Thoughts on this Battery Mount?

[DunaganM]

Hey CD! In anticipation of tall robots, here’s a bottom loaded battery mount for fun and center of gravity reasons. Onshape Link: Onshape

The original goal with this design was to create a solution for putting the battery as low and out of the way as possible in the robot, while still being easy to change (sub 30sec).

The basic design is as follows.

In the ready position, 6 standoffs and the side walls of the polycarbonate tray fully constrain the battery.

To change a battery the latch rotates and disengages from the round standoff allowing the tray to rotate until it hard stops against the frame tube. Once in this position, the connector can be accessed and unplugged. The battery can then slide over the lip of the tray and out of the robot, from between the tray and the belly pan(not pictured). The process is reversed to load a fresh battery.

Another cool feature of this mount is the way it ensures the connector stays plugged in throughout the match. After a few unfortunate mishaps in elims, our team usually zipties the connectors together for every elims match. In this design, as the battery is latched into position, the face of the battery slightly compresses the 6 AWG wire running from the lugs to the connector, against the two top standoffs. This creates a strain relief at the connector by effectively shortening the length of loose wire in this area so that the connector cannot be unplugged without first overcoming the small stiff section of wire behind it.

Lastly, special consideration was taken to mitigate the possibility of the latch coming loose during a match resulting in a dead robot. When the assembly is in the latched position, due to the geometry of the pivot points, the weight of the battery actually acts to force the latch further into the latched position. Furthermore, any forces pushing up from underneath the robot will do the same.

If this design gets built there will likely need to be multiple revisions to the latch geometry at the latch standoff interface to get the right amount of actuation resistance over the bump, since I just took a WAG at the right geometry.

Thoughts and feedback would be appreciated, thanks!

[Nick.kremer]

While this is pretty cool, I guess my biggest question is why? This seems really complex for something that could be achieved with a bellypan, a plate, and some standoffs.

Like this, but with the battery flat:

[DunaganM]

A few reasons:

1. This idea popped into my head and I just had to finish it.
2. Maybe if we design ourselves into a corner we can pull this out of our library instead of putting the battery in the middle of the robot where it’s a pain to change from the top like we did last year.
3. It puts the CG just a little lower
4. (This is probably the biggest reason) the bottom of the robot gives you a lot of freedom in where you put the battery which allows you to better balance out the weight of other mechanisms.

[cadandcookies]

Honestly, looks super cool.

I think the biggest thing that strikes me is that if you’re doing this from the bottom, and don’t have access to the top, plugging in and unplugging the battery strikes me as particularly painful.

I’d definitely recommend playing around with the latch geometry using a 3D printer or laser cutter before the full fabrication. Linkages and latches like that can be a bit finicky. Worst case you add something to hold the hook in position I think.

[BrendanManning]

Just a general question, but how would you access the bottom of the robot to change the battery?

[DunaganM]

[BrendanManning]

I think that’s a red card

[DunaganM]

Access to the connector may be a pain. I’m hoping it won’t be but we’ll have to test it. There is about a half inch to get you fingers around in and get a grip, which isn’t much. After measuring our battery cable lengths though they may be enough room to partially remove the battery off the back corner and let it drop a bit which would make it easier.
Just gonna have to build it and find out.

[DunaganM]

How about if it’s self inflicted?

[BrendanManning]

That decal/paper is great!
The real question is if you are your own opponent in this situation. Only tipping an opponent robot is a penalty.

[DunaganM]

Certainly! The decal served a dual purpose that year with our climb.

[Jonathan_L]

@DunaganM

[Pushing a van over]

Awesome answer! LOL!

–

I would revisit having the wires go between the battery and the battery holder (particularly conductive material). We had a sensor wire rub against our frame or something until the insulation wore off; caused the sensor to not work and the frame to become become electrically connected I think. No one wants a firework show right next to your battery!

Also keep in mind what happens to the latch if when your robot gets turned over on the field intentionally or not. Or when defense causes negative Gs to your robot.

[TFM110]

Realistically, you will be changing the battery in the pits/on your robot cart, so tipping the robot to change the battery isn’t a red card since its not during a match

[BrendanManning]

Yeah, just making a joke. Sorry for any confusion.

[TFM110]

ah, no worries lol

[troy_dietz]

Underside battery mounts are fairly common, I could list off at least a dozen that I remember seeing in-person. The vast majority of those used a bottom plate that attached with screws to rivnuts or similar on the bellypan. That also usually involved a student using an impact driver and lifting the battery above their head while under the robot on the cart, or a multi-student process involving tilting the robot on its side.
Underside batteries are a solid meh for those reasons, but assuming you’re sticking with that, this doesn’t seem like an inherently bad implementation.

