I’m looking to make a device that would power a robot from mains voltage, to power on a robot (while it’s sitting on a cart, wheels in the air of course), so when working on the robot for long periods we wouldn’t need to worry about charging and changing batteries.
When looking through 12V power supplies, I see many with high amperage ratings, like 50A for $35, advertised for charging lithium batteries or running long LED strips. Should I worry about the power supply being able to react to sudden spikes? I don’t care much about the efficiency as I’m not paying the electricity bill, but is there a rating or a spec measuring this behavior? And is there anything else I should look out for when making this?
The first thing I’m going to say is: Why?
As in: Why are you working on the robot while it’s powered on? I can’t think of a single reason to be in and working on the robot while on, when you have access to mains voltage, that can’t be handled by a battery. Safety is a thing around these parts…
Second: Don’t use a battery charger as a power supply. If you MUST use a power supply, use an actual power supply. Battery chargers (especially lithium ones) tend to supply more voltage than the rating–SLA chargers can do 13V easily (at least they can leave the batteries reading 13V); lithium chargers often run closer to 14.5V. A power supply will run at voltage as long as the amperage is high enough. Remember that you’re dealing with a 120A main breaker. (And yes, I’m sure you can get supplies that high.)
Or you can do it like every other team and use your batteries, swapping them out for recharge on a regular basis.
2 Likes
Hi Eric, thanks for the answer. I have a few answers as well
1 - Working on the robot, meaning working on the arm or something like that. I’m not talking about drilling stuff over it… the programmer and mechanics guys take turns, when there’s a bug or something the mechanics team gets a few minutes to do whatever.
2 - Will something of this sort be good for this use case? https://a.aliexpress.com/_om8sAA1
3 - we’re always looking to innovate and improve😉
Not at the cost of major safety concerns, of course, but tossing out an idea just because everyone else is not doing it is not a reason I’d drop an idea for😊
In this case the advantage of a power supply is not wearing out the batteries and battery charger, and manpower of changing them.
1 Like
Not an advantage. The batteries will get worn out by their regular use just as fast. Battery charger, they don’t wear out noticeably.
Power. Off. At least when the mechanics are working. (Obviously the programmers need power.)
At least go to Amazon… I’m not sure I fully trust Alibaba. Thing is, for one power supply plus shipping and any import duties, you’re talking not all that much less than one pair of batteries from AndyMark, and it’s known that directly going to the manufacturer can be cheaper than AMAndyMark.
First thing you should ask: If we’re the only team doing this, is it because we figured out something new? Or is there something we should know with the rules (there isn’t, at least at your own shop, in this case)? Or is it just not worth it? (By the way, doing this at an event? If seen, 90% chance the LRILead Robot Inspector makes you stop.)
There are way to do this but they would either be very dangerous or dangerous and expensive.
I would recommend making a robot on a board (Rio, PDH/PDP, VRM/rev mini power thing, radio, etc…) and plugging your arm/lift/etc… into that for testing, while the rest of the robot is being worked on. If you use Rev controllers you can plug into them with USB-C and control them as you will.
Echoing Eric, I don’t feel that it is worth the time, effort and expense to set up a 12VDC power supply for doing your testing. Depending on what you are testing, the current can be well over 100A, while the mechanism is running. The starting surge current is likely to be several times higher. DC power supplies almost always have self-protection features that will cut off the output within microseconds when the output current exceeds the rated output current by something like 10-20%. It will be difficult to start motor-type loads unless the motor current is a fraction of the power supply current rating. DC power supplies rated for 12VDC and 100’s of Amps tend to be rather large and expensive. You would also have to run thick cables from the power supply to your mechanism. The reality is that an FRC battery should be able to run something like an arm during software testing for several hours since it will really only be running intermittently. It would only run down the battery quickly if you were doing durability testing where it is made to travel back and forth, continuously.
