[ 3 / a / adv / an / asp / cgl / ck / co / diy / fa / g / gd / int / jp / k / lit / m / mlp / mu / n / o / out / p / po / sci / sp / tg / toy / trv / tv / v / vg / vp / vr / w / wsg / x]

/diy/ - Do-It-Yourself - DIY Electrified floor

<< back to board
[Delete this thread]

File: Bumper car floor.jpg-(59 KB, 733x550)
DIY Electrified floor
DIY Electrified floor Anonymous 05/17/14(Sat)12:49 UTC+1 No.639446 Report

>tldr: What's the best cheap, corrosion resistant, conductive material.

Hi /diy/
I'm planning on starting a robot wars of sorts with a few friends from uni (we're all mechatronics students), and I'm deciding what to use for power. Being uni students, we need the most cost effective method possible. And we're in Australia, so most things are more expensive than they should be.
We want to avoid batteries, because batteries suck.

What I've come across is the system newer bumper cars use - strips of metal, alternately positive and negative with insulators in between, and with contacts on the underside of the cars wide enough to always be in contact with both a positive and negative strip, with rectifiers (Like in the pic).
What I'm wondering is, what cost-effective material would be best to use for the flooring and contacts?
For the contacts I'm planning on using gold or silver plated copper, as diy plating is pretty cheap and easy and I won't need much copper - but the floor is going to be a pretty big area, and sheet metal is expensive.
From what I can tell, current systems use stainless steel floors with graphite for lubricant. Copper and aluminium seem to be a no go, due to tarnishing/oxidisation (and I can't afford to plate all that area) - brass tarnishes slower than copper, but still does.
Anyway, the best I can think of is stainless steel foil/tape, which I can find for $10/metre at 100mm wide and 0.01mm thick.

While I'm here, I'm just planning on using a bunch of shitty computer psu's for the power supply and running 12v - though this will likely result in somewhere around 170 amps running into the whole arena, so any advice in this regards would be lovely.
Sorry for the wall of text, it's late and I've been drinking, and staring at material properties sheets for hours.
Anonymous 05/17/14(Sat)13:45 UTC+1 No.639459 Report

bump for interest
Anonymous 05/17/14(Sat)14:27 UTC+1 No.639468 Report

as the floor plate thins its resistance increases and its current capacity decreases
increased resistance means more power loss, at the far end you may see performance drop off.
increase the voltage to overcome this, you could do 24v comfortably, 50v is typically largest voltage you can use in commercial products before safety ratings so as a guideline you could use 48v comfortably.

to reduce costs you could look at decreasing width of floor conductors whille increasing seperation whil compensating on the car conductors.

with careful car conductor choice you can clean the floor as you drive over it but thats more helpfl on scalextrix where you go over the same bit, less helpful when you have free roam.

consider just surrounding the entire area with a huge coil of wire and pump it with 1000A and use 'wireless power' type deelio

bottom line is you wont manage to do this cheaply
Anonymous 05/17/14(Sat)16:44 UTC+1 No.639512 Report

I think I'd use pennies or the equivalent copper coin (I like the circular close packing arrangement instead of "stripes") and carbon brushes for the motile's electrical contact with the floor. oxidation shouldn't be much of a problem unless you don't use it for long periods. if you think about it, what do they use in electric motors? the commutator doesn't suffer from oxidation because of the slightly abrasive action of the brushes.
I think I'd run AC to the floor and let the motile rectify it onboard, as you already need diodes for polarity protection why suffer the 0.7v drop needlessly
Anonymous 05/18/14(Sun)00:15 UTC+1 No.639644 Report

Yeah I did some actual calculations one I had some sleep, the stainless steel tape I was looking at would have about 0.8 ohms/metre, I'd be looking at barbecued robots (or at least crap efficiency)
With 0.5*100mm brass sheet I'm looking at closer to 1.3 milliohms/m, much better.
Do you think gold plated copper contacts would be sufficient? (Shaped like in the pic). With the sort of current running through them they should clean themselves off pretty well.
Also, if I can remember how circuits work, if I connect 4 psu's in series I should be able to get 48V - anything wrong with doing this? Say they provide 500W on the 12V rail each, I should be able to draw 2kW, shouldn't I?

I considered just plugging the floor into the wall, but I'm not too hot on the idea of having 240VAC in the open, I'd like to keep the killing for the robots :)
I'm thinking of doing this for my final year project, I've worked with the uni before so I'm pretty buddy-buddy with their marketing department (I've got the uni a world record, they'd better love me), so if I have any chance for them to fund me I'm probably going to need to keep it at least a little bit safe.

I'm hoping that the contacts brushing over the surface will ensure good conductivity, I guess I'll have to do some prototypes.