No offense, but this design initially strikes me as something overengineered that could be used to dupe judges into giving out a design award. I’ve done the same thing a few times. I’d recommend figuring out an easy way to show it off if it does end up on a robot. Presented with the same hinge concept, I think a lot of people would go for a piano hinged door with a fastener securing it on the opposite side.

As Nick notes, you’ll want a cotter pin or something for a negative-g event, and the battery cable is more annoying that I think you take it to be. IRL grabbing a battery and connector is important to figure out how annoying cable routing and unplugging it can be.

Ideally, the only thing in the yellow box is the battery connector and cables. That box also extends up and down like an inch. Some people have different shaped hands and any restriction will get old fast.

[archer_tiki]

Replying to this I would say this is a really neat prototype and well thought out idea. I don’t know if it’s entirely practical. Typically in First history batteries are hard enough with a whole bunch of subsystems in the way and I don’t think it would make it any easier if to remove the battery you had to pull in up well angled down. Also would it be raised from the belly-pan since the latch mechanism is dropping down. Additionally I would have to look at specific dimensions but your power wires running under the top standoffs might need a little more clearance especially to attach them to the connector. Still it’s a pretty sweet locking mount.

[philso]

You should mock this up in something simulating the size and weight of a real robot and make sure it is possible to get the battery in and out with too much trouble and that you have sufficient access to unplug the connector. You may also want to build one, make sure it is working properly, then intentionally bend it in different ways to see if it still works before using it in a competition robot.

It looks like the two pocketed pieces that are stationary are only attached at one end. If they twist relative to each other, the latch may only work on one side.

[DunaganM]

So, after a season of use, 5 comps including TRI, and many practice hours, I figured I should write a follow-up post to share our experiences with this design so far.

Before the season started, we built a V1 of the mount and ran it through an extensive testing program in the hopes of finding any problems before we dropped the battery on the field in finals. Initial impressions were good as far as ease of use was concerned, though I ended up using some gaff tape to increase the compression on the latch, this change was later reincorporated into the design. In order to test the mount, I built the “bonking cage”, a collection of 2x4s screwed together in a manner that allowed us to apply very large shock loads to the system from any direction by tipping the contraption, or as we later started doing, picking up the whole mess and chucking it as far as we could manage. Anecdotally I would estimate that the bonk applied to the system in each load cycle was somewhere in the ballpark of a 2022 robot falling off the hanger. After many load cycles stuff finally started breaking.

https://drive.google.com/drive/folders/1IasreC4c_0N2sZtIpv8AZ__ixfVObXNO?usp=drive_link

We identified a failure case that caused the battery mount to open under impact. Upon closer inspection, this appeared to be due to two causes. First, the main picot posts had deformed over time and lessened the compression in the system. Second, a negative g impact that pushed the battery up in the mount would allow the latch to move upwards a small amount, for a brief period. The geometry of the latch then forms a ramp that pushes the latch open. With these problems identified, we designed a V2 of the mount that upped the main pivots from 3/8in hex to 1/2in hex and changed the geometry of the latch to remove the ramping issue. Furthermore, we beefed up certain hole locations where we had begun to see cracking appear.

At this point, the project was shelved until the season began, at which point we immediately decided on a design with a dumb battery placement inherent in its function. Out came the bottom loading mount design which we ended up integrating into the shoulder joint support.

During the season we made multiple revisions to the latching nub trying to find a happy balance between secure holding, and easy unlatching. In the end, we ended up erring on the side of easy unlatching and added a set of spring detent pins as a secondary safeguard, though they probably weren’t necessary, this added about 5 seconds to the process of changing a battery. During the season we only had 2 structural failures on the mount, one during practice and one during TRI. Both seemed to be caused by fatigue, and neither resulted in a dropped battery. In the future, it would probably be wise to replace the polycarbonate parts of the mount between events.

As far as ease of use goes, we are quite happy battery changes take about 45 seconds, which is more than the target but still quite acceptable. Though ideally changing the battery on the cart is a two-person process, due to the design of our arm this year we needed a third person to manage it during changeouts. While we were initially concerned about access to the connector, it ended up being very easy to use and when the mount is latched shut it is neigh impossible to unplug the connector. Another unanticipated benefit of the design is the great viewing angle it gives on the connector which makes it very easy to identify a connector that is only partially plugged in.

Overall we are very happy with the design for the instances in which it makes sense to use it. While designs like the one Nick mentioned at the top of the thread are superior when the space is available to use them, this mount filled a niche that was very useful to us this season.

-Dunagan