Pre-charging pneumatic air tank

[magnets]

Originally Posted by pfreivald

First, the relief valve must be properly installed in order to do its job.

Second, relief valves bleed, they don't gush--it's entirely possible to overpressurize a system by adding air faster than the relief valve can relieve it.

Third, I'm 100% positive that you don't have personal knowledge of all the "incidents" mentioned, and so cannot speak authoritatively on what did or did not cause them.

Fourth, rationalizing unsafe or disallowed practices with shoddy handwavium arguments and invented "facts" is not useful. So please stop.

1). Yes, it does require that my relief valve is connected and calibrated. But I've already done that so that I can pass inspection.

2). Please show me an FRC compressor filling faster than the relief valve can bleed.

3). The two incidents mentioned in this thread were at normal operating pressure. Please find me the one that isn't.

4). Please show me the rule that prevents me from running the compressor for an extra couple of seconds after the pressure switch tells the compressor to stop. Also, can you show me my "fact" I invented?

[Jon Stratis]

Originally Posted by magnets

4). Please show me the rule that prevents me from running the compressor for an extra couple of seconds after the pressure switch tells the compressor to stop. Also, can you show me my "fact" I invented?

I would argue R88 part C prevents that. Since you are choosing to continue running the compressor after the switch tells you to stop, I would say you are over pressuring the system - you're pressuring it past the stop point for the switch, which is the only legal feedback for controlling the compressor.

R88
The pressure switch requirements are:

It must be connected to the high-pressure side of the pneumatic circuit (i.e. prior to the pressure regulator) to sense the “stored” pressure of the circuit.
The two wires from the pressure switch must be connected directly to a digital input and ground pin on the Digital Sidecar.
The cRIO must be programmed to sense the state of the switch and operate the relay module that powers the compressor to prevent over-pressuring the system.

[magnets]

Originally Posted by Jon Stratis

I would argue R88 part C prevents that. Since you are choosing to continue running the compressor after the switch tells you to stop, I would say you are over pressuring the system - you're pressuring it past the stop point for the switch, which is the only legal feedback for controlling the compressor.

R81 “Stored” air pressure on the ROBOT must be no greater than 120 psi.

My pressure switch stops at 114 psi. So, if I run for 8 extra seconds, I'll get to 119. It's not over the pressure limit.

[Alan Anderson]

Originally Posted by Tristan Lall

It's not an unreasonable interpretation to postulate that the "one and only one" compressor treats the original and/or any replacements as the same (as it would be for motor usage, cost accounting, etc.).

I believe the critical word here is "replacement". If you leave the onboard compressor in place while you use an offboard one, you're not using a replacement. You're using a second compressor.

[cgmv123]

Originally Posted by magnets

4). Please show me the rule that prevents me from running the compressor for an extra couple of seconds after the pressure switch tells the compressor to stop.

R34 does not list compressed air not generated in compliance with pneumatic rules as an allowable source of energy on the robot.

R88 says the cRIO must be able to control the compressor relay based on the state of the pressure switch. Shorting the pressure switch doesn't change its state.

[cgmv123]

Originally Posted by magnets

R81 “Stored” air pressure on the ROBOT must be no greater than 120 psi.

My pressure switch stops at 114 psi. So, if I run for 8 extra seconds, I'll get to 119. It's not over the pressure limit.

There is a way to calibrate your pressure switch to stop at 120psi.

[magnets]

Originally Posted by cgmv123

There is a way to calibrate your pressure switch to stop at 120psi.

How?

[cgmv123]

Originally Posted by magnets

How?

I don't know, but see this post:

http://www.chiefdelphi.com/forums/sh...77&postcount=5

[FrankJ]

You can recalibrate your pressure switch. Remember you don't have to use a specific pressure switch. You really don't want to stop at 120 psi because with scatter it might be 121 psi the one time the inspector is watching it. Those little pressure gauges aren't that accurate anyway.

With all the exploding air tanks, regardless of reason, expect the inspectors to be particularly diligent and narrow in their interpretation of the rules. To get the sticker you will have to conform to their interpretation, not yours. That is just is how life works.

