Jump to content


drsteve

Member Since 20 Nov 2005
Offline Last Active Oct 20 2014 02:11 AM
-----

#337965 New Housing "Airlock" Vacuum from Backscatter

Posted by drsteve on 09 October 2013 - 02:10 PM

The rule of thumb for static face seals is to design around a nominal 20-30% compression with a worst case including tolerance stackup of  >10%.  Reciprocating seals on shafts (i.e. buttons) are typically designed around 15% compression with a minimum compression  >10%.  It varies somewhat for different materials.

 

I measured my 5D housing since it has a traditional rectangular o-ring groove.  The vacuum system is on my 5D3 housing which uses a dovetail groove.  The calculations are a bit tricker for the dovetail, but I pump my housing down by 10inHg = 4.9psi.  The cross sectional area of the rear door is roughly 50 in^2, so the vacuum is applying an additional 250lbs of force.  Jean said that they set the latches to apply 45lbs, so an additional 250lb should be safe ;-)




#337929 New Housing "Airlock" Vacuum from Backscatter

Posted by drsteve on 08 October 2013 - 09:40 PM

Jean, you inspired me to measure my 5D housing to see how well it conforms to the standard mechanical engineering face seal guidelines.    Attached is the face seal design from the Parker o-ring catalog.  The first thing to notice is that it assumes sufficient clamping pressure that the surfaces are pressed together so that the two metal surfaces touch and the o-ring is completely trapped within the groove.   My 5D housing uses a standard 1/8" ring.  The terminology is unfortunate because a 1/8" o-ring has an actual thickness of 0.139 +-0.004.  The guidelines specify that the groove should be between 0.101 and 0.107 deep, which would give a compression ratio of 20% to 30%.  The width for a standard o-ring should be 0.177 and 0.187 to give room for the o-ring to expand laterally when compressed.

 

The groove in the housing measures 0.100 deep and 0.136 wide.  The depth conforms to the standard  depth assuming enough clamping force to compress the two halves until they touch.  However the cam latches do not provide enough force to do this.  When closed, there is a gap of 0.032 so that the o-ring compression is only 0.008 or 5.7%  This is seriously under compressed, which is the point I made in the original message.

 

I believe that you test housings to insure that they seal reliably, however since the initial seal is seriously under compressed it will work only if everything is perfectly clean and greased and nothing gets bumped in the rinse tank.  Of course everyone everyone should keep their o-rings perfectly clean and greased, but there are plenty of stories of someone getting a stand of hair or a piece of lint under their o-ring causing it to leak. 

 

The if o-ring were designed per the guidelines, it would be much more robust.  A hair or a spec of lint wouldn't cause it to fail.  The good news is that the vacuum system provides enough clamping force to make the seal much more reliable.

Attached Files




#337476 New Housing "Airlock" Vacuum from Backscatter

Posted by drsteve on 25 September 2013 - 04:18 PM

Adam, you are partially right, but you have confused the hardness of the o-ring material (durometer) with the o-ring resiliance, which is the tendency to become permanently deformed.  The durometer rating is related to the clamping pressure required to make the seal, whereas the resilience depends primarily on the material used and the percent compression. If you have enough clamping pressure (ie bolts) you don't need to use a soft material.  For example Buna-N o-rings are inexpensive and have good resilience, which is why they are so commonly used.  A face seal using a 70 durometer Buna-N o-ring with a 15% compression is good for over 3000psi without even trying.  If you use bolts, you get plenty of clamping force.  You don't need a torque wrench, you can simply tighten the screws until you hit the hard stop of the two surfaces touching.  The problem is that the suitcase clamps can't do this, so the housing vendors rely on an incomplete face seal and let the water pressure complete the seal.  To me this is a poor trade-off and explains the tendency for housings to flood in the rinse tank.  Frankly, the pressures experienced in recreational diving are trivial for a good o-ring design, so unless the o-ring is physically damaged, issues like compression set should be almost irrelevant.  If you are interested, there is more than you ever wanted to know about designing seals at  http://www.logwell.c...er_Handbook.pdf

 

I too need to give this thread a rest.




#337404 Aquatica port assembly

Posted by drsteve on 23 September 2013 - 09:44 PM

drsteve, you mention that your housing was still under a PARTIAL vacuum.  Do you have a leak??


No I didn't have a leak. When you pump it down, you only partially evacuate the housing. I started with a partial vacuum (-10inHg) and ended with the exact same partial vacuum (-10inHg). Voila!


#333913 Great White Shark cage diving & system to shoot

Posted by drsteve on 05 July 2013 - 11:15 AM

My experience is different.  The water at Guadlupe is often bright and clear.   I have done this trip on the Nautilus Explorer three times with my Canon 5D classic.  I got the best results with my 17-40 zoom.  The 15mm fisheye is a little too wide unless the sharks come VERY close.   Of course when they DO come close the 15mm is awesome, but you don't want to go on a trip and come back with images tiny sharks in the blue.  I typically shot the 17-40 for the first couple days until I had some keepers in the bag and then switched to the 15mm on the last day to play.  YMMV.  Here are the results

 

Enjoy!