Shaft seal in split tank

Greetings all,

I am looking for any help you can offer in my problem. I have to run a turning 5/8" shaft out of an aluminum tank. The tank has a 1 1/2" hole in it and is split in half down the centerline of this hole. My first question is what kind of bearing and seal mountings could I use? How would that change if I wanted to seal the tank good to a high pressure gas such as Helium or Hydrogen and what would be a practical (cost effective) pressure limit of that method. I'm thinking of a flange recessed in the hole with an O-ring around it, but I don't see a foolproof way to seal the tank's O-ring with the bearing housing's O-ring.

The tank wall is currently 1/2" thick and is sealed by a groove with a recessed O-ring. There would be very little radial load and no axial load on this shaft and the rated RPM is nominally 1800 but up to 2500 max.

Thnks in advance.

and... Have a NICE day! Todd (use the actual number in my email address to reply

Reply to
tmckissickTWO
Loading thread data ...

You will _never_ get intersecting o-ring seals to not leak.

I'd plug the 1-1/2" hole with an aluminum plug, brazed in place or secured with serious Loctite, and remachine the tank o-ring groove through it, so that groove would at least be planar and uninterrupted.

Then I'd drill a new 1-1/2" hole, somewhere away from the tank seam, and run the shaft through that. At modest pressures, a simple lip seal will leak less than the tank's o-ring. At higher pressures, maybe a carbon face seal would be okay.

Me, I'd weld in an aluminum cup instead of the 1-1/2" hole, and adapt the coupling from a magnetic drive pump, and weld the tank seam.

I don't know what you mean by 'high pressure', but this split tank with static o-ring seal and intersecting rotary shaft seal sounds like a student design, and frankly the student needs a bit more study before anyone pressurizes that thing, especially with a gas.

Have it analyzed by someone who understands the ASME Code. Hydro test it with water. Install a relief valve. At least install a NHRA style scattershield. Your future children will thank you. Think about it.

-Mike-

Reply to
Mike Halloran

Mike,

Thanks for your thoughts. They have given me lots of valuable insight.

I like your idea of moving the hole away from the seam. It will complicate some other stuff, but that's going to be balanced out by this problem being easier. However, since I can't move the shaft, I think I will move the tank seam to the outer edge of the seal/bearing housing area.

I may have mislead you in calling it a tank. That was for simplicity of conveying the situation. It _is_ a tank, but in actuality it has a bolt pattern spaced on 2" centers all the way around. The _static_ O-ring seal is modeled after the seal on an automotive air conditioning compressor where the lines' manifold block mounts to the compressor body. Originally this system held 2500 PSI and we realize that without some serious expense, that pressure is long gone. And to your next question, No, this seam was not in the tank when it held 2500 PSI. However, with different seal techniques for both the seam and the shaft, we would like some kind of idea of the pressures we will be able to end up with. If that pressure is not high enough, the entire project will not be considered.

The magnetic coupling path has been considered and was rejected because of power loss due to eddy currents and cost.

Could you get me started in finding the ultimate max pressure that a setup like this could reasonably sustain? Without this guestimate, my company won't consider spending any money for the formal engineering services that you suggested. I'm not a mechanical engineer (or a student) but I just happen to have the most experience in the circle at this time. I wish I could go back to school for this knowledge, but that was too many decades ago.

And... Have a NICE day! Todd Use actual number in address to reply.

Reply to
tmckissickTWO

This didn't read like a typical "look at me, I'm smarter than you" response. There was a ring of truth to it. When I read about leak proofing a hyrodgen container through a split seal - I was agog for responses. Hydrogen can leak through solid metal., after all...

Brian Whatcott Altus OK

Reply to
Brian Whatcott

Brian,

Did I give off that impression? I didn't mean to. I am very aware of the issues of storing hydrogen which is why helium is our #2 alternative. I am merely looking for the general consensus of how much pressure we _can_ expect to hold. Whether that ends up 3 PSI or 300 or 3000 will determine the viability of our project. If there's a method that can be applied to this situation that will increase the pressure it can hold, I am asking for that suggestion as well as a rough pressure that would be appropriate. I think the break-even point would be around 400-600 PSI of He or 300-450 H2.

I'm guessing that in the end we will have to use some kind of a graphite/teflon gasket and a carbon graphite bearing/seal assembly mounted in a hole that's separate from the tank's split. Unfortunately, I don't know so that's why I came to the real world experts.

And... Have a NICE day! Todd (use actual number in address to reply)

This didn't read like a typical "look at me, I'm smarter than you" response. There was a ring of truth to it. When I read about leak proofing a hyrodgen container through a split seal - I was agog for responses. Hydrogen can leak through solid metal., after all...

