Is a shifting transmission really necessary?

Technical Technical Discussion
#1

There have been a lot of great discussion threads regarding how to design/build the best shifting transmission. These discussions almost imply that it is a foregone conclusion that shifting is better. I personally do not agree with this point of view and think the topic merits discussion.

Maybe you don’t need to shift at all.

I would argue that for the last two years shifting was not necessary. If you can attain your desired top speed and still have enough low end torque to break loose all of your tires/tracks in a pushing match you do not need to shift. Electric motors unlike internal combustion engines make their max torque at stall, which is where you want the most torque and pushing force. For max speed you need enough torque to go only as fast as you can drive. For us this is about 10 ft/sec. At neither extreme do you need to be at the max power output of your motors.

In 2001 and 2002 we could pick up movable goals. In 2002 a robot could pick up part or all of the two 90# goals at once and drag them back to their scoring zone while dragging another robot backwards. This was a season to have a shifting robot. For the past two years we did not have the ability to transfer any additional weight onto our robots during a match. We will make the shifting /non-shifting decision after the game is announced.

What are the risks rewards of having a shifting transmission?

Risks: Reliability, robustness, weight, packaging space, cost, complexity, time and energy required, etc.
Rewards: Maximum low end torque / top end speed.

A transmission is sexy and lots of fun to design and build. We have built three different two speed transmissions and a CVT. We have never put a shifting transmission on the field. In the last two years we have played over 200 matches (in season and post season) and have never lost a match because of a transmission problem.

Many teams do not really have the resources to put a shifting transmission on their robot but will proceed to attempt it anyway. This can divert scarce human and financial resources away from the other robot systems that need them: ball grabbers, hanging devices, etc. In many cases, I feel that this should not be done.

#2

I agree :slight_smile:

#3

i think that is why people like andy baker and mark koors are building the “AM Shifter” so people don’t have nearly as many problems as you describe…to each his own, but to me is it necessary? no…is it nice to have? absolutely.

#4

This is a debate we’ve had MANY times on these boards before.

Jay,
You’re logic is absolutely infallible. I agree with your points 100%.
But, there are certainly arguments (within your own logic) that show it can be important to shift. (I’d like to go over those quickly, for our viewers at home.)

My calculations prove it is ABSOLUTELY necessary for the functions we’d like.

Design Parameters:
-130 lb robot, no weight added.
-4 motor drivetrain (2 drills + 2 chips)
-80% efficiency
-1.4 Coefficient of Friction Wheels (experimentally determined)
-Max Desired Motor Loading @ continuous push = 50 amps.

Desired:
-12+ fps speed

My calculations show with the above setup, in order to design your robot to not only slip your wheels while still maintaining the 50 amp continuous max current draw… one must gear their robot at about 8 fps.

So… if we’re using the parameters listed above, I must design our transmission at about 8 fps (I ran some quick calculations).

Granted, I could either load my breaker a little more (maybe I’ll risk 60 amps continuous draw), or decrease the traction of my tires… but I don’t want to.

Another option is to over-gear my robot, and just trust my driver not to get into extended pushing matches, or possibly even implement a current limiting circuit to protect my breakers… but these options just don’t cut it for me.

OR… I could put some more motor in my drivetrain.
Maybe 6 motor drive is the way to go. Right now, we’ll say it’s not.

In this case, I need to ask myself, is 8fps good enough?

Not for our team, no.
We’d like to be faster. We’re also not willing to trade off our pushing power.
This means, we NEED to shift.

It’s all about what you’re willing to give up.
We feel (for most FIRST games) it is beneficial to go the extra mile and build a shifting transmission. However, maybe next year’s game will be the exception.

What it boils down to for us is:
-There is always a time you want to push through someone else.
-There is always a time you want to outrun someone else.

Either we put enough motor power into the drivetrain to do both, we sacrifice one for the other (in some capacity), or we shift gears.

Again,
Only try it if your team can handle it.
To be argumentative… last year we never lost a match due to a transmission problem either. In fact, we had NO problems last year with our drivetrain (the arm was a different story!) There were definitely times where the transmissions proved themselves a worthy investment of time and energy.
Depending on the game, we will strongly consider shifting again.

$.02
John

#5

The transmission is a means to and end…
It is the engineers answer to a set of functional requirements.

