Andy Mark Super Shifter
 

Would anyone be able to tell me around how long it takes the Andy Mark Super Shifter to shift. It is not on the Andy Mark site that I know of and one advisor says .15 seconds and another says 1.5.

Any help would be appreciated, thank you.

Re: Andy Mark Super Shifter
 

well i believe that it varies because i have played with our super shifters and the gear has to spin usually for the dog gear to fall in place. So it depends on how fast the gears are spinning for the dog to fall in place and sometimes you could be shifting at the right time and they are lined up and they fall right in. So my answer is it varies but i am sure you could get an average time.

Re: Andy Mark Super Shifter
 

It depends on what you are using to shift with...

If you are using a servo shifter, it's going to take you a second or two to shift because servo is slow and you are only using about 5 lbs of force to shift with.

If you are using a cylinder to shift with, it's going to shift in spits of a second because a cylinder shoots out or in fast and you are also using 35-40 lbs of force to shift with.

Good luck.

Re: Andy Mark Super Shifter
 

We don't have any experience with the servo shifting models but with the pneumatic shifting the shift is almost instant.

If you are moving and you trigger the shift you won't notice any lag at all unless you are moving very slow. On the fly at speed it's bang and go!

Re: Andy Mark Super Shifter
 

to clarify, our team is wondering how long it takes to shift using the servo

Re: Andy Mark Super Shifter
 

Andrew, read my post that I made earlier in this thread.

Re: Andy Mark Super Shifter
 

We used Gen 2 servo shifters last year. Once we "broke in" the transmissions, they shifted very smoothly. Very. Still do. Not as fast as pneumatic, but I'd guess below a second.

We have no experience with the SuperShifter servo shifting, but I'd expect similar performance. I feel the simplicity as well as the weight and space savings of the servo shifting setup (goodbye solenoid valve, cylinders, tubing, Spike Relay) balance the drop in shifting performance relative to pneumatics.

Just keep the backup battery fresh. Buy enough to keep them charged and swap out every 1-2 matches, just to be safe. We rotated four backups last year, but you could probably get away with only two. Dying backup battery = weak servo holding force = shifters coming out of gear = bad.

You could also choose to employ an onboard backup battery charging circuit to keep the battery fresh.

Re: Andy Mark Super Shifter
 

Our brand new servo shifters (not broken in at all--less than 20mn run time) shift in a little less than a second. However, it is really important to oil the dog well and set good points for the servo to shift to. The shifting slows down a bit when the backup battery wears down.

Re: Andy Mark Super Shifter
 

we switched from pneumatic shifting to servo shifting this year. The servo shift seems to take roughly the same amount of time as the pneumatic. I was worried that the servo shift wouldn't be as reliable, but from what I've seen so far (1.5 hrs or so on the drivetrain), they seem to work just fine.

Edit: oh, by the way, we use gen2's, not supershifters, but i would venture to guess the shifting is quite similar

Re: Andy Mark Super Shifter
 

Team 40 used Andy's Servo shifters last year and a vex 7.2 battery for a backup battery. They worked great, though I would recommend servo savers as we broke about 8 servo's using them. The gears in the servo kept breaking so we just bought replacement gear sets. We used a shift on the fly as a turbo button so the robot ran in low unless "turbo" was pressed and the shift was instantaneous. We didn't cut power to shift which may also have had a negative effect on the servos.

Re: Andy Mark Super Shifter
 

Are teams using the shifter transmissions backing off on the power and then shifting or are you shifting under power?

Re: Andy Mark Super Shifter
 

Shifting under power. These are designed for shift on the fly operation.

We also still have the original servos on our Gen 2's. They lasted 6+ events' worth of matches with no problems last year. We overdrove the servos a bit past each of the shift positions to help maintain the dog gear mesh (you command the servo to keep going to a position a bit past the shift point but it can never make it there because of the mechanical interference of the gears, so there's always a bit of error-correcting holding force applied in the direction of the shift), but we didn't peg the servos to their limits. You will find you'll have to experiment and dial in your servo shift positions to obtain the best, most repeatable performance.

Re: Andy Mark Super Shifter
 

I asked because your shift times seem slow. A programed (x)millisecond coast command to the motor controller should allow quicker shifting.

Re: Andy Mark Super Shifter
 

I believe that "coasting" would actually be bad. The system is designed to shift while the gears are moving, which allows the gears to match up and engage properly.

Simply your code and remove any coasting involved. Even more important, probably, is to add some code to PREVENT shifting if the motors are not being powered (you should probably set some minimum level of power required to allow a shift, which would take a little experimentation).

