Well-used Haskel checkout, cleaning, setup and suggestions?

[tmassey]

Say hello to the newest member of my family!



What do you mean I have an ugly baby?

It's a well-used Haskel AGD-7 that I just received from eBay. Supposedly it was "tested and working". I guess we'll see. And yes, it's pretty ugly.

I've got zero experience with boosters. I've seen one in operation exactly twice. So, before I do something stupid and break it (or me!), I'm here for advice. I own a SCUBA compressor and have some small experience with high-pressure gas so I don't overly need the most basic generic safety information, though I'm happy to be reminded, too. Anything Haskel/booster specific, though, and you should assume I know nothing.

Let's start operationally: what do I need to check out before I hook up gas to this thing? Any tests I should perform? Any suggestions to do -- or not do -- to start this guy operating? How can I verify that it's working like I could or should expect it to?

One specific question: the drive gas outlet has the elbow, but seems to be missing whatever device goes on the end of the elbow. What is that, and do I need it? I'm hoping it's something like, say, a muffler that I might want to have but I'll be OK today without it. But please let me know!

Then, let's talk cleaning. This will only pump inert gas, so I'm not worried about O2 clean. What I *am* worried about is breathing-gas clean. What can be done to make sure what it pumps won't kill me? I don't mean air-quality testing: I know about that. I mean is there something I can do to visually/otherwise check on this thing, or is there a way to proactively clean it?

While we're talking clean: the outside is pretty sad-looking. Any suggestions about how to safely and effectively spruce this thing up? (I'm hoping the inside isn't quite as sad-looking...)

I know it's a baby booster, better for rebreather tanks than OC doubles. The 2500 PSI maximum and 7:1 boost ratios are both tricky. But doing the math, it seems it will just barely meet my worst-case, which is pushing 1200 PSI of helium on top of 1200 PSI of 18/45 to let me make 3600 PSI of 18/45, from an equalized supply pressure of 1600 PSI. I realize it will do so slowly and using a fair amount of drive gas: from the chart at Haskel model AGD-7 Double-Acting, Single-Stage Gas Booster it seems that drive gas will be roughly 120% of boosted gas volume, so something like 100 CF.

I can live with that. I doubt I'll be doing that worst-case more than two or three times a year *tops*. The nice thing is that I use a set of SCUBA doubles as my helium storage bottles, so I don't need to worry about supply pressures lower than 1200 PSI. When the source gets to that point, I add EAN32 on top and go dive them!

Here are some thoughts on how I plan on setting this up. Initially, I plan on using a 444 ft3 bank bottle filled with air (from my SCUBA compressor), connecting a SCUBA first stage and using a a LP port to connect to the booster. Initially, I plan on putting a needle valve in there for initial testing and operation. Once I have a better idea of how this works, the intention would be to investigate a shop air compressor instead of using my poor SCUBA compressor to supply glorified shop air. But as I don't own a shop air compressor at this time, I'll start with what I've got. Given the frequency that this thing will be used at this time (not much), I'm not in a rush right now.

The source/destination ports would be plumbed with via Parker ST quick connects (like the rest of my compressor configuration uses). Both the source and the destination would then terminate in hoses and DIN filler adapters: one connected to the doubles that contain pure helium, and the other to the set of doubles I'm blending trimix into. (Same type of setup I use with my O2 (Argox) concentrator/compressor, nitrox stick, deco bottles, etc. Works really pretty well.)

Honestly, my idea with this was largely to get started with Haskel ownership and experience. I made a fairly low-ball offer and was surprised when it was accepted. But I certainly don't want to have to sink a bunch of money into it, which is why I'm looking for advice. I've got 30 days to determine if it's what I was told it was -- a "tested and working" booster. If it is, I'm OK with the limitations. If I need more down the road I can always re-sell it and upgrade to something bigger then.

Thank you for any help you might be able to provide! I really appreciate it.

 

[TrimixToo]

Yes, a muffler goes on the 90* fitting. Put "something" on there to keep crud out of the LP section.

With unknown internal condition, I'd want to rebuild it or have it done, and O2 cleaned while it's apart. Then you can use it for anything, assuming clean gas in. But for testing, just hook it up to a couple of tanks and see what happens. A shop air compressor with a valve in the LP side will let you control the cyclic rate. Use a water filter on the LP gas. Use a filter on the input gas, too (5 micron, IIRC...always a dangerous assumption, so check!). Cleaning the HP side tank afterward is a good plan.

Parts are easily available but they ain't cheap. I'm sure Haskel would be happy to supply new nuts and bolts, for example, but cleaning them up and painting them is a lot cheaper and new bolts and nuts of the right grade can be had commercially. The tubing and HP fittings are relatively cheap (I use Swagelok and even *that* is affordable) New valves are not *too* bad. I understand the cylinders are really expensive, so if it's leaking in either the LP section of the HP sections, exercising the option to return could be a good idea unless you can bore and sleeve them yourself.

