pic: 222 3 speed closeup/cutaway

[Nuson]

That's absolutely beautiful. What power source are you using to shift?

[team222badbrad]

We use a pair of 3/4 inch air cylinders. With a throw of about an inch.

They can be seen in this picture:

http://www.chiefdelphi.com/forums/pi...&quiet=Verbose

One air cylinder shifts between high and low and the other shifts the transmissions in 2nd!

[jskene]

You mentioned that the force required to shift the transmission is quite light. Would it be possible to use the torque produced by a servo motor to shift?

I have drawn a sketch below of the concept. Basically, the inner shaft shifter rod is pushed and pulled back and forth by a pushrod attached to a wheel that is rotated by a servo, in the same way that you would push up/down an aileron on a model plane. The servo can be directly controlled by any PWM output on the robot controller. The servo wheel can be set to any desired rotational angle by varying the PWM output duty cycle. Certain angles would correspond to 1st, 2nd 3rd gear, as well as the neutrals in between each gear.

Do you think a servo would have sufficient power? It would certainly be a smaller, lighter-weight solution compared to using pneumatics, and could probably provide control of 3-4 gears. Because the electric motor in a servo is not particulalry strong, it may be possible to eliminate the springs in the original design. If the transmission is not aligned properly for the shift to occur, the servo motor just stalls, and waits the fraction of a second required for the next gear to line up. It would then rotate just the right amount to engage the ball bearings on the next sequential gear.

[CJO]

Do you think a servo would have sufficient power? It would certainly be a smaller, lighter-weight solution compared to using pneumatics, and could probably provide control of 3-4 gears. Because the electric motor in a servo is not particulalry strong, it may be possible to eliminate the springs in the original design. If the transmission is not aligned properly for the shift to occur, the servo motor just stalls, and waits the fraction of a second required for the next gear to line up. It would then rotate just the right amount to engage the ball bearings on the next sequential gear.

What about using one of the rotary pnematic actuators. They include magnetic position sensors which turn them into high powered pnematic servos. The positioning is very precise, plus, you could still use a spring, which would save wear and tear on your turning device, as you would not be continually stalling it out.

On another note, how does the CIM work? We built a transmission over the summer, where the CIM was drawing huge amounts of power, and ended up for this year with something similar to the two motor design in the white papers. Does a single CIM work? And if so, doesn't manual shifting get confusing during a fast paced match?

Back to the earlier SCS thing. It might be possible to design a "steptronic" type control that normally shifts automatically, but could be overidden for climbing, &c.

[henryBsick]

Originally Posted by CJO

What about using one of the rotary pnematic actuators. They include magnetic position sensors which turn them into high powered pnematic servos. The positioning is very precise, plus, you could still use a spring, which would save wear and tear on your turning device, as you would not be continually stalling it out.

On another note, how does the CIM work? We built a transmission over the summer, where the CIM was drawing huge amounts of power, and ended up for this year with something similar to the two motor design in the white papers. Does a single CIM work? And if so, doesn't manual shifting get confusing during a fast paced match?

Back to the earlier SCS thing. It might be possible to design a "steptronic" type control that normally shifts automatically, but could be overidden for climbing, &c.

Have to check on the rotary pneum.'s thanks for the idea.

Yes, it probably is possible to shift with a servo but we wanted to get away from them. We had troubles on the drill transmisions with the number being in the right spot to tell the servos where to stop. They worked every match but they occasionaly caused us grief.

We checked the power curve for the CIM's to figure in on a good ratio.

I drive it and it isn't bad. I shoot over some where quickly and then just get into a lower gear to push, pull, climb. It really isn't bad. I was talking to our programmer about adding a new switch and a current sensor to make it auto. We might we might not. It isn't really neccesary for this game to much.(for us anyway). But back ont to topic, no it isn't hard on the driver (me).

I am lost with the whole SCS thing since I saw the first post about it.

-Henry

[CJO]

Originally Posted by Henry_222

I am lost with the whole SCS thing since I saw the first post about it.

Correct me if I'm wrong, somebody, but I thik the whole SCS thing was to add sensors to the transmission so that the computer would keep it in the most advantageous gear at all times, similar to the transmissions on most cars.

[henryBsick]

Originally Posted by CJO

Correct me if I'm wrong, somebody, but I thik the whole SCS thing was to add sensors to the transmission so that the computer would keep it in the most advantageous gear at all times, similar to the transmissions on most cars.

I understand that something would be done to monitor the rpm's or another measurable function that would contribute to the horsepower. They would then shift up/down to get max. horsepower. The only thing was the specifics of the sensor. Where to put, what kind etc.

[tkwetzel]

Correct me if I'm wrong, somebody, but I thik the whole SCS thing was to add sensors to the transmission so that the computer would keep it in the most advantageous gear at all times, similar to the transmissions on most cars.

I understand that something would be done to monitor the rpm's or another measurable function that would contribute to the horsepower. They would then shift up/down to get max. horsepower. The only thing was the specifics of the sensor. Where to put, what kind etc.

The SCS was to be able to synchronize the gear and shaft. Automatic transmissions already put it into the most advantageous gear at all times. The SCS would lower or increase the power to the motors so that the gear selecting shaft would be running at the same speed as the next gear it is about to select prior to actually engaging the gear. It would cause less stress on all the components and give you smoother shifts.

[henryBsick]

Originally Posted by tkwetzel

The SCS was to be able to synchronize the gear and shaft. Automatic transmissions already put it into the most advantageous gear at all times. The SCS would lower or increase the power to the motors so that the gear selecting shaft would be running at the same speed as the next gear it is about to select prior to actually engaging the gear. It would cause less stress on all the components and give you smoother shifts.

I see now. Thankyou. There would be two set of sensors, and they would match rpm of the input shaft to that of the output shaft.

Great system but I see one probem with that. The shift wouldn't be exactly insantaneous, it would wait for shaft speed to = then shift. If a robust enough tranny is built i see no need for the matching of speeds, other than smooth shifting. Ours now is pretty smooth though thanks to the springs on the shifting rod. I have slammed it from 3rd into 1st many times (just to make sure) and it has skidded, I wasn't on carpet all the time, but it could take the throw of the ratio change. I still like both ideas though . / thinks to self/ there is an offseason for a purpose /done thinking/

[tkwetzel]

The shift wouldn't be exactly insantaneous, it would wait for shaft speed to = then shift.

That is a very good thought, but without any load on the motors, the speeds would match very quickly and it would still shift rather quickly.

[henryBsick]

Originally Posted by tkwetzel

That is a very good thought, but without any load on the motors, the speeds would match very quickly and it would still shift rather quickly.

Yeah, oops, forgot about the neutral positions. It would require servos or some other sort of multi psitioning shifter instead of our 1,2,3 pneumatics.
/Learns more and more each day. /

[tkwetzel]

/Learns more and more each day. /

That is what FIRST is about!

[CJO]

Team 254 used the same sensors on their drive shafts this year as are used on cnc mills to determine table position (I forget what they're called)

~Christopher

[team222badbrad]

It would be cool to have an automatic shifter but I do not see the need for it.

I doubt a servo would last very long if it were used with our trannies. If it was stalled while the holes were not aligned it might hurt the little motors. The servos would also require very accurate maching methods for the shift rod and also require very good programing to align the shifting rod with the ball bearings.
However a servo could probably handle the job.

[team222badbrad]

Just like to let everyone know that you are more than welcome to come take a look at our transmissions. I hope to have a handout and a powerpoint presentation running. We will also have a display board at our pits.

Everyone will be more than happy to talk to you about it.

Good luck!