Brian Whatcott Altus OK

Reply to
tmckissickTWO

Hi Todd:

How about a packing gland?

Don

Reply to
eromlignod

Hi Don,

That's another good way. I don't know any sort of pressures that could be sealed with that though. Does anyone have a rough estimate of a reasonably expected pressure I could count on? That's the whole goal here. That way we can decide if this is a move we want to undertake. The gas will probably be He and I know that like H2, it leaks out of very small holes easily so what material would need to be used in the packing?

And... Have a NICE day! Todd (use actual number in address to reply)

Hi Todd:

How about a packing gland?

Don

Reply to
tmckissickTWO

Dear tmckissickTWO:

They are used in hydraulic cylinders up to 10,000 psi. Either of these gases are going to leak through the seal material anyway, if not the metal. Your efforts will only reduce this to (perhaps) an acceptable level. How long do you want to hold this "charge"?

David A. Smith

Reply to
dlzc1.cox

Hire an engineer to scope out your particular situation and make the guesstimate. Plenty of skilled engineers are out of work now, so it doesn't have to cost a fortune. Unless you elect not to do it. High pressure gas is dangerous stuff.

Spend money on engineering services, or on funeral services and legal services, your choice.

If I had _two_ such tanks, pretty much identical, I'd pressurize one, WITH WATER, until it failed, then use the other at a lesser pressure.

If the project's business viability hinges on adapting existing hardware at zero cost, then it must not have sufficient upside to offset even one of the glitches that you don't know about yet. Save time and abandon it now.

-Mike-

Reply to
Mike Halloran

WOW! You are not a PE or PEng and you are attempting to design something that possibly might end your life. You are not yourself an expert... but rather than hire someone tangible that you can physically see and touch... you've decided anyone on this list is good enough for you.

The responses I've read are good... especially the advice to hire an Engineer... and I'd add: have him stand beside it during the testing phase.

Reply to
Alan O'Neill

And ask him about the fatigue strength of aluminum, and have him tell you how many times you get to fill the tank... before it fails.

David A. Smith

Reply to
dlzc1.cox

For a " quick and dirty " assessment of the danger...

Energy approx. = Pressure X Volume

Convert energy to " hand grenades " ( 1 hand grenade = 600 kJ ).

N.B. I find this unit very useful for reports to non engineers.

-- Jonathan

Barnes's theorem; for every foolproof device there is a fool greater than the proof.

To reply remove AT

Reply to
Jonathan Barnes

David,

Thanks for your response. It's nice to see one that's on topic. Obviously hydraulic fluid and He are two different critters. Is there a place to reliably determine this property of a material? In one place, a given material lists gas permeability or other qualities related to this question, but I can't seem to find any info like that for assemblies such as a specific packing material or a certain cartridge lip seal. Carbon graphite gives a 'little' info on some of the pages for how much pressure they are rated for with a certain gas, but I want to compare apples to apples. As I've said numerous times, we haven't finalized the gas to be used. I'd _like_ to use the lightest one at the highest pressure possible, but I am willing to settle for the lightest one we can seal. If that ends up being no gas at all but molasses at 10 PSI, then so be it. My long term goal is to hold 2500 PSI of He for 20 years, but I know that's impractical. Could you give me some practical numbers that fit in this equation? Am I even close to think it's possible to seal this tank to 400 PSI of He and make that charge last for a year or more? Is that possible without spending $1,000,000?

Another consideration is friction. Wouldn't a packing gland tightened down enough to seal this scenerio produce an enourmous amount of drag compared to some other alternatives?

And... Have a NICE day! Todd (use actual number in address to reply)

Dear tmckissickTWO:

They are used in hydraulic cylinders up to 10,000 psi. Either of these gases are going to leak through the seal material anyway, if not the metal. Your efforts will only reduce this to (perhaps) an acceptable level. How long do you want to hold this "charge"?

David A. Smith

Reply to
tmckissickTWO

Boy some of you really take the cake. I ask a question and I get jumped on. I am not the one who will design the end setup. I will not build it. I will not test it. I am only trying to find out if it is worth spending money to pursue it. You guys are giving me hand grenade safety equivalents, critiquing my education and experience level, and insinuating that I am capable of outsmarting (or the opposite) a foolproof device. I don't see any answers to the original question yet. If I had asked if it was possible to send a person to Uranus in 3 days for under $100, you'd all have an answer and it wouldn't have centered around the dangers of living in a vaccuum. If I'd have asked if it was possible to power my son's skateboard to 5 MPH for under $10,000, I'd get more answers. That's because those lines in the sand are very clear to everyone. All I'm asking is if it's possible. If so, I will hire the appropriate engineering services to ACCOMPLISH it. Only after that step will I worry about the steps involved in testing it.