All that said, if you got this at a really low price, any needed repairs might be a really cost effective option. At first glance it looks like they go for $7K+ new.

Good luck! I'm sort of envious.

HTH.

 

[tbone1004]

@tmassey that is anything but a baby booster, it is certainly rated for continuous duty for large volumes. Since it's an AGD, you get a much better stall formula than AGT's or AG's. 7*135=945psi on top of whatever the inlet pressure is which is ideal for this. Shop compressor will be a bit lower around 850, but not bad.

Static outlet stall pressure formula: 7 Pa + Ps (Pa = air drive pressure, Ps = GAS supply pressure)

As far as testing, what @TrimixToo said. Just hook it up to a pair of scuba tanks and see what happens. On the receiving side, I would plan on cleaning it afterwards, so if you have one that you need to O2 clean and/or VIP, it's not a bad idea to use that one. It's obviously not complete without the muffler that is also used to create backpressure to control the cycle rate. You can get around that with a valve on the drive gas side, but they should be working in tandem. If you have a shop compressor running it, then you won't ever have to worry about cycle speed though.
Keeping the LP side clean and dry is critical, and yes a 5um filter on the supply side is what is recommended.
I would hook the receiving side up first and pressurize it. That will let you check the check valves on the outlet side which are quite critical for proper operation. Not dangerous if they don't work and it won't break anything, it just won't actually boost. Will cycle normally, but don't get anywhere with pressures.

After that, since it looks that rough, I would take it fully apart and overhaul it with a full O2 clean, TLC with some steel wool on the chromed bits, clean/sand/paint the frame, etc. Probably looking at $1500 or more for the rebuild kit from Haskel though. That is the major downside to them, built like tanks, but the kits are not cheap

 

[tmassey]

Thank you both for the replies!

First, the bad news. It doesn't at this second run. But I may have hooked it up wrong/incompletely.

I've got a full HP80 hooked up to a reg. I've got a BC inflator hose from that connected to a QR to 1/4" NPT, through an inline valve (a typical 6000 PSI brass panel valve like used on every fill board you've ever seen), then into a 1/4" to 1/2" bushing, into the booster drive gas inlet as supplied (which included a ball valve on the inlet side). Nothing on the drive gas outlet side, and nothing on the working gas inlet or outlet.

I closed all valves, opened the inlet ball valve, opened the tank valve, and then slowly opened the inline valve, eventually all the way. I heard a small hiss of gas from the drive outlet. But nothing else: no movement of any kind anywhere. The hiss of gas was less than I expected: easily less than you'd hear when you hit your power inflator. I know the BCD QR is a high-restriction fitting, but my understanding is that low volume to a booster like this will cause it to run *slowly*, but I thought it would still run.

Like I said: maybe I've got it hooked up wrong. You both implied having a source and destination cylinder connected. I was going to see if things would move before I bothered just yet -- if for no other reason than it's gonna mean dirtying some of my breathing air hoses and DIN fillers *and* a tank and valve... But maybe it's required that there be some sort of pressure on the working side?

Or is this thing just broken?

OK, brass tacks time: I paid $1000 delivered, turns out from India. I figure it's from some sort of scrap metal buyer who's trying to sell anything of value before it gets melted down, and they have no idea of whether it runs or not. Returns are free, and I've got 30 days. If it works, I would not plan on returning it. If it doesn't, $1000 + $1500 + lots of labor = slightly more than I could buy a larger, *much* better looking, actually-tested unit from here in the US. (Of course, that one's not O2 clean, either.) I don't mind lots of labor (I could see the photos the same as you could), but $2500 plus lots of labor is too much.

So: first and most urgently, is this broken, or do I need to hook things up differently/more completely? And second, is this a good enough value to continue down this path?

Edited to add: A third item: is there any kind of service diagram for this thing? Something that would help me to be *able* to disassemble and clean this thing?

Thank you very much for your help!

 

[tmassey]

Mistake 1...

I decided to take off the inlet/outlet tubes and see if I could get an idea of the cleanness of the inside. In taking an elbow off, I managed to take the large fitting perpendicular to and at the end of the cylinder off. Makes sense: turns out it's an ORB fitting, not an interference fit like the NPT parts. Unfortunately, when I did so, a bunch of *other* parts came out, too! And I didn't see how they went together!

Good news is, the springs and ball bearing are bright and shiny! Are they exposed to drive or working gas? I would imagine they would have to be, but what do I know. Either way, I still don't know how they go together...



So.... anyone know where I can get that service diagram?

 

[TrimixToo]

I can't be sure you understand how to hook it up from your description.

