HELP: Internal Solenoid Leaks Killed Us!

[Jim E]

We had some issues with the Festo valves this year. One worked fine, but the other had an open relay coil. The valve wouldn't fire.

We picked up 2 more from the FIRST spare parts table and they also had an open relay coil.

Finally a team donated one to us and it worked.

It weight isn't an issue, go with the SMC valves. They will be more reliable and easier to repair if something does go wrong.

[de_]

team number 1310

All SMC hardware. (not Festo). Got the SMC engineers on it. However, the leak rate could be the norm in industrial environments and we are expecting too much using a 12v compressor that has to be off for 5+ minutes between when the robot is placed on the field and when the balls are randomized and announcer finally finishes his lengthy intros in the round robin matches. By then 1/2 the air may be gone with the leaks !!!

Thats why I'm looking for input on teams that have used SMC solenoids for years in non-trival pneumatics on their robots. We have 8 cylinders but with careful design (use gravity a lot to retract cylinders) and our initial minimal leak practise days, we had enough air to be very competitive.

Thanks for the heads up about potential leaks in the regulator.

[Bochek]

It wasn't your guys fault we lost... Feces Occurs

Get it fixed up for GTR and we'll take down 1114.

If you need any help conserving air Ive been known to come up with some fairly nuts ideas.

- Bochek

[Chris Fultz]

We have used a significant number of pneumatics for several years, with very few problems, but this year we have replaced several of the SMC double solenoids. (we also failed a regulator last year.)

The SMC valves seemed to leak around the seals (between the top and bottom) and also 2 had some sort of internal issue where one exhaust port stayed in a venting mode and let the system pressure drop. We replaced 2 valves before shipping the robot, and at least one at each competition so far.

The replacement valves SMC sold us seem to be less reliable than the ones in the kit. the kit part numbers were sold out everywhere - this is a slightly better valve, but has not performed as good as the less expensive ones in the KOP.

As far as the overal system, you should be able to get all connections such that you lose less than 5 or 10 psi in 30 minutes.

[Mark McLeod]

We've always used pneumatics, with both Festo and SMC solenoid valves.
It does help to have a supply you can swap out suspect parts with (older robots end up with any leaky parts).

Our system this year holds pressure seemingly indefinitely, I'll have to send you our pneumatics lead. I think he'll fit in a crate...

At one time or another I've seen just about every possible part arrive with a manufacturing flaw: regulators, pressure release valves, solenoids, pressure sensor, damaged actuators, damaged fittings. It helps of course to assemble your pneumatics system outside the robot first making it easier to locate leaking sections, but it's probably too late for that unless you have a lot of spare parts.

• Square tube cuts (there's a special cutter for this), also rough, scratched tubing ends won't seal well either
• Use as few fittings as possible (I wouldn't follow the Pneumatics Manual example layout)
• Check for SMC gasket damage or flaws
• Push the tubing all the way in until it's firmly seated
• Don't use teflon tape twice. If you unscrew a fitting at all then completely remove the old tape and replace with fresh
• Debris such as flakes of teflon tape (don't ever wrap threads all the way to the tip) will lodge inside the solenoid valves to cause a leaky valve
• Check for barely visible nicks in long (and short) lengths of tubing
• Look for damaged threads from cross-threading, leaving them to rattle around in a large bag of loose metal parts
• Push-on fittings can be damaged by yanking the tube out by force, rough tube ends

Isolate the pneumatics system a section at a time working from upstream to downstream and eliminate all leaks in each section before moving on. Just temporarily dead-end the tubing with a plugged brass fitting or a loopback T-fitting. You should be able to isolate the problem to a specific part that way.

[Racer26]

1075 has had issues in the past with a SMC valve jamming because of very small flecks of aluminum that found their way into the air system... Not really sure if we ever got those out of that robot...

[ayeckley]

At the Buckeye Regional we went through three (3!) brand new, in-the-box Festo valves and did not find a single one that worked. We had good LED indications, and sufficient pilot valve pressure, but we never got a single stroke out of them. I'm confident it wasn't user error - our 2007 KOP Festo worked just fine in the same installation. If it weren't for a more favorable Cv from the Festo, we'd have gone right to the SMC valves in the initial design. Ultimately that's what we wound up doing anyhow. In the future we'll just save ourselves the hassle and go right to the SMC.