The more important questions are:
-Is it really necessary to go 10+ fps?
-Is it really necessary to push with 200 lbs of force?

If yes, then… the transmission is one perfectly viable solution.
Obviously, some sort of solution is “necessary”.

Physics isn’t just a polite suggestion, it’s the law.

John

#6

I believe that shifting is a nice luxury to have. Should all teams attempt to build one? This would be a yes. The issue is when should they try. I am with Team 188 and this year we had a shift on the fly transmission with 3 motors per side. This was the 1st year that we have successfully put one on the field. Last year we attempted one and failed. It was an excellent learning experience. It ended up teaching us not to wait so long before leaving the idea and that a gearbox can be built in 2 nights. The fact that we tried at this point shows that we did not jump headlong into it. Our team has been learning and growing and this was just another step in our growth curve.

Every year a team should look at it’s resources and decide what they can and cannot do. This is not limited to just transmissions. Our team tried to do everything this year and we had the manpower to try. Everything was not perfect but we were proud of our team and robot. That is what FIRST is all about. I helped teams that did not attempt nearly as much and yet when they were finished they were just as proud of their accomplishments as we were.

I don’t believe that there is any right or wrong here, but dreaming, that is what makes FIRST great.

#7

I agree with Jay TenBrink. 2 speed is always nice because you have speed and the power at the same time. you can have speed and torque with one speed transmission, but you can have more advantage from a 2 speed transmission. Is shifting necessary? not really, it really depends on the game, shift in fly is nice to have since you dont waste time to stop, shift and then go. for example, 2003 Stack Attack, a lot of teams have used 2 speed transmission in order to have speed and torque at the same time. some team went with 1 speed… like Baxter Bomb Squad, they were really fast and reliable. so what i say is, it really depends on the game. :slight_smile:

#8

Well, for what it’s worth, 1293 ran one speed last year. The end result wasn’t the speediest thing on two wheels and casters (see www.billfredinthenighttime.com/windowmotor_hi.wmv for video), but we could easily push around an office chair with a high school junior (now senior, I think) on it, which runs about the weight of a mobile goal. And we could juke and jive with the best of them.

This year, a two-speed is one of the things we’ve spoken of. Of course, <Dave-speak>things could multiply at any given time.</Dave-speak>

#9

we ran onespeed last year and felt it wasnt enough…
right now for OCCRA, ive designed our chain ratio so we can change our ratios through the drill transmission. we leave it in 2nd gear most of the time.(our “standard” gear) and when we feel our opponents might push us around or we want to be aggressive, we switch to 1stgear… and then of course when we want to zoom around we go to 3rd gear. (oh i love this new dewalt tranny)
of course we cant shift on the fly… more like cant shift for a whole match, but if we watch our opponents it should help us some. :smiley:

#10

Obviously, but what I was referring to was the shifting, thus the topic. Reminder (for our viewers at home): Is a shifting transmission really necessary?

necessary? no (no you do not need a shifting transmission to do well in this competition, Team 25 anyone?) …is it nice to have? absolutely (I would like to shift given I can sacrifice whatever weight, time, etc…necessary to produce such a shifting transmission)

thx :rolleyes:

#11

Sparky has never used a shifting tranny yet, although we have thrown together various ideas and have one in blue sky session.

As a principal of FIRST that has been reiterated again and again by Dean, you DON’T need Andy Bakers or other great engineers to build a robot, so a shifting transmission isn’t needed to function in the game. We have never used it and we have won two regionals in our 5 year history, plus we made it to the eliminations in Champs.

As Johnny said before, if you want to do some things, you might want to look at a shifting transmission. Back in 2002 it was a good idea to use one to push around the goals, especially if you were like Sparky 3, MOEhawk, or Beatty that year and wanted to tackle all three at once.

#12

John,

We design for a higher max current draw. With the addition of the little blue battery this year it is possible to increase the max current draw of the drive system without putting the controller to sleep if you momentarily drop under 8 volts.

Our efficiency is such that we can achieve 10 ft/sec. max speed (which is all we feel we need) and we are also able to break all four knobby pneumatic tires loose at the low end.

In 2003 we had a 4 motor system with a robot weighing 115#. This year we weighed 130# and had a 6 motor drive.

You are extremely correct: It is important to get there fast. Once you are there, you want to be able to stand your ground. We have been able to accomplish both with a one speed gearbox.

