Chief Delphi

Chief Delphi (http://www.chiefdelphi.com/forums/index.php)
-   Technical Discussion (http://www.chiefdelphi.com/forums/forumdisplay.php?f=22)
-   -   Shifting with servos (http://www.chiefdelphi.com/forums/showthread.php?t=125171)

Micah Chetrit 24-01-2014 16:35

Shifting with servos
 
Hello,
our team has never used a shifting gearbox on our robots before, and we are considering using an andymark 2774 to have a shifting system similar to how RI3D released their shooting mechanism. We don't plan on having pnumatics on our robot because the only place we would use them would be on the shifter, so we thought about using a servo to actuate it.
Can a servo handle it well? I doubt that the force required is as simple as the force required to move a lightweight dog gear since there are other forces acting on the gear, so I don't know how much resistance to expect. Are there any issues you foresee with using a servo to release a launching mechanism?

RonnieS 24-01-2014 16:40

Re: Shifting with servos
 
Quote:

Originally Posted by Micah Chetrit (Post 1331793)
Hello,
our team has never used a shifting gearbox on our robots before, and we are considering using an andymark 2774 to have a shifting system similar to how RI3D released their shooting mechanism. We don't plan on having pnumatics on our robot because the only place we would use them would be on the shifter, so we thought about using a servo to actuate it.
Can a servo handle it well? I doubt that the force required is as simple as the force required to move a lightweight dog gear since there are other forces acting on the gear, so I don't know how much resistance to expect. Are there any issues you foresee with using a servo to release a launching mechanism?

I don't have any math on hand to back this up but I will venture to say no. Remember, you might shoot a max of 10 times a match maybe more...might be able to just have storage with off board compressor..of course go by the rules but it shouldn't be ruled out (pneumatics that is)::safety::

Isaac501 24-01-2014 17:35

Re: Shifting with servos
 
In my experience, servos are poor choices for shifting mechanisms on FRC robots.

I've observed them to have a difficult time shifting under load, and noted delays between left/right gearboxes when shifting.

Micah Chetrit 25-01-2014 10:26

Re: Shifting with servos
 
Thanks for the input guys!

KamalRC 25-01-2014 12:32

Re: Shifting with servos
 
This video of Anymark's sonic shifter shows pneumatic and servo switching.

http://www.youtube.com/watch?v=cK3PqMj1TeM

Willyspu 25-01-2014 14:34

Re: Shifting with servos
 
Quote:

Originally Posted by Micah Chetrit (Post 1331793)
Hello,
our team has never used a shifting gearbox on our robots before, and we are considering using an andymark 2774 to have a shifting system similar to how RI3D released their shooting mechanism. We don't plan on having pnumatics on our robot because the only place we would use them would be on the shifter, so we thought about using a servo to actuate it.
Can a servo handle it well? I doubt that the force required is as simple as the force required to move a lightweight dog gear since there are other forces acting on the gear, so I don't know how much resistance to expect. Are there any issues you foresee with using a servo to release a launching mechanism?

We haven't tested it yet, but we are looking at using one of these https://www.servocity.com/html/hs-805mg_servo.html
to move an arm that will pull & push the dog gear mechanism.


Jim

EricH 25-01-2014 15:10

Re: Shifting with servos
 
Quote:

Originally Posted by Willyspu (Post 1332151)
We haven't tested it yet, but we are looking at using one of these https://www.servocity.com/html/hs-805mg_servo.html
to move an arm that will pull & push the dog gear mechanism.


Jim

You sure that's within the servo spec for FRC? 4W, free speed*stall torque is the max.


With respect to the video, that's a non-loaded drivetrain, which will behave differently than a competition drivetrain, and there was at least once when the servo stopped in the middle of a shift. And the servo in general was slower than the cylinder. Both of those can easily be deal-breakers for a drivetrain servo shifter.

aciarniello 25-01-2014 16:36

Re: Shifting with servos
 
We are using an AM supershifter for our prototype winch mechanism. With a full load on our catapault, we could not get the dog to disengage even at 60 psi. I would be VERY doubtful that a servo could get it done.

Willyspu 26-01-2014 16:13

Re: Shifting with servos
 
Quote:

Originally Posted by aciarniello (Post 1332210)
We are using an AM supershifter for our prototype winch mechanism. With a full load on our catapault, we could not get the dog to disengage even at 60 psi. I would be VERY doubtful that a servo could get it done.

that begs the question . . . did you find a solution? :)

Willyspu 26-01-2014 16:42

Re: Shifting with servos
 
Quote:

Originally Posted by EricH (Post 1332167)
You sure that's within the servo spec for FRC? 4W, free speed*stall torque is the max.

Hmm, I guess 830mAx6v=4.98watts won't fly. Thanks!

Micah Chetrit 26-01-2014 19:06

Re: Shifting with servos
 
Quote:

Originally Posted by aciarniello (Post 1332210)
We are using an AM supershifter for our prototype winch mechanism. With a full load on our catapault, we could not get the dog to disengage even at 60 psi. I would be VERY doubtful that a servo could get it done.

Interesting. thanks for the heads up. do you have a work around yet?

Micah Chetrit 26-01-2014 19:23

Re: Shifting with servos
 
Would this help? http://www.andymark.com/product-p/am-2584.htm
It has about twice the torque of a normal servo but only has 2 positions which apparently gives it more torque. Power drain isn't any more than other servos.

Tom Line 26-01-2014 20:13

Re: Shifting with servos
 
Quote:

Originally Posted by Willyspu (Post 1332798)
Hmm, I guess 830mAx6v=4.98watts won't fly. Thanks!

