Hi there. We’re in the middle of building and configuring our BlueRobotics2. I purchased the vacuum testing rig you recommended. I’ve tested the test rig as instructed and it holds vacuum just fine. However, when I put the vacuum test plug into the vent port to pull vacuum on the watertight electronics enclosure, it doesn’t seem to fit very tightly at all. It’s a little wobbly compared to the actual vent plug that screws into the vent penetrator.
I used the o-rings included in the bill of materials. But they seem a bit small, perhaps.
The vacuum plugs are a little loose compared to the plugs, but they are tight enough to seal. Could you post a picture of your set-up so I have a better idea of what you are talking about?
Here’s what I’ve got set up, Jonathan. If the vent plug isn’t the problem, then I probably need to unfasten all my penetrators to check the o-rings, as well as the o-rings on the bulkheads…
In the bag that the vacuum plug you are using came in, there should be a tee, another vacuum plug, and 3 lengths of tubing. I attached a picture of how it should look assembled and during testing.
If you just test the electronics enclosure, air will slowly transfer from the battery enclosure to the electronics enclosure through the power cable and make it look like your ROV failed the vacuum test. However, if you use the vacuum plug on both enclosures, you will be able to see the ROV hold a good vacuum.
I got a really stupid question … why the vacuum testing? The reason I ask that is the O-ring seal designs are for “hydrostatic” pressure and a seal design that you would use in a vacuum environment is different than what you would use to keep water out. In addition, you wouldn’t be using a porous material like BUNA-N O-rings either.
Also, how much of a vacuum are you testing for? I see the gauge goes to a full vacuum; however, I seriously doubt that setup can pull a try one. My Leybold pumps at work (monster piggies) do a pretty good job but that rig looks like some piston driven unit. If you want a cheap, real no crap pump … look at the ones that your HVAC techs use for pulling down your systems. They are good to double digit milliTorr ranges.
The vacuum test is more-or-less a “sanity check” that everything is in the right place, there are no hairs or particles caught in seals, and there are no missing pieces. We only go to about 10-15 inHg below ambient pressure - no where near a true vacuum.
This is standard operating procedure with most ROVs.
I reconfigured my vacuum apparatus to conform with the layout you sent in the pic. The vacuum pump does a nice job of holding a vacuum. Unfortunately, the enclosures still don’t hold a vacuum. Wherever the leak is, it’s not small. I can’t pump the handle fast enough to get past 1 or 2 in hg. on the pump gauge.
I pulled the battery enclosure out and checked all the penetrators on the bulkhead for orings. They are all there and clean. I’m almost certain the penetrators on the electronics enclosure are the same, but removing them at this point for inspection will require several hours of work.
I tried connecting one of the vacuum plugs to the battery enclosure and plugged the other with the rubber stopper. Same result - no vacuum. I’m guessing the issue might be in the power cable connecting the electronics enclosure to the battery enclosure? The power wires on both ends appear to be well potted. I guess I could remove the battery cable penetrator from the battery enclosure and replace it with a plug penetrator from the electronics enclosure bulkhead and see if I can pull a vacuum on the battery enclosure with no battery cable. That way at least I’d know the problem is in the electronics enclosure.
I hate to disassemble the electronics bulkhead. All my wires in the enclosure itself are groomed as well as the cables to the thrusters and lights. Ugh. They look so tidy at the moment. Hate to think about snipping all those zip ties.
This problem highlighted a potential improvement in the documentation. It might be helpful to insert a vacuum test at some point into the assembly manual. Perhaps move the section where the thruster and light penetrators are installed into the bulkhead to precede the actual connection of all the wiring to the boards in the enclosure. That way, you could do a vacuum test by simply stuffing all the unconnected wires into the enclosure and doing a vacuum test at that point - before all the connecting and grooming occurs. It would be relatively simple to diagnose a problem at that point. It’s a thought.
Any ideas for me at this point before I start snipping…?
I just tried vacuum testing a battery enclosure connected to an electronics enclosure without the vent plug in, and I was pretty easily able to pump up to 10 in/Hg with little effort. Could you check the following things:
Visually inspect the radial O-rings. Just make sure there are 2 on each side and that they are compressed between 1-2 mm. Also check that there is no debris between the O-ring and the tube.
Take off both end caps from the O-ring flange and inspect the O-rings for damage.
Re-install the end caps.
Attempt to hand loosen the penetrators. Tighten them with a wrench of you can hand loosen them.
Vacuum test just the battery enclosure again.
Generally when I have a ROV fail a vacuum test it is one of the following problems:
I just tried vacuum testing a battery enclosure connected to an electronics enclosure without the vent plug in, and I was pretty easily able to pump up to 10 in/Hg with little effort. Could you check the following things:
Visually inspect the radial O-rings. Just make sure there are 2 on each side and that they are compressed between 1-2 mm. Also check that there is no debris between the O-ring and the tube.
