RS-395 safe loading

I’m trying to figure out more ways to cut weight in my swerve drive, and am considering using an RS-390 or RS-395 for the module turning motor in place of the usual RS-550 (1/4lb savings).
According to JVN calc, I’m gearing it down 74:1 and that results in a 210RPM free speed. It draws 4.1A under typical load, that is, all 4 wheels on the ground on a 150lb robot. When sitting on two wheels it’s loaded at 7.6A.

Is this a safe current to run through an RS-395 through a match for a swerve drive, or should I stick with a 500-series motor?

Not worth it, stick with the heavier motor.

I haven’t used the RS-395, so I’m not familiar with its durability, but applying a bit of common sense:

Under typical steering, you’re drawing about 1.5x the current that the motor draws at maximum efficiency. That should be golden, especially for match durations. If almost all of your drive time is in a straight line or gentle curve, probably all day without a break.

Under “peak” conditions you’re essentially running the motor at its maximum power point. Unless you’re planning to recreate Tumbleweed for a near-circular race game*, you should be good. If regular peak conditions are anticipated, you’re on the bubble of common sense and will need some definite experience or testing to determine if the motor can run at this current for a full match.

So your savings overall would be 1lb?
It really doesn’t sound like it would be worth it.

1717 used 395s (I think, could be 390s) with undeniable success. I don’t know about the gearing or issues they had with them but they definitely work.
http://wiki.team1640.com/index.php?title=DEWBOT_VIII_Drive-train_Team_Page#FRC_Team_1717.27s_awesome_pivots

1640 used a 395 in 2012 and upgraded to a 540 in 2013 (both using a Banebots 133:1 planetary) according to their BOM. I don’t know the reason for the change.

I wouldn’t bother.

edit: 550s were used instead on 1717 swerve.

Which year did 1717 use 395s?

2012 at least. The linked pic library shows what I think is a 395 (maybe the 390- I can’t distinguish them visually). I vaguely remember them saying it was a 395 on their 2013 and 2014 bots but I honestly can’t remember.

edit: am mistaken-550s were used instead.

Well it’s basically a free pound, so why not?
I think I’ll design for both for now, and try them out as needed.

Steering motors can be loaded in lots of unexpected ways. Trying to turn when your wheel is up against an obstacle or imperfection in the floor…when dynamics shift most of your normal force onto one or two wheels…on slopes…during intermittent bouncing contact with the floor…

Also, the 300-series motors have not been used nearly as often in FRC as the 500- and 700- series motors. There is less (not none, but less) collective community experience in knowing their limits and endurance under FRC loads. Spec sheets seldom tell the whole story (especially when you are talking about a fairly generic trade sized motor from China).

Given the severity of losing a steering motor, my personal inclination would be to accept the 1 lb penalty and look for opportunities to save weight elsewhere.

EDIT: Also, out of curiosity, what method did you use to calculate the load on each motor in a steering application?

Yes, 1640 used RS395 motors 1 year. Decision was based on motor availability. The 395’s performed OK that year but we did burn up a few. That summer active cooling for them was explored. Heat was a problem. Since then RS540’s have been used with great success. We also added a quick change feature to the Banesbot trans and motor mount. the motor, pinion and base plate can be removed and replaced without total transmission removal. 2014 was the most abusive year for our swerve. The 540’s held up but were routinely hot after matches. The 300’s would not have worked in that game. Do not under estimate the amount of power needed for future games. RS540’s are probably the minimum. 2015 put absolutely no stress on our swerves. 300 would have worked.

I used the JVN calc “rotary mechanism” tab. To calculate load I used
wheel width * 1/4 bot weight * CoF. In my case that was 0.7537.51.3, or about 36lbs. Then I added a ~1.5x factor of safety and decided on 50lbs as a good estimate for loading.

Based on what I’m seeing here I guess I’ll just suck it up and use a 500. Kind of unfortunate, but I’ll take what I can get.

EDIT: Whoops, that calculation is wrong. The load on the wheel is equal to 150lbs/4 wheels * 1.3 CoF, or 48lbs. The lever arm is 1/2 the wheel width, or 3/8".

