View Full Version : exploding plastic storage tanks
Jon Stratis
10-03-2014, 12:23
This past week at the Lake Superior regional we had one of the plastic air tanks from Clippard explode on the field, sending pieces all the way up into the stands (at LSR the stands start about 10 ft above the field, with plenty of space between them and the field), and I heard there was a tank that blew in week 1 as well.
At LSR, we believe the failure was caused by damage to the tank, not a manufacturing defect (like a void in the plastic wall). The team used hose clamps to attach the tank to their robot - the screws from the clamps bit into the sides of the tank, and when reassembled the tank was slightly deformed from being squeezed by the clamps. As far as we can tell, the initial fracture point was right under the clamp, near the screw. Further, the tanks were mounted in a very exposed position, and took some hits from other robots.
I want to strongly caution teams to mount these tanks properly (with the provided plastic clips or through some other means that holds them securely without applying undue pressure on the sides of the tanks) and to protect them inside the frame perimeter. The tanks were designed for a much gentler use than we're putting them through (mounting to a static mechanism that doesn't take impacts or damage, and doesn't need firm mounting to hold the tanks in place), and rough handling will eventually lead to failure!
scaryone
10-03-2014, 12:27
Maybe wrapping the tank with a sleeve to catch pieces if they do fail would work?
jijiglobe
10-03-2014, 12:32
Our team uses velcro
Jared Russell
10-03-2014, 12:34
I have witnessed this in person twice (both times in 2013) and thankfully nobody was hurt either time. But they probably could have been.
In one case, the problem was a student over-tightening a fitting to the plastic threads. In another case, we believe that Loctite was used on the fitting threads.
Last year I proposed a mandatory sleeve in this (http://www.chiefdelphi.com/forums/showthread.php?p=1269092#post1269092) post. I think it would be a cheap, easy, and effective harm reduction measure (though it would certainly need to be well-tested)...and it would let teams color their air tanks however they see fit :cool:
I would definitely second the mention of being careful with your mounting. Any harsh mounting (hose clamps) that deforms the tank or otherwise marks it is changing the mechanics of the system rather dramatically.
The polymer used to make these tanks seems fairly brittle, that combined with the high stresses involved can lead to failure with little to no warning.
Makes me comforted to know that on Neutrino our air tanks are placed in our 3x3 6061 drive tubes so in the event of a failure it should all be contained.
MrForbes
10-03-2014, 12:46
Would a "sleeve" contain the shrapnel, or direct it (at higher velocity) in two directions?
We have several old metal tanks, we'll continue to use them.
JamesCH95
10-03-2014, 12:52
Would a "sleeve" contain the shrapnel, or direct it (at higher velocity) in two directions?
We have several old metal tanks, we'll continue to use them.
A properly designed containment sleeve would contain shrapnel, but a poorly designed one could certainly direct shrapnel out either end at a high velocity.
FRC3883FRC
10-03-2014, 12:56
Our team had our tanks on our intake a very vulnerable position. We were highly highly HIGHLY recommended by the inspectors (almost to the point of not passing) to cover our tanks. We simply added bumper material over both of the tanks. They still fit right in the same c claps they some with and they dont look all that bad. A couple zip ties and the bumper fabric and we were good to go. You can see in the video that our robot has two red tubes at the top of the intake and around 9 seconds in you can see why they highly recommended covering them up.
Here is the video.
http://www.youtube.com/watch?v=nvAFLz-S9Rw&list=UUgokhIYOmu8USGMhWY2qgkw
If you are using these tanks make some covers or make sure they are in a very safe location.
JamesCH95
10-03-2014, 13:06
Our team had our tanks on our intake a very vulnerable position. We were highly highly HIGHLY recommended by the inspectors (almost to the point of not passing) to cover our tanks. We simply added bumper material over both of the tanks. They still fit right in the same c claps they some with and they dont look all that bad. A couple zip ties and the bumper fabric and we were good to go. You can see in the video that our robot has two red tubes at the top of the intake and around 9 seconds in you can see why they highly recommended covering them up.
Here is the video.
http://www.youtube.com/watch?v=nvAFLz-S9Rw&list=UUgokhIYOmu8USGMhWY2qgkw
If you are using these tanks make some covers or make sure they are in a very safe location.
Do you think that you are actually providing any meaningful protection by wrapping the tanks with bumper fabric and zip ties?
Tom Line
10-03-2014, 13:09
A spandex sleeve (or even bumper material...) over each tank would be an elegant and easy way to safeguard folks from flying debris, and the porous nature of the fabric would prevent the air pressure from 'exploding' the sleeve.
Hey ANDYMARK! Are you listening?
mman1506
10-03-2014, 13:12
Our team had our tanks on our intake a very vulnerable position. We were highly highly HIGHLY recommended by the inspectors (almost to the point of not passing) to cover our tanks. We simply added bumper material over both of the tanks. They still fit right in the same c claps they some with and they dont look all that bad. A couple zip ties and the bumper fabric and we were good to go. You can see in the video that our robot has two red tubes at the top of the intake and around 9 seconds in you can see why they highly recommended covering them up.
Here is the video.
http://www.youtube.com/watch?v=nvAFLz-S9Rw&list=UUgokhIYOmu8USGMhWY2qgkw
If you are using these tanks make some covers or make sure they are in a very safe location.
1310 had an air tank explode in their pit and the pieces went right through the FRC Ball puncturing 2 layers of fabric and rubber without slowing it down. I doubt the bumper fabric would be able to contain it.
donkehote
10-03-2014, 13:15
and I heard there was a tank that blew in week 1 as well.
One of 1310's tanks blew at GTR west. One of the mentors was tightening a fitting into the tank with the tank inflated to 120 psi try to eliminate a leak. The tank exploded in the mentors hands, (thankfully no injury's) and destroyed the laptop that was below the tank, aswell as popping one of their balls. The shrapnel was sent flying all the way to the roof of the fieldhouse, and was quite a shock to those at the event.
PLEASE ensure that if you are using the tanks, that you do not over tighten the fittings. I highly recommend pipe dope (thread sealant paste) for the plastic tanks. It allows you to tighten the fittings less while maintaining a good seal.
something like this
http://www.henkelna.com/product-search-1554.htm?nodeid=8797874651137
The tanks are both a blessing and a curse, please treat them with care, as both inexperienced students and "experienced" mentors have fallen victim to these tanks.
In one case, the problem was a student over-tightening a fitting to the plastic threads. In another case, we believe that Loctite was used on the fitting threads.
A good point. A good many liquid pipe thread sealers are not intended for plastic pipe. Only use them on the material listed on the can.
JamesCH95
10-03-2014, 13:18
A spandex sleeve (or even bumper material...) over each tank would be an elegant and easy way to safeguard folks from flying debris, and the porous nature of the fabric would prevent the air pressure from 'exploding' the sleeve.
Hey ANDYMARK! Are you listening?
No, no it wouldn't. See other posts in this thread about shrapnel puncturing ball covers without issue.
I think I posted something similar last year... but biaxial kevlar sleeve or similar would be required to provide even modest protection. Multiple layers would be required IMO, and a good way to keep the ends of the sleeve closed.
Edit: what velocity air flow is needed to exert a significant pressure on a fabric? Said fabric doesn't need to be air-tight to sustain significant loading in the event of a storage tank's catastrophic failure. Not to mention cutting and tearing resistance to the sharp tank fragments.
Tytus Gerrish
10-03-2014, 13:26
this makes me want to test one to failure just to see that it takes to bust em
this makes me want to test one to failure just to see that it takes to bust em
Don't do it. It's seriously scary, especially when you're 4ft away from it. I couldn't hear from the ear facing the tank for about 15 minutes, and it hurt for the rest of the weekend. The shrapnel went flying into a pit about 30 ft. away, and killed our driver station computer.
Don't do it.
So it seems that this has happened at least four(?) times in the last two years. If FIRST was serious about safety these would be outlawed.
thefro526
10-03-2014, 13:29
In one case, the problem was a student over-tightening a fitting to the plastic threads. In another case, we believe that Loctite was used on the fitting threads.
Thankfully, the new (for 2014) Black Clippard Tanks that AM sells have a push to connect fitting that should (in theory) eliminate any failures that the connection point, or at the very least cause the hose to pop off before the tank fails. Otherwise, over tightening of threads, especially NPT threads into plastic is an issue that's difficult to make 100% fool proof.
1310 had an air tank explode in their pit and the pieces went right through the FRC Ball puncturing 2 layers of fabric and rubber without slowing it down. I doubt the bumper fabric would be able to contain it.