I am certainly not going to trust ANY opinion found anonimously in here. I came here to get some advice and direction. I will do my own research once I have suggestions in hand. You don't have to be an hypocritical PE or official engineer to understand concepts and create designs. There are people out there that have enough experience and education to do things that the 'professional' person is supposed to do. If I'd have been stopped by that rule, I would have had to pay an average of 4 times as much to rebuild my house, pour my driveway, wire my house (all the way to the pole), plumb it, rebuild my transmission and my friend's hydrostat in his 250,000 tractor, automate some of the numerous steam plants that I've done, set up some of the automation lines I've done or implement some of the innovative boiler changes that have been very successful. I'll bet you would have driven all the great inventors of the past away.

I've seen rants in here about jobs going to China. I'll bet I know why. There's no focus in this country. Everyone seems to want people to keep jobs here, but rather than try to help someone out, you'd rather be critical to support your own personal ego. THAT is precisely why I have to do this on a shoestring. I am the direct result of that downsizing seeing as how my income has dropped to 1/3 of what it was 3 years ago. I currently have 2 offers to design and build this prototype for NO money up front, but I'd rather keep every bit of work as local as I can. I'm suppporting Nebraska first, and the US second. Unfortunately those offers are overseas and come with an ownership price. Without a working proof of concept of this design, I can't find enough funding to build a scaled down model for a lousy $2000. The end design has been praised by a number of trusted engineers as well as the team leader of the NASA team attempting this a few years ago. He commented that this is a fair amount better than what they came up with after 13 years and millions of dollars. All my eggs are in this basket so I'm taking it very seriously. It's even taken the ultimate domestic toll. Therefore, I have no resources but to do as much as I can. The closer to the final design I get on my own, the better the proof of my concept with my limited funds and the cheaper (translated sooner) the engineering services it requires. I'm not stupid in the arts of pressures or mechanical, but I'm sure starting to feel that way in this crowd. Unfortunately, I've found that if a person tries to cross the bounds of what's been done before (even just over the lines), then it's very easy to get categorized with the over-unity / perpetual motion / something for nothing crowd. Why don't some of you put your money (or effort) where your mouth is and support someone trying to make a difference. I'm not going to let this one go like some of the others... I was told then to get it patented before I could get funding. When I came back with the patent in my name, those same people wouldn't touch it because 'there might be conflict of interest'. Financial support would sure be nice, but in its absence, one would hope to get some kind of advice type support somewhere. By the way, the business prospectus supports about 120 employees... 4 of which are engineers.

Too the rest of you, I sincerely appologize and do not include you in the above. I'll be a good little conforming sheep from now on. I'm sure I'll get a little heat for this one, huh?

And... Have a NICE day! Todd (use actual number in address to reply)

(others wrote)

For a " quick and dirty " assessment of the danger...

Energy approx. = Pressure X Volume

Convert energy to " hand grenades " ( 1 hand grenade = 600 kJ ).

N.B. I find this unit very useful for reports to non engineers.

Barnes's theorem; for every foolproof device there is a fool greater than the proof.

Reply to
tmckissick2

Reply to
eromlignod

Thanks very much. That's pretty much exactly what I was looking for. I didn't realize it would be even that high a pressure.

Is there any kind of de-rating factor to use for different working fluids? How can I compare it with air vs. Nitrogen vs. He. I'm assuming that the rating is for just air. Am I correct?

And... Have a NICE day! Todd (usa actual number in address to reply)

Reply to
tmckissickTWO

Dear tmckissickTWO:

It is a rating for gas pressure, taking no account of

- diffusion of the gas through the material

- reaction of the gas with the material

- diffusion of gas across a rotating or reciprocating shaft (and its non-uniformities)

The gas contents *will* go away slowly over time. The amount and type of material will just determine how fast.

I've forgotten what this shaft is supposed to do. Is is rotating something inside the tank? If so, are you familiar with the operating principles of a "mag-drive" pump?

David A. Smith

Reply to
dlzc1.cox

OUCH! boy that really hurt me. But your still breathing, eh? (your welcome)

Reply to
Alan O'Neill

I strongly suggest you all hold on the the e-mail where he said he'd hire a PE because if he doesn't and something blows up and they recover your messages from his computer "you are legally liable."

Every response to any e-mail here should probably contain a professional's caviate, don't you think?

Reply to
Alan O'Neill

Oh, nonsense.

Reply to
eromlignod

PolyTech Forum website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.