We ran our swerve steering on 550s through 49:1 Versaplanetaries and then to an 84:24 reduction for a total reduction of 171.5:1 and they got REALLY burning hot. It was kind of surprising and somewhat concerning. It may be due in part to inefficiencies in the Versaplanetaries. I think the frequent direction reversals and PID position holding are also a contributing factor.

1/4 bot weight per wheel is a best case scenario. Expect to routinely get pushed onto one or two wheels during pushing matches and having more weight on certain wheels when accelerating. 1/2 or 3/4 * bot weight seems like a far safer number.

Drivetrain is a terrible place to try to skimp out on. I want my drive to be as reliable as possible. I’d be especially concerned with the intensity of pushing matches and the small breaks between matches during eliminations, where reliability is the aspect that differentiates between teams the most.

In my mind, reliability trumps weight in this situation given that 540s are already pushing limits because they aren’t very well suited to the needs of your steering motor.

Scott I think your off here. 1717 has for sure used 550s over the the past few years. For 2012 in that wiki it says as well:

2012 - Unicorn Drive, CIM+Custom Shifters & 550s
In 2012, our drivetrain consisted of four independent wheel modules. Each wheel module’s drive was powered by one CIM motor with a two-speed custom gearbox. Our wheel modules’ turning was powered by a single Banebot RS-550 with a custom transmission. All of our gears were made from aluminum and they were cut in house.

whoops, I feel like an idiot. previous posts edited.

David,

Thats odd, we used a BAG with a 75:1 total reduction (way less than you guys). Ours remained cool after 30 minutes of driving. We really like using the BAG’s on steering this year, especially since they can handle the heat.

The 2012 Neutrino CIM in wheel swerve was steered using banebot 395 motors, didn’t have any problems. I don’t remember the final gear ratio off the top of my head.

-Aren

The JVN calculator may not be good for steering calculations on co-axial steering. The use of miter gears cause an additional load on the steering motor. If the top pulley turns clockwise, there is a counter clock wise torque on the wheel cage that the steering system must overcome. The more load put on the driven wheels, the greater this forces is. During hard accelerations and intense robot to robots interaction this force can be quite large. I have no idea how to set up an equation for it. Our team has never quantified it. We could have used the new current measuring capability of the PDM. We need to look at this current draw because if we have a more normal FRC game next year we will have to worry about total current draw and power management. In 2014 our current logs showed that there were a few matches the Roborio would have entered brown out.

Having never built a swerve I can only relay what information has been given to me by those who have. John Novak (16) remarked that they wished they could use the 395 without it using their allotment of BB motors (this was prior to 2015 motor rules) His reasoning was that they have had issues with starting off in a direction other than where their wheels are currently pointing due to current draw. (4x 550s and 4x Cims starting up at once is a fair bit of current draw) and by switching to smaller motors their current draw would be lower. Based on this being John, I assume he realizes the tradeoffs with that.

My theory on why 16 could get away with it is less based on mechanical construction and more based on software and driving style. A lot of swerves operate as swerves all the time (rapid direction changes) but 16’s swerves are typically driven more like 254 drives their tank drives (long smooth arcs) This is, obviously, less rough on the steering motors than rapidly changing directions all the time. This theory is backed up by the behavior of 91 (16’s rookie team) at Ozark Mountain Brawl in 2014. The drivers had minimal drive practice and tended to rely on frequent direction changes. Their motors were hotter and, in their last elims match, they managed to pop the main breaker on the bot.

Of course, John, JT, of Jeff would know more than I would.

This will be slightly off topic as we didn’t use 395’s for steering, but to rotate our turret. We did this in both 2009 and 2012. We used the same approach both times with great success.

The basic configuration was a 395 driving a planetary transmission removed from a Globe motor. Globe motors were no longer legal for use and we had dozens of them sitting around.

Here is a little pictorial of how we interfaced them.


Notice the small aluminum sleeve between the motor shaft and the sun gear. That was the trickiest part to make. So, as you can imagine, these were quite simple to create.

Globe transmissions are approximately 90:1 reduction. These assemblies never even got warm after running tests and calibrations for hours.