SR2015 flag design/prototyping

[Rich Barlow]

Hi all,

I've drawn up a design for the SR2015 flags. This can be found in
arena.git[1]. For reference, a 'flag' consists of a 250x250x250mm cube
on castors that weighs 2kg. The README should explain everything, but
here's a brief summary:

The design consists of panels of 3mm laser cut MDF. The majority of
these are glued together. The lid of the cube is screwed in place, to
allow access to the inside with the intention of adding ballast. Inside
the cube there is a small compartment that can hold dry sand.

The castors used are 15mm diameter ball castors from OmniTrack[2]. These
will give a ground clearance of 10.8mm (the rules specify 10mm +- 5mm).

I am going to prototype this design ASAP. Hopefully, if all goes well,
we can publish this design to teams so that they can construct it
themselves.

This will cost about £40 to get cut at RazorLAB[3]. The castors will be
around £12 per flag. The cost of the other stuff is negligible. I think
it should be possible to get the laser cutting+material cost down to
£15/flag elsewhere. This gives a total cost of ~£30/flag. Assuming we
make 15 of them (5 per arena, 5 spare), that's £450 total.

Rich

[1] https://www.studentrobotics.org/cgit/arena.git/tree/comp15/flags
[2] http://www.omnitrack.co.uk/lightduty/saturn-type-ball-units-LD-m.php
[3] http://www.razorlab.co.uk/

[Peter Law]

Rich wrote:
> I've drawn up a design for the SR2015 flags. This can be found in
> arena.git[1]. For reference, a 'flag' consists of a 250x250x250mm cube
> on castors that weighs 2kg. The README should explain everything

Awesome, thanks for picking this up. Having read your email first I
was going to suggest that the pieces be interlocking (I've had issues
when only glueing MDF in the past), but I see you've already done
this, yay.

I've linked to this thread from the ticket for creating the flags (#2772).

From a quick look over the pieces, the only bit I'm confused by is the
pairs of slots along one edge of the side pieces - could you clarify
what these are for?

> This will cost about £40 to get cut at RazorLAB[3]. The castors will be
> around £12 per flag. The cost of the other stuff is negligible. I think
> it should be possible to get the laser cutting+material cost down to
> £15/flag elsewhere. This gives a total cost of ~£30/flag. Assuming we
> make 15 of them (5 per arena, 5 spare), that's £450 total.

Having had a quick look at the budget's master branch, there doesn't
yet seem to be any allocation at al for the SR2015 competition -- from
what I recall this is in another branch somewhere. Do you know how
this expenditure compares to what has been budgeted?

Thanks,
Peter

#2772 https://www.studentrobotics.org/trac/ticket/2772

[Rich Barlow]

On Fri, 2015-01-02 at 20:42 +0000, Peter Law wrote:
> From a quick look over the pieces, the only bit I'm confused by is the
> pairs of slots along one edge of the side pieces - could you clarify
> what these are for?

These are to hold the 'sr2015_flag_top_fixing_plate.dxf' parts. There is
one in each corner of the cube towards the top. They are to allow the
lid of the cube to be non-permanently fixed. The 1.6mm hole cut into
them acts as a pilot hole for a woodscrew.

> > This will cost about £40 to get cut at RazorLAB[3]. The castors will be
> > around £12 per flag. The cost of the other stuff is negligible. I think
> > it should be possible to get the laser cutting+material cost down to
> > £15/flag elsewhere. This gives a total cost of ~£30/flag. Assuming we
> > make 15 of them (5 per arena, 5 spare), that's £450 total.
>
> Having had a quick look at the budget's master branch, there doesn't
> yet seem to be any allocation at al for the SR2015 competition -- from
> what I recall this is in another branch somewhere. Do you know how
> this expenditure compares to what has been budgeted?

This budget is in the 'sr2015-comp' branch of Rob's Gitorious repo[1]
(note that this is synonymous with the moto-app-2014 branch). The reason
this is currently not merged into master is that we are still waiting
for the Moto funding to clear/arrive. Obviously the amount allocated to
tokens isn't enough to cover this. I will try to get some quotes from
real manufacturers.

Note that I haven't actually made this design yet, so it could be crap.

Rich

[1]
https://gitorious.org/sr-budget/budget/source/db9e61a0d5231de275fc810a8badb46906889de5:

[Lila Fisch]

could we profit from students having lasercutting access at a university?

Admittedly I neither have the training yet nor do I know the cost here,

but I am supposed to get that training anyways.

cheers

lilafisch

--
You received this message because you are subscribed to the Google Groups "Student Robotics Development" group.
To unsubscribe from this group and stop receiving emails from it, send an email to srobo-devel...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Richard Barlow]

On 17/01/15 11:27, Rich Barlow wrote:
> I will try to get some quotes from real manufacturers.

I've now got a couple of quotes for cutting, materials and delivery:

www.cut-tec.co.uk: £300.00 inc. VAT (£20.00/flag*)
www.cutlasercut.co.uk: £341.64 inc. VAT (£22.78/flag)

* I actually asked for 18, the price might be a little less if we only
get the 15 we need.

Lilafisch wrote:
> could we profit from students having lasercutting access at a university?
> Admittedly I neither have the training yet nor do I know the cost here,
> but I am supposed to get that training anyways.

