch and tries to turn on.

If it struggles but won't start... First of all, don't let it try for too long, or you may damage the motor or condenser.

Second, it could be the motor/condenser, or it could be the catch valve.

Now with the compressor unplugged, let all the air out. Once the tank is empty, try to start it again. If it starts right up, it is most likely not the motor/condenser, but instead most likely the catch valve.

Guntrader wrote:Less load on the motor and pump while delivering more air with less back pressure.
Pump spends more % of time at a lower pressure.

Selador wrote:
It takes slightly over twice as long to fill.

Why only twice as long when it is 3 times the capacity... I have no idea.

That makes sense.

So it is delivering cubic feet at a faster rate, for a longer time. Making up that less than a third of a difference during that time.

Thank you.