There are two LP inputs. One, for LP gas, gets attached to a fitting that runs to the shiny HP heads on the ends of the smaller-diameter cylinders on the booster. It gets connected to the gas to be boosted to a higher pressure. The other gets attached to a fitting on the body of the large (central) cylinder. It gets connected to shop air, after a water trap.

The shop air volume controls the cycle rate. The gas to be boosted should not be restricted. I'll disagree with @TBone slightly by saying the cycle rate can be too high for O2 boosting without control over the shop air volume at some combinations of input pressure, output pressure, and shop air pressure.

Now on to what you took apart...more pictures would be good, here. But you should be able to find a parts manual that shows what goes where via your favorite internet search engine. Did you disconnect the input side or the output side? The balls and seats should all point "upstream," away from the output gas connection and toward the input gas connection. The spring goes on one side and the seat on the other.

Finally, if thing is rebuildable without major replacement parts, it's a bargain even if you need a kit costing $1.5K if you can DIY. This is a $7K+ booster new, I think, and even a good rebuilt will run a healthy fraction of that price.

 

[tbone1004]

@tmassey you NEED to have something hooked up to the inlet to get it to cycle. It has reversing valves, but it needs the inlet gas in there to cycle the piston, and it needs something on the receiving side *can just be a hose with a line valve on it*, but it needs to build pressure before it will cycle. That's why there is a minimum inlet pressure
You likely took the check valves off based off of the springs. Haskel has repair videos on youtube *from Haskel themselves*, so that'll show you how to put it back together.
You aren't going to get a "bigger" booster unless you get an 8" piston, but those are ridiculous. Maybe a higher boost ratio *physically smaller boost chamber, identical drive gas section*. A new AG30 is about $4500, a new baby booster is $2200 but that's practically restricted to 3000psi anyway and is VERY tiny in comparison. For reference, your AGD-7 moves 26.4 cubic inches per cycle. The ubiquitous for our industry AG30 does 3.1, and the baby booster does 1.23. You have a BIG booster.

@TrimixToo you certainly need some control, but I was referring to maintaining less than the 1 cycle/second that the booster is rated for. It's unlikely that someone actually has a 20cfm+ shop compressor in their garage. Mine gets hooked up to a 2200cfm compressor at work when I need to boost oodles of gas and that certainly needs restriction, but when it runs in my garage compressor/s *the one in the pelican case, plus another one in tandem*, it still can't get anywhere near the 1 cycle/second. That is often far too fast for booster CCR bottles though and the bottles get hot. Probably should have clarified that.

 

[tmassey]

Thank you for the replies!

So, I put it all back together. I found an exploded diagram in one of the rebuild kit's manual which let me put together correctly the check valve I accidentally disassembled (hopefully). It's tool-tight, but I didn't bother to ape down on everything or re-PTFE-tape: at this point, it can leak.

I put a couple of small cylinders on the In and the Out. I then opened the In so that it was flowing, and opened the drive gas. At this point, every 4-5 seconds I heard a small and very quiet puff within the hiss of the exhaust gas. I played with different things with the In and Out cylinders, etc. Eventually I found that even if I shut off the In cylinder, I'd still get the puffs every 4-5 seconds.

From my *very* limited experience with hearing a Haskel, the puffs should be a *lot* louder. They're definitely not much louder than the exhaust hiss. I'm sure they should be faster, too, but that's probably because I'm starving the drive gas with my BC QR connection. I've got video, but it's too big for SB, and I'd rather not try to upload it to Youtube just for this. It's boring, but if you wanted to hear it "run", you could.

So, does that mean that it's at least partially working?

At this point, I'm in full agreement: the first thing to do is to tear it apart into tiny bits and clean it within an inch of its life. But I can't in good conscience tear it apart until I decide to keep ownership of it (I already went too far with disassembly as it is...), and I don't *want* to take ownership of it unless I can have a reasonable degree of likelihood that it works even as-is. It doesn't have to work especially well, but I don't want something that's starting out broken.

The only thing that makes me think it's worth it is how clean the check valve was when I accidentally dumped it out. It looked as-new. Which is odd: because the external tubes and fittings I removed did *not* look new, outside *or* inside. But if the inside is as clean as it should be, and if all of the parts are there (except the muffler), there's a good chance I can have a working device.

Should I be worried about something much more serious? Maybe that the cylinder walls could be worn or scored that might not allow it to work correctly when cleaned? Or something else of that nature?



One other oddity: I've uploaded another photo, showing some of the other side. At the top you will see the drive gas inlet that ends in the ball valve. Below the elbow is a grey/silver box that leads to a silver tube. That grey box has a small round insert in it. My son happened to notice that when the booster makes its puff sound, that insert goes in, then on the next puff pops back flush, then back in, back flush, etc. I can't seem to find that item in any of the (*very* limited) diagrams I've been able to find. What is that?