Our currency uses Cupronickel - would this be a good material?
Anonymous 05/18/14(Sun)00:19 UTC+1 No.639647 Report


Aluminum sheet is probably going to be your best bet as far as price/performance goes. I wouldn't immediately dismiss it out of concern for corrosion. I'd hazard a guess that the reason that would be a problem in most electrified floors is that a significant portion (if not most) of them are going to be both outdoors and need to be pretty hard wearing.

As far as the stainless tape goes, I think you're going to have issues with foil that thin. 100x0.01mm is only 1mm^2. This is close to the equivalent of a single 17AWG wire per 100mm section.

Even in 500mm sections (which is far too wide for anything but a super heavyweight robot), you're looking at a roughly 10AWG equivalent conductors. You couldn't shove 200A though a 10AWG conductor even if it was made of copper, let alone stainless steel. Even if they weren't overheating, they'd have a huge voltage drop and wouldn't be able to actually deliver much power at 12V.

This is also assuming the thickness of the actual foil on the tape is 0.01mm, rather than that being the total thickness of the foil/adhesive.

You're going to need to find something thicker. It doesn't have to be much, even 0.1mm would work, but 0.01mm just isn't going to cut it.

I would also suggest using a 48V system, since that would drop amperage requirements. Use carbon contacts (like they use in commutator brushes). This thing is going to arc at least a little, and that will eat away plating fast.

Honestly, however, I would suggest NOT using this kind of floor for fighting robots because of shorting issues (seriously, metal-clad robots bashing each other around on a conductive floor?).

Exactly what's your beef with batteries? RC lipos aren't that expensive (especially compared to the cost of a fighting robot), reasonably light, and have extremely high power delivery capability. Granted you can't run them forever like you can for a conductive floor, but it's not like robot fights generally last very long in the first place...
Anonymous 05/18/14(Sun)00:31 UTC+1 No.639652 Report

Aluminium oxide is ridiculously good electrical insulator, I'd need something constantly carving through the surface to get good conductivity.
I came across carbon brushes in some earlier googling, what sort of conductivity do they have? I couldn't find material properties on them.
And yeah, I reconsidered most of the stuff you said once I actually got some sleep.
We're thinking of having robot races as well, kind of an IRL mario kart, if we want to have consecutive races lipos just aren't going to last long enough.
I'm also thinking we'll use MDF or some other cheap, non conductive material for the outsides of the robots, hopefully they won't short.

Oh, forgot to say, induction would be cool, but we want them to be wirelessly controlled, I imagine there'd just be too much interference.
I can get 1.5mm stainless steel for about $165/square meter, I've got like a year to save for it so I could probably wrangle it, unless there's a better option :)
Anonymous 05/18/14(Sun)01:21 UTC+1 No.639672 Report


It is, but the natural layer that forms is extremely thin, and functions very poorly as an insulator. Go on, take a multimeter and touch it to an aluminum something or other. On a years-old aluminum heatsink, mine's showing 50m? of resistance per contact with just the miniscule contact area the probe tips provide.

No idea what kind of resistance carbon brushes have, but, considering they're made and used exactly fort this purpose (contacts for mechanical commutators) I'm inclined to think it's pretty low.

1.5mm should be about right. That's 150mm^2 per 100mm section; higher than even 5/0 AWG, which will carry 1000A or more in copper. Granted, the resistance of stainless is a little over 40 times that of copper, so you'd have to derate that significantly. Should still be workable, but I'd really try and be sure that aluminum wouldn't cut it, because aluminum, while not as good as copper, is still about 21 times better at conducting than stainless.

Not to mention it's quite a bit cheaper by volume. My local metal yard sells aluminum surplus at $2.40/lb, making a square meter of 1mm aluminum under $15.
Anonymous 05/18/14(Sun)02:20 UTC+1 No.639687 Report

Damn that's cheap, I probably won't find it that cheap around here, but so long as it's comparable that's hard to ignore.
Ok, I'll give it a shot, my work has Al everywhere - I'll go and test everything I can. I think I have some carbon brushes leftover from a drill somewhere, I'll have a play around with them.
Actually they have silver plated contacts everywhere too (we make pretty heavy duty relays), I'll pinch some to test as contacts to compare to carbon brushes.
I'll need a better multimeter to test them properly (suggestions? I don't know which brands are good).
Anonymous 05/18/14(Sun)03:09 UTC+1 No.639700 Report


As far as I'm aware, there aren't really any multimeters that will reliably test lower than 100m? or so. My cheap-ish one ($20 "Extech" thing) can't even go lower than that since it doesn't have a m? range.

You'll need a milliohm meter. These are kind of expensive, starting out at like $100, but you can improvise one if you have access to a constant current source that can do at least 10A. Just hook that up to the contact circuit and measure the voltage drop with a regular multimeter's mV range.

The silver contacts should work better than the carbon brushes in terms of electrical conductivity, but the reason carbon brushes will be preferable is because of their wear characteristics. Consider, after all, that a high speed rotary tool will engage the brushes more times during 5 minutes of use than your average relay will within its entire service life
Anonymous 05/18/14(Sun)03:40 UTC+1 No.639712 Report

>> robot wars of sorts
so what happens when you get the inevitable conductive metal debris on your floor?
Anonymous 05/18/14(Sun)04:41 UTC+1 No.639742 Report

1) I'll sweep it
2) The floor will have graphite on it anyway to lubricate it, shouldn't be an issue

Ah I think I might even have one those at work too, they're at least accurate to a milliohm. If not, current source it is.
And true, they'll form to the tracks surface won't they, I won't even have to adjust anything.
So, does anyone know if the 4-psu's-in-series thing would work?
Anonymous 05/18/14(Sun)04:43 UTC+1 No.639746 Report

I think they were talking about it shorting the floor, not scratching it up or whatever you're talking about.
Anonymous 05/18/14(Sun)04:45 UTC+1 No.639747 Report

Yeah I worked that out right after I submitted it.
Most of the frame should be mdf - just in case, i'll have a circuit breaker at the source.
Anonymous 05/18/14(Sun)05:09 UTC+1 No.639753 Report

Plus, you could defeat a robot just my putting a wedge under it to break contact with the floor.

And you'll get a short if a robot's metal body shell, or any metal debris, makes contact with the floor while spanning across a gap.
Anonymous 05/18/14(Sun)16:31 UTC+1 No.639903 Report

Just use fucking batteries, they really aren't that expensive. Probably a lot cheaper than your current plan, and far more useful,

Just get a couple of these.

I'm pretty sure hobbyking is owned by an austrailian too.
Anonymous 05/18/14(Sun)16:36 UTC+1 No.639906 Report

As soon as someone puts a wedge on the front of their robot its going to rip the conductors off the floor.

Aluminium is soft as butter and won't last a month if you are running any form of robot combat.

Requiring combat robots to have a continuous connection to the floor is retarded.

You will need supercaps in each robot to compensate for gaps anyways, which will probably cost just as much as batteries would.
Anonymous 05/19/14(Mon)06:24 UTC+1 No.640206 Report

It's not just about the cost, I'm happy to spend more if it doesn't mean waiting around for batteries to charge - that said, I wouldn't be opposed to putting small capacity batteries in the robots, in case they do lose contact.
It's more the cost of the raw materials I wanted to reduce, as I don't want to spend more than I absolutely need to.
From where we currently stand it looks like we'll be focussing on racing more than just all out fighting, so the weapons will be designed more to knock people out of the way rather than tear chunks off (probably). If bumper cars can do it, I don't see why we couldn't.
Anyway, thanks for the input guys, maybe this'll just end up being too difficult and I'll end up going back to batteries, but I'll pursue it as longa s I can
Anonymous 05/19/14(Mon)06:33 UTC+1 No.640213 Report

>Deathracing robots
Fuck yes, I love you.

I hate to be that guy that outright denies what the OP has requested in place of another method, but...

One possible floor (pun intended) in the plan could be: If the floor is conductive, I'd just have my robot charge massive caps then dump them, frying all the other cars, then finish the race on a tiny reserve battery.

You mentioned bumper cars, why not do exactly as they do, and have a pole extending from the robot and making contact with a "roof" of conductive mesh?
Anonymous 05/19/14(Mon)07:38 UTC+1 No.640237 Report

It's fun being an idealistic engineering student :)

It's bumper cars that got me onto this idea, this seems to be what newer systems use (You can see them in this pdf: http://www.iepark.com/wp-content/uploads/2012/11/BUMPER-CARS.pdf - warning, bigger than it should be).
They're going to be wirelessly controlled, I'm worried there'll be too much interference from the mesh - then again, there could be anyway with just the floor, I'm going to find out one way or another.
Also I guess we won't be able to see as well with a mesh over the top, that was our first idea though until I stumbled across the newer systems.

Well I might just have to put a clamping circuit in each robot now that you've said that, 'cause that's a good idea.

If all this doesn't work we'll likely go with batteries, and we'll just charge half while using the other half or something. I'm hoping to be able to use this for my final year project as I mentioned and want to make it as flashy as possible, the teachers seem to eat that shit up - an electrified floor should fall into that category.
All the content on this website comes from 4chan.org. All trademarks and copyrights on this page are owned by their respective parties. Images uploaded are the responsibility of the Poster. Comments are owned by the Poster. 4chanArchive is not affiliated with 4chan.