Peter Wallers Robot Ramblings

All things antweight

Moderators: administrator, BeligerAnt, petec

Post Reply
User avatar
voorsk
Posts: 167
Joined: Tue Jan 02, 2018 8:11 am
Location: Stockport, UK
Contact:

Re: Peter Wallers Robot Ramblings

Post by voorsk » Mon Oct 01, 2018 3:09 pm

peterwaller wrote:
Sun Sep 30, 2018 10:08 am
I tried the yellow material above against a drum spinner and it failed quite quickly
That's a shame - that colour is amazing! 8)
Team BLEEP Suspicious Houmous / Sprouting Potato / Fermented Melon / Black Carrot

User avatar
peterwaller
Posts: 3699
Joined: Fri Feb 15, 2002 12:00 am
Location: Aylesbury Bucks
Contact:

Re: Peter Wallers Robot Ramblings

Post by peterwaller » Tue Oct 30, 2018 7:33 pm

The new circuit boards eventually arrived from Ragworm (29 days).
These use the larger 5 amp drivers and the new board combines the receiver, motors, power switch and servo and power connectors.
Image
Image
The boards I ordered from PCBWay in China arrived in less than a week and were cheaper.
These use the DRV8872 drivers which are 3.6 A but much smaller. The board doesn't include the motors but you can piggyback the 20 A Afro brushless controller.
Image
Image
This second version I have incorporated into the newer version of Combatant which I have had to reduce in size to get it to fit the cube.
Image
I have also replaced the 3D printed disc with one I cnc'd in 3.2 mm fibreglass.
Image

User avatar
peterwaller
Posts: 3699
Joined: Fri Feb 15, 2002 12:00 am
Location: Aylesbury Bucks
Contact:

Re: Peter Wallers Robot Ramblings

Post by peterwaller » Thu Nov 15, 2018 7:16 pm

Doing a bit of a rework on my drum spinner. I have modified the casing for my new 5 A controller to take the Afro 20 A brushless board.
Image
I am redoing the drum in fibreglass sections which I have made on my small CNC mill.
Image
I have also redesigned the chassis which was too flexible in Nylon . I have tried printing it in a combination of nylon with PLA stiffening sections but although in prints nicely in the two materials the nylon does't really adhere to the PLA.
I have just ordered some new filaments from France ABS with Polycarb and ABS with Kevlar I will let you know how I get on when they arrive.

User avatar
peterwaller
Posts: 3699
Joined: Fri Feb 15, 2002 12:00 am
Location: Aylesbury Bucks
Contact:

Re: Peter Wallers Robot Ramblings

Post by peterwaller » Tue Nov 20, 2018 1:54 pm

The ABS with Kevlar prints really nicely at 260 C on a table at 90 C. I have reprinted the new design for my drum spinner in it but only time will tell how resilient it is. I have reduced the belt drive pulley which has halved the drum speed and improved the balance a bit but there is still vibration. The only down side I have found so far for the new filament is the colour a sort of yellowish off white but you can't have everything.
Image
Hopefully will get to test it on Saturday in darkest Dorset.

User avatar
peterwaller
Posts: 3699
Joined: Fri Feb 15, 2002 12:00 am
Location: Aylesbury Bucks
Contact:

Re: Peter Wallers Robot Ramblings

Post by peterwaller » Sun Nov 25, 2018 10:14 am

The drum spinner went out before it faced a spinner but as a test I had printed the flipper on my elastic powered flipper in both the ABS with Kevlar and the ABS with Polycarb and after the main event put them up against a really nice vertical spinner. The first thing that became apparent was that the flipper wasn't quite getting fully down which meant the spinner could get under the front which was a definite problem as not many 1.5 mm thick plastics withstand a decent spinner. Initially it appeared that the ABS with Polycarb (the white in the photos)was the more resilient as it appeared less damaged,
Image
But on closer examination the flipper plate in ABS with Polycarb survived because it delaminated from the supports and was able flex more. The ABS with the Kevlar showed absolutely no delamination and so is the better material.
Image
The green line on the kevlar was supplied by Mr Shakey.

