Thick plate materials for underwater housing in harbour 1

[JeffFry]

Howdy,

I need to migrate the design of a large underwater instrument housing from Al 6061-T6 (hard anodised, with sacrifical anodes) to a near-maintenance-free material for permanent underwater installation in a harbour environment. I would appreicate any suggestions for this replacement material.

From my reading to date, my first guess is Copper-Nickel (90/10 or 70/30) or Ni-Al Bronze. I am, however, limited by availability of plate size, as the instrument housing is ~40 inches in diameter and almost 8 inches thick (think big hockey puck ). My aluminum prototypes to date have been hogged out from from 8" plate, which I'm finding difficult to source in the above materials. I also need to weld several structural pieces onto the housing (similar material).

FYI, I am looking into casting this part as well if the cost makes sense, so if anyone has any comments on corrosion performance of these materials in cast form I would appreciate them as well.

Thanks a bunch!

 

[MikeHalloran]

What's the net wall thickness after you've hogged out the aluminum you don't need?

OR,

How deep is the harbor?

Just an odd thought: You could make the housing out of relatively thin plastic, if the interior is filled with something like jellied silicone oil at ambient water pressure. Is that an option, or do your electronics absolutely require a gaseous environment at sea level-ish pressure?





Mike Halloran
Pembroke Pines, FL, USA

 

[JeffFry]

Thanks for the quick reply!

Just to help clarify, the cylindrical part of the housing is essentially in the form of a hoop/annular ring, with some reinforcement. The end caps are essentially large flat plates, with some weight saving mill-outs. I also use an internal custom chassis to reinforce the top and bottom end caps near the center.

Wall thickness minimum on the hoop is just over 1/2 inch, but the hoop profile has an integrated flange that would demand a 3 inch wall thickness if I wanted to fabricate from pipe/tube. I'll admit it may seem kinda silly to hog out the hoop from an 8 inch plate, perhaps, but I was not able to readily get a 40" OD pipe/tube section in appropriate materials and with the wall thickness needed. Maybe I'm not looking in the right places.

Note that I'm fairly confident about my structural analysis, but it was completed with Al 6061-T6 specs.

The nominal operating depth for the device is about 50 feet. I'd love to be able to pressure balance this beast and use delrin or something similar, but I need near 1-atm air pressure inside for a huge amount of tightly packed off-the-shelf electronics.

Thanks again.

Jeff Fry
NL, Canada

 

[MikeHalloran]

Hogging out thick plate makes perfect sense these days; it's the sort of thing you can do by running a CNC machine unattended.

Unlike aluminum, CuNi can be welded without significant loss of strength, so you could fabricate your housing from a rectangular piece of sheet rolled into a short tube, with flanges cut from available plate.



Mike Halloran
Pembroke Pines, FL, USA

 

[JeffFry]

Thanks a bunch, Mike. Rolling a sheet and welding on the flanges is probably my best bet. Going with thinner plate certainly opens up the availability of materials. I'll probably oversise the plate so I have some to machine off. I don't suspect post-roll stress relieving is overly necessary.

Would you recommend the 90/10 or 70/30 (or the Ni-Al Bronze for that matter)? Also, I wasn't planning on using cathodic protection or any finishes, in order to maximise the biofouling performance of the Cu alloys. Does that make sense? Im going to run my structural analyses again with the new parameters once I decide on a material, but I don't see too many structural design mods required when compared to 6061-T6.

Hope you're not seeing too much fallout from the gulf storm(s). You southern US folks certainly have the sympathy of us Canucks through this hurricane season from hell.

Jeff Fry,
NL, Canada

 

[MikeHalloran]

We were lucky; just lost a few shingles.

This page may help you decide:

http://www.copper.org/applications/cuni/txt_swater_corrosion_resistance.html

Applying a finish would be counterproductive.





Mike Halloran
Pembroke Pines, FL, USA

 

[JeffFry]

Great link - copper.org has a wealth of info on their site. I'll let you know how it all turns out. Thanks again - you've been helpful to this poster noob.

Cheers,


Jeff Fry
NL, Canada

 

[EdStainless]

Stay with 70/30 if you can. It really will stand up to more abuse.

= = = = = = = = = = = = = = = = = = = =
Corrosion, every where, all the time.
Manage it or it will manage you.

http://www.trent-tube.com/contact/Tech_Assist.cfm

 

[JeffFry]

Thanks, EdStainless. 70/30 is the way I was leaning - I'm replacing 6061-T6 with Cu-Ni in my underwater housing design and the somewhat higher strength of 70/30 versus 90/10 is attractive in order to minimise any wall thickness design changes.

That brings up one other question if anyone can comment. Are the M20/25 (hot-rolled) temper effects consistent throughout the thickness of a 4 inch thick plate and in weld HAZs, and is it necessary to stress relief or re-temper after substantial welding?

Thanks!


Jeff Fry
NL, Canada

 

[MikeHalloran]

You haven't priced the 70/30 plate yet, have you?

;-)



Mike Halloran
Pembroke Pines, FL, USA

 

[JeffFry]

Mike,

LOL - I have quotes coming from 2 mills (Alaskan and HM Hillman), but don't have them yet! I'm expecting both back by tomorrow, and am a little concerned about $, yes. I know there will be a heavy premium for the thicker plate at least (it looks I can get the 4 inch and 8 inch, that is if I can afford it!), and I am looking at other ways of fabricating the system from 1 inch or thinner plate, such as your suggestion for rolling the external wall. Mind you, the 8 inch AL 6061-T6 plate I am currently using isn't cheap!

Price will drive the direction, no doubt. At least I don't have to use Titanium!

Jeff Fry
NL, Canada

 

[DAVIDSTECKER]

If you want a casting look into Aluminium Alloy 357.
If you want to machine and fabricate look at 5086.