Filter Stack Plans/Design

[GKaplan]

Nice Blending System!

Mike Casey (Our Filtration Expert) Has a few concerns about filling a Trimix with your current filter system. I tried to find a contact email on your profile, no luck. Please contact Mike directly at mcasey@lawrence-factor.com make sure to remind him what this is regarding.

Thanks Jason,
Gabe

 

[divingtactics]

Okay, thanks Gabe - email sent to both of you.

 

[iain/hsm]

Apologies for butting in here but I cant resist a good ID drawing. How I wish every dive fill system supplied drawings like this If you don't mind how long did it take, what drawing program.

Could you mark up or post the maximum working pressure on the LP bank and booster drive gas cylinder. Also cylinder sizes (in litres water capacity) LOL Also the pipework OD diameters of the tubes otherwise us old folk get confused what is HP and what's LP in the system run outs

First view IMHO from the lay out the Haskel should have the LP relief valve upstream of the 1/4 turn valve (not downstream)to protect it from a HP creep from the HP reg when in the closed position. Also a blind tee and an additional 2nd ball valve on the OPV would be useful for an LP air source to the Haskel in the future as IMHO the AGT 15/30 wont run long from your LP supply if it goes through the AE regulator. The female outlet Din is I assume for back fill to one of the Male Dins

On the Undersea filter why such a small filter? and why the by pass? Are the Swagelok plug valve brass Nupro 3000psi max WP or stainless? Also the 1/4 turns on the Bank and drive gas cylinders what is the tube diameter.

Then again on HP why not consider using the cheaper 6000psi Sherwood line valves?

On the oxygen diffuser from the GA drawing the oxygen inlet look very close to the LP air intake filter if I'm reading the GA correct. Again IMHO the distance between the oxygen intake position and the muffler is too close

I would fear loss of too much expensive oxygen wasted into the surrounding room atmosphere. While it wont affect the accuracy percentage or performance IMHO it's just a waste of resource.

However if the intake filter was put on the end of a short extended say 10 inch tube.
You would use all the oxygen flow and not waste some to atmosphere. The potential of the reciprocating compressor to push swept volume out with your oxygen would be reduced.

An air compressor only takes displacement against charging rate into account because as ambient air is considered free, pulsation dampening is not normally considered, and the extension tube gives you this.

Cracking work though mate.

Just goes to show that in gas transfer the more you put in the more you get out. Iain Middlebrook

 

[divingtactics]

Hi Ian,

All input and feedback/advice is welcome.

The drawing is in AutoCAD and is not to scale. It has been a work in progress and probably took about 20 hours I suppose. I have a few versions as the panel has grown and it is now called Frankenpanel as it keeps morphing.

LP Banks are rated to 200 Bar working pressure, I normally pump them to 220 and they cool back a bit to around 210/205 depending on how empty they are.
The HP bank is rated to 350 Bar but I cut it off at 330 and don't let it drop below 250 Bar.
All the tubing is 316 SS 1/4 inch x .9mm rated to 5000 PSI working pressure.
All banks are 50 litre.

I never leave the Haskel running by itself and turn the regulator down to zero and turn the drive bank off at the valve when not in use. The OPV is there to stop the booster running over 175psi if I turn something the wrong way by mistake.
I don't have a big enough LP compressor to keep up with this hungry Haskel and the single bank keeps up fairly well. I do however have two 3000PSI Swagelok plug valves between the booster, so I can supplement the drive gas with the two LP banks if need be. If I do this, I always turn the drive gas bank off and just run the other two through the cross and into the booster. Please clarify what you mean by the booster won't run long if being driven by the AE regulator? Are they dirty, are they troublesome...??

Yes, the DIN on the output is there so I can boost back through the panel by connecting the short whip in there. The Tee off the Oxygen main line and the cross off the Helium main line is so I can boost straight from the banks of Oxygen and Helium without changing any fittings and the digital gauge is there so I know exactly what is left in my banks after boosting those gasses. It works well. The analogue gauges are really panel bling and there so I don't have to turn my head and look at the bank regulators, which are just welding reg sets.

