2 motors on 1 Andy Mark transmission

[Japper]

During last year's competition I saw a few teams use 2 motors on 1 Andy Mark transmission and their bots were fast and a couple of them were only 2 wheel drive....

What is the benefit of using 2 motors on 1 Andy Mark transmission?

Does this allow a higher gear ratio without popping the circuit breakers?

Is it better for pushing and torque or climbing with the right type of wheels?

Do issues arise with thie configuration if each of the two motors has a slightly different output speed per the PWM output fed to them or is this balanced out in some way?

If you used this type of configuration on your robot last year, what was your rationale?

Thanks

[JesseK]

Originally Posted by Japper

During last year's competition I saw a few teams use 2 motors on 1 Andy Mark transmission and their bots were fast and a couple of them were only 2 wheel drive....

What is the benefit of using 2 motors on 1 Andy Mark transmission?
...
If you used this type of configuration on your robot last year, what was your rationale?

Thanks

The total current used on the drive train is roughly the same whether it's 2 motor or 4 motor. The difference is that with 4 motors, the current to each individual motor is half of a 2-motor drive train. This also means that each motor in a 4-motor drive train has half of the load that a 2-motor does. Upon viewing the motor curve graphs, you can see that at the lower load the motor has more torque and an overall higher efficiency.

In the end, 4 motors have the same maximum speed as 2 motors. However, 4 motors have much more acceleration and can take a beating whereas two motors have much shorter lifespans. The advantage of using only two motors is that you can use the 2 other CIMs somewhere else on your bot -- most commonly on elevators/telescopic arms.

This is all easy to see when using JVN's calculator. The biggest use of this spreadsheet I've found is to design a drive train that's capable of pushing someone while also having enough current leftover in the overall system to use other motors on the bot.

[JVN]

Originally Posted by Japper

What is the benefit of using 2 motors on 1 Andy Mark transmission?

More motors = More Power.
More Power allows (among other things):

• Faster acceleration (at your chosen speed)
• Faster top-speed gearing (at your chosen acceleration)
• More Motor Loading = More Output Torque (at your chosen speed)
• Less Current Draw (at a given loading)

If you understand the concept of "Power" you will understand immediately the various benefits multiple motors will provide you. Research this concept (check out a physics textbook, talk to a physics teacher, talk to a mechanical engineer, etc) and you'll be all set. This is one of those FUNDAMENTAL concepts for drivetrain design.

Honestly,
I'm floored by the number of people who design FIRST drivetrains with seemingly no knowledge of this concept.

Regards,
John

[=Martin=Taylor=]

This was a major issue for us this season.

We decided to use two CIMs in the elevator and two CIMs in the drive train. We used a two-speed transmission with the idea that we could use the two gears to accelerate; similar to the way a car accelerates, by "shifting up."

The result was a lightening fast elevator (It could hurdle as fast as most shooters), and a somewhat slow drive train.

Actually, the drive train was very speedy, it just lacked acceleration. Without any defense slowing us down we could move very quickly around the track. But as soon as we started being defended (Finals on Curie) we lost all our "momentum" and were unable to accelerate back up to speed.

Another problem was that our robot always slowed down significantly when turning, which hurt our laps.

For the offseason (CalGames), we're going to use "Super Non-Shifters." We basically removed the shifting mechanism and gears, remade the side plates, and added the FP motors. We're gearing it for the same speed, but now it will accelerate more quickly.

I wish the JVN calculator showed some acceleration figures, like “distance to reach top-speed.” We’ve been using the “Tentacle Calculator” for this. We’ve never quite been able to comprehend the connections between torque, acceleration, and current draw.

[JesseK]

Originally Posted by Hachiban VIII

...
We’ve never quite been able to comprehend the connections between torque, acceleration, and current draw.

Seems easy enough -- f = m * a --> a = f / m

Convert the torque (moment force) output of the wheels into a linear force in the same coordinate system as "foward" for your robot and voila! It doesn't seem to be of much use to have the actual number if you're deciding on 2 CIMs vs 4 CIMs. Yet if you're into more exotic gearbox setups (a la 118, Robonauts) then it could be useful to see what motors aren't worth their weight for their contribution.