Could you please work out how much this would cost so that we can
compare it to the above quotes? Note that I'd like to see a sample flag
cut first just to verify the quality as there's no point in doing it if
the parts don't fit together properly due to an issue with the cutting
itself.

Thanks,
Rich

[Lila Fisch]

Hi Rich,

the laser cutter I could use:

Epilog Laser Cutter

various materials

457 x 305 (mm)

up to 10mm thickness

its an ECS cutter, and a friend already has access for it.

It is open for all ecs students, so we could get blueshirts trained.
https://secure.ecs.soton.ac.uk/notes/3D/epilog/index.php

the list of materials includes wood, but not explicitly mdf.

there is no cost for cutting, so the cost comes down to how much flag we can get on one mdf plate that fits the cutter.

I won't get around to looking at your designs before Monday.

cheers

lilafisch

--
You received this message because you are subscribed to the Google Groups "Student Robotics Development" group.

To unsubscribe from this group and stop receiving emails from it, send an email to srobo-devel+unsubscribe@googlegroups.com.

[Lila Fisch]

Hi,

I finally looked at the .dxf files; with the current laser cutter we could only cut one side at once, for the six sides of a cube that would make almost 100 cuttings. Unless I can find a laser cutter with a larger area this is not a pleasant solution.

There is one large one in the EDMC stores, I'll try to find out more.
cheers
lilafisch

[Lila Fisch]

ok,

if I can get EDMC to teach me (I'll find out monday)

we should be able to fit the razorlab panels.

then we'd just have the material cost.

It might be easier to use wood though, since we could pre-cut that by hand
without health risk.

cheers

lilafisch

[Peter Law]

Hi,

lilafisch wrote:
> <ECS have a cutter, let's use that>

Even in the event that the ECS cutter isn't big enough to really do
the final job, could we look at using their cutter to build the
prototype?

This would at least prove that the design works, and allow us to
publish the detailed specs that the competitors so desperately need?

Thanks,
Peter

[Lila Fisch]

Hi,

This would at least prove that the design works, and allow us to
publish the detailed specs that the competitors so desperately need?

I have no experience how much the design depends on laser specific properties.

I will try to get an ECS student to get acces to the lasercutter, so maybe we can have a first prototype by the active day already. Since we will have to buy smaller sheets/cut down sheet ourselved this might get more costly.

I am currently arranging for a laser introduction with the bigger machine, which should occur in the next three weeks. 

The cost for the material should stay below GBP 150, based on cost for smaller 3mm mdf plates.

cheers

lilafisch

[Peter Law]

Hi,

lilafisch wrote:
> <ECS have a cutter, let's use that>

I wrote:
>> This would at least prove that the design works, and allow us to
>> publish the detailed specs that the competitors so desperately need?

lilafisch wrote:
> I have no experience how much the design depends on laser specific
> properties.

I've only ever used one laser in the past, though I'd hope that they
were close enough to being reproducible -- otherwise it'd be rather
hard to actually design things for them.

Either way, as I understand it we don't even know if the designs
themselves work (ie, is there a tab or hole missing somewhere). While
I have confidence in Rich's design abilities, actually making a
prototype seems the next step.

Rich: did you end up getting a prototype from either of the companies
you got quotes from?

> I will try to get an ECS student to get acces to the lasercutter, so maybe
> we can have a first prototype by the active day already. Since we will have
> to buy smaller sheets/cut down sheet ourselved this might get more costly.
>
> I am currently arranging for a laser introduction with the bigger machine,
> which should occur in the next three weeks.

Cool. Given that the active day was a couple of weeks ago did this happen?

Do we have any further information about whether using the ECS cutter
is possibly viable, or should we be looking at a different supplier?

Thanks,
Peter

[Samson Danziger]

Do we have any further information about whether using the ECS cutter 
is possibly viable, or should we be looking at a different supplier? 

Freddie, Andy and I have been on a laser cutting induction for ECS. 

At the moment they have said no to cutting MDF, but they are going to do some experiments and get back to me. So the only available material is acrylic.

With regards to sizing, looking at the files in the arena.git repo, we would need 4 sheets of 600x300mm (the max size the laser cutter can take) for one flag.

Tony Gardner in the workshop suggests buying supplies from Hobarts, which for MDF is £0.96 per sheet [1], so it would come to £3.84 per flag. But this it is 3.2mm not 3mm as specified. 

I think Freddie was going to, or has purchased some acrylic for a test flag. 

Samson

[1]: https://hobarts.com/sheet-materials/wood/wood:-laser-grade-mdf_124_15_55/2mm-laser-grade-mdf-p-87.html

[Lila Fisch]

Hi,

> I will try to get an ECS student to get acces to the lasercutter, so maybe
> we can have a first prototype by the active day already. Since we will have
> to buy smaller sheets/cut down sheet ourselved this might get more costly.
>
> I am currently arranging for a laser introduction with the bigger machine,
> which should occur in the next three weeks.

Cool. Given that the active day was a couple of weeks ago did this happen?

Do we have any further information about whether using the ECS cutter
is possibly viable, or should we be looking at a different supplier?

I got the introduction last week, but was down for most of the week thanks to the flu.

Planning to cut one flag this week, but there are still a