Are two speed transmissions worth it?

[thefro526]

After numerous discussions with team members and doing a bit of research I'm still wondering are two speed transmissions worth the money, weight, time, etc... I am speaking more along the lines of the AM gen 2 shifters not a custom one off application. It seems that in most cases it depends on stratagies but in a overall statement are they worth it?

A few people on my team believe that there is no point to having them. I always say to them that we've always had robots in the mid speed range ~7-8 fps and I'd love to have two speeds somewhere around 5 fps in low and 10 fps in high. It seems that we always try to tell our alliance partners yeah we are offensive primarily but we can play defense but it always seems to be that we never have the required power to push other robots out of the way. Not saying that we need super pushing power or light speed but I would love know that I have a little extra power and a little extra speed.

Also are my speeds a bit off? Would the speeds that come out of the transmission be higher or lower with normal gearing (KOP gears)? And what experiences have you had with your teams transmissions? My GMP (Grand Master Plan) is some kind of six wheel or eight wheel drive (like 1270) with IFI high traction wheels. Any advice on 6 or 8 wheel two speed drive trains would be greatly appriciated

-Dustin

[AdamHeard]

Two great examples of efficient use of a 2-speed shifter this year are 968/254 (custom) and 330 (AM).

Both had a low speed around 5-6 fps and a high of 15-16 fps.

It may seem unnecessary, but I have a great perspective on these drive trains as I have played defense on 330 in about 5-6 matches (including 3 in finals) and in their shop on their practice bot. I also played defense on 968/254 in at least 4 matches.

Their low speed gave didn't allow me to out push them, and their high meant that they could easily break away. All of these teams used this advantage very well.

For reference, we had a single speed at 10 fps with a small CIM, big CIM and FP that was traction limited.



I would recommend the AndyMark.biz shifters for getting started. They take all the work out of it, and even 254/968 are planning on using the guts of them next year. My team is currently doing a prototype of the previously mentioned base with them integrated.

Overall, shifters can be very useful if you have the weight. You will never get a group to agree on this though, as it is a hot issue.

The value of shifters varies directly depending on your driver. If your drive uses them a lot to their advantage (take 254 for example), then yes, they are worth it. If your drive almost never uses it, then no, it's not worth it at all. However, if you plan to train your drive to take advantage of, or otherwise build your drive around the robot, then yeah, I think they're totally worth it.

[Cody Carey]

If they work, then yes. We had 3 speed this year, and The bottom one, (2.6 fps) was the only way we could score with heavy defense (Yeah, you 48 )

But before we perfected them, That thing on our control board wasn't a shifter at all... It was a "Get stuck between gears and Lose the match" lever.

[Joel J]

Something I'm curious to investigate is some sort of current-based shifting "algorithm." Have the drive default to a reasonable high gear (9-10 fps), then shift to low whenever the current draw grows beyond a certain value for a certain period of time. Then shift back to high when the current draw falls below a certain value for a certain length of time. Then have a manual override of some sort. Pretty much, the system only shifts to low to push.

[AdamHeard]

Originally Posted by Joel J.

Something I'm curious to investigate is some sort of current-based shifting "algorithm." Have the drive default to a reasonable high gear (9-10 fps), then shift to low whenever the current draw grows beyond a certain value for a certain period of time. Then shift back to high when the current draw falls below a certain value for a certain length of time. Then have a manual override of some sort. Pretty much, the system only shifts to low to push.

You could do this with encoders.

If the measured speed of the robot is significantly less than what it should be, it must be hitting something and then it shifts down. If the robot is in low and going 95-100% speed, then it isn't hitting anything and will shift back up.

I made a vex robot with that exact functionality.

[Joel J]

Originally Posted by AdamHeard

You could do this with encoders.

If the measured speed of the robot is significantly less than what it should be, it must be hitting something and then it shifts down. If the robot is in low and going 95-100% speed, then it isn't hitting anything and will shift back up.

I made a vex robot with that exact functionality.

Yep. I knew about this option, but measuring current draw appeals to me more. It seems as though it would be "simpler" to get a definitive answer as to which gear the robot should be in. 1126 and 33 have succesfully done auto-shifting on a 4 speed transmission using the encoder approach.