Take off both end caps from the O-ring flange and inspect the O-rings for damage.
Re-install the end caps.
Attempt to hand loosen the penetrators. Tighten them with a wrench of you can hand loosen them.
Vacuum test just the battery enclosure again.
Generally when I have a ROV fail a vacuum test it is one of the following problems:
Interesting that you can leave the vent plug out of the electronics enclosure and still pull a decent vacuum on the battery enclosure. That’s actually what I’d expect, given how well the wires are potted in the battery cable penetrator. I’ll keep focusing on the battery enclosure for now.
Here’s a little trivia :
10" of mercury equates to 136" of water, or roughly 11.3’ of depth in (presumably fresh) water.
15" of mercury equates to 204" of water, or 17’ of depth.
Not bad for sanity checks. Certainly worked in my case…
I’ll get to work on the checks you described above.
I spent a little quality time with my battery enclosure this evening. Here’s what I can report. All orings in both bulkheads are present, accounted for, and well-seated. This includes the penetrator orings, the faceplate orings, and the barrel orings. All the fasteners were seated snugly - nothing loose.
I did take a closer look at the battery cable penetrator on the enclosure bulkhead. I noticed a void on the outside surface of the penetrator. I also noticed cracking of the potting material around the wires on the inside of the penetrator. Not sure if this is the problem, but have attached photos for your review.
If this potting material is failing (and I’m not saying it is), you might consider 3M 5200 Fast Cure Marine Sealant as an alternate potting material. We use in on the boat to permanently attach things to the hull. Typically it would be a Starboard backing plate for a brass thru hull in a fiberglass hull. It’s indestructible and bonds very well to the materials we’ve used it on. It stays slightly flexible as well. I’ve never seen it crack like the potting material in the penetrator photos attached. Not sure how it would work on the battery cable penetrators…
Rusty and I have discussed the issue of the Loc Tite Marine Epoxy cracking, several times on the forum. I recommend the West System g/flex as a substitute. Look at my posts under username Richard. I have also used 5200 on my boats. It is flexible and indestructible and may make a good potting compound. It certainly works well in a non-pressure environment for thru-hull sealing.
May I make an observation based on my ROV chassis that uses two 4 inch WTCs. One WTC for the navigation controller and one for the batteries.
I vacuum tested each WTC without any interconnecting power cables between them. I was able to pump down each individual WTC to 560 mm Hg (~22 inches) and had only a 10 mm vacuum loss over a 24 hour period.
I then connected the navigation controller WTC to the battery WTC though a single power cable. I then proceeded to pump down just the navigation controller and found that there was air passing from the battery WTC to the navigation controller WTC though the battery cable. Therefore I had to do multiple pump downs of the navigation controller WTC until the vacuum in the battery WTC was equivalent to that in the navigation controller. After both WTCs were equalized to a 560 mm Hg vacuum I found that it took around 24 hours to lose 10 mm of vacuum just like with the single WTC vacuum test.
After trying to duplicate your problem, I agree with your assessment that it is probably the power cable. Would you mind trying swapping a blank from the 14-hole end cap to the battery enclosure where the power wire goes in to confirm. If this turns out to be the case, we’ll get a new power cable shipped to you first thing Monday.
Sorry about the frustrating problem. I hope we have this figured out.
Today was a trying day. I’ve been trying to isolate a leak in the electronics watertight enclosure on my BR2. Here’s what I found:
I determined that the leak is indeed in the electronics WTE. I isolated the battery enclosure and was able to pull and maintain a 15 in hg vacuum just fine.
I removed and both the umbilical penetrator and the battery cable penetrator and replaced them with blank penetrators. Still unable to pull at vacuum on the electronics WTE.
I completely disassembled the entire electronics chassis, removing it from the bulkhead so I could test each of 14 bulkhead penetrators one by one, replacing each one in turn with a blank penetrator and attempting to pull a vacuum. Still could not pull a vacuum.
As a last resort, I submerged the whole damn thing in a big tank to see if I could determine where the leak is. As it turns out, the leak is in the dome somewhere. It started leaking water from where the dome subassembly is captured on the bayonet assembly. The bayonet o-rings are fine.
I’ve already disassembled/reassembled the dome side of the electronics WTE. The o-rings and machining seem fine. At this point, I suspect a leak in the dome subassembly itself, which I cannot disassemble any further.
I highly suggest the vacuum test be moved up in the assembly instruction. It is a royal pain in the butt to connect all the wiring, groom and re-groom with zip ties to get everything just right, only to have to remove everything to trouble shoot a leak. If the test was moved up, it would still be frustrating, but way less than it was today.
