[DFTF] Motors... ...Drive Motors...

[Joe Johnson]

This is part of a series of posts called Drinking From The Firehose on getting Dr Joe back up to speed on All Things FIRST.

Today's topic:
Motors... ...Drive Motors...

Okay, so I am reading the rules minding my own business and BAM! I am hit with more Watts that any robot chassis has a right to expect.

In my day, there were two drills in the kit with about 200W of peak power apiece. After that, the motor power fell like a rock to the sub 30W range (window and seat adjusters motors mostly the Globes would come later...).

There was literally no reasonable option but to drive with the drill motors.

Fast forward to last year.

• 4 "Coke Cans" from CIM @ 340W each
• 2 550's from FP @ 290W each
• 4 775's from BB @ 270W each (the 18V version scaled to 12V)

That's a lot of Watts.


Sooo... ...I suppose that the most common drive system is still 2 CIM's per side (4 CIM total). Because they are awesome motors in so many ways, not the least of which is that their size allows them to get rid of a lot of heat fast (and we all know that we pour a lot of heat into them -- FYI a curling iron is typically about 20W!)


But it seems that there are enough high power motors in the kit that other options may make sense as well.


Are there folks that used the FP's for driving (either as a 3rd motor per side or to use a CIM somewhere else or to save weight or because your team just felt like it)? How'd that work out?



On thing that worries me about the FP is that they have a circuit breakers in them now a days (at least they did 2 years ago when I was helping a team in MD. That fuse caused my winch to wig out -- not a happy event when you are trying to hang). Did people who drove with the FP's have breakers trip (not the Power Distribution Breaker but the one in the motor itself).



Same questions about the 18V 775's from BB.

One thing I always say is if you don't move, you don't play after lunch on Saturday. To that end, I am very cautious about drive train failures. So... if the FP and BB motors are not up to the task, I want to know about it before we design our robot drive train around them.



Tell me what traps to avoid.


On a closely related topic, let me know if the CIMple Box trannies from AndyMark are as awesome as they look on paper (I love the name, I love the compactness, I love that they do exactly what they need to and no more -- looks pretty sweet to me... ...but if it turned out to be a trap it won't be the first time FIRST has shipped a rookie killer in every kit ;-)


The only thing that I get a little worried about is that output shaft and how it's supported. Can it really stand up to a 2 CIM motor stall with a #35 12T sprocket hanging out their on the end of the shaft? Andy and Mark are clever guys so I suppose it can, but I would like some testimonials if you have "FIRST hand" experience.



Before I go, I want to caution folks to be nice. Drive systems are a touchy subject. Let's not start unnecessary battles (for example the shifters vs non shifters - we can discuss it yes, but let's just agree that we are never going to reach total agreement. In any case, let's keep the flame wars to a minimum).


Joe J.

[EricH]

4 CIMs in AndyMark gearboxs will more than suffice for most drives. The only thing you have to worry about is whether you want to shift or not. There are very, very few complaints about AM products in general--if there are, usually it's because a ton of teams didn't get a part they were supposed to (and the result there is that they get that part in their next order or in their mail, their choice).

The FPs have changed--last year there was only one.

There were reported problems with the BaneBots 775s case shorting last year. I don't know the entire details, but it involved debris inside the motor.

I've never used any motor other than a CIM on my team's drivetrains. CIMs are so popular among teams because they come in the kit, and fit with the kit gearboxes. Aside from that, they are also very nice motors with nice power and can fit into a lot of other gearboxes if needed.

Also, the drive train is usually low to the ground. Have you seen any CIMs high up in the air before? They may be nice, but they are a tad bit heavy (perfect for drivetrains, especially with all the support down there), and they are pretty decently sized, compared to other motors in the kit.

Hope that answered your question!

Keep posting these kinds of threads! I love them!

[sanddrag]

In 2008 (Overdrive, oval laps), 968 originally used two CIMs and one Fisher price in each side of the drivetrain. It was also a two speed. It was geared for about 18 feet per second in high gear. When we later reconfigured things and used the FPs elsewhere, we there was a noticable top speed drop and lag in acceleration. Also, we began to burn out the CIMs without the FP there helping. Unless the game is something like oval laps again or offers some way to get a ton of normal force on the vehicle (like 2002), I don't see a whole lot of benefit to a 6-motor drive. I suppose it could be beneficial in a single-speed gearbox that is geared properly. But usually the Fisher price motor and BaneBots motors are prime candidates to use elsewhere.