From there it's a simple use of the 3 kinematic equations to find anything else you want (disregarding energy losses, but you could account for that in the JVN calc). This part really is easy.

[JVN]

Originally Posted by JesseK

Seems easy enough -- f = m * a --> a = f / m

Convert the torque (moment force) output of the wheels into a linear force in the same coordinate system as "foward" for your robot and voila! It doesn't seem to be of much use to have the actual number if you're deciding on 2 CIMs vs 4 CIMs. Yet if you're into more exotic gearbox setups (a la 118, Robonauts) then it could be useful to see what motors aren't worth their weight for their contribution.

From there it's a simple use of the 3 kinematic equations to find anything else you want (disregarding energy losses, but you could account for that in the JVN calc). This part really is easy.

The original 2004 version of my calculator had some acceleration modeling features (using a simple non-calculus based approach):
http://www.chiefdelphi.com/media/papers/1469

I determined when I did the re-release that most users wouldn't need these features, and they would just confuse the majority of users.

If you're interested in playing around with them, check out the old paper I linked above.

Honestly, I don't use them anymore. They were really great when I was starting out playing with drivetrain calculations and they REALLY helped me gain an understanding of the physical concepts at work here. Now I just tend to use my experiential bench-marks. This is a total cop-out, but it is pretty effective for MOST games. Last year (Overdrive), I ended up doing a whole bunch of calculations, but that game is more drive-intensive than most are.

There is a lot to be learned by "digging deeper" into these concepts and models. Once you get a good understanding of how a FIRST drivetrain can be modeled as a system, it will then be intuitive to you to begin removing layers of complexity to find a simpler model which is more practical for everyday use. You will also be a MUCH better designer because of the work you've done.

Take the time, it is tough, but it can be a lot of fun if you dig that sort of thing (I do!) and the payoff is TOTALLY worth it!

-John

[waialua359]

Originally Posted by JVN

More motors = More Power.
More Power allows (among other things):

• Faster acceleration (at your chosen speed)
• Faster top-speed gearing (at your chosen acceleration)
• More Motor Loading = More Output Torque (at your chosen speed)
• Less Current Draw (at a given loading)

If you understand the concept of "Power" you will understand immediately the various benefits multiple motors will provide you. Research this concept (check out a physics textbook, talk to a physics teacher, talk to a mechanical engineer, etc) and you'll be all set. This is one of those FUNDAMENTAL concepts for drivetrain design.

Honestly,
I'm floored by the number of people who design FIRST drivetrains with seemingly no knowledge of this concept.

Regards,
John

John, thanks for the info. Didnt really consider the current draw and its effects.
I can attest that you will definitely get more power using 2 instead of 1 motor per tranny. But specifically for our situation, we were forced to use a 1 CIM, 1 FP with planetary gearbox setup since we needed the CIM for other functions and due to weight issues. We gave up on the idea at championships because with 2 CIMs per tranny, the response and power was noticeably greater especially around the turns of the field. I can only imagine what it would have been like if we only had just 1 motor per tranny to do this.

Glenn

[=Martin=Taylor=]

Originally Posted by JesseK

Seems easy enough -- f = m * a --> a = f / m

Convert the torque (moment force) output of the wheels into a linear force in the same coordinate system as "foward" for your robot and voila! It doesn't seem to be of much use to have the actual number if you're deciding on 2 CIMs vs 4 CIMs. Yet if you're into more exotic gearbox setups (a la 118, Robonauts) then it could be useful to see what motors aren't worth their weight for their contribution.

From there it's a simple use of the 3 kinematic equations to find anything else you want (disregarding energy losses, but you could account for that in the JVN calc). This part really is easy.

Yeah I get THAT part.

I only know the torque of the CIM at stall.... and it wouldn't be moving at stall would it? So I have to calculate the torque based on kT (happens to be 3 oz-in/A) which is dependent on amp draw. hmmm... I could figure that out using the JVN calculator

[vivek16]

Here's how I think of it:

Each motor is a biker. Would you rather have two bikers pulling a load or 4?

If you have 4, you will be able to pull twice as much and go faster easier (faster acceleration).

-Vivek

[AndyB]

I have to recommend JVN's app. Works like a charm and should tell you everything you need to know.