Assuming the proposed sleeves all are wrapped reasonably tightly around the tanks, it shouldn't take much to contain a tank failure pretty well. In the instance you mention, it sounds like the tank burst and the pieces of shrapnel were allowed to accelerate to some speed (becoming projectiles) and then they punctured the ball. Assuming you're containing the failure as it happens, the chances of getting projectile type shrapnel will be significantly lower (since they'll be slowed as they contact/break through the sleeve) and if anything does mention to get past the sleeve, it will be traveling significantly slower than if the sleeve were not there.
------
We're actually running two of the black clippard tanks on our robot now, and haven't had any issues. Our mounting is fairly simple, we ran a strip of adhesive backed rubber down one of our metal structural members and then ziptied the tank to the member. The rubber does a pretty good job of holding the tank snug while also allowing it to have a bit of cushioning.
No, no it wouldn't. See other posts in this thread about shrapnel puncturing ball covers without issue.
I think I posted something similar last year... but biaxial kevlar sleeve or similar would be required to provide even modest protection. Multiple layers would be required IMO, and a good way to keep the ends of the sleeve closed.
Edit: what velocity air flow is needed to exert a significant pressure on a fabric? Said fabric doesn't need to be air-tight to sustain significant loading in the event of a storage tank's catastrophic failure.
I realize what happened at GTR West and 1310's exploding tank puncturing their ball and killing a laptop.
I can't help but wonder, though, if even a bumper-fabric sleeve would eat a significant amount of the kinetic energy, when applied at the failure point (thereby partially inhibiting the extreme acceleration of the plastic shrapnel). Agreed that kevlar would be better, though.
One of 1310's tanks blew at GTR west. One of the mentors was tightening a fitting into the tank with the tank inflated to 120 psi try to eliminate a leak. The tank exploded in the mentors hands, (thankfully no injury's) and destroyed the laptop that was below the tank, aswell as popping one of their balls. The shrapnel was sent flying all the way to the roof of the fieldhouse, and was quite a shock to those at the event.
I spoke to the mentor involved within a minute of the explosion. He told me that he had tightened the tank and was testing the tank when it exploded. I may be wrong but if he was tightening when it exploded he would have had shrapnel in his hands or body. There was enough force to blow the top off of a laptop and throw plastic everywhere and he had only a small minor cut on one of his hands. There are many "stories" going around about what happened. Please don't pass around hearsay.
That said, air pressure is nothing to play around with. As we have seen an many posts about exploding tanks this year, safety MUST be taken when using pneumatics. Many people are trying to find a way around the Safety Rules by saying they are not rules. In FIRST we do not want to see people injured. We want FIRST to be a positive experience for everyone. Losing a part of your body or an eye is NOT a positive experience. I know, it won't happen to you, but we don't want to take a chance. Some rules are maybe a bit overboard but I would rather have that than have someone hurt because of lack of rules.
The tank exploded in the mentors hands, (thankfully no injury's)
Maybe not visible ones. I am concerned about his hearing. Voice of experience, 43 years later...
Tytus Gerrish
10-03-2014, 13:36
as a person whom has had PVC explode in my hands I would like to testify that you won't necessarily get pvc shrapnel damage to your body
Not over tightening the fittings is important as well. The threads are tapered, so the tighter they get, the more force is exerted on the tank. Brass vs. Plastic, plastic tends to get stressed and deform more.
As soon as someone gets seriously injured, I conjecture that FIRST will require metal tanks again. Those little tanks from Clippard are a pain in more ways than one, but FIRST HQ may have their hands bound by insurance should it happen.
I think the biggest mitigation moving forward is a pneumatic safety campaign. Perhaps we could come up with some standardized signage to post at events? Things like
Don't over-tighten fittings Do not use Loctite on fittings DO release pressure before adjusting Unofficial Pneumatic Safety Consultants: ________ Teams who are willing to help with any pneumatic situation: _______
Any takers?
Tom Line
10-03-2014, 13:45
Don't do it. It's seriously scary, especially when you're 4ft away from it. I couldn't hear from the ear facing the tank for about 15 minutes, and it hurt for the rest of the weekend. The shrapnel went flying into a pit about 30 ft. away, and killed our driver station computer.
Don't do it.
So it seems that this has happened at least four(?) times in the last two years. If FIRST was serious about safety these would be outlawed.
I STRONGLY disagree that the only avenue of improving the potential safety issues around plastic pneumatics tanks is outlawing them. I would much rather see a numbers-driven problem solving method applied.
However, I can see vendors stepping away from supplying a 'safety' item like a sleeve since it becomes a question of liability if someone gets injured.
mman1506
10-03-2014, 13:53
Whether a sleeve is an effective solution or not wrapping a piece of bumper fabric around a tank in a bad location and deeming it safe without any proper testing is ignorant.
Lil' Lavery
10-03-2014, 13:55
After observing a crack in one of our plastic tanks in our shop last year, 1712 straight up refuses to use them at this point. We'd much rather sacrifice the weight and continue to use Clippard's aluminum tanks. It's simply not worth the risk, imo. I'd hope FIRST's safety rules would follow suit, and either institute a ban on these tanks or a safety procedure to ensure that exploding tanks cannot cause injury.
Mark Sheridan
10-03-2014, 14:00
Looks like I am gonna have to relocate my air tanks.
Just a crazy idea that is not FIRST legal:
Could we make our own storage system with large OD high pressure tubing and barb fittings?
http://www.mcmaster.com/#5632K43
This stuff is ductile and could be snaked around, cut to length and should not explode when pierced. Has anyone used this stuff at 120 PSI, the most I ever used is 90 PSI, same with barbed fittings. I am not a fan of barbed fittings but I think this could be safer.
The rules would have to change in order for this to happen. Thought? I am not sure this is the ideal tubing, but this is something I used before that sort of works.
I STRONGLY disagree that the only avenue of improving the potential safety issues around plastic pneumatics tanks is outlawing them. I would much rather see a numbers-driven problem solving method applied.
However, I can see vendors stepping away from supplying a 'safety' item like a sleeve since it becomes a question of liability if someone gets injured.
In an ideal world every safety precaution should be taken using all the advice given by the manufacturer and of that said here on Chief Delphi, but that simply does not happen. This is MUCH more dangerous than a HP sticking their hand into the field for half of a quarter of an eighth of a second, yet there are no strict rules on these tanks.
As said earlier in this thread, tanks hose clamped down passed inspection! There was a team at GTRE this weekend with the black Clippard tanks (which are better in terms of not needing fittings) on the bottom of their intake arms which extended beyond frame perimeter and were the first point of contact in high speed crashes with other robots! They passed inspection too.
The black tanks have their place in FRC if mounted correctly in the appropriate location, but the white tanks need to go. They've a hazard to everyone within 30 ft. of them. When teams (including my own) use them incorrectly, it's just a recipe for injury, and it's only a matter of time before someone seriously gets injured.
I've been hit by moving robots, smacked by robot arms, and had my fingers pinched in more robot mechanisms than I care to count, but none of these are even close to the potential danger of these tanks.
For anyone who doesn't know, these are the terrifying pieces flying through the air that we're dealing with.
DanielPlotas
10-03-2014, 14:13
1310 had an air tank explode in their pit and the pieces went right through the FRC Ball puncturing 2 layers of fabric and rubber without slowing it down. I doubt the bumper fabric would be able to contain it.
The ball was punctured by pieces already moving at considerable speeds, but the fabric would be able stop them before they get moving.
For anyone who doesn't know, these are the terrifying pieces flying through the air that we're dealing with.
Moving with enough kinetic energy to do lots of damage. Luckily so far, the serious damage has all been to inanimate objects. I agree HQ needs to do something, though I'm not sure an outright ban is necessary.
Mark Sheridan
10-03-2014, 14:16
For anyone who doesn't know, these are the terrifying pieces flying through the air that we're dealing with.
"You'll shoot your eye out"
Those are scary. Are the tanks breaking in a localized area or is the whole thing exploding?
Duncan Macdonald
10-03-2014, 14:17
One of 1310's tanks blew at GTR west. One of the mentors was tightening a fitting into the tank with the tank inflated to 120 psi try to eliminate a leak. The tank exploded in the mentors hands, (thankfully no injury's) and destroyed the laptop that was below the tank, aswell as popping one of their balls. The shrapnel was sent flying all the way to the roof of the fieldhouse, and was quite a shock to those at the event.
Could a 1310 representative give us an official story from someone who was there?
This story is spreading like a game of broken telephone. Link (http://www.reddit.com/r/FRC/comments/1ztzd3/some_technical_difficulties_in_duluth_at_the_lake/cfx3y0i)
Have all of the exploding tanks been the white clippard tanks from 2013 and before? Are these (http://www.clippard.com/part/AVT-PP-35) the same tank in a new colour?
