"Transmission Design Assistant"

[Greg Perkins]

Hello,
Most of you remember JVN's old Gear Ratio Calculator. I LOVED it, yet at times found it hard to use and to follow. So I am traveling the road of creating my own, and would like to have some feedback to as if this will be useful to teams for their pre-season drive concepts and etc. I would like to know what you would like to see in it (*NOTE* this is not going to be able to tell you how to manufacture, machine, or even tell you what your final design will look like, just mathematically being able to do Ratio's and etc.) IE, wheel selection and sizes, or gearing/sprocket choices and motor selection.
Note when i am done with the first version of it, I will ask for beta testers to see if they have found any glitches. ALSO, THIS IS SOMETHING I WILL KEEP UPDATED FOR THE REST OF MY CARRER IN FIRST UNTIL SOMEONE ELSE TAKES CONTROL OF IT. So please help me make it great for teams to take advantage of.

thanks for all of your cooperation!

[JVN]

My advice to you...
The toughest part in creating one of these is finding a balance between versatility and simplicity while maintaining functionality. The math isn't that bad. Spend a lot of on organization. Take the time to think through in advance how it will function, how it will flow, and what you want it to look like. Decide early what features you want to include, and what assumptions you are going to make.

If you need help with any of the math, check out the seminar my father and I gave at the 2005 Championship.

Good Luck!
If you do this right, you're going to gain a deeper understanding of this topic than you probably every wanted to know. It is definitely a great learning tool. My previous one was made over the course of about 2 years. I used it as a means of condenscing all the knowledge I took piecemeal off Chief Delphi, and from other sources.


It is funny you bring this up...
I have actually been recently going through and updating my old one so I can release a new revision for the 2006 season. Mostly organizational changes (I always knew the old one didn't cut it, when it came to ease of use) but there are a few functional enhancements as well. Expect it before the end of the year.

-JV

[Swampdude]

Frank Saladino from team 353 ( white paper ) made a great tool that I've used quite a bit in the past. I think some neat upgrades would be to:

1. Plug in the standard kit motor data, and give a pull down selection that would fill in all the factors for you.

2. Make a selection list of material coefficients of the commonly used wheel treads. i.e. wheel chair, skyway pneumatics, wedgetop, some belts etc. Include flooring coefficients i.e. HDPE or carpet.

3. Give a place to key in the weight of your Bot.

4. Give results on a power curve where the wheels would break loose or if you'll go to stall etc. This would need to account for all motors and driven wheels etc. but could be helpful.

5. Subtract from your top speed for efficiency losses, friction etc.

That's all deluxe edition stuff..

BTW here's the whitepaper to JVN's calculator

[team222badbrad]

Not sure if you may want to include other information other than just ratios, but here is something to think about.

One suggestion would be to have a flow chart.

I suppose you could also include drivetrains, etc...

I just think a flow chart may be handy to help the rookies or veterans decide what they want kind of drivetrain they want to build.

Once they decide 2wd, 4wd, the list goes on..., the flow chart could then decribe what type of transmission is best for that specific drivetrain.

So for example if you have a tank drive you can't easily have a high top speed.

Also another thing hand to know would be the pro's and con's of shifting transmissions and the pro's and con's of multiple motor transmissions.

Good Luck!

I will help you look the "rough draft" over.

[sanddrag]

Well, here I go deviating from the flow of the thread again.

My personal opinion is that in this case it is one of those "calculators make you stupid" scenarios. I think some teams may rely on magical spreadsheets so much that they don't know what is actually happening to the numbers behind the scenes. While they are great for some, my personal opinion is that sometimes these spreadsheets can be detrimental to the technical/design ability of some people/teams. I prefer to "run the numbers" the good ol' fashion way because I actually know what is going on and have a better understanding of why the numbers are what they are.

I say this with nothing against the creators of these spreadsheets; I want to make sure that is certain. These tools are wonderful resources for some people, just not for me I guess.

[Greg Perkins]

Sanddrag I understand whole-heartdly.
Included in each section (ie: 1 speed 2 speed) will be the mathematical equations as well as a simple autocad sketch showing the different gears and how they are relating to the calculator.
I want to stress that this is going to be easy to use and will give you all the data *i hope* that you are going to need to design your own box. probably by the beginning of next week i will release a "virgin-copy" on here so y'all can see how its going to look. It will not be done, nor looke pretty yet, it will just show you what you can do with it!

keep the comments coming

[Cory]

Originally Posted by sanddrag

My personal opinion is that in this case it is one of those "calculators make you stupid" scenarios. I think some teams may rely on magical spreadsheets so much that they don't know what is actually happening to the numbers behind the scenes.

If you know how to do the math, why waste time and effort if you can put it into a spreadsheet and get the answer with a fraction of the time and effort required to do all the math by hand?

[Cygnet_Info]

Originally Posted by Cory

If you know how to do the math, why waste time and effort if you can put it into a spreadsheet and get the answer with a fraction of the time and effort required to do all the math by hand?

I am also feeling the same.