While I'm at it, I will renew my plea: does anyone have *any* kind of exploded diagrams or service information for this thing? I know you mentioned YouTube videos, TBone, and I'm gonna go look for those. But I *hate* videos as a way to learn technical information, and I would *really* like some diagrams to go *with* the videos...

Anyway, thank you once again for the help you've provided me so far. I'm not looking for validation or absolution, but if there's any way I can have an idea if this is going to be a money pit I sure would like to...

 

[tbone1004]

@tmassey the box is likely the pilot reversing valve. Can't see it in the picture. I'll get to the pm's later tonight or tomorrow, been a weird couple of days.

Videos here
Haskel High-Pressure Products

 

[tmassey]

Thank you *very* much to those of you who are sticking with me and my everly-long posts! I really appreciate your help.

TL;DR: It doesn't work: there's missing o-rings. I've attached some pictures: what do you think about the internal condition?


OK. I've carefully hooked everything back up as I believe it should be, but it's not working -- it's actually working even less!

With everything connected, there is a constant, large rush of drive gas out of the drive gas exhaust (as there was when I first hooked it up), and no more sounds of operation (the very quiet puffs I was hearing), and my drive gas supply cylinder (an HP80) drops by about 100 PSI per second or so! When I compare to YouTube videos of ACD-7's properly running (like the one below), they do not have a constant rush of gas: they *only* have *strong* puffs of gas as the device cycles.


In searching for examples of videos with working AGD-7's, I stumbled across the Haskel cycling drive rebuild video and I watched it. It answered a few questions!

I had asked what that box at the bottom with the insert that popped in and out was. The video says it's the cycling valve. And within the first 45 seconds of that video, it states that if your booster is exhausting too much gas, it's a problem with the cycling valve.

Great. So after watching the video, I pulled the cycling valve spool. The video says there's 8 same-size o-rings across its length (and one smaller one at the end). Guess what? Mine only has 6: the two in the middle are completely missing.

So, here's my guess as to what's happening. When I first hooked it up, I had no working gas supply pressure. So the device didn't cycle -- but the missing o-rings allowed the drive gas to bleed right out the exhaust. When I added supply pressure -- but no destination back pressure -- there was enough pressure to cycle the port back and forth (while *still* bleeding lots of gas out the exhaust). But when I had both supply *and* destination back pressure, the device did not cycle anymore.

That's my guess, anyway.

I found that if I use my finger to manipulate the cycling valve spool, it would make the tiny cycling puffs I was hearing before without a destination load. But again, really quiet compared to the videos, and still with the constant rapid rush of gas escaping out the exhaust.

How could the two o-rings be completely missing like that? Are they still somewhere inside the unit? Don't know, but I'm a bit worried about that...

So, I've messaged the seller and told him that I can't make his unit work. I've told him how I've hooked it up, and what it's doing. Not about my discoveries, however... We'll see what he comes back with.

I've attached some photos of the spool, and the sleeve inside the booster. What do you think of the condition? The bright orange rust-looking stuff on the spool rubs off easily: it's no doubt rust from somewhere, but it's not the spool itself.



Again, thank you very much for your attention! I'm going to put some more details below: @TrimixToo wasn't sure how things are hooked up, so this might help him and others. Don't read it unless you're bored or have a question about my setup. But please let me know what you think of the internal condition!


Here is how things are currently configured and how I'm trying to operate it:

DRIVE GAS:
I have a SCUBA cylinder connected to a regulator, which has an LP port adapted to 1/4" NPT. This is connected to a hose, to a ball valve, to the intake of the Haskel drive section (which is 1/2" pipe).

WORKING GAS SUPPLY:
I have a SCUBA cylinder connected to a DIN filler, connected to a hose, connected to the IN side of the Haskel, which as you can see from the photo in post #1 is a tee, then plumbed to a fitting on the outer end of each piston, both on the same side.

WORKING GAS DESTINATION:
Across from the inlet on each piston is the outlet. Again, from the photo in post #1, you can see each are plumbed to a tee. That tee is now plumbed directly to a tee that contains a pressure gauge and a valve.

INITIAL STATE:
The Drive Gas SCUBA cylinder valve is closed. The Drive Gas ball valve is closed. The Working Gas Destination outlet valve is closed. The Working Gas Source SCUBA cylinder valve is closed. Everything is de-pressurized. I open the Working Gas Supply SCUBA cylinder. It has 1000 PSI right now, and it equalizes. The Working Gas Destination pressure gauge shows 800 PSI (close enough for now). There are no audible leaks.

STARTING OPERATION:
I open the Drive Gas SCUBA cylinder valve. I then slowly open the Drive Gas ball valve. I hear a reasonably loud, steady rush of air from the exhaust. The Haskel does not cycle. The outlet pressure gauge does not change. I can visibly watch the pressure gauge on the supply tank drop: 100 PSI/second or so from an HP80.