Thanks for the comments all,

Jay

#13

It depends on the game, your strategy, and your machine. We have never had a shifting tranny. However in 2002 we wish we had.

Team 25, in 2002, comes to mind. They would fly out to the goals, grab them and then shift gears.

In 2003 it would have been nice for certain situations, but certainly not needed. We compensated for single speed with very high friction tires. We feel that our tires helped us win many matches.

For 2004 I really don’t see any use for a shifting tranny at all.

But again, it depends on YOUR situation. Although, it is very cool to have one. :smiley:

#14

Note: much of the theory here, and the debates between 4 motor vs 6 motor vs shifting were discussed in depth last year in this thread: http://www.chiefdelphi.com/forums/showthread.php?t=22604

Jay,
Would you be willing to share more details about your gearbox?
Ratios and such?

Based on experience running these numbers, I know how the power output from a 6-motor drive would be enough to slip wheels at 10 fps. (As I discussed in my initial post.)

This is really the “other” option.
Again, it’s all about tradeoffs – either you build a shifter, or you add more power (more motors). 494 has taken the Tim Allen approach, and it has suited them well.

I’m just curious about where exactly you’re loading your motors.
Obviously, I know how well it works (I’ve seen you in action several times now).

How far can you really push the envelope?

Probably further, than I am willing.
Perhaps my reluctance to toe the line is hurting me, I know I’ve been accused of this before.

I’m never averse to taking pointers from a national champ,
John

#15

Jay,

Good question.
This thread also discussed the same issue. As I stated in the other thread, we have shifted gears every year since 1999. While you point out the obvious rewards, there are more. Robot driving control and voltage conservation are two major rewards that come from shifting from a high gear to a low gear. These are two benefits that you did not mention.

A robot with 2 speeds can be driven more precisely than one with just a high speed. From my experience, FIRST robot drivers who have 2 speeds to work with can have more control over their robot as opposed to drivers with only a high speed. Also, if a driver wants to move their robot into a tight position, it is easier to do while in a lower gear.

Also, a typical FIRST robot that is geared to go above 9-10 ft/sec gets hot when running multiple matches in an elimination scenario. If a robot is using a 4-motor drive setup, it is good to have a low gear to switch over to, in case motors get hot and circuit breakers begin to trip.

I do think that this explains much. The 2003 game was unique in that it did not require much precise positioning, and allowed for less control of robots. The 2004 game required more precision, and I can see why a single-speed, 6 motor drive base was successful.

I still like the 2-speed design. There are good 2-speed designs that are reliable, robust, and lightweight… and they don’t cost too much either.

It really does depend on the game. If a game comes around like 2002, a 2-speed will be a hot commodity.

Andy B.

#16

Just to add my two cents…
I understand the up side to dual speed but (a big “but”) is it really needed on a 48 foot long playing field with other robots and field pieces in the way. The difference between 8f/s and 12 f/s is only 2 seconds from end to end. As to the design current draw, the circuit breaker although rated for 50 amps will require a current 6 times that great for more than fifteen seconds before the circuit breaker will trip. Add in the influence of the wiring resistances, the “on” resistance of the speed controllers, internal resistance of the battery, and all of the other factors and it is impossible to deliver full rated stall current to any motor in a functional competition robot. Given that max stall current will occur when motor brushes are in contact with two commutator segments (two motor windings in parallel) a multi motor drive will likely place more than one motor in such a condition during a standoff. Depending on slip in the transmission(s) motors are likely to rapidly alternate between two and one winding while in a standoff thereby changing (lowering) the load current. Granted a six motor drive will have more motors in this condition and therefore would get closer to max trip current than a four motor drive. I think that anyone who spent anytime in the Martians pit and watched the results of the dyno test or the Matians superior pushing ability would have to concede that there is an alternative (and a good one) to multispeed transmissions.
Also, what is the risk to add speed change decisions to the drivers? It seems that adding another variable into the driver equation would tend to slow down response times. Although speed changes on the transmissions is fairly rapid, is it the right thing to do in a 105 second match? In rough guess of perhaps four speed changes in a match, you may have lost 2-4 seconds of match time in decisions and speed changes. As we all know, that length of time could make or break a match.
So the question still remains and decisions to weigh the alternatives still exists. I for one think dual speed transmissions are cool but adding up all the factors (time, current, weight, complexity) I still favor single speed.