My recollection from this past year is that we were having to switch more often then that. We were also running a bit too weak of a gearing so that may have been part of the problem, but we’d switch when the arm started to have trouble moving.
Is putting a battery into the circuit that will give that extra power possible?
When they’re greasing/tightening belts or whatever (this type of work, of tuning things and such, is something we see a lot in the weeks leading up to the first comp) it’s usually fine…
Possible? Anything is possible.
But it’s not going to be easy.
You MIGHT want to look at acquiring a 12V marine battery that can handle the current. They tend to have a larger capacity. Mount it on the cart for shop work, leave it home for events. Fairly “standard” parade setup for some teams with a bit of robot modification.
Just because nobody’s gotten hurt YET doesn’t mean it’s fine.
3 Likes
I’m guessing that’s expensive
Wdym? Why is doing such activities when the robot that dangerous in your opinion?
Seeing as you have to ask…
You are working on the robot, let’s say you’re tensioning a drive chain. Programmers decide to test auto, don’t give a warning because they have a brain fart, and enable.
Your fingers WERE in the chain. Now they’re on the floor, but your hand isn’t. I hope I don’t need to make myself any clearer?
4 Likes
Imo demanding a turned off robot before allowing working on it is egregious.
In some cases, that may be true but you should NEVER work on belts or chains of a powered up system. As EricH said, that’s how you lose a finger.
1 Like
A DC power supply will not respond the same way as a competition battery will – it will try to hold a constant output voltage with its own internal feedback loop (minus a bit of resistance in the wiring). The competition-legal batteries have significantly-variable output voltage under load, due to their electrochemical nature (and ridiculous current draw in FRC applications).
If you tune mechanisms or control loops while running off of a DC power supply with a constant output voltage, they won’t respond the same way when powered with a battery. This is especially true if you are trying to stay clear of the power supply’s overcurrent limits, whatever type they may be.
Furthermore, the robot can send some electrical energy back to the battery when mechanisms decelerate – regenerative braking. The permanent-magnet motors (both brushless and non-) used in FRC are also pretty potent generators. When you use one of them to decelerate a mechanism, some of the kinetic energy gets converted back to electrical energy and a current is forced into the battery. Most DC power supplies aren’t reversible – they can’t convert the energy back to AC and send it out the power cord. Some can handle a little bit of dissipation/“braking”, but, in all likelihood, you’re going to fry it, and possibly some other devices on the robot – all of the bulk capacitors in the 12V system will experience a voltage rise, because there’s nowhere else for the current to go.
6 Likes
Having to power cycle the robot and wait for it to boot again is an extremely small (and absolutely necessary) price to pay to ensure nobody, let alone a student under your supervision, is seriously injured…
I can’t believe we’re actually having this conversation (dare I say debate) right now…
4 Likes
Thanks for the info.
Do you think a cheap 10A power supply would be suitable for powering an idle robot or a motorless circuit used for testing out vision?
You are entitled to your opinion.
The rest of us are entitled to say “we told you so” (or some equivalent) when your opinion gets you or others hurt.
I’m not going to say that you have to follow our collective opinions. With that said, if you continue to say things like “it can’t be that bad”, around here that has a tendency to turn into a piling-on.
2 Likes
Absolutely. It’s pretty common to have an independent test system mounted on a board for this sort of development – roboRIO, vision co-processor, etc… If you’re going to use an old robot, consider disconnecting everything you don’t need.
It’s still important to follow good electrical practices (securing wires, insulation, overcurrent protection in the form of breakers or fuses) on a test board like this. Even if you don’t have mechanisms, it’s still possible to be injured, or to zorch an expensive component like a roboRIO or a LimeLight.
1 Like
You can always add precautions. You can wear a helmet while driving a car. You can set a 4 meter radius of safety while enabling the robot. You can demand a mentor inspection of the code before deploying it. All of these are valuable precautions, but don’t expect everyone to take them. And sure, you can “I told you so” all day, I don’t see how that helps anyone.