[IndySam]

Sometimes new valves wont work unless you manually cycle them a couple times first.

[ayeckley]

Originally Posted by IndySam

Sometimes new valves wont work unless you manually cycle them a couple times first.

Tried that also, I'm afraid...

[Alex Cormier]

Originally Posted by ayeckley

At the Buckeye Regional we went through three (3!) brand new, in-the-box Festo valves and did not find a single one that worked. We had good LED indications, and sufficient pilot valve pressure, but we never got a single stroke out of them. I'm confident it wasn't user error - our 2007 KOP Festo worked just fine in the same installation. If it weren't for a more favorable Cv from the Festo, we'd have gone right to the SMC valves in the initial design. Ultimately that's what we wound up doing anyhow. In the future we'll just save ourselves the hassle and go right to the SMC.

I also suggest to check them out on what voltage they run. I have noticed numerous 24 volt valves laying around the 1930 shop. and we actually put one on the robot by accident. So, check them out to see if they are the supplied 24v or 12v. I don't have a clue why they actually send us 24V ones anyway.

[ayeckley]

Originally Posted by Alex Cormier

So, check them out to see if they are the supplied 24v or 12v. I don't have a clue why they actually send us 24V ones anyway.

I know SMC sends us some 24V solenoids, but I was unaware that Festo did the same. Is there a way to tell from the P/N on the valve body? I verified that we had a good 12V input to the solenoid pins.

Sorry to hijack the thread away from the internal leak issue...

[Alex Cormier]

Originally Posted by ayeckley

I know SMC sends us some 24V solenoids, but I was unaware that Festo did the same. Is there a way to tell from the P/N on the valve body? I verified that we had a good 12V input to the solenoid pins.

Sorry to hijack the thread away from the internal leak issue...

It states right on the solenoid what voltage it takes.

[IndySam]

Originally Posted by ayeckley

Tried that also, I'm afraid...

I find it very difficult to believe that three brand new one were bad right out of the box? Festo makes great products that many teams use with great success, including us. After three I would question my installation. How were you switching them + to - or on and off?


I am not just trying to defend Festo, my dad retired from SMC.

[ayeckley]

Originally Posted by IndySam

I find it very difficult to believe that three brand new one were bad right out of the box?

I do too, which is why the problem was so consternating. Usually a pattern like that points to user error (translation: my error), but I just haven't been able to figure out what we could have done wrong, especially given that we did have one working Festo, albeit from the 2007 KOP. There's just not that much that we could have done wrong, and I think we've worked through all the possibilities other than infant mortality.

My initial guess when we identified the first problem was that we'd somehow overtorqued the mounting screws and distorted the plastic case, causing the pilot to bind, but we rejected that theory after we had the same problems with the unmounted replacement valves.

If we'd wired up things wrong electrically, then we shouldn't have had a good LED indication, nor should we have had a good 12V measurement on pins 1 and 2. The silkscreening on the valve body indicated that it was a 12V part, not a 24V part, so I think we have to reject that theory as well, unless it was mislabeled.

If we'd plumbed things wrong pneumatically, then we should have had the same issue with the 2007 KOP Festo, but we didn't. When the valves were swapped, the problem followed the valves.

If the infant mortality rate of the Festos in the 2008 KOP really is 50% (which it honestly seems to be, based on the reports I'm reading from other teams) then it isn't really surprising that we'd get three bad ones in a row from the parts desk at the regional. The real question is how could the IMR be that high? The most obvious possibility is that the 2008 KOP valves were units that had issues on the production line, and rather than scrapping them they were donated. I'm not prepared to make that assertion just yet since it has some pretty negative connotations, but right now it's the best explanation I've got.

[ebarker]

Originally Posted by ayeckley

If the infant mortality rate of the Festos in the 2008 KOP really is 50% (which it honestly seems to be, based on the reports I'm reading from other teams) then it isn't really surprising that we'd get three bad ones in a row from the parts desk at the regional. The real question is how could the IMR be that high? The most obvious possibility is that the 2008 KOP valves were units that had issues on the production line, and rather than scrapping them they were donated. I'm not prepared to make that assertion just yet since it has some pretty negative connotations, but right now it's the best explanation I've got.

Unfortunately that thought has crossed more than a couple of peoples minds.