[Tristan Lall]

Originally Posted by Alan Anderson

I believe the critical word here is "replacement". If you leave the onboard compressor in place while you use an offboard one, you're not using a replacement. You're using a second compressor.

I think I see the problem: when I read "Disconnect on-board compressor (as though to remove it and replace with a spare)", I assumed that meant "take it off the robot (like you would to replace with a spare)", not "unplug its air hose".

As the former, it's essentially two spare part swaps in quick succession. As the latter, I can see why it might be more problematic.

[Alan Anderson]

Originally Posted by Tristan Lall

I think I see the problem: when I read "Disconnect on-board compressor (as though to remove it and replace with a spare)", I assumed that meant "take it off the robot (like you would to replace with a spare)", not "unplug its air hose".

Considering that the goal is to end up with a fully-pressurized pneumatic system, I would expect that unplugging air hoses would be problematic. My interpretation of the "disconnect" proposal was that it was talking about the electrical connections, but since it seemed very unlikely to meet the requirements of the rules I didn't pay a lot of attention to the details.

As the former, it's essentially two spare part swaps in quick succession. As the latter, I can see why it might be more problematic.

I can't see any benefit in doing (or pretending to do) such a double swap, so I doubt that was the intent.

[Al Skierkiewicz]

The pressure switch is not adjustable. If your switch senses pressure at 114 psi, and you are sure your pressure gauge is accurate then replace the switch. It is likely defective and about to fail. I naturally would assume your gauge is faulty as over the years, this is more often the case with this complaint. Therefore running your system beyond the cutoff sensed by the switch is likely over pressurizing your system. Know when your data is lying to you.

[Tristan Lall]

Originally Posted by Alan Anderson

Considering that the goal is to end up with a fully-pressurized pneumatic system, I would expect that unplugging air hoses would be problematic.

I was envisioning a variant of the poor man's tubing plug: bend the tubing over sharply, and zip-tie around both legs of the bend to stop the air. I admit I've never tried that with an already-pressurized tube (and don't particularly recommend it either way).

Originally Posted by Alan Anderson

I can't see any benefit in doing (or pretending to do) such a double swap, so I doubt that was the intent.

To charge the tanks right before the match, yet still have a cool compressor for use during the match?

[Retired Starman]

Many of the robot construction rules may not make sense to the robot builders. But think not of "rules" but of "specifications". I wish FIRST would call the robot construction "rules" ,"specifications" instead.

When real engineers build products for the customer, it is the customer who gives them the specifications. The engineer doesn't argue with the specifications, but builds the product to meet them, because that's what the customer is PAYING FOR.

The customer may have a very valid reason for the specification, even if the engineer doesn't know what they are. For instance, specifying the model motor to be used on the robot. Say a customer is buying a mail-delivery robot to place in each of its 5000 business locations. They already inventory a particular motor at each of these locations, so it makes sense for them to specify this motor for use on the robot you have been contracted to design. It saves the customer from having to spend thousands to stock a different motor when you can design around their existing motor. Motors go bad, and having the right motor in inventory is nice and saves the customer money.

Another example from the same mail-delivery robot. The specifications may call for dimensions to not exceed 28 in. If you design larger than this, the robot won't go through all the doors in the customer's many locations.

Stop trying to get around FIRST's robot specifications Design your robot to meet them, even if you, in your infinite wisdom, think these specifications are dumb!

Dr. Bob
Chairman's Award is not about building the robot. Every team builds a robot.

[DampRobot]

Originally Posted by Tristan Lall

I was envisioning a variant of the poor man's tubing plug: bend the tubing over sharply, and zip-tie around both legs of the bend to stop the air. I admit I've never tried that with an already-pressurized tube (and don't particularly recommend it either way).

Theres a guy on our team that really loves to do this, especially when the tube is already pressurized. It leaks a bit while you're folding it over, but holds much better than a push to connect fitting once it's folded over and zip tied.