#17

Is a shifting transmission really necessary?

The better educated engineers above me have pointed out most of the pros and cons of shifting or not. That doesn’t really address the question.

The correct answer is, it depends on the game and your game play strategy. Its been mentioned before, but it has been buried inside justification-by-feature.

In 2002, it was absolutely necessary for some teams to shift to get to the goals faster. The difference between 8fps and 12 fps was a full second of personal time with the goals. However, Beatty didn’t need a shifting transmission because they thought outside the box and engineered another solution to the problem. Now, because of rule changes, their file card trick won’t work anymore, but I am POSITIVE that there is another way to pull off a similar outside-the-box solution to a movement problem.

Is a shifting transmission really necessary?

I’d say no, not because the features are unnecessary, but because limiting yourself to one solution is a bad idea in FIRST.

Wetzel

#18

That means you can get to the center of the field a full second earlier. In the 2004 game, beating the opponent to the bonus ball tees and mobile goal in autonomous mode was critical to many teams’ success. Two seconds is not terribly important in itself, but in competition against another robot every advantage helps.

A well-designed shifting transmission, controlled by an experienced driver (or a clever bit of software), is an advantage.

#19

I agree with most that you need to first work out your functional requirements before making a decision. There are a lot of possibilities that are not really considered by the topic.

2 motors versus 2 motors + shift?

2 motors + shift versus 4 motors?

4 motors + shift versus 6 motors?

6 motors + shift versus … ?

Many people have been going down the 4 motors in a gear box + shifting for a while. This seems to be a pretty wide area in functional space for the general constraints that FIRST has been imposing.

Having a shifting gear box design in your tool box is probably a pretty good idea. In 2004, as an example, if you used a hanger to pick your robot up, you would probably want a shifting gearbox. Hanger Up = FAST, low torque, Hanger Down = SLOW, high torque. In other words, having a shifting gearbox in your tool box does not mean you will only use it on your drive system.

What appears to be another wide area in functional space is the six motor, independently driven wheel design.

This design gives both low end torque and high speed, without a transition point that would come in a transmission. One thing which has not been talked about is getting up to speed. Unless you have a shift-on-the-fly transmission, you have to tolerate low acceleration to get to high speed. This may account for 1-2 seconds delay for a 12 fps relative to a multi-driven-wheel design.

Therefore, the multi-driven-wheel design has an attribute which I will call “quickness.” This translates into an ability to maneuver (which can also be accomplished by an omni-wheel design).

The other advantage to a multi-drive-wheel design is “failure tolerance.” If one motor blows out, freezes, over-heats, you still preserve the majority of your function. As an example, in 2004, one of our drive motors burned out. We noticed that the robot “pulled” to that side, but it took two matches before we realized that we had lost a motor.

If you only have one drive axle per side, a single failure can eliminate half your drive system. In our case in 2004, with three driven axles, it would take a triple failure on one side to take us completely out of commission.

My experience in FIRST has taught me that, whatever you do, make sure that no single failure can eliminate half your drive system. Whether this is a resetable breaker, drive shaft, drive chain, speed controller, pneumatic pump, etc., make sure that there is no single path to failure. The worst feeling in the world is having your robot driving around in circles when there is critical business to be done on the playing field.

#20

Just to throw out there my thoughts on the matter. A shifting transmission is a hot commotity, it’s nice to have, wether it be a 2 speed or up to a 6 speed, it doesn’t matter. Looking at the whole thing from an automotive standpoint it’s just a quicker way to get to top end, but that is with a combustion engine, I know. A quote comes to mind when thinking about the way we build robots, we bash them, we crash them, we abuse them. Our drive shafts can get buggered up and torqued. And a quote comes to mind:

ROBUST, ROBUST, and ROBUST if you want to win - dan 322

A shifting transmission can be awsome, you get multiple speeds and torque, but you can have problems, gears grind and teeth break, dogs can crack, and all kinds of bad things can happen. You can get wobbel and warp, and then you are in trouble. Being on Team 384 for 4 years, and now I’m a mentor, and from the great mentors I learned from, the more simplifed your robot, the more robust you can create it. I’m not saying it’s not nice to have a shifting transmission, it is, but i don’t think it’s truly necessary for a competative robot.
mark