The Undersea filter is so small because it is about the same capacity as the P21 on the Oceanus that is supplying it and is my first attempt at better filtration. Most likely I'll put a 33-inch tower on there in place of the Undersea one and take the dessicant and activated carbon out of the P21 repackable filter, so it just acts as a separator. I have fitted solenoid controlled timed auto dumps to the 2nd stage separator and the final separator/filter (P21) and I have them set to dump every 10 minutes. The bypass is so I can save filter life when filling only the Booster drive gas bank, but I will probably do away with that part and the drive gas bank in favour of two more LP air quality banks and filter everything with the 33-inch tower.
Yes, the valves on the banks are Swagelok 3000 PSI plug valves with harder 90 duro o-rings and they seem to be okay to date if I move them slowly and not under full pressure. I have had to rebuild them a couple of times when they had softer o-rings - mainly the small o-ring, which is a Parker 902 odd size, a little smaller than an 010.

All the valves that are running HP are Sherwood 6000 PSI models.
Oxygen are Hoke 5000 PSI needle valves, apart from the one feeding the booster, which is a Sherwood as it has a much higher Cv = Flow Coefficient.

On the nitrox stick, yes the intake is close, but there are about 3 practice golf balls as baffles between the filter and the intake. I only pay $28 for a 50 litre cylinder of oxygen, so I'm not overly concerned about losing a few litres here and there at .0028 cents a litre But you are right, it could be improved with a length of tube. Since I got the booster, I tend not to bother with the nitrox/trimix sticks anyway. The drawing shows the intake of the compressor permanently connected to the sticks, but they are not normally permanently connected, I just plug them in if I want to use them.

I will also add a second Oxygen bank in there too in the not too distant future.

Thanks for your nice comments and advice, I know that you know your stuff.

Kind regards,

Jason

 

[iain/hsm]

Thanks Jason.

Over here we tend to class LP low pressure at 5 to 16 barg and 200 barg as HP (I'm easily confused.)
makes more sense now. Also why you by-pass the undersea 2nd filter air for just Haskel drive air.

Still think its a waste of good HP filtered air, the HP compressor and power just to produce drive air for the booster.
You guys should be able to get a cheap oil-free low pressure 10 barg nail gun compressor, and save your HP some grief.

At a guess premature service on the HP air compressor will cost you more than purchasing a low pressure drive air unit.
(well it would with us lot LOL)

With your future plans also consider using the Haskel to boost medium pressure air/nitrox/ or helix from say your
200-220 bar storage mark to the maximum bottle pressure of 350 barg. You can also blend this first into a flask with
your HP oxygen by the partial pressure method and use the booster to root it up to storage pressure.

This is a most effective way of producing storing and dispensing gas and air.

Your little HP air compressor runs into a medium outlet storage pressure cylinder not much more than the BPR setting say
150 barg. Its at its least worked output for rod load, heat dissipation and wear tear and lubrication (tribology) of the compressor.

Most little 3 stage recip compressors are over clocked anyway so mean time between service and service life is extended.

From a compression ratio point of view the Haskel takes this 150 bar and at a CR of 2.33 compresses up to your 350 barg
to the final storage banks. Now for filling scuba cylinders you have both decanting and dispensing options.
The HP air compressor life is longer and the Haskel run within its own CR design.

When the HP oxygen cylinders are lowish you make nitrox by PP and use the booster to jack it up to storage pressure and I agree
it almost negates the need to add oxygen to an oil lubricated air compressor. Something I wish we were all aware of the risks
and perils at the ppm level of exotic chemical concentrations oxygen and lubricant synthetics create. Iain Middlebrook

 

[divingtactics]

Thanks for your advice Iain,

Hmmm, an interesting concept using the booster as you suggest, it does make sense. At present, I only have one 350 bar bank and the other ones are all 200/220. The 350 bar one is hard plumbed into the panel via tubing, so I'd have to look at putting a tee and valve in there and then hooking another separate whip up to the outlet of the Haskel and into the bank. I have all of this equipment already, so it would be reasonably easy to do.

