Compressor start up, shut down and running procedures?

[tmassey]

With all of my efforts on my compressor today, I've been giving some thought to the best way of running a compressor. I would apprecaite any suggestions.

I've got a few questions, so I'll start separate threads for each. This one is abut starting, stopping and running a compressor. I also have questions about oil changes, as well as replacing the compressor filter stacks.

A bit about my compressor: 6CFM Poseidon PFU-150. It's a 5HP, 3-stage compressor with a block made by Bauer. It doesn't have an oil pressure gauge, so I assume it's splash-lubricated. It's got a single two-part moisture separator, plus a moisture drain in the filter stack on the back of the compressor. It's supposed to have auto drains, but it leaked pressure so I've removed it and simply use three needle valves with tubes to drain the water. Past the compressor I have a fill board with a large single-part gold Bauer separator (which is always bone-dry, of course) and another 32" filter stack, then the back pressure regulator (set to 2000 PSI) and out to my tanks.

My current startup process is this: I have a checklist that I follow before I start the compressor. It includes checking that everything is properly connected, that all the valves are either open or closed as needed, and checking the oil level (and other things like make sure I'm wearing ear protection, etc.).

I then close the valve right at the output of the compressor and turn the compressor on. This is to allow it to build up some pressure as quickly as possible. I've noticed that the compressor rattles a bit when it first starts, and I believe this has to do with the compressor needing pressure to get everything into proper position. This shortens the rattle to about one second. (Am I being paranoid here? Or am I misunderstanding something?) I then crack the valve until the pressure builds up downstream enough (500 PSI or so), which takes less than a minute, and then open it fully.

While running, I manually drain the compressor every 8 to 10 minutes. There are three valves I manually crack, and I can see what's coming out in the clear tubes attached to them. I usually only have them cracked for 2-3 or so seconds: usually the water only takes 1-2 seconds to clear, and an extra second to make sure.

For shutdown, I tend to start the process by performing a drain, but I leave the drain valves cracked slightly. Not enough to drain the pressure quickly, just enough for the pressure to slowly drop. The purpose of this is to try to flush out as much moisture as possible, and to give the compressor a chance to run at a lower (and hopefully cooler) pressure for a bit.

I'll usually let it run like that for 5 or so minutes. Usually, the pressure doesn't drop below 2000 PSI during this time. I then stop the compressor.

One thing I've noticed is that when I start my compressor the next time, I tend to open the drains very early -- likely while I'm waiting for pressure to build up before I fully open the valve. During that drain, a fair amount of water comes out, and the compressor hasn't been running long enough to generate that water. So that's there from before.

So, here are my questions:

1) Any thoughts about my process? Is there something I'm missing? Or something that I'm doing that's unnecessary (like building up the pressure)?

2) What can I do to get the moisture out of the compressor when I'm done, so I don't have a bunch in there when I start it the next time? That can't be good...

3) Is there anything else I should be doing to minimize the wear and tear on my compressor? I have read recommendations that you should run it at *least* once a month, and more frequent is better, and that you should run it for at least an hour to drive as much moisture out of the oil as possible. Do you have a better recommendation?

I only run my compressor less than 20 or so hours per *year*, so managing the running of my compressor is something that I have to be very deliberate about. As you can see from my usage, I *really* should not have this compressor, but I do have it, and I really like the convenience, so I'm trying to do what I can to maximize its life.

And if you're in the area and want a deal on gas fills, get in touch! I've got air, nitrox, O2, you name it!

Thank you very much for your help! I appreciate it.

 

[tbone1004]

back pressure-you should have check valves and PMV's on the filters so everything before the whips needs to stay pressurized. This is crucial for the life of the filter stacks which don't like pressure cycling.
Building pressure on the compressor ASAP is ideal and if you want to make that a manual process, you can put another PMV on there set to anything you want above a couple hundred psi.

drain everything as soon as you are done filling, can do that with the compressor running then shut it immediately off, or can shut the compressor off, open the drains, and as soon as they're done, close them to maintain system pressure.

like any engine, pump, etc. cycling them is what kills them. Once it is on, keep it on, and make sure it gets up to temperature. Similar to your car where a 2 mile commute to work is more damaging than a 20 mile commute to work due to start up wear despite the other being much higher with miles. If filling any big bottles, it will be happier, albeit negligibly filling them in series instead of in parallel. I.e. if filling cascade bottles, fill them one at a time so it is not operating above 3000psi for excessive amounts of time.

From your use case, it sounds fine, just try to fill a couple bottles at once vs. firing it up to fill a single AL80.

 

[tmassey]

drain everything as soon as you are done filling, can do that with the compressor running then shut it immediately off, or can shut the compressor off, open the drains, and as soon as they're done, close them to maintain system pressure.