I also feel safer around tanks with the push-in fitting standard. One less source of failure.
Looks like I am gonna have to relocate my air tanks.
Just a crazy idea that is not FIRST legal:
Could we make our own storage system with large OD high pressure tubing and barb fittings?
http://www.mcmaster.com/#5632K43
This stuff is ductile and could be snaked around, cut to length and should not explode when pierced. Has anyone used this stuff at 120 PSI, the most I ever used is 90 PSI, same with barbed fittings. I am not a fan of barbed fittings but I think this could be safer.
The rules would have to change in order for this to happen. Thought? I am not sure this is the ideal tubing, but this is something I used before that sort of works.
Cut to length sounds appealing, but "Snaked around" with a 36" bend radius (6' diameter !?) shouldn't be promoted.
"You'll shoot your eye out"
And that is why you wear safety glasses in the pit.
Could a 1310 representative give us an official story from someone who was there?
This story is spreading like a game of broken telephone. Link (http://www.reddit.com/r/FRC/comments/1ztzd3/some_technical_difficulties_in_duluth_at_the_lake/cfx3y0i)
Have all of the exploding tanks been the white clippard tanks from 2013 and before? Are these (http://www.clippard.com/part/AVT-PP-35) the same tank in a new colour?
I also feel safer around tanks with the push-in fitting standard. One less source of failure.
Gregor is a member of 1310 and has made numerous comments about his experience 4ft from this incident in this thread.
Mark Sheridan
10-03-2014, 14:21
Cut to length sounds appealing, but "Snaked around" with a 36" bend radius (6' diameter !?) shouldn't be promoted.
Hmmm...I need to go straighten something out, brb
team222badbrad
10-03-2014, 14:26
Could a 1310 representative give us an official story from someone who was there?
This story is spreading like a game of broken telephone. Link (http://www.reddit.com/r/FRC/comments/1ztzd3/some_technical_difficulties_in_duluth_at_the_lake/cfx3y0i)
Have all of the exploding tanks been the white clippard tanks from 2013 and before? Are these (http://www.clippard.com/part/AVT-PP-35) the same tank in a new colour?
I also feel safer around tanks with the push-in fitting standard. One less source of failure.
I believe the the only tanks that had an issue were from 2013. Clippard offered to replace them for free last year with a different model after the issue was found.
http://www.chiefdelphi.com/forums/showthread.php?t=112417&highlight=clippard+replace
If anything FIRST should ban the usage of the white threaded tanks since they obviously still have the casing cracking at the threads from over tightening and/or poor manufacturing and can easily be replaced with the new push to connect model.
I have yet to hear of a failure with the new black "push to connect" storage tanks which are much better, but you should still locate them in a relatively safe spot on your robot to prevent damage to them. ::ouch::
Jon Stratis
10-03-2014, 14:27
Have all of the exploding tanks been the white clippard tanks from 2013 and before? Are these (http://www.clippard.com/part/AVT-PP-35) the same tank in a new colour?
From the research we did while up in Duluth, the white 2013 tanks have been discontinued. While they were rated at 150 psig working pressure, their replacements (which at least look identical on the surface, although I can't speak towards any manufacturing changes or non-visual/structural changes that may have been made) are only rated for 125 psig.
JamesCH95
10-03-2014, 14:29
I STRONGLY disagree that the only avenue of improving the potential safety issues around plastic pneumatics tanks is outlawing them. I would much rather see a numbers-driven problem solving method applied.
However, I can see vendors stepping away from supplying a 'safety' item like a sleeve since it becomes a question of liability if someone gets injured.
You're doing it wrong. The question should not be "how do we make plastic storage tanks safe?" The question should be "how do we make safe, light, air storage tanks?" Because teams only use plastic storage tanks because they are lighter than commonly available metal tanks.
Here are some numbers:
Polypropylene: ~37MPa TS, 0.94g/cc density, 39MPa/(g/cc) strength:density ratio
304 Stainless Steel: 515MPA TS, 8.0g/cc density, 64MPa/(g/cc) strength:density ratio
The reason why teams use plastic tanks is because they are lighter than the stainless tanks from Clippard. Unfortunately, this low weight comes at the price of strength and durability. Stainless steel has a significantly higher strength:weight ratio than polypropylene.
Plastics tend to have brittle failure modes in dynamic situations, which is why they generate shrapnel, and why they are generally a terrible choice for pressurized gas applications. Not to mention their sensitivity to temperatures, certain chemicals, nicks/scratches/gouges, over-tightening, etc.
Have any of these discussed failures been with the new black tanks?
Duncan Macdonald
10-03-2014, 14:41
Gregor is a member of 1310 and has made numerous comments about his experience 4ft from this incident in this thread.
My fault, poor phrasing....
Could someone from 1310 describe:
-What was the source or compressed air (Through the robot or a separate compressor)
-If the tank had been on a previous robot
-How they mount their tanks. (supplied connector, zip ties, loose inside the robot, etc.)
-Is there any chance the tank had been dropped or mishandled.
Jared Russell
10-03-2014, 14:43
Have any of these discussed failures been with the new black tanks?
Both of the failures that I referenced in my post were with the white tanks without push-in fittings.
team222badbrad
10-03-2014, 14:52
From the research we did while up in Duluth, the white 2013 tanks have been discontinued. While they were rated at 150 psig working pressure, their replacements (which at least look identical on the surface, although I can't speak towards any manufacturing changes or non-visual/structural changes that may have been made) are only rated for 125 psig.
There might be other changes, but the two main visual changes to the tanks are:
Color change. (This may mean a different grade of material or nothing at all.)
Plastic welded push to connect fittings. (This eliminated the failure point due to threading push to connect fittings into the threaded tanks.)
For those who haven't see the ends of the new tanks available from AndyMark:
http://farm4.staticflickr.com/3202/13066142304_c8f5011a3e_z.jpg
Tristan Lall
10-03-2014, 14:58
Plastics tend to have brittle failure modes in dynamic situations, which is why they generate shrapnel, and why they are generally a terrible choice for pressurized gas applications. Not to mention their sensitivity to temperatures, certain chemicals, nicks/scratches/gouges, over-tightening, etc.
Exactly right about this: the ductile vs. brittle failure modes are a critical distinction between metal and plastic tanks. It would take an extremely (and unrealistically) high strain rate to cause an unhardened metal tank to fail in a brittle manner.
The reason why teams use plastic tanks is because they are lighter than the stainless tanks from Clippard. Unfortunately, this low weight comes at the price of strength and durability. Stainless steel has a significantly higher strength:weight ratio than polypropylene.
Incidentally, the carbon fibre skin on the Boeing 787 is (in some places) thicker than structurally necessary because it needs to be resistant to penetration by objects like baggage carts. Physical damage tolerance is often an application-specific requirement, and may not be adequately regulated in the FRC rules.
One can conceive of a similar situation with conventional ductile metal tanks: you could make them very thin and still strong enough to withstand the pressure load, but any damage could cause them to fail (albeit in a ductile way). (Contrast this with a hypothetical hardened (i.e. not ductile) metal tank: paper-thin and very strong, but if damaged, it shatters. That would be a bad design in this application.)
protoserge
10-03-2014, 15:00
I haven't seen it yet, so I'll reiterate this point: NEVER WORK ON A PRESSURIZED SYSTEM.
Tightening brass fittings in a pressurized plastic tank has a higher probability of causing cracks, which ultimately leads to a rupture.
I don't know the details of the GTR tank rupture, but all of the evidence in past explosions points to user error, not a design flaw.
JamesCH95
10-03-2014, 15:04
I haven't seen it yet, so I'll reiterate this point: NEVER WORK ON A PRESSURIZED SYSTEM.
Tightening brass fittings in a pressurized plastic tank has a higher probability of causing cracks, which ultimately leads to a rupture.
I don't know the details of the GTR tank rupture, but all of the evidence in past explosions points to user error, not a design flaw.
If any of the mentioned failure mechanisms had been applied to a metallic tank would there have been a failure? Or a brittle/catastrophic failure?
Lil' Lavery
10-03-2014, 15:10
I don't know the details of the GTR tank rupture, but all of the evidence in past explosions points to user error, not a design flaw.
In a competition designed for high school students where safety is assessed by volunteers whos professional background may have no connection with pressurized gas systems, the barrier between "user error" and hazardous explosions should be pretty darn high. This isn't about getting a refund for busted tanks, it's about preventing injury. If the current scenario has resulted in repeated failures, it's obviously not adequate.
In a competition designed for high school students where safety is assessed by volunteers whos professional background may have no connection with pressurized gas systems, the barrier between "user error" and hazardous explosions should be pretty darn high. This isn't about getting a refund for busted tanks, it's about preventing injury. If the current scenario has resulted in repeated failures, it's obviously not adequate.