[Cuog]

Like has been said above it really depends on the game. Also it depends on how well your team can make them. If you cannot shift and get stuck between gears you will only hurt yourselves so if you are thinking about doing it, make some prototypes in the offseason see if you can pull it off.

For shifting maybe I'm old fashioned but I would prefer a manual selection of gears to give the driver exactly what they need/want.

[Goldeye]

Originally Posted by Joel J.

Something I'm curious to investigate is some sort of current-based shifting "algorithm." Have the drive default to a reasonable high gear (9-10 fps), then shift to low whenever the current draw grows beyond a certain value for a certain period of time. Then shift back to high when the current draw falls below a certain value for a certain length of time. Then have a manual override of some sort. Pretty much, the system only shifts to low to push.

I tried to do that this year. More to enhance our pushing power (ie prevent spinning out) than to shift gears but that was the next step. Never had time to collect data and get it working, though.

[newton418]

Originally Posted by Joel J.

Something I'm curious to investigate is some sort of current-based shifting "algorithm." Have the drive default to a reasonable high gear (9-10 fps), then shift to low whenever the current draw grows beyond a certain value for a certain period of time. Then shift back to high when the current draw falls below a certain value for a certain length of time. Then have a manual override of some sort. Pretty much, the system only shifts to low to push.

We tried this but found that usually, if we were in 2nd gear and all of the sudden had a current spike, the motors were already close to stalling, or they were pulling so much current they would reset the RC. Well, in either of these cases the DeWalts we used weren't able to shift.
Basically, I'm sure it can be done, but keep in mind that the motors might not be turning when the robot is told to shift, so design your shifter accordingly.

As a driver, I think I liked the control over the system. 1st gear (5 ft/s) is painfully slow in a 2 min match, and getting pushed around is so frustrating you have to downshift.

[Doug G]

Originally Posted by Joel J.

Something I'm curious to investigate is some sort of current-based shifting "algorithm." Have the drive default to a reasonable high gear (9-10 fps), then shift to low whenever the current draw grows beyond a certain value for a certain period of time. Then shift back to high when the current draw falls below a certain value for a certain length of time. Then have a manual override of some sort. Pretty much, the system only shifts to low to push.

We experimented with the allegro current sensors in the off season of '05 and had it working exactly how you said. We integrated it into our '06 bot with AM pnuematic shifters, but it annoyed our driver and so he always used the override mode and manually shifted. Perhaps with more practice a driver could get used to this autoshifting behavior. I say current based autoshifting is a cool thing to look into - make sure to set it's thresholds high enough so that it's not shifting when turning if you use high friction wheels and skid steer.

[Billfred]

<insert standard "it depends on your needs" text here>

1618 used Gen2 AndyMarks this year. The result was about the fastest thing on six wheels at Palmetto, while still enabling us to push as needed. (Sorry, 342.)

I was actually a little surprised to find out that our driver was using them backwards from what I imagined--apparently, the cylinders were rigged to shift to low when the joystick trigger was pressed, instead of the other way around. Acceleration in high gear was still surprisingly good.

I'm not saying anything definite, but they're certainly on the short list for next season.

[=Martin=Taylor=]

I am the only one to have voted "no" so far...

We used our 2 speed AM shifters 1 out of every 2 matches this year. They helped us dart across the field, but we never used them for an extended period of time.

I'll admit they weren't useless... but they weren't worth the weight of the pneumatics.

If you're going to use pneumatics anyways I would say "go ahead." But otherwise save the weight and put it where it counts - in the scoring device.

[Alpha 997]

Originally Posted by Hachiban VIII

I am the only one to have voted "no" so far...

We used our 2 speed AM shifters 1 out of every 2 matches this year. They helped us dart across the field, but we never used them for an extended period of time.

I'll admit they weren't useless... but they weren't worth the weight of the pneumatics.

If you're going to use pneumatics anyways I would say "go ahead." But otherwise save the weight and put it where it counts - in the scoring device.

Yeah but there are shifting device that does not require pneumatics thought. Forgot what they are called but they run on electricity and weights like nothing.

[Alex Cormier]

Originally Posted by Alpha 997

Yeah but there are shifting device that does not require pneumatics thought. Forgot what they are called but they run on electricity and weights like nothing.

servos.

I love transmissions and like the 4 speed auto's better then anything else. They give you the speed and pushing power you need and a mix in the middle to not lose power.