You've mentioned this a couple of times here. I've got a PMV at the end of the chain. Everything builds up to 2000 PSI and stays at least there through the entire run process. The pressures *will* fluctuate up and down above 2000 PSI, say as I switch from one cylinder to another. But even if the cylinders were empty, it won't go below 2000 PSI because of the PMV. I'm pretty sure that's as good as that gets, right?

*But*, the system *will* lose pressure once I turn the compressor off. I'm sure it's due to tiny leaks, etc. in the system: if the compressor is off, the pressure will drop. I'm not talking quickly or even perceptibly, but over hours or days the pressure will drop.

Is this a problem?

like any engine, pump, etc. cycling them is what kills them. Once it is on, keep it on, and make sure it gets up to temperature. Similar to your car where a 2 mile commute to work is more damaging than a 20 mile commute to work due to start up wear despite the other being much higher with miles. If filling any big bottles, it will be happier, albeit negligibly filling them in series instead of in parallel. I.e. if filling cascade bottles, fill them one at a time so it is not operating above 3000psi for excessive amounts of time.

I'm usually somewhat neurotic about this. I have two banks of 4 bottles and one of 3. I usually fill them in parallel until about 3000-3500 PSI or so, then fill them sequentially up to 4500 PSI. Mainly to give the compressor a bit of a break from running above 3500 PSI. I kinda picked those numbers out of the air, and I'm not even sure how much of a difference it makes. But it's what I've been doing.

Of course, at places like Amigos, those compressors spend their *lives* filling a dozen or more bank bottles from 3000 to 4500... multiple times a day... and they seem to do OK...

And this year I'm going to see what it's like mostly filling without the bank: mainly to take more opportunities to run the compressor.

From your use case, it sounds fine, just try to fill a couple bottles at once vs. firing it up to fill a single AL80.

That's not a problem: I only dive doubles... And I rarely run my compressor for less than an hour, and even then it's gonna be at least 45 minutes.

Thank you for the information!

 

[Wookie]

Your rattle on startup is a floating piston on the final stage. Nothing to worry about.

I would hunt down those leaks that drain your filter stack over time. That really shortens the effective life of your filters.

 

[tbone1004]

@tmassey system shouldn't leak... If it does, go find and fix them. Having it drain down over time isn't the worst thing in the world, but it is not ideal.

Personally I think 2000psi is a bit low on the PMV, but I don't know what the "balance" pressure on that compressor is. It needs some back pressure to seat the floating final stage, but they also tend to run a bit happier at a certain pressure. PMV should be set to that. Can fill in parallel up to that pressure, then fill in series above that. Higher the pressure on the PMV the better, to a point.

On the compressors at Amigos, those are still filling in series, just not individually, but those compressors are also much higher CFM. You were also asking about how to maximize your compressors life and minimize wear which is not something that those guys are terribly concerned about. With your bank size you are well within the compressors capability to fill them at once, but it will be happier filling them in series. Will also ultimately be happier filling the high pressure banks first, then the low pressure banks so it finishes at a lower pressure.

On the PMV setting. At what pressure do you usually fire the compressor up? I.e. if filling from banks, what is the low pressure that usually triggers the compressor to start? If that is higher than 2000, you may as well increase the PMV setting to increase filter life.

 

[tmassey]

My lowest bank bottle will usually get to 1000 before the pressure in my highest bottle gets below 3500, and if I'm going to start filling bottles directly (to run the compressor more), then yeah, under 1500 is not exactly unusual. The value I've usually seen for PMV is 2000, and it's what the system had when I bought it. In general, I guess I want the setting that is the best balance of efficiency for the compressor and performance for the filters -- and I'd lean toward the compressor, because I'm so over on filter capacity.

I'll look more closely into leaks. There are enough valves and gauges in place that I can divide the parts up into separate sections and see where pressure may be leaving. What is a realistic expectation for the system to hold pressure for? Going from 4k to 1k over days or weeks time? Or truly like weeks without change? I wouldn't have thought that the compressor itself would have been able to hold pressure like that, and things would leak backwards through the compressor.

I'll fire up the compressor today for an hour or so and then shut it down with the following procedure:

1) Get to where I'm ready to shut down (bottles full or whatever)
2) Use the manual valves to crack all three drains and let the compressor run for a bit (a few minutes minutes) to make sure everything is well and truly drained and cool down the compressor by lessening its work
3) Close the drain valves and let the pressure build up for 60 seconds or so (in preparation for leak testing)
4) Shut down the compressor
5) Note the time and the pressure gauge settings

Then we'll see where the pressure goes.