Likely a PSA of some sort would be needed, or harsher restrictions in order to prevent future problems. Unfortunately, these tanks are probably perfectly adequate in the other environments where they see a static loading. I recollect Clippard testing tanks last year at several times their advertised maximum pressure without failure in a presumably static low load environment. The environment of FRC is typically anything but static and warrants more care.
I would argue that user error in play would be utilizing the tanks improperly in an environment that they aren't made for. The problem is, teams see other teams successfully doing it and don't take the time to fully educate themselves on the different protocols necessary to make them safe.
If you want to have tanks in an area that risks being hit or needs to be strapped down rigidly, go with metal tanks. Using plastic tanks necessitates finding a location where they will be under minimal load, mounting which will result in no damage, and ideally if one does explode shrapnel will be contained or redirected in a safer direction (i.e. at the ground).
Despite maintenance being a chore that is one of the reasons I now prefer putting plastic tanks in our 3x3 1/8" wall drive tubes. In the event of a fracture shrapnel would be stopped by the walls of the tubes and the drive modules at either end. Additionally I make a huge emphasis of informing any of our students who may work on our plastic storage tanks that if it is leaking you are better off removing the fitting and redoing the teflon tape than continuing to tighten it if it is leaking.
Tom Line
10-03-2014, 15:31
Whether a sleeve is an effective solution or not wrapping a piece of bumper fabric around a tank in a bad location and deeming it safe without any proper testing is ignorant.
Who suggested that someone should do that?
Take some heavy socks and make tank cozies! At the very least the fabric will help contain the fragments while allowing the air to escape.
JamesCH95
10-03-2014, 15:37
Who suggested that someone should do that?
You, and someone on 3883 said they covered up their tanks with bumper fabric.
A spandex sleeve (or even bumper material...) over each tank would be an elegant and easy way to safeguard folks from flying debris, and the porous nature of the fabric would prevent the air pressure from 'exploding' the sleeve.
Hey ANDYMARK! Are you listening?
Our team had our tanks on our intake a very vulnerable position. We were highly highly HIGHLY recommended by the inspectors (almost to the point of not passing) to cover our tanks. We simply added bumper material over both of the tanks. They still fit right in the same c claps they some with and they dont look all that bad. A couple zip ties and the bumper fabric and we were good to go. You can see in the video that our robot has two red tubes at the top of the intake and around 9 seconds in you can see why they highly recommended covering them up.
Here is the video.
http://www.youtube.com/watch?v=nvAFLz-S9Rw&list=UUgokhIYOmu8USGMhWY2qgkw
If you are using these tanks make some covers or make sure they are in a very safe location.
Tom Line
10-03-2014, 15:38
Does anyone happen to know if the Andy Mark white tank that is sold is Clippard, or Pneuaire?
Can someone confirm that the tanks that exploded are only the Clippard brand 'white' threaded version?
Jim Wilks
10-03-2014, 15:44
Is Clippard the only maker of plastic tanks? Are the ones from PneuAire also from Clippard, or does PneuAire make their own?. Anyone else?
Are all of the failures so far on Clippard tanks?
thefro526
10-03-2014, 15:55
Not to add another reason to not use (or be wary of) plastic air tanks but, has anyone paid attention to the temperatures they're being exposed to, either while in operation or storage?
The published temperature range for the New (2014) Black AVT-PP-35 Tank is 35* F - 100* F (1.5*C - 37.5*C), and I believe (but don't quote me) that the tanks from previous years have the same or at least similar temperature range.
I can think of a handful of different ways that tanks could be exposed to conditions outside of that window, which makes me wonder if environmental conditions play some role in the tank ruptures...
Wren Hensgen
10-03-2014, 15:57
Team 225 has managed to make one of the new black plastic tanks fail, but the failure actually occurred on the push-to-connect fitting, which snapped out under pressure. This might be an intentional design, as losing a hose is probably less dangerous than throwing shrapnel around
JamesCH95
10-03-2014, 16:01
Not to add another reason to not use (or be wary of) plastic air tanks but, has anyone paid attention to the temperatures they're being exposed to, either while in operation or storage?
The published temperature range for the New (2014) Black AVT-PP-35 Tank is 35* F - 100* F (1.5*C - 37.5*C), and I believe (but don't quote me) that the tanks from previous years have the same or at least similar temperature range.
I can think of a handful of different ways that tanks could be exposed to conditions outside of that window, which makes me wonder if environmental conditions play some role in the tank ruptures...
The compressor definitely outputs air warmer than 100F, evidenced by the compressor routinely getting "too hot to touch." I don't know if plastic tanks have ever gotten to that temperature, but the possibility is definitely there.
Personally I'm for a ban of the plastic tanks. It is not that they are inherently unsafe, but that there are so many ways to unintentionally compromise their safety that are likely to occur when used in FRC.
Not to add another reason to not use (or be wary of) plastic air tanks but, has anyone paid attention to the temperatures they're being exposed to, either while in operation or storage?
The published temperature range for the New (2014) Black AVT-PP-35 Tank is 35* F - 100* F (1.5*C - 37.5*C), and I believe (but don't quote me) that the tanks from previous years have the same or at least similar temperature range.
I can think of a handful of different ways that tanks could be exposed to conditions outside of that window, which makes me wonder if environmental conditions play some role in the tank ruptures...
1310s tank would have definitely seen temperatures (at least briefly) outside of that range, if only when transiting between their school and haul vehicle, and then into the venue. If the haul vehicle sat outside not running anywhere for a while, the tank could have conceivably seen temps in the neighborhood of -20C.
Team 225 has managed to make one of the new black plastic tanks fail, but the failure actually occurred on the push-to-connect fitting, which snapped out under pressure. This might be an intentional design, as losing a hose is probably less dangerous than throwing shrapnel around
Well, its a failure, but at least its not a catastrophic failure of the part, resulting in explosive decompression and flying shrapnel.
Is Clippard the only maker of plastic tanks? Are the ones from PneuAire also from Clippard, or does PneuAire make their own?. Anyone else?
Are all of the failures so far on Clippard tanks?
The PneuAire tanks are distinctly different in design from the Clippard tanks. Unfortunately I don't have a pair handy to take photos of, but I have handled both at the same time this past build season.
thefro526
10-03-2014, 16:16
Team 225 has managed to make one of the new black plastic tanks fail, but the failure actually occurred on the push-to-connect fitting, which snapped out under pressure. This might be an intentional design, as losing a hose is probably less dangerous than throwing shrapnel around
Oddly enough, this may be due in part to a manufacturing defect, as we've had one of the fittings come out as well, and AM has a note about it on their site. http://www.andymark.com/product-p/am-2649.htm
The compressor definitely outputs air warmer than 100F, evidenced by the compressor routinely getting "too hot to touch." I don't know if plastic tanks have ever gotten to that temperature, but the possibility is definitely there.
Next time I have the chance, I'll take some temperature readings of the tanks themselves after a few compressor cycles and see if there's any sort of noticeable increase in temperature. I wouldn't be surprised if the right setup (one where the tanks are essentially next to the compressor) was able to pump a bit of heat into the tanks...
1310s tank would have definitely seen temperatures (at least briefly) outside of that range, if only when transiting between their school and haul vehicle, and then into the venue. If the haul vehicle sat outside not running anywhere for a while, the tank could have conceivably seen temps in the neighborhood of -20C.
.
That's what I was thinking as well, but didn't know enough details about the situation to really say. I'm also not sure of how well the plastic tanks would handle heat cycling, specifically being exposed to temperatures outside of the operating range, then being brought back into the operating range before use.
JamesCH95
10-03-2014, 16:22
Also consider that PP's CTE is an order of magnitude more than steel or aluminum, if a plastic tank and mounting bracket were warmed up together the PP would strain against any metal mounting components, perhaps significantly.
AllenGregoryIV
10-03-2014, 17:38
Is Clippard the only maker of plastic tanks? Are the ones from PneuAire also from Clippard, or does PneuAire make their own?. Anyone else?
Are all of the failures so far on Clippard tanks?
Before we call for a ban on all the plastic tanks, I think we really need to answer these questions. As far as I know PneuAire made the plastic tanks before Clippard started doing it. It was my understanding that Clippard released the 2013 plastic air tanks in response to teams using the PneuAire ones in the 2012 season.
We have used the PneuAire tanks (bought from PneuAire) without any failures for the past 3 years (only 2 in competition).
Safety is very important but we also shouldn't overreact.
Daniel_LaFleur
10-03-2014, 18:17
Before we call for a ban on all the plastic tanks, I think we really need to answer these questions. As far as I know PneuAire made the plastic tanks before Clippard started doing it. It was my understanding that Clippard released the 2013 plastic air tanks in response to teams using the PneuAire ones in the 2012 season.