Re: Andy Mark Super Shifter
 

All,

I'll post my opinion here about pneumatic vs. servo shifting.

If you have ANY pneumatics on your robot, then you should use the pneumatic shifting setup for 99% of the applications in FIRST Robotics. The only reason that servo shifting is better than pneumatic shifting IF pneumatics are already present is that servo shifting can provide a neutral gear. This seems to be a trivial feature, and therefore I suggest using a pneumatic shifter to anyone who asks me, unless they have no pneumatics on their robot.

Here are the advantages and disadvantages as I see them:

Pneumatic Shifting:
Advantages: 20-25 pound shifting force; very, very fast shifting (under 0.5 sec), will shift while under a pushing or accelerating load, mechanical adjustment available so that dog gear is not bottomed out
Disadvantages: requires a pneumatic system* on the robot, no neutral gear

Servo Shifting:
Advantages: no pneumatic system needed, provides a neutral gear
Disadvantages: weak shifting force (5 lbs), slower shifting (1-1.5 seconds), shift may hang up under a pushing or accelerating load, software adjustment needed to set high and low positions, requires a charged backup battery, wears out servos if the servo is constantly pushing on dog when software is not telling the servo to go to the correct position.

We did not offer a servo shifting solution until 2007, after the 2006 FRC game that required little or no use of pneumatics. We only offered it after MANY teams said that they really wanted a servo solution.

I wish that FIRST would let us use solenoids on FRC robots so that we could use a solenoid instead of a servo to shift.

* - for a pneumatic shifter, an entire pneumatic system is not required. I have run systems that use 2 accumulator tanks and an off-board compressor. These systems can shift for more than 20 times during a match if no other pneumatics are being used on the robot.

For what it's worth....

Andy B.

Re: Andy Mark Super Shifter
 

We just installed our SuperShifter this weekend and had trouble with the servo shifting.

Our mentor found out that you have to ease up on acceleration to shift with the servo while the pneumatic you can do under full power.

Re: Andy Mark Super Shifter
 

When you speed shift a motorcycle, some will just bang gears by shifting without letting up on the throttle. But on a motorcycle it will shift faster when you release the torque (on the trans) from the engine and allow it to disengage before the rear wheel starts to drive it.

Drag bikes are one example of this, with a press of a button, their pneumatic shifters kill the engine first, then it shifts.

The bike restarts instantly because it's still rolling. This method works on all motorcycles and cars with manual transmissions.

Some "DOG" style transmissions use an undercut on both dogs (in this case it would be on the slider and the free-wheeler gear). The under cut is approx. 7' to 14' degrees and it helps pull the gears together when shifting. But as you might imagine, a clutch or letting off on the throttle becomes more important.

(Directed at Andy Baker)
I am very curious if you looked in to a Hodaka style transmission before you built the super shifter?

Bill

Re: Andy Mark Super Shifter
 

Bill,

222 has used a 70's hodaka ball shifting type transmission since 2004 and we don't even need to think about backing off one bit to shift using pneumatics. However I will say that Dog shifting has been around FIRST much longer and has many more hours of run time than ball shifting. Both systems have their merits, but I think the dog shifting transmission is probably less costly to produce and doesn't require quite as tight of tolerances. I cannot say this for certain because I have only designed and built ball shifting transmissions.

OK back on topic, I know in the past 222 used servos to shift the drill transmissions and we weren't happy with the results because of the amount of fine tuning they took and also the amount of time they took to shift. To solve this problem we programmed in a 'jog' (cut power, jog motors, then resume operator control) whenever the driver shifted. This method took approximately 1.5 seconds to do, but it ensured as good of shift as we could get.

I have seen and played with the SuperShifters by hand and they feel like they have a very smooth throw. So I have to think it's a servo problem more then anything else. I am of the same belief as Andy Baker that 99% of the time a pneumatic shifter is going to be better on a FIRST robot.

Re: Andy Mark Super Shifter
 

quick question... Did teams that used the Super Shifters this year support the output shaft or did you leave it cantilevered. If cantilevered, did you experience problems?

Re: Andy Mark Super Shifter
 

we use the super shifter's and yes we did support the shaft.

Re: Andy Mark Super Shifter
 

We (973) put the longer output shaft on, and supported it a few inches off the face of the gearbox with cantilevered wheel on the outside of the support.

294 used the normal output shaft cantilevered.