I'd appreciate thoughts on step 2. I've had people recommend letting the compressor run for a bit without load to let it cool down: the only air that's moving over the compressor is generated by the flywheel... If you don't like my process in #2, what would/do you do? I'm not certain how necessary it is -- most dive shop compressors just seem to hit the pressure limit and shut down.

 

[rjack321]

8 to 10 mins between manual drains seems really short to me. I use a kitchen timer and go 18mins between dumps. But in winter when humidity is highest its cold here and when its actually warm here humidity is low so overall my coalescers don't seem to work very hard.

I would never run a compressor without a load. That's brutal on the 3rd stage floating piston which is going to bang around and wear way faster without a load.

 

[tbone1004]

@tmassey our big filter stacks typically don't go more than 1-2 weeks without being run *over breaks, summer, etc*, but there is never any appreciable drop in pressure *down to the PMV setpoint*. There are leaks in the system after the PMV's due to long runs of hose, multitude of valves, etc etc. but the PMV and check valves don't leak. Leaking after the PMV is obviously not ideal and will cause wear on the hoses, but other than that is pretty harmless. If it gets down to ambient you'll want to purge the lines before filling though.

Do not run the compressor without a load, you'll murder your floating pistons like @rjack321 said, but it also can't knock moisture out without pressure so you don't want those coalescers to drop too much in pressure. More frequent, but shorter openings are better than infrequent long openings. I.e. a 1sec purge every 5 minutes is better than a 3sec purge every 15 minutes. Same total open time, but because the interval is shorter the pressure drop is less. What I was talking about was while it was running opening the drains like you normally would to make sure they're completely blown out. Whether that is a couple of 1sec purges, a single 2-3 sec purge, etc. but make sure it's drained all the way out.
For letting it run and "cool down" you can let it run with an open fill whip which will let it run at PMV pressure, probably not the worst idea if you just got done filling a cascade bottle to 4500psi, but probably not going to make any significant difference in longevity.

 

[rjack321]

I wouldn't have thought that the compressor itself would have been able to hold pressure like that, and things would leak backwards through the compressor.

my little on compressor filter leaks backwards through the third stage valves fairly quickly, but I also drain that as I shutdown anyway so its at 0 pressure within a minute or within a day doesn't matter much (valves are probably dirty but not going to mess with them until they break)

There is a check valve between my little P0 primary filter and my secondary tower, the PMV is downstream of that 2nd tower and there's pressure in that tower for a months.

I monitor %RH in real time with a spyglass between tower1 and tower2 and replace tower 1 when the %RH is greater than 15-20%. I get a reasonable amount of hours from a P0 sized primary filter so I don't see a need to change what I'm doing. 10hrs summer, 20hrs winter

 

[tmassey]

I think you guys misunderstood 'cracking the drain valves and running the compressor'. By cracking I mean *cracking*. The pressure stays above half of max operating pressure. Output pressure is above 2000, interstage pressures above 400 psi and 80 (which is basically where it runs when the output is at 2000). So it's never "no load". Just low-load. And there's never any knocking or rattling: I know what to listen for because I hear it when the compressor fires up.

I'm not married to this. If I'm being an idiot, you can tell me. Sounds like I'm being needlessly creative. What is the best (or reasonably best) way to handle shutdown to end with a cool, dry compressor? @tbone1004 says drain it, shut it down and move on with my life... Anyone else have any thoughts?

As for drain frequency and duration: the solenoid timer for the old auto drain is set for 8 minutes and 8 seconds drain. I basically reproduced the time length, but because I'm a human I can simply watch and see when the liquid is out and close it. It's usually 2 seconds or less. I'm usually keeping an eye on things and futzing around so every 8 minutes is not a burden, any more than any other time with manual drains would be.

I try to pick and choose when to run it and try to keep intake RH below 60% where possible. Air temp is usually in the 80's. I have no way of knowing how much is in the coalescer, so it's hard to know what a reasonable time is. Is there a way to know what the status of the coalescer is to know what a more reasonable frequency would be? What is a sign you're letting the coalescer go too long? What is the risk if you do? It's not like an engine so you're not going to get hydro-lock, but can it do damage? I'm sure it could ruin a filter if liquid makes its way there...

I have a moisture eye in mine as well, but the glass is dirty and it's in a really awkward spot. Besides, it's a 20/40/60 IIRC, so it's not real proactive anyway...

Power was out so no running the compressor today. Thank you for the information on what you're seeing leak-wise. I'm going to temporarily throw a valve on the output of the filter board. That will give me a valve right at the output of the compressor (which is after the first stack), as well as one right at the exit of the filter board. I'll see what the pressure does. As soon as I can I'll let you all know -- I'm sure you can't wait!

Seriously, thank you for all of the information. I appreciate it. I'd like to protect my multi-thousand dollar investment...