We have used the PneuAire tanks (bought from PneuAire) without any failures for the past 3 years (only 2 in competition).
Safety is very important but we also shouldn't overreact.
Here I feel the opposite.
Considering the failure mode (sharp plastic shards flying a high velocity) I would err on the side of caution. Until we can prove that the plastic volumes are safe in a FIRST environment (High impact, low technical knowledge, large swings in temperature) I would want to ban the plastic volumes for foreseeable future starting next year.
P.S. Has a metal volume ever failed in a dangerous way in FIRST (they have a much longer history)
AllenGregoryIV
10-03-2014, 18:32
Here I feel the opposite.
Considering the failure mode (sharp plastic shards flying a high velocity) I would err on the side of caution. Until we can prove that the plastic volumes are safe in a FIRST environment (High impact, low technical knowledge, large swings in temperature) I would want to ban the plastic volumes for foreseeable future starting next year.
P.S. Has a metal volume ever failed in a dangerous way in FIRST (they have a much longer history)
I'm not a plastics or pneumatics expert but I do know we don't ban all car seats when a single model is found to have manufacturing defects. Again I may be wrong and maybe the pneuaire tanks have a similar failure mode but lumping them together because they look the same seems like we are jumping to conclusions. I think we need more information.
Daniel_LaFleur
10-03-2014, 18:45
I'm not a plastics or pneumatics expert but I do know we don't ban all car seats when a single model is found to have manufacturing defects. Again I may be wrong and maybe the pneuaire tanks have a similar failure mode but lumping them together because they look the same seems like we are jumping to conclusions. I think we need more information.
Allen, I understand where you are coming from but as someone who has had a 1 liter PVC pneumatic volume explode within 1 foot of me (thank goodness it blew out away from me) I will respectfully disagree here.
The dangers to anyone near one of these if it explodes, especially to bystanders, some of which may be children, is too high for my liking. Thus I say 'ensure it is safe and then we can use it'.
Jon Stratis
10-03-2014, 19:34
I'm not a plastics or pneumatics expert but I do know we don't ban all car seats when a single model is found to have manufacturing defects. Again I may be wrong and maybe the pneuaire tanks have a similar failure mode but lumping them together because they look the same seems like we are jumping to conclusions. I think we need more information.
I would agree... however the case I saw up at LSR (and prompted me starting this thread) was not due to a manufacturing defect - it was due to improper use and handling by the team.
From what I've seen of plastics in general, I'm not aware of any that could be use in this application that wouldn't have a similar failure mode. Then again, I'm a software engineer and not a chemist or material scientist. Still, it worries me.
AllenGregoryIV
10-03-2014, 19:47
I would agree... however the case I saw up at LSR (and prompted me starting this thread) was not due to a manufacturing defect - it was due to improper use and handling by the team.
From what I've seen of plastics in general, I'm not aware of any that could be use in this application that wouldn't have a similar failure mode. Then again, I'm a software engineer and not a chemist or material scientist. Still, it worries me.
I understand that point, I also understand that we need to better train teams to the dangers of pneumatics and for them to properly care for their equipment. The team in this case was operating illegally since the screw had to damage the air tank and thus modify it. There is a reason for this blue box in the manual.
Do not, for example, paint, file, machine, or abrasively remove any part of a pneumatic COMPONENT – this would cause the part to become a prohibited item. Consider pneumatic COMPONENTS sacred.
My point is that the PneuAir tanks have been in use on FRC robots longer than the Clippard tanks and I have never (doesn't mean it hasn't happened) heard of one of them failing do to misuse or defect.
I would agree... however the case I saw up at LSR (and prompted me starting this thread) was not due to a manufacturing defect - it was due to improper use and handling by the team.
I agree. User error is significantly more dangerous than any potential manufacturing defects. I don't think there is any substantial hazard from plastic tanks that are used appropriately. This includes not over tightening the fittings, not deforming the body and minimizing stress on the entire tank.
Chris is me
10-03-2014, 21:00
Allen, I understand where you are coming from but as someone who has had a 1 liter PVC pneumatic volume explode within 1 foot of me (thank goodness it blew out away from me) I will respectfully disagree here.
The dangers to anyone near one of these if it explodes, especially to bystanders, some of which may be children, is too high for my liking. Thus I say 'ensure it is safe and then we can use it'.
A pressurized, homemade PVC tank is incredibly different from an industry designed and rated part made of polypropylene. I get that these are scary, and I even wouldn't oppose phasing out / banning the white tanks with the screw fittings, but a blanket ban on anything that reminds you of something dangerous you were once near is a bit extreme.
As an aside, I'm interested to see what wrapping these tanks in gaffers tape would do for their failure mode. Obviously the pieces would rip through it, but the added resistance right as they start to gain velocity may be of more help than you'd think.
AllenGregoryIV
10-03-2014, 21:06
I even wouldn't oppose phasing out / banning the white tanks with the screw fittings, but a blanket ban on anything that reminds you of something dangerous you were once near is a bit extreme.
Are you talking about all white plastic tanks with screw fittings or just the Clippard ones from 2013 that had a known failure and an optional recall (http://www.usfirst.org/roboticsprograms/frc/blog-plastic-air-tanks-important-safety-notice)? I really think we need to know more before we start making recommendations. This seems to be a pretty confusing issue, we don't know a lot about the differences between the different tanks (clippard, PneuAire, Andymark, etc.).
Pat Hart
10-03-2014, 21:09
Another means to address the safety concern of a light weight plastic tank would be a design that included either an integrated pressure release valve or a fail safe. Something that would release pressure in a known manner before catastrophic failure. This would allow lightweight tanks to continue to be used and eliminate the risk of a "user error" causing significant harm.
Dave Campbell
10-03-2014, 21:09
For those teams still using the white Clippard volume storage tanks from last year, you should read the FRC Blog linked here (http://www.usfirst.org/roboticsprograms/frc/blog-plastic-air-tanks-important-safety-notice) dated January 30, 2013.
There are very specific instructions included on the safe installation of the fittings on these tanks. I'm not sure if Clippard will still replace them, however I personally know the Clippard family and company and I know they want to protect the safety and health of every FIRST participant, spectator and volunteer.
We currently use two of the new black vst from Clippard and are very pleased with them.
edit - Allen beat me to the post above...
Bald & Bearded
10-03-2014, 21:26
First this is a great discussion with a lot of points on all sides of the issue.
I do have to agree with JesseK that if a simple and reliable solution is not identified FIRST will have no option but to go back to the metal tanks.
I would be great if we could get consumer reports, mythbusters or someone of that quality to test various sleeves/coverings.
I am not a physics person but from an intuitive point of view a sleeve/shroud of some kind should reduce the danger.
User error is the real issue.
Another means to address the safety concern of a light weight plastic tank would be a design that included either an integrated pressure release valve or a fail safe. Something that would release pressure in a known manner before catastrophic failure. This would allow lightweight tanks to continue to be used and eliminate the risk of a "user error" causing significant harm.
A relief valve won't solve the issue causing failures. Failures aren't caused by over pressurization in most cases (assuming in competition and with passed inspections). The problem is a degradation in pressure capacity due to mishandling the tanks (over tightening, improper mounting, etc.). What do you do when a tank fails at 100 PSI? 80? 60? If you put a crack in the right spot or damage a wall on a tank, any of these pressures could cause a failure in the right circumstance. The white Clippard tanks are great tanks, they just have very specific instructions for their use that are tough to follow perfectly. But they have a really phenomenal warning alarm when used improperly in the form of an explosive blast that will wake the neighbors and potentially kill you.
Lil' Lavery
10-03-2014, 21:45
This isn't about whether or not these tanks are quality products, it's about risk. Even if the failures are 100% caused by "user error," the tanks' failure mode is an explosion that sends shard traveling at high speeds. There are two attributes to risk, probability and severity. While we obviously don't have access to FIRST's safety risk matrix, I know that regardless of how low we drive the probability of the risk, the severity is still to high for these tanks to be acceptable for use on 1712 (and I would hope other teams). I feel much more comfortable using a tank with a failure mode that doesn't involve high speed, sharp projectiles.
Having the failure point be at the push in fittings instead of at the sealed thread should overcome the explosiveness of the plastic tanks. I am not sure but the 125psi rating is likely given to the push in connector, and not the body of the tank itself.
Just like fuses are meant to to open an electrical circuit when current flow exceeds a predetermined range, this fail safe connection should the weakest link in the pneumatic system. This should make for a much safer alternative to the white pvc tanks.