[Billfred]

Originally Posted by Joe Johnson

This is part of a series of posts called Drinking From The Firehose on getting Dr Joe back up to speed on All Things FIRST.

Today's topic:
Motors... ...Drive Motors...

Okay, so I am reading the rules minding my own business and BAM! I am hit with more Watts that any robot chassis has a right to expect.

In my day, there were two drills in the kit with about 200W of peak power apiece. After that, the motor power fell like a rock to the sub 30W range (window and seat adjusters motors mostly the Globes would come later...).

There was literally no reasonable option but to drive with the drill motors.

Fast forward to last year.

• 4 "Coke Cans" from CIM @ 340W each
• 2 550's from FP @ 290W each
• 4 775's from BB @ 270W each (the 18V version scaled to 12V)

That's a lot of Watts.


Sooo... ...I suppose that the most common drive system is still 2 CIM's per side (4 CIM total). Because they are awesome motors in so many ways, not the least of which is that their size allows them to get rid of a lot of heat fast (and we all know that we pour a lot of heat into them -- FYI a curling iron is typically about 20W!)


But it seems that there are enough high power motors in the kit that other options may make sense as well.


Are there folks that used the FP's for driving (either as a 3rd motor per side or to use a CIM somewhere else or to save weight or because your team just felt like it)? How'd that work out?



On thing that worries me about the FP is that they have a circuit breakers in them now a days (at least they did 2 years ago when I was helping a team in MD. That fuse caused my winch to wig out -- not a happy event when you are trying to hang). Did people who drove with the FP's have breakers trip (not the Power Distribution Breaker but the one in the motor itself).



Same questions about the 18V 775's from BB.

One thing I always say is if you don't move, you don't play after lunch on Saturday. To that end, I am very cautious about drive train failures. So... if the FP and BB motors are not up to the task, I want to know about it before we design our robot drive train around them.



Tell me what traps to avoid.


On a closely related topic, let me know if the CIMple Box trannies from AndyMark are as awesome as they look on paper (I love the name, I love the compactness, I love that they do exactly what they need to and no more -- looks pretty sweet to me... ...but if it turned out to be a trap it won't be the first time FIRST has shipped a rookie killer in every kit ;-)


The only thing that I get a little worried about is that output shaft and how it's supported. Can it really stand up to a 2 CIM motor stall with a #35 12T sprocket hanging out their on the end of the shaft? Andy and Mark are clever guys so I suppose it can, but I would like some testimonials if you have "FIRST hand" experience.



Before I go, I want to caution folks to be nice. Drive systems are a touchy subject. Let's not start unnecessary battles (for example the shifters vs non shifters - we can discuss it yes, but let's just agree that we are never going to reach total agreement. In any case, let's keep the flame wars to a minimum).


Joe J.

From our experiences:

-We used the Fisher-Price motors through an AM Planetary on drive in 2010 in a six-motor two-speed setup. It was probably overkill for the game, but our design didn't call for their use anywhere else and they caused us no problems across five robot-events (1398 and 2815 collaborated; both attended two regionals and the former attended Championship).

-Call it lingering memories of 2007's kit gearboxes, but I have been leery of using any of BaneBots' products in mission-critical applications. Indeed, a case short on one of those 775s sidelined 781 in the worst possible place: Einstein finals. We used their smaller motors (540s? 550s? Been so long...) on our roller claw without a problem...but that's something where we could've come up with a Plan B if they were bad. Doing that on drive is a little tougher.

-For all our faults this year, the CIMple Boxes never gave us a lick of trouble. I wasn't sure about the math without that second stage of reduction in the gearbox, but it works out fine once you crunch the numbers (cue Mark Leon...)

-A four-CIM kitbot built properly with the tweaks obvious to you (grippier wheels*, 6WD, sprocket-based gearing choices to match the speed of the game) will usually be an 80th-to-90th-percentile performer on drivetrain at most events. For us, that's enough that we'll stop there and start focusing on manipulator; for one of the pioneers of swerve, well...

*And for that, I really can't beat the bargain of skinning the gray rubber off the kit wheels and riveting your own tread onto them. Especially if you're on a budget: http://twitpic.com/4afp59

[JamesCH95]

I would vote CIMs in the drivetrain every day of the week simply because they have far and away the most OTS drivetrain oriented gearboxes of any of the motors.