Paulmchurd
Posts: 223
Joined: Sun Dec 10, 2017 4:39 pm

Re: Peter Wallers Robot Ramblings

Post by Paulmchurd » Sun Nov 25, 2018 3:56 pm

How much worse of a state would a PLA print of these be?

User avatar
peterwaller
Posts: 3699
Joined: Fri Feb 15, 2002 12:00 am
Location: Aylesbury Bucks
Contact:

Re: Peter Wallers Robot Ramblings

Post by peterwaller » Sun Nov 25, 2018 4:14 pm

When I have used PLA in the past it has shattered. It is also a lot to do with thickness and angles had the flipper been low enough to get under the spinner it would have survived much better.

User avatar
peterwaller
Posts: 3699
Joined: Fri Feb 15, 2002 12:00 am
Location: Aylesbury Bucks
Contact:

Re: Peter Wallers Robot Ramblings

Post by peterwaller » Fri Dec 14, 2018 3:56 pm

Both my spinners did very badly at the Dorset event.
Combatant was really nicely balanced and set up but the hardened steel lathe tools snapped and were useless.
I have now reverted to a 2mm thick spring steel bar which also seems nicely balanced and is hopefully unbreakable.
Image
Anticyclone had balance troubles which didn't seem too bad when held but became undriveable in the arena.
After much modifying and testing I gave up on the drum made from 3D printed segments and bought some thick walled Aluminium Tube that was large enough to go around the motor. Still I had terrible balance problems and after much faffing around discovered the wall thickness of tube is not very consistent. Having bored the tube out on the lathe I eventually managed a reasonable balance and it is now fitted with 3 M2.5 high tensile socket head screws spaced at 120 degrees.
Image

User avatar
peterwaller
Posts: 3699
Joined: Fri Feb 15, 2002 12:00 am
Location: Aylesbury Bucks
Contact:

Re: Peter Wallers Robot Ramblings

Post by peterwaller » Mon Dec 17, 2018 2:39 pm

Although the balance was much improved it was still far from perfect and when sat in hard contact with arena floor it started to bounce more.
I had decided to have a front support that was sprung so printed one that could be attached using double sided foam tape and this reduced the bounce.
Image
At the same time Ant Ipodean one of our forum members from New Zealand contacted me to point out although having 3 blades at 120 degrees gives a static balance it needs also to be balanced along the drum. Below is his second attempt to explain it to me as I was a bit slow on the uptake.
I was having difficulty putting the concept into words, but thinking about it a bit more, try imagining this:
Make drum with one tooth at one end and a second tooth 180 degrees away at the other end.
It will still balance statically, but be massively unbalanced at speed, like 2 separate unbalanced flywheels.
This accounts for the alternating side to side vibration.
Your setup is a milder version of this.

As I had drilled and tapped all three holes in each tooth position along the existing drum so I could easily fit new teeth if any got damaged I was quickly able to fit all 9 screws to test the theory out and it works brilliantly.
https://youtu.be/AtjnFo-RfVA
If you look at the printed side of the robot you can see the kevlar fibers in the ABS they can be easily removed to make it look pretty.

User avatar
peterwaller
Posts: 3699
Joined: Fri Feb 15, 2002 12:00 am
Location: Aylesbury Bucks
Contact:

Re: Peter Wallers Robot Ramblings

Post by peterwaller » Fri Jan 04, 2019 9:01 pm

Having had second thoughts about how well the fleaweight spinner would stand up as part of an antweight cluster I deided to design a new cluster from scratch with a slightly more even weight distribution. Here is FlipAntFlop a 97.5 gmflipper and a 45.2 gm double ended wedge both printed in ABS with Kevlar.
Image
Image
https://youtu.be/SiR_W4e_Q2g
https://youtu.be/MyLNvlK5w2c

Post Reply