It will be interesting to see how the black polypropylene tank's safety record will fare in comparison to the diy fitting pvc tanks. Any air tank with a crack will explode sooner or later, In my opinion any sort of plastic air tank with sharp interior corners (threads) should be banned from use in FIRST. All of the boom'ed air tanks are to my knowledge were a result of user error. Whether it be over tightened fittings or hose claps, these pressure vessels demand respect.
This might sound harsh but when you actually witness one go, I bet you would have a change of heart. Keep in mind this is an event geared towards children.
FIRST should step it up and add the mounting and placement of all air tanks to the robot inspector check list. I have inspected at two events so far, and am amazed at how many inspectors just ask the questions in a manner that all a team has to say is "yes" or "no". I understand that there are a lot of robots to inspect, but the job of the robot inspector is to ensure that the robot is legal and above all safe to compete.
This is a problem that Teachers, Mentors, and Officials have to look at more closely. You can't expect a bunch of high school kids to be the responsible ones here.
This thread is quite interesting and explanatory in several ways. I am reminded of the recent incident involving the painting of one of the tanks. As I recall, the painted tanks were declared illegal as modifications presumably due to chemical alteration of the tank by the paint.
The other think of interest is the operating temperature range for the tanks. I know certain plastics become very brittle by exposure to cold temperatures. Mailboxes, plastic fencing and furniture suffer adversely from sub-freezing temperatures. I do not know if the manufacturer has advised about how durable their products are in cold weather, but it seems irresponsible to sell an obvious outdoor object for use in a northern climate.
Back to the tanks. If the temperature range pertains to operating conditions and not "storage" then we may be OK with outdoor transport of the tanks. There's a robot with tanks in my unheated garage right now. Maybe I should be concerned if merely getting cold will alter the properties of those tanks.
On the other end of the heat range, there was some concern about heated air being stored. I hope someone can give some observations about this. I suppose the black color of the newer tanks will allow the re-radiation of heated air's temperature more readily than the white ones. That will help keep the system within operating temperature range. I wonder if there are some 1/4 inch i.d. finned heat sinks I can thread onto the exit tubing from the compressor. Will it be worth dealing with a metal tube as the direct output from the compressor? I know I can get finned tubing somewhere.
In any event, our robots will not go back into competition with the white tanks. Another reason to thank CD posters for the heads up.
this makes me want to test one to failure just to see that it takes to bust em
You wish :P
That seems very fun [and dangerous]. Do not try this at home!
I think the best sleeve would be a kit of the tank and sleeve so the strongest parts of the tank are the places uncovered. The rest should be contained. We are using mostly metal tanks, but we might (I can't properly recollect) have 2 plastic tanks!
What are these tanks made of? "Polymer" is quite wide of a material definition. It feels like PVC, but I doubt PVC is that strong!
I think the best sleeve would be a kit of the tank and sleeve so the strongest parts of the tank are the places uncovered. The rest should be contained. We are using mostly metal tanks, but we might (I can't properly recollect) have 2 plastic tanks!
What are these tanks made of? "Polymer" is quite wide of a material definition. It feels like PVC, but I doubt PVC is that strong!
The old white tanks are made of pvc, a quick google search reveals the new tanks are made of polypropylene.
Rupturing tanks at 120 psi is not fun.. not in the slightest. While there have been (luckily) no serious injuries from exploding tanks, it can leave you quite shaken. The sound is nothing to sneeze at either, at GTRW many students had momentary hearing loss and ear pain that lasted into the next day just from being in close proximity of it going off, and this was in a large open area.
Polypropylene, not PVC.
From the FIRST blog last year:
Over the weekend, it was brought to our attention that a failure of the white polypropylene air tank, donated by Clippard (Clippard PN: AVT-PP-41, AndyMark PN: am-2007, FIRST Choice PN: fc13-100), had occurred.
The old white tanks are made of pvc, a quick google search reveals the new tanks are made of polypropylene.
The old tanks were also made of polypropylene: http://www.chiefdelphi.com/forums/showthread.php?t=112396
As are Pneuaire air tanks.
Jon Stratis
10-03-2014, 22:58
The old white tanks are made of pvc, a quick google search reveals the new tanks are made of polypropylene.
Rupturing tanks at 120 psi is not fun.. not in the slightest. While there have been (luckily) no serious injuries from exploding tanks, it can leave you quite shaken. The sound is nothing to sneeze at either, at GTRW many students had momentary hearing loss and ear pain that lasted into the next day just from being in close proximity of it going off, and this was in a large open area.
Correction. The old white tanks, Clippard PN AVT-PP-41, are polypropylene as well, not PVC. I am not aware of a single manufacturer that rates PVC for use in gas storage - PVC is generally only rated for liquid pressure.
wow I was corrected 4 times over super quick! thanks!
Jon Stratis
10-03-2014, 23:00
wow I was corrected 4 times over super quick! thanks!
Everyone makes mistakes :)
Moon2020
10-03-2014, 23:09
Additionally, plastic becomes brittle with age.
It is why infant seats for cars have an expiration date.
We use Consequence vs. Likelihood in our 5x5 risk matrices.
All components that see 125 psi should be rated for 125 psi or above. There is a factor of safety already built into their rating. Make sure the PRV is calibrated properly. Is the gage calibrated properly?
The hazard is from the shrapnel created. The cause can be tank rupture due to: brittle plastic, over pressurization (not properly calibrated PRV), manufacturing flaw, design flaw, workmanship issues (assembly), damage from improper handling, mistreatment, environmental factors (temp, humidity, etc.), etc.
So that makes me think, if these tanks can easily become such a hazard, why are they legal? We have our tanks outside, right under our catching mechanism so if something fails, the tanks are in risk of being damaged. However, those are the metal tanks
Just doing some rudimentary comparisons with CES EduPack (materials database).
PP seems to generally have higher impact toughness (absorbs more energy when fracturing) than PVC but is slightly lower yield strength. Tensile strength differences are pretty marginal.
Probably the biggest draw for PP over PVC is the comparative glass transition temperatures, PP is below freezing while PVC is nearly the boiling point of water. This is a reversible transition that is accompanied by a change in elastic modulus (stiffness), addition when above this temperature the polymer should be more ductile which correlates with PP taking more energy to fracture.
So that makes me think, if these tanks can easily become such a hazard, why are they legal? We have our tanks outside, right under our catching mechanism so if something fails, the tanks are in risk of being damaged. However, those are the metal tanks
The tanks themselves are perfectly safe. It's using them properly that's the issue. Putting stress on the tank increases the likelihood of it bursting.
Daniel_LaFleur
11-03-2014, 08:34
The tanks themselves are perfectly safe. It's using them properly that's the issue. Putting stress on the tank increases the likelihood of it bursting.
... and this is the core of my issue with the tanks.
You have students that have very little understanding of the forces and failure modes and 'how to use the tanks properly' in an environment where the public is being exposed to that risk.
All it takes is one serious injury. Safety (for our students, mentors and the general public) needs to be the top priority, always.
If it's possible that through ignorance of safe protocol the tanks can injure the public, then we must look at ways to mitigate that danger, including possibly banning their use.
We are having an LRI conference call this evening. I will make sure that it is brought up. I know that Al is aware of these threads so he may already be working behind the scenes. I will post back tomorrow.
Dave Campbell
11-03-2014, 08:58
... and this is the core of my issue with the tanks.
You have students that have very little understanding of the forces and failure modes and 'how to use the tanks properly' in an environment where the public is being exposed to that risk.
Isn't this the purpose of FIRST? To teach, read inspire, students about the forces and how to use the tanks properly? Each team should find a Fluid Power Engineer or a mentor properly trained in the safe use of pneumatics, or at least follow the manufacturer's instructions for safe installation and use of the parts.
Of course, I don't want any person injured from mis-use or improper installation, so there must be some adult involved in the process. A key tenant of FIRST - students need mentors!
JamesCH95
11-03-2014, 09:08
Here is statement from the Plastic Pipe Institute: http://plasticpipe.org/pdf/recommendation-b_transport_compressed_gas.pdf
Notably:
The Plastics Pipe Institute recommends that thermoplastics piping intended for the transport of compressed air or other compressed gases should be installed by burial, encasement in shatter resistant material, or other appropriate means, to prevent or minimize the possibility of mechanical damage.
PP is a thermoplastic. Seems like a bad plan to use PP tanks from the get-go.
Allen, I understand where you are coming from but as someone who has had a 1 liter PVC pneumatic volume explode within 1 foot of me (thank goodness it blew out away from me) I will respectfully disagree here.
The dangers to anyone near one of these if it explodes, especially to bystanders, some of which may be children, is too high for my liking. Thus I say 'ensure it is safe and then we can use it'.
To the best of my knowledge PVC is not rated by any organization to contain compressed gases. Any use of PVC to do so is at your own risk and against industry practices.
This isn't about whether or not these tanks are quality products, it's about risk. Even if the failures are 100% caused by "user error," the tanks' failure mode is an explosion that sends shard traveling at high speeds. There are two attributes to risk, probability and severity. While we obviously don't have access to FIRST's safety risk matrix, I know that regardless of how low we drive the probability of the risk, the severity is still to high for these tanks to be acceptable for use on 1712 (and I would hope other teams). I feel much more comfortable using a tank with a failure mode that doesn't involve high speed, sharp projectiles.
Exactly!
The tanks themselves are perfectly safe. It's using them properly that's the issue. Putting stress on the tank increases the likelihood of it bursting.
The plastic tanks are not what I would call "perfectly safe." They are safe when used in exactly their prescribed manor with little room for error. They are not robust or fault-tolerant. Some of these tanks will go to teams where no mentors or students know how to properly handle them, and FIRST has no way to stop this until the robot is at an event.
Isn't this the purpose of FIRST? To teach, read inspire, students about the forces and how to use the tanks properly? Each team should find a Fluid Power Engineer or a mentor properly trained in the safe use of pneumatics, or at least follow the manufacturer's instructions for safe installation and use of the parts.
Of course, I don't want any person injured from mis-use or improper installation, so there must be some adult involved in the process. A key tenant of FIRST - students need mentors!
They should, but not every team will.
When inspecting our robot's pneumatic system this year our ERF was slightly out of adjustment and only started venting at ~140psi. The plastic tanks are rated for 125psi (the average Clippard metallic tank is rated to 250psi). If our robot had these tanks, they would have been subjected to pressures outside of their rated range. I wonder how many teams this has happened to...
techhelpbb
11-03-2014, 09:11
Isn't this the purpose of FIRST? To teach, read inspire, students about the forces and how to use the tanks properly? Each team should find a Fluid Power Engineer or a mentor properly trained in the safe use of pneumatics, or at least follow the manufacturer's instructions for safe installation and use of the parts.
Of course, I don't want any person injured from mis-use or improper installation, so there must be some adult involved in the process. A key tenant of FIRST - students need mentors!
I agree with the idea of finding a mentor to help with this.
However pneumatics and hydraulics can kill even when properly engineered and suffering from damage from wear and tear.
Sure these are smaller systems but even a syringe can kill you with just air in it.
If we are going to allow plastic tanks with several examples of unpleasant failures FIRST ought to think carefully about enforcing the safety element through inspection and published recommendations about safety sleeves.
Accidents are preventable - all it takes is recognizing the risk.
A few explosions are a demonstrable risk hard to ignore.
I also want to say as a CSA - I am around all of your fielded robot all of the time.
If I get hurt because of something like this don't think you have legal protection.
FIRST you know you have an issue - address it and don't be negligent.
What I wrote here might seem quite blunt - realize I've worked on hydraulic systems that deform building steel.
These things are unforgiving.
... and this is the core of my issue with the tanks.
You have students that have very little understanding of the forces and failure modes and 'how to use the tanks properly' in an environment where the public is being exposed to that risk.
All it takes is one serious injury. Safety (for our students, mentors and the general public) needs to be the top priority, always.
If it's possible that through ignorance of safe protocol the tanks can injure the public, then we must look at ways to mitigate that danger, including possibly banning their use.
You had better make some changes for the rest of this season and every season from now on. With that attitude it would be downright crazy to allow a high school student near a welder, drill, table saw, chop saw, band saw, drill press, mill, lathe, plasma cutter, water jet, can of spray paint, battery, rotating component, belt drive, spring, bungee, surgical tubing, sander, robot cart, any heavy item, shearing press, sheet metal brake, or anything else that could cause an injury if someone doesn't follow the directions. Just because there is an issue with an item doesn't mean it should be banned. There are better suited items on the market, and maybe we should all use those. But, I for one, like threaded fittings that let me change my push to connect fittings because I know they wear out. After working at Parker Hannifin Brass Products for a few years after college I am very familiar with the design of those fittings and I know that they don't last forever. I don't necessarily want to buy a new tank when my o-ring gets a tear. I can just be careful when assembling my tanks.
With that said, we only use the stainless tanks. Besides, they look cooler and I like shiny things...
JCharlton
11-03-2014, 09:58
Here's a suggestion, require teams to install shields (e.g. Lexan) around tanks, and for teams tight on weight don't include the shield weight in the weight of the robot. Sleeves might work, but small pieces could still escape - at high velocity.
This is a very dangerous failure and should be addressed by teams and by FIRST as soon as possible.
JamesCH95
11-03-2014, 10:08
Here's a suggestion, require teams to install shields (e.g. Lexan) around tanks, and for teams tight on weight don't include the shield weight in the weight of the robot. Sleeves might work, but small pieces could still escape - at high velocity.
This is a very dangerous failure and should be addressed by teams and by FIRST as soon as possible.
Why would you do that when a metallic tank leaves nothing to chance?
How are teams going to provide engineering-based proof that their blast containment scheme works?
This is still a band-aid fix to mitigate an inherently dangerous failure. The real solution is to use tank material that does not exhibit brittle failure in the first place.
MrBasse has a point. With the mechanisms on the robot for today's game there are just as much if not more chance of people getting hurt. Surgical tubing can cause great hurt when attached to metal parts. On my teams robot there are 4 springs, 2x60lb and 2x90lb. That's 300lb force on the arm. I have seen and inspected even stronger mechanisms.
There are many tools that we use that are dangerous. This does not stop us from using them. We must take precautions and make sure safety is prime in our thoughts, second nature if you will. If we find that a saw blade breaks and has issues we don't continue to use those blades we find better quality ones. The same would go for pneumatics, tanks prove to have problems, find better. I am all for safety and agree we must be doing something but banning is not the issue.
It's Tuesday, maybe we will see something from FIRST today. After all, there are a lot of bright minds at FIRST that want the best for the students.
Has anyone seen a black tank fail?
I cut a black tank in two because it was damaged by being rubbed by a drill chuck. Doesn't hold air now. But that probably not what you were asking. :)
JamesCH95
11-03-2014, 12:01
I've never seen or heard of a black tank failing, yet. Perhaps this is a result of the design improvements and the heightened awareness of plastic tank failures.
Has anyone seen a black tank fail?
Here are examples of the black Clippards failing, albeit a different failure mode than the "exploding":
Team 225 has managed to make one of the new black plastic tanks fail, but the failure actually occurred on the push-to-connect fitting, which snapped out under pressure. This might be an intentional design, as losing a hose is probably less dangerous than throwing shrapnel around
Oddly enough, this may be due in part to a manufacturing defect, as we've had one of the fittings come out as well, and AM has a note about it on their site. http://www.andymark.com/product-p/am-2649.htm
We broke the fitting on one of our black tanks, but that was the fitting. I'm fairly certain the tank is still able to hold air just fine.
Lil' Lavery
11-03-2014, 13:02
The tanks themselves are perfectly safe. It's using them properly that's the issue. Putting stress on the tank increases the likelihood of it bursting.
What evidence do you have to back your claim? What kind of testing or observational data can you share to support your hypothesis that these tanks are "perfectly safe" and that the only issue is "using them improperly?" What kind of impact or thermal testing can you provide about using these tanks in an FRC environment?
It's not that I disagree with your general point that user error is a factor in the failures (though I also don't think that's an acceptable excuse), but if you're going to make an assertion like that (repeatedly, no less) you need to have something to back it up. Especially when safety is involved.
So, how do you know these tanks are safe?
What evidence do you have to back your claim? What kind of testing or observational data can you share to support your hypothesis that these tanks are "perfectly safe" and that the only issue is "using them improperly?" What kind of impact or thermal testing can you provide about using these tanks in an FRC environment?
It's not that I disagree with your general point that user error is a factor in the failures (though I also don't think that's an acceptable excuse), but if you're going to make an assertion like that (repeatedly, no less) you need to have something to back it up. Especially when safety is involved.
So, how do you know these tanks are safe?
The identified main source of failure is from user error in over tightening the fittings. Maybe since Clippard has discontinued production and sales we should all just call it a day and stop using them. However, it is a product defect as much as a car has a defect when a person claims they thought they could take a nap while driving because they had cruise control on...
http://www.clippard.com/part/AVT-PP-41
Alex Cormier
11-03-2014, 13:19
We are having an LRI conference call this evening. I will make sure that it is brought up. I know that Al is aware of these threads so he may already be working behind the scenes. I will post back tomorrow.
I hope it's not too drastic. I just placed an order for 6 of those black tanks.
I hope it's not too drastic. I just placed an order for 6 of those black tanks.