Power = Voltage * Amps is one way to calculate power, but it is NOT the way the servo industry calculates it, and it is not the way FIRST calculates it. Please look at Eric's post that explains what the industry standard formula is. That servo is much much higher than 4.98 watts.

yash101 26-01-2014 23:47

Re: Shifting with servos
 
Actually, FIRST goes by the output power, not the input power.

Servos are a very bad choice because you will be shifting at top speeds. Servos have very little output torque, as compared to a piston. Also, designing the system with pneumatic pistons would be very easy!

You just need to have a pneumatic tank, valve, solenoid, all the basic components. You may be able to run off an external compressor, and precharge the cylinders!

MichaelBick 27-01-2014 17:03

Re: Shifting with servos
 
Quote:

Originally Posted by yash101 (Post 1332978)
Actually, FIRST goes by the output power, not the input power.

Servos are a very bad choice because you will be shifting at top speeds. Servos have very little output torque, as compared to a piston. Also, designing the system with pneumatic pistons would be very easy!

You just need to have a pneumatic tank, valve, solenoid, all the basic components. You may be able to run off an external compressor, and precharge the cylinders!

Some teams don't like using pneumatics because of the weight. If you are just running shifters than it may not be worth it.

67 has been one of these teams in the past. They have used the window motors instead of servos to successfully shift. You cannot shift on the fly, but it can also be lighter than a pneumatics system.

magnets 27-01-2014 17:21

Re: Shifting with servos
 
Quote:

Originally Posted by yash101 (Post 1332978)
Actually, FIRST goes by the output power, not the input power.

Servos are a very bad choice because you will be shifting at top speeds.

You just need to have a pneumatic tank, valve, solenoid, all the basic components. You may be able to run off an external compressor, and precharge the cylinders!

Servo's aren't a very bad choice. They're a great solution if you're only looking to shift a few times a match while you're not moving.
Quote:

Servos have very little output torque, as compared to a piston.
This is false, as pneumatic cylinders have no torque. You can also gear up your servo or use an alternate motor.

Quote:

Also, designing the system with pneumatic pistons would be very easy!
Yes, but a servo is easier. The super shifter comes by default with a servo. You'd need to spend more money and assemble an extra cylinder mount. Also, you'd have to put tanks, valves, regulator... on the robot. I'd argue that servos are a lot easier.

Andrew Schreiber 27-01-2014 18:47

Re: Shifting with servos
 
Quote:

Originally Posted by MichaelBick (Post 1333287)
Some teams don't like using pneumatics because of the weight. If you are just running shifters than it may not be worth it.

67 has been one of these teams in the past. They have used the window motors instead of servos to successfully shift. You cannot shift on the fly, but it can also be lighter than a pneumatics system.


Or because they haven't got anyone with experience using them. 67 has not used pneumatics since 2004 from what my brother was saying.

Bear in mind that when you are disengaging the dog gear you are fighting friction. If you don't have something else taking the load from the winch you will be fighting a LOT of friction. You can overcome it but chances are it won't be easy. A simpler solution might be to use a ratchet style system that takes the load. You'd need to figure out the math behind your particular system to tell if a servo or a motor with a cam could achieve the desired results.

RobotDoktor 28-01-2014 21:40

Re: Shifting with servos
 
I hope you are geared pretty low, considering the tension levels needed to launch the ball well. A ratcheting mechanism of some type will also be needed to avoid stalling the motor. Is using electrical solenoids an option for shifting?

Gregor 28-01-2014 21:50

Re: Shifting with servos
 
Quote:

Originally Posted by magnets (Post 1333296)
Servo's aren't a very bad choice. They're a great solution if you're only looking to shift a few times a match while you're not moving.

When would anyone ever want to do that?

pfreivald 29-01-2014 08:35

Re: Shifting with servos
 
Quote:

Originally Posted by Gregor (Post 1333947)
When would anyone ever want to do that?

Well, one would hope the catapult winch isn't driving while you shift into neutral.

As to the original question, I have two points to consider:

1. Using a ratchet to take the load off of the gearbox entirely is a good idea. We built a custom dog system for a winch, and once we get it into place the motor drives back a little and the ratchet/pawl takes the entirety of the load, so we can use a tiny little pneumatic cylinder to disengage the winch. (We have a fatter-but-not-insanely-so cylinder then disengaging the pawl, which takes about 12 lbs of force.) Given that you probably need a ratchet of some kind to keep the system from back driving anyway, this is a good idea to consider.

2. If you're not going to have any other pneumatics on your robot and a servo just won't do, consider using a screw drive. They're as strong/fast as you need them to be (based on motor power, of course--in 2010 we lifted our robot using a CIM-driven screw), accurate, easy to control, and for your application perhaps the best thing is that they're motor driven.

What they're not is efficient, but you can use this to your advantage to prevent back drive in some applications where that's desired.

Andrew Schreiber 29-01-2014 10:39

Re: Shifting with servos
 
Quote:

Originally Posted by Gregor (Post 1333947)
When would anyone ever want to do that?

Heavily depends on your strategy. Shift on the fly is nice but if your ratios are spread too wide it can actually be jarring. Perhaps the game has multiple modes wherein one speed is beneficial and another wherein a vastly different speed is useful. OR in the case of a power takeoff.

I've also run bots that servo shifted dewalts and when a shift was requested we backed off the power to the drive motors momentarily while the shift happened. Not the best, but it worked.

Course, I'm loving the simplicity of a 6CIM single speed right now.


All times are GMT -5. The time now is 16:43.

Powered by vBulletin® Version 3.6.4
Copyright ©2000 - 2017, Jelsoft Enterprises Ltd.
Copyright © Chief Delphi