[Greg Needel]

Originally Posted by Cygnet_Info

I am also feeling the same.

i think he is making a broad statement about the application of these spreadsheets for general use. for example i know alot of people at my school that can't do simple math in their Head's because they have been so Dependant on their calculators for so long. IMO this calculators are a great tool but if you are making a new one you should also include a step by step 2nd document explaining what does what and why....so people arn't just blindly punching numbers.

and a cool feature you could do is a drivetain generator, people punch in the speed they want and give you other factors like what wheels they have. and it might generate "best options" from the motors in the kit, using presets options. this might be good for teams with limited resources that can only use what they have lying around.

[Hieb]

Originally Posted by sanddrag

My personal opinion is that in this case it is one of those "calculators make you stupid" scenarios.

As a math teacher, I would argue that it is not the calculator (or spreadsheet) that "make[s] you stupid", but the way in which it is used. Both the calculator in the classroom or the spreadsheet for motor calculations can be a very powerful learning tool if used appropriately. Prior to last season, I had very little understanding of electric motors and how to choose an appropriate motor. Exploring the JVN Motor Calc spreadsheets and playing with the various equations helped me gain a much better understanding that I could then pass on to the students.

By the same token, just using the spreadsheet with no understanding of what the results really mean or how they were obtained can have the same negative effect as handing a calculator to an elementary student to do math. With no understanding of the output, it is impossible to judge if the results make sense.

I look forward to seeing this new design assistant, as well as the updated JVN spreadsheets, to give me more tools to help teach the students.

[JVN]

"Design is Iteration."
When you go through multiple iterations and revisions of a design, and it continues to evolve, the end-product will be much better than the original.
More revisions = more improvements. (This is not a linear trend, eventually you need to shoot the engineer and just build the thing).

It is much easier to iterate drivetrain numbers if you are able to plug and play.

I have a strong understanding of the underlying concepts, I've lived these numbers for 3 seasons now. Using this understanding of how a change here or there effects the overall outcome (be it speed, acceleration, or torque) I am able to change/tweak various factors in my spreadsheet to help "optimize" a design.

I would HATE to have to recalculate by hand every time I wanted to make a change. This would greatly slow down, and possibly impede the design process.

A tool like this doesn't make you dumber.
To take an extreme analogy: "I have a nailgun but I don't use it for building my house, because I want to make sure I'm really good at hammering."

If you have a true understanding of these numbers, a design spreadsheet is an incredibly helpful resource. If you don't know what you're doing... you can mess around and maybe come up with something passable. Most people will take the time to learn the underlying principles. (There is definitely no shortage of opportunities to learn these within the FIRST community.)

• I know how to hand-draft, but I use CAD.
• I know my algebra, but I use Excel.
• I know my DiffEqs, but whenever I can use a numerical approximation that works, I will.
• I am a lazy, productive engineer.

Just another opinion, good debate happening here.
-JV

[Greg Perkins]

Well here it is...a V.1.1 of my calculator..

Note whereever it says "please choose gears/sprockets from the gear/sprockets tab" dont look for that tab because i havent gotten that far yet..
Please note that this is still very much unfinished i still have to add all the torque , coeffiecnient and torque curves. so i just would like for people to comment and see what you like about it and what needs to change

[John Wanninger]

Originally Posted by JVN

"...To take an extreme analogy: "I have a nailgun but I don't use it for building my house, because I want to make sure I'm really good at hammering."

If you have a true understanding of these numbers, a design spreadsheet is an incredibly helpful resource. If you don't know what you're doing... you can mess around and maybe come up with something passable. Most people will take the time to learn the underlying principles...-JV

Now John, not only can you hammer, but you could probably design and build a (better) nail gun as well. As for others, I would probably make sure they could hammer before I would hand them a nail gun. (And if they didn't, but wanted to learn, I'd be happy to teach them).

I agree with you whole-heartedly that spreadsheets are WONDERFUL for what-ifs and iterations. That said, I don't think I would trust the answer of anyone using them if they did not know the calculations behind them, unless they could validate their work by some other means.

Regardless of how one comes up with a solution, one should ALWAYS ask "Does this answer make sense?", and then verify it. I always like to double-check my calculations using another tack (i.e. virtual work, solving it backwards, etc.) , just to see if it does. Sometimes it's a quick run through in my head, other times it involves using a spreadsheet as a "printing calculator", to verify my numbers and have an additional audit trail for assurance.

[Greg Perkins]

So any pros/cons to the calculator? anybody see any blatently wrong answers? or glitches?

[Madison]

There's a link on the spreadsheet for a "Read Me" file, but no such thing was included -- so it's hard to know if I am using the spreadsheet correctly. That said, it seem a bit simplistic in its representation of drivetrain design and offers little flexibility for varied designs.

As best as I can tell, it does not allow for the user to independently change multiple reduction stages within a gearbox. It also lacks any information at all about torque output. It's good to know how fast your robot is going to move, though it's more important to know if it'll move at all.

If this is simply an early iteration of the design, it may be wiser to wait until there's considerably more functionality before asking us to validate it.