I think black tanks should be OK.
AllenGregoryIV
11-03-2014, 15:22
I still think people are grouping the tanks into two categories when in reality there are many more than that. Color is not the only distinguishing factor.
2013 Clippard White Tanks - As far as I know are no longer being sold after the manufacture discovered problems and offered to replace them for free.
Only place I have found a photo is in the 2013 pneumatics manual http://www.usfirst.org/sites/default/files/2013FRCPneumaticsInfo_RevB.pdf Teams have had many issues with these tanks some of which have been posted in this thread.
Pneuaire - http://www.pneuaire.com/reca44cuin.html These have been in use for 3 seasons and as far as I know have not had any problems.
AndyMark- http://www.andymark.com/product-p/am-2479.htm I'm not sure who manfactors these tanks but they look similar to the Pneuaire variety and again I don't know of teams having problems with these tanks.
Clippard 2014 Tank - http://www.andymark.com/product-p/am-2649.htm (http://www.andymark.com/product-p/am-2649.htm) Uses ress in hose fitting and has had at least one failure but not explosive failure.
It's my understanding that Clippard just started making plastic air tanks for the FRC market in 2013. I never saw them on their website before then. Where Pneuaire has been producing them with a burst pressure rating of 500 psi and an operating rating of 125psi for several years before the FRC community started using them.
This is a safety matter but we need to make sure we are getting the facts right.
Pneuaire - http://www.pneuaire.com/reca44cuin.html These have been in use for 3 seasons and as far as I know have not had any problems.
AndyMark- http://www.andymark.com/product-p/am-2479.htm I'm not sure who manfactors these tanks but they look similar to the Pneuaire variety and again I don't know of teams having problems with these tanks.
This is a safety matter but we need to make sure we are getting the facts right.
While Pneuaire and AndyMark look eerily similar, Pneuaire does not sell a 36 cubic inch tank on their website.
Additionally, somehow the Andymark tank is physically larger (Diameter 2.7" vs ~2.5" and Length 12.1" vs 9.9") than the Pneuaire 44 CI tank, but manages to hold less (36 vs 44 cubic inches). Unfortunately I do not have an Andymark tank to compare physical dimensions, weight, volume etc.
Jon Stratis
11-03-2014, 15:59
I agree we need to keep our facts straight. I do wonder though how much of the failures we've seen so far are influenced by usage statistics. For example, which tanks are used the most? Which tanks are used more by knowledgeable, experienced teams versus those used by inexperienced teams? Statistics like that could provide an explanation for why we've seen failures in the white 2013 tanks but not the others:
- The white 2013 tanks have been around longer than the black 2014 tanks
- The white 2013 tanks might be more prevalent in FRC than the Pneuaire tanks, due to their availability on FIRST Choice and AndyMark last year.
- The Pneuaire tanks might be used less by inexperienced teams - 3 years ago it was only a few teams pushing the boundaries that used the plastic tanks, and you would expect those teams to be more knowledgeable and having done more research than the teams that just used what everyone else used.
Obviously, the above statements are made with some big assumptions, and I'm not aware of any collected data to show whether those assumptions are correct or not. But without knowing some of those statistics, I also don't think we can limit our concern only to the white 2013 tanks - if there are orders of magnitude more white 2013 tanks in use than the other options, then our sample size for the others simply might not be big enough.
I would tend to agree that the Pneuaire tanks were used more extensively by experienced teams which extensively researched alternative lighter options to the conventional metal tanks.
If I had to guess I would say the Clippard tanks were used 5-10x as much if not more. Had I not read threads on the Pneuaire tanks on Chief Delphi when people first started using them I probably wouldn't realize they were a "viable" option.
I would also suspect to market for small tanks such as these is relatively small outside of FRC as compared to large 1L+ metal tanks.
Paul Copioli
11-03-2014, 16:09
My point is that the PneuAir tanks have been in use on FRC robots longer than the Clippard tanks and I have never (doesn't mean it hasn't happened) heard of one of them failing do to misuse or defect.
^this.
We have also used the pneuaire tanks with absolutely no failure. What would be nice is if anyone has experienced a failure with the pneuaire tanks that they share their results here.
I think we should not over react. We should also make sure tanks are agency approved before they are competition legal.
Paul
I think that people are being too quick to assume that user error is the end of the cause of the failures. I have yet to see someone ask why this user error is occurring.
After pondering this, it does seem to be a design flaw. Likely this design flaw is in the documentation and the lack of safety knowledge. If there is a danger caused by stress from over-tightening the threaded fittings, there should be a sticker around the fitting that warns against over-tightening. The same should be true for any other safety issue with any device.
Also for any COTS device, read the manual before use. While the device may not pose an imitate safety hazard from a user failing to read the manual, reading the manual for the product will increase safety while using the product and prolong the life of the product.
Mark Sheridan
11-03-2014, 18:04
I think that people are being too quick to assume that user error is the end of the cause of the failures. I have yet to see someone ask why this user error is occurring.
After pondering this, it does seem to be a design flaw. Likely this design flaw is in the documentation and the lack of safety knowledge. If there is a danger caused by stress from over-tightening the threaded fittings, there should be a sticker around the fitting that warns against over-tightening. The same should be true for any other safety issue with any device.
Also for any COTS device, read the manual before use. While the device may not pose an imitate safety hazard from a user failing to read the manual, reading the manual for the product will increase safety while using the product and prolong the life of the product.
In any case we need to gather more info and be careful until we know for sure whats going on.
I am using the 2013 clippard tanks, I thought its failure mode was user error too. Am I mistaken, was there a product defect that lead to its recall?
I am using the 2013 clippard tanks, I thought its failure mode was user error too. Am I mistaken, was there a product defect that lead to its recall?
I do not believe that there is a mechanical flaw with the design, just a lack of world knowledge which led to improper use.
team222badbrad
11-03-2014, 18:16
I do not believe that there is a mechanical flaw with the design, just a lack of world knowledge which led to improper use.
Or lack of instruction/documentation included with the product?
Or lack of instruction/documentation included with the product?
The only problem with simply including documentation is that likely only one person will read the documentation and some people will think they do not need too. This is especially important with temperature issues and material safety.
You don't get safety information provided on a sticker on any other pneumatic components. If I crank a cast iron fitting into a stainless steel tank to far the tank won't hold air anymore, but there is no warning for that. I have yet to buy a cylinder that came with documentation about anything. Everything is online if you want it. Clippard had installation instructions on the product page that specified the torque for fitting to seal the connection. When FIRST posted this:
Clippard recommends using 2-4 layers of Teflon tape and tightening the ¼” NPT fitting to 10-12 inch pounds (do not to exceed 15 inch pounds).
last year they didn't make it up, it came from the instructions Clippard provided.
...
Pneuaire - http://www.pneuaire.com/reca44cuin.html These have been in use for 3 seasons and as far as I know have not had any problems.
...
I believe this is the 4th season of the Pneuaire tanks. Here (http://www.thunderchickens.org/index.php?option=com_content&view=category&layout=blog&id=30&Itemid=41) is a little evidence.
AllenGregoryIV
11-03-2014, 21:59
I believe this is the 4th season of the Pneuaire tanks. Here (http://www.thunderchickens.org/index.php?option=com_content&view=category&layout=blog&id=30&Itemid=41) is a little evidence.
Thanks, I didn't know that.
Has anyone used the air tanks from Robart (http://www.robart.com/store/air-systems-components/air-tanks)? I stumbled across them when looking for an alternative lightweight air storage solution.
Tristan Lall
01-04-2014, 13:05
Has anyone used the air tanks from Robart (http://www.robart.com/store/air-systems-components/air-tanks)? I stumbled across them when looking for an alternative lightweight air storage solution.
They have a rather ugly connector on them—like a bayonet fitting, without the tapered ribs. I wouldn't use them for FRC without substantial testing. They work fine on model aircraft for the landing gear.
AdamHeard
01-04-2014, 13:17
Has anyone used the air tanks from Robart (http://www.robart.com/store/air-systems-components/air-tanks)? I stumbled across them when looking for an alternative lightweight air storage solution.
I've used them, that fitting has a very small orifice so I'm curious if that would show up as an issue in FRC.
They are not robust in the FRC sense if exposed to direct impact (very thin walls), but would be completely fine if not capable of being hit.
Alex Cormier
01-04-2014, 13:18
Has anyone used the air tanks from Robart (http://www.robart.com/store/air-systems-components/air-tanks)? I stumbled across them when looking for an alternative lightweight air storage solution.
Team 174 has a few on the robot. They seemed to really like them.
vBulletin® v3.6.4, Copyright ©2000-2017, Jelsoft Enterprises Ltd.