Regarding a LP compressor, it would need really good filtration and ideally a drier for me to feel happy about using it in the Haskel and I also prefer to keep to single phase power. I may find it difficult to get something that will do the job within that criteria. The other reason I use the Banks to drive it is that I like to hear the booster cycling, so if something goes wrong with it, I can shut it down before I blow the house up. I suppose with every system, there are compromises that we have to make and filling that extra bank may cost me more in the long run. It is pretty much just me who uses the system, so my HP compressor doesn't get too much use. I think it has about 65 hours on it since I bought it two years ago - yes, I need to get out and dive more Soon, I'll be buying a JJ-CCR, so it will get even less use in years to come. If it needs fixing, I'll pull it apart, source the bits and put it back together again. I'm changing the oil every 50 hours or two years and at the moment, it seems to have no trouble getting up to 300 bar. I think if I do adopt the concept of boosting LP air to HP, then I would definitely need to look at driving the booster with a shop compressor.

On the filtration side of things, I was going to get a machine shop to make me a filter and element from 7075T6, but I priced up a length of billet today and it isn't economically viable to bother doing it myself, so I'll most likely get a LF 33-inch tower, subject to what Mike and Gabe suggest at LF. What are your thoughts on removing the dessicant from the P21 and just using it as a final separator? Of course, I'll be adding a PMV set to 150 bar downstream from the 33-inch tower also and I think I'll use hopcolite in the filter, even though I use an electric motor on the compressor.

Great discussion, I appreciate your input.

Jason

 

[divingtactics]

Thought I'd post a couple of pictures of Frankenpanel FYI.

The thin whip is actually a Miflex HP 7.5 metre one, I wouldn't trust them underwater but they seem to be good up to 345 bar on the panel. The scuba tank on the lower picture with the plumbing into it is a 240 bar rated ally tank reservoir for the auto drains off the compressor and then the open valve feeds into a 100mm plumbing pipe with a rocker cover breather on it - this secondary one is just a breather. All liquid is captured in the scuba tank and then there is a ball valve on the bottom of it to get the residue out after a few hours of use. I've virtually rebuilt the compressor and have fully rebuilt the booster from scratch. It's been a fun little project and I'm not done yet, still a few things to add in there. Any suggestions, modifications, additions are welcome. Some things I have done for specific reasons, others because I couldn't think of a better way at the time. When it is finally all finished, I may have a stainless panel cover made for it, but I don't think so, I like looking at all the fittings and stuff - panel bling.


Panel, booster, compressor, banks and Nitrox/Trimix sticks



Frankenpanel



Booster AGT15/30 Haskel



Bauer Oceanus compressor with auto dumps

 

[iain/hsm]

Quite agree on you not making your own filter shell. Not withstanding the vessel design aspects, thread tolerances, concentricity. You then have to hard anodise it and have ring and plug gauges to gauge the threads. Although a dollar and cents call I would much rather have someone sell you one cheap than have you make yourself. That said a self made stainless repack cartridge to fit inside is a different matter.

Glad you liked the booster idea, gas routing from medium pressure is a good trick to have and a great addition to a panel. While not totally negating the need for an ambient air compressor, but in your case anything that reduces putting oxygen and helium throughout that little Bauer is a help.

For the amount of use you do IMHO I wouldn't automatically go for a 33 inch tower, frankly you just don't push enough gas through it and any and all carry over, emulsified oil and water from that compressor will simply hang on in there and stagnate the system, further lack of use will create a C02 build up as the synthetic zeolite acts as a pressure swing adsorber and the next time you use it a slug of C02 will occur.

