Team MedBots

All things antweight

Moderators: BeligerAnt, petec, administrator

Post Reply
Rapidrory
Posts: 1160
Joined: Sun Nov 24, 2013 9:54 pm

Re: Team MedBots

Post by Rapidrory »

You tried using Design Spark PCB? It's what I use for all my PCBs as it's free and relatively straight forward to use..
Rory Mangles - Team Nuts

Robots: Nuts 2 and many more...

NanoTwo Motor Controllers: https://nutsandbots.co.uk/product/nanotwodualesc
User avatar
BeligerAnt
Posts: 1872
Joined: Wed May 15, 2002 12:00 am
Location: Brighton
Contact:

Re: Team MedBots

Post by BeligerAnt »

Try moving the parts around. You can't route the layout that you currently have on a single layer (unless you route traces between pins - not recommended).
If you absolutely need the most compact layout possible you will probably have to accept some jumpers on a single-layer board. However, only one jumper isn't too bad :)

+1 for Design Spark, it's about the best free PCB tool I've found. Eagle has a *lot* of support but the free version is limited in some way (don't remember the details). OK for very small designs like this but probably won't handle your next project so you'll have to learn something else...
Gary, Team BeligerAnt
Max
Posts: 105
Joined: Tue Oct 08, 2013 3:14 pm

Re: Team MedBots

Post by Max »

Ok I'll have a look at that.
Here is the latest project:
http://youtu.be/tRFs32l8ZUU

It's (meant to be) a spring powered flipper, at the moment it os using a peice of glass fibre sheet which gets bent to act as the spring but that needs some work. It has a 250:1 polulu motor to pull down the flipper arm and uses a one way bearing so that once it gets past the equilibrium point the arm is free to move up (due to the spring) and move faster than the motor.

I need to fit the switch at the bottom so that the motor stops just before the equilibrium position (at the lowest position)

Let me know what you think!
Max
Posts: 105
Joined: Tue Oct 08, 2013 3:14 pm

Re: Team MedBots

Post by Max »

Made some modifications:
http://youtu.be/PnzzIJ72xyM

Still need to find a good way of putting a switch or sensor in to stop the arm at just the right time.
Rapidrory
Posts: 1160
Joined: Sun Nov 24, 2013 9:54 pm

Re: Team MedBots

Post by Rapidrory »

Looking really good!
Rory Mangles - Team Nuts

Robots: Nuts 2 and many more...

NanoTwo Motor Controllers: https://nutsandbots.co.uk/product/nanotwodualesc
User avatar
Shakey
Posts: 1119
Joined: Sat Mar 13, 2010 8:38 pm
Location: Reading

Re: Team MedBots

Post by Shakey »

That looks awesome! I was wondering what you'd need a one way bearing for. :P
Nuts And Bots - For all your components and ready built antweights!

Alex Shakespeare - Team Shakey / Nuts And Bots / Team Nuts:
AWS 44, 45, 49, 51 & 55 Winner - Far too many robots!
Max
Posts: 105
Joined: Tue Oct 08, 2013 3:14 pm

Re: Team MedBots

Post by Max »

Thanks guys, I've fitted more elastic bands and now its even more powerful!

I've re-jigged the board layout and am now much happier with it:
motor controller board v2.png
motor controller board v2.png (178.06 KiB) Viewed 7485 times
Can anyone see any potential problems or have any comments before I see if I can get it printed?
Thanks in advance!
User avatar
BeligerAnt
Posts: 1872
Joined: Wed May 15, 2002 12:00 am
Location: Brighton
Contact:

Re: Team MedBots

Post by BeligerAnt »

Really, learn a proper PCB layout package if you intend to do many more boards! :)

* Get rid of the acute angles. The corners will get over-etched and can cause problems.
* Check spacings between tracks and pads carefully. A proper PCB tool will do this for you :) Smallest gap should be at least as big as thinnest track.
* Keep the tracks/gaps as big as you can. It makes the board easier to manufacture. Whilst 0.1mm track/gap is possible these days it's very expensive because the yield is bad. I would suggest 0.5mm track/gap for a low-density through-hole design like this. Wider tracks also save you from having to worry about current-carrying capacity so much.
* Check the size of the pad vs the hole you are drilling in it. The annulus (width of the copper you have to solder to) is given by
(Pad dia - hole dia) / 2 so make sure this comes out at something sensible. Generally bigger holes have (proportionally) bigger pads. I'd suggest aiming for an annulus of 0.5mm. Anything less than 0.4mm is going to get difficult to solder to.
* If you're drilling the board yourself make the pads as big as possible in case the drill doesn't quite hit the centre of the pad.
* Keep routes to horizontal, vertical and 45 degree angles. The layout will be neater, there will be no acute angles and you will generally fit more in assuming you are placing parts on a grid.

You can ignore the last one if you like but it's considered good practice.
Gary, Team BeligerAnt
Max
Posts: 105
Joined: Tue Oct 08, 2013 3:14 pm

Re: Team MedBots

Post by Max »

Thanks I'll make all of those changes. I did have a go with design spark but I found it confusing and hard to use but I imagine I would get used to it with practice, the nice thing about Fritzing is it lets you design the circuit on a breadboard then automatically generates the PCB then you juts have to manually drag the components around.

Here is a CAD model showing how the rubber band powered flipper works:
https://www.youtube.com/watch?v=b-4uOvd ... e=youtu.be

The problem is I need some way of sensing when the flipper arm is at its minimum position so I can then stop the motor and be ready for the next flip. The original idea was to have some sort of microswitch but I fear it would have to be positioned so precisely it would be almost impossible to do and would probably move over time. Next idea was to use a potentiometer so I could adjust the endpoint in the software; but my problem is finding somewhere to attach it, there is no obvious point on the blue cam shaped thing and the only other point I could think of was at the pivot point of the flipper arm but then it is very tricky to not end up using the potentiometer as a pivot/bearing and I fear it would break under the load on the arm due to the rubber bands.

Has anyone got any ideas for how to get position feedback?
User avatar
BeligerAnt
Posts: 1872
Joined: Wed May 15, 2002 12:00 am
Location: Brighton
Contact:

Re: Team MedBots

Post by BeligerAnt »

All CAD packages are necessarily complicated and take some learning. :( Follow the tutorials.

You've covered most of the bases when it comes to position feedback. The only other options are optical or magnetic, all of which are likely to be too big/heavy/complex for use in an antweight weapon.

You're right about the problems of using a microswitch as a limit switch. I did try once, a long time ago, but it was impossible (with my limited level of technology) to get it working repeatably. The real downside is that it *needs* to work, if it fails you usually lose the weapon :( It might be possible to get a microswitch system to work with a 3D-printed chassis to hold everything accurately in place, although there is still the danger of parts flexing/vibrating in use.

A potentiometer would be a good solution if you could couple it to a shaft such that it doesn't get destroyed!
Gary, Team BeligerAnt
Post Reply