Re: Andy Mark Super Shifter
 

our team use the super shifters w/ servos this yr and it work excellent no problems as far as the actual shift when we were driving the robot the it took bout .5 seconds to see a differenct in speed hints the shift.

Re: Andy Mark Super Shifter
 

We did NOT support the end of the shaft and it turned out okay. We never had any issues with them.

Re: Andy Mark Super Shifter
 

did you use the standard length shaft or the extended? I would guess that the shorter "standard" shaft would be fine but if you got the extended one it would be best to support it.

Re: Andy Mark Super Shifter
 

Short, we ran one sprocket off of it, down to the center wheel, then used the center wheel to run chains to the front and rear wheels. If you use the shaft as a direct drive onto a wheel, you probably do want to somehow anchor the end of the shaft, to avoid bending or misshaping.

Re: Andy Mark Super Shifter
 

Hello All-

Wouldn't supporting the output shaft (presumably with a bearing) mean that this shaft had three support points? I thought this was a no-no.

For those who did support the output shaft, how did you do it?

-Mr. Van
Coach, Robodox

Re: Andy Mark Super Shifter
 

1618 ran the standard shaft cantilevered with two sprockets. No problems whatsoever.

Re: Andy Mark Super Shifter
 

1816 got the longer output shafts and two bearing blocks and supported as shown in these pictures:
http://www.edinarobotics.com/photos/photo_detail.php?recordID=838&GalleryID=100&positi on=8
http://www.edinarobotics.com/photos/photo_detail.php?recordID=837&GalleryID=100&positi on=7
These pictures were from early in the build season, a lot has changed, included parts of the drivetrain, but we kept the shafts supported this way and had no problems so far.

Re: Andy Mark Super Shifter
 

These were my thoughts exactly. I'd presume that teams who supported the shaft built it into their drive train design and also have a tested/proven frame. Power loss due to a misaligned support bearing would be much larger than the standard efficiency loss.

For the record, we did not support the shaft. The 2 sprockets running off the shaft were 1/2" and 7/8" away from the transmission side plate.

Re: Andy Mark Super Shifter
 

I can't speak for other teams who cantilever their wheels, but 254 and 968 avoid this by only having one bearing in the gearbox itself for the output shaft. Our bearing housing for the center wheel extends through the 1x2 frame member and into a slip fit hole on the gearbox side plate.

Re: Andy Mark Super Shifter
 

inspired by the same design concept as 254/968, we did the same thing this year.

Re: Andy Mark Super Shifter
 

We went the other way (some will say the wrong way) and had each shaft supported at FOUR points, but adjustments were built in...

We machined our own extra-long output shafts (the AM long shafts were about 2 inches too short) to direct drive our half-track treads (centrally mounted on each side, omnis at the four corners, one AM SS per side). Two roller bearings, one built into either tread plate, supported the far end of the shaft and the keyed drive wheel. We then had an additional 1+" gap between the inner tread plate and the gearbox plate to allow for an optional sprocket set to drive the rear omni wheels (which we never added). Two beefy custom brackets with oblong mounting holes allowed us to adjust the exact position of the gearbox w.r.t the bearings in the tread plates.

With the gearboxes locked down, we could remove the outer tread plates for belt changes without worry of misalignment. No problems with the gearboxes or bearings through 3 regionals and the Championship.

Re: Andy Mark Super Shifter
 

This spreadsheet may be useful in looking at the difference in loading of the shafts. http://www.chiefdelphi.com/media/papers/1998

It would seem, based upon the math that the AM long shaft can handle ~85 lbf at the end with a safety factor of 2 and the short shaft ~175 lbf with a safety factor of 2.

We ran them cantilevered driving a chain.

Re: Andy Mark Super Shifter
 

Andy we used the servos on the Supershifter and for the most part they were great. We burned one out because it was pushing hard and wore out. The other weird problem we had was that when we turned on the robot and were set for autonomous mode to kick in the robot controller seam to give a command to the servos so there was a slight jitter and would cause our servos to jump out of gear. (especially on on left side) Once the autonomous mode would start then the servo would move to the correct position. The robot wouls not stay in gear when we set it on the field for the match. Our programmer said it was the act of the controller intiallizing that caused it. I guess if you relocate the position of the servo mechanically then the jitter might not cause the gear to go to neutral. This problem reeked havoc with our autonomous, the robot would lurch to the left until the gear popped in and then we went straight. We had to put in a delay so the servos would have time to lock into position before the robot would roll. I think we will go with pneumatics next year. Less problems
We love the Supershifter!