I know the smaller towers say 16 inch or 18 inch are cost for length more expensive but consider your use and plan on filter changes every three months regardless of use. Small may be neater and cheaper, again another option is go with the long 33 inch tower but use a short double stacked repack filter system.

Thanks for the photos I have leant that the Undersea filter you have is bigger than I first thought, but again would suggest you consider taking the by pass line directly to the Haskel storage cylinder directly from the plug valve and not routed up through the distribution board that you share with that HP oxygen line.

In effect dirty air contaminates the distribution block every time you fill the Haskel cylinder. IMHO that bit need a re think.

From the CO removal point of view I have another suggestion, Over here in UK we are pretty mean when it comes to CO catalyst chemicals at around £60/Kg
and all the throw away filter cartridge suppliers add them to the normal drying oil adsorbing filter cartridge chemical length reducing its effectiveness for water vapour absorption and you throw the lot away each and every time.

We simply use a mini filter at the end of the heat exchanger/filter tower run and that filter contains only hocalite, You change it every 1 or 2 years, its protected by a visual indicator 10/20/40 RH and is under pressure with the BPR at all times. Good news is you have already that pretty useless at the moment Undersea filter going spare make a stainless repack cartridge and your good to go.

OK dwell time due to the high gas velocity because or the small diameter of the filter cartridge and its short length is going to be much too high, if we knew the internal diameter of the filter bed and bed length you could work out an effectiveness but it wont be very good but its a thought. Iain Middlebrook

 

[divingtactics]

Thanks Iain,

About the only other gas now that is fed through the Bauer is the last 10 bar out of a 50 litre Helium cylinder when I transfill from the hired one to my own and a little bit of 32% that I may be moving from a series of single tanks into doubles after a cave diving expedition. I can't think of the last time I used the sticks.

Yeah, that's what I thought about the 16-inch versus the 33-inch, the bigger one and the cartridges wasn't much more in cost, so I naturally thought the bigger one would be better. If I was to keep the cartridge at 150 bar pressure all the time with the final PMV, would there be that slug of CO2?

I'm going to ditch the dedicated Haskel drive bank and put two more 200 Bar 50 litre cylinders in there and separate the last one for drive gas but still have it cascaded to the other two like it is now. This means that all input gas will be double filtered and the bypass of the secondary filter will also be ditched.

I have had a filter cartridge made for the little Undersea unit, but it was made out of standard aluminium and is not anodized - do you see this as an issue? The internal dimensions are 28mm diameter and it is 160mm between the top and the bottom of the two felt inserts. It is currently about 50/50 of activated carbon and MS. So I could convert it into only Hopcolite as you suggest, but should I get the repackable cartridge remade in stainless steel? I thought ally may be okay, but maybe not and it is probably cheaper for me to have a stainless one made than to get this one anodized if that has to be done. I haven't put the aluminium one into service yet as I still have an Undersea one in there.

Thanks,

Jason

 

[iain/hsm]

The more I think about it the more I would go for the 33 inch, cheaper by length and offers some options.
It gives you the opportunity to experiment later on making two short cartridges double stacked, with the only downside you have to fill the darn thing with a small compressor first and then maintain the storage to BPR set pressure

To date you will only see this on helium gas reclaim and recovery systems used offshore and on pure air missile guidance compressors. But I like building stuff.

The down side is even with a single short tower or even the 30" stack you still are faced with the fact your not pushing much gas through it, you should overcome this by flushing after a long lay off, but maintain at least the BPR pressure at all time while flushing.


Regarding standard aluminium cartridges, if you look at the Bauer range they are thin walled aluminium, we have discussed before on SB I think modifying the standard tube into a repack

Also the fact that wet or damp activated alumina, and commercial molecular sieve will corrode out a Bauer tube in a year or two. So the answer I guess is no, but its in reality just a simple inspection call. More later. iain