Regarding AndyMark Products, I have used many and never had an issue with anything. I would not hesitate to buy anything from their catalog.

SuperNerd256 brings up a good point with weight, if the motor is going on an actuator somewhere higher up on the robot (the second DOF of an arm for example) a 13oz RS775-18 would make life a lot easier than a 45oz CIM with nearly the same power.

Last year I became an instant fanboi of the RS775-18 motor. It's designed to handle 18v, so the 12v we run it at is a cake walk. It has plenty of power. It has a bigger can (i.e. more thermal mass) and better cooling than any 550 sized motor I've seen. We used two (but could have used one) as the main shoulder joint actuator on our robot, stalling them to hold arm position, and never had a single problem.

The trap I see with any of the BaneBot's motors is that the BB transmissions can be had in ratios such that the attached motor will over-load the transmission if not properly controlled.

For example, the BB P60 transmissions are rated at 35 lb-ft of torque. The RS775-18 has a stall torque of 166oz-in = 0.86lb-ft. If a 256:1 transmission is used, it would be possible to push 220lb-ft of torque through the transmission. Clearly not a good idea. In my opinion the most fool-proof way to get around this is to not use any transmission of a ratio greater than 64:1 and use sprockets or gears for further reduction. The transmission could still see 55lb-ft at stall with a 64:1 ratio, but a little bit of current limiting in the control code should prevent damage.

[Brandon Holley]

I consider 4 CIMs in the DT a minimum as far as motor power is concerned. I would start there as a baseline.

In 2008 we built our drive system to use 6 motors in total, 4 CIMs and 2 FPs. We did not shift, but we were still able to move pretty quickly around the track. I've seen a gradual increase in the number of 6 motor drives over the past couple of years, but its still rarely used.

I cannot speak to using 775s in a drive system, but I will tell you that they caused NOTHING but problems for us in our elevator this past season. I think we went through 8 or 9 motors and each and every one of them was case shorted. This in turn caused widespread electrical issues such as causing servos to twitch (launch minibot prematurely) and smoking Jaguars. We will avoid them as best we can moving forward.

The FP has changed several times over the years, and I think the newest iterations are much more robust than their smoking older siblings. I wouldn't hesitate using a FP in most applications now.

-Brando

[Jim Zondag]

Joe,
In the testing that Isaac and I have done, both in simuation and on real chassis, we have found that there is a "scrappiness threshold' at about 1kW for 150lb GVW FIRST Robots. We did most of this work on a standard 6x6 dropcenter configuration chassis. Obviously the exact gear ratios and wheel placement play into this, but there is a big noticeable difference in the dynamic performance of the machine when you drop below about 1kW, simply due to the Power to Weight ratio. Hence most teams will use a quad CIM drive which keeps you comfortably above this zone, even with a pretty inefficient execution. Gone are the days of 2WD caster machines we all built in the 90s.

The CIMs have been great additions to the FIRST KOP, and the selection of off the shelf parts available from AndyMark, VexPro and Banebots make many of the difficult, precision things that only the elite teams could once do available to all teams.

[JamesCH95]

Originally Posted by Brandon Holley

I cannot speak to using 775s in a drive system, but I will tell you that they caused NOTHING but problems for us in our elevator this past season. I think we went through 8 or 9 motors and each and every one of them was case shorted. This in turn caused widespread electrical issues such as causing servos to twitch (launch minibot prematurely) and smoking Jaguars. We will avoid them as best we can moving forward.

How were the motors setup? Transmissions, loads, control, etc. Like I said my team had a 100% positive experience with them, so I'm curious as to why your experience was 100% negative.

[Brandon Holley]

Originally Posted by JamesCH95

How were the motors setup? Transmissions, loads, control, etc. Like I said my team had a 100% positive experience with them, so I'm curious as to why your experience was 100% negative.

We used 2 to run our elevator winch. The winch was close to the bottom of the robot, and powered a ~12" long aluminum drum that would wrap up the sailing line in our elevator.

The winch itself had each 775 powered through a "CIM-U-LATOR" and then further reduction inside a gearbox where the motors were paired. I'm not sure of the overall ratio off the top of my head, I can report back later. We did quite a bit of math to make sure we would be geared for an appropriate speed. Additionally about 75% of the weight of the elevator (and claw system) was offset by spring force. To say this thing was overpowered would be an understatement.

When we had functioning 775s, the elevator ran up and down with few issues. Unfortunately we never had one 775 that functioned flawlessly. Even if we confirmed there was no case short when installed, the motors tended to develop them over time. It wouldn't take long either. We checked the case short after each and every match to basically gauge how long we could continue running that set of 775s before needing a different pair. After each match we would watch the resistance drop until finally we would have issues with the minibot deployment which signified an end of life for the motor.

We tried to isolate the motor from the robot by placing dielectric sheet between the face of the motor and the gearbox plate. We also tried wrapping screws in teflon and then installing them. The short went through the motor pinion, into the CIMULATOR gearbox, through the CIMULATOR output shaft into our custom secondary gearbox, through the output shaft, through the bearings pressed in the plate that the shaft ran through, into the gearbox plate and the gearbox plate mounting screws, and finally into the frame of the robot. It was a nasty problem to have.

"Zapping" the motors was a common practice teams used to fix this issue, and we did so as well. On some motors it would definitely enhance their lifetime, on others it didn't do much of anything. It was almost kind of sad how good we got at pulling the motor off the robot, "zapping" it, and then replacing it. We often did this process, or a full replacement of the motors in-between matches.

If Banebots can show us that this isn't going to be a reoccurring issue, then I would say we'd be open to using the motors again. Like I said, when they worked, they worked perfectly fine. However, it was my understanding that Banebots essentially said the issue was minor and recommended the "zapping" process.

I will add that we used a pretty well tuned PID loop to control our elevator. We did a pretty good job about handling stall though. There were several timeouts built in that would cut power to the motors after a couple seconds of stall and ensure we weren't just driving around with the motors humming.

Just our experience with them. I know others who have had success with them, and I know others who dealt with the same issues we did. It seemed to be luck of the draw this past year.

-Brando

[Jon Stratis]

Originally Posted by Brandon Holley

The FP has changed several times over the years, and I think the newest iterations are much more robust than their smoking older siblings. I wouldn't hesitate using a FP in most applications now.

Interesting... our experience this past year with them is exactly opposite. I can't even count the number of FP's we burned up with our elevator this year. In the off season here, we stopped buying the "new" ones that were legal this year, and instead threw on an old one from 2008, as it was just sitting there. Haven't had a single problem with it since, and that includes a full day of playing matches at an off season event (only 10 teams were there too, so we were on the field almost every other match). This years FP's would last maybe 2 rounds in the eliminations at North Star this past year.

[Brandon Holley]

Originally Posted by eagle33199

Interesting... our experience this past year with them is exactly opposite. I can't even count the number of FP's we burned up with our elevator this year. In the off season here, we stopped buying the "new" ones that were legal this year, and instead threw on an old one from 2008, as it was just sitting there. Haven't had a single problem with it since, and that includes a full day of playing matches at an off season event (only 10 teams were there too, so we were on the field almost every other match). This years FP's would last maybe 2 rounds in the eliminations at North Star this past year.

That is interesting.

In all honesty I don't know a lot about the actual difference between each years FP motor. I do know that it has tended to be different almost every year for the past few years.

If I recall correctly we were given only 1 of the FPs this year, and it was a significant step up in power.

We placed it in our roller claw and absolutely no problems with all year. In the past we've used them in various systems (elevators, hopper systems, drive trains), and as long as you didn't stall them, they worked fantastically.

-Brando

[artdutra04]

Originally Posted by Brandon Holley

If Banebots can show us that this isn't going to be a reoccurring issue, then I would say we'd be open to using the motors again. Like I said, when they worked, they worked perfectly fine. However, it was my understanding that Banebots essentially said the issue was minor and recommended the "zapping" process.

This is a perfect scenario that would benefit from loosened motor rules. The Banebot RS775 motor is an exact clone (specification wise) to the RS-775WC-8514 motor produced by Mabuchi, with the only difference being the cost/quality of the motor construction.

[Ether]

The Banebot RS775 motor is an exact clone (specification wise) to the RS-775WC-8514 motor produced by Mabuchi, with the only difference being the cost/quality of the motor construction.

Does anyone know who is the manufacturer of the Banebot RS775 ?

[JamesCH95]

Originally Posted by Ether

Does anyone know who is the manufacturer of the Banebot RS775 ?

I'd bet the lowest Chinese bidder, not a real brand. I wouldn't be surprised if the variations in performance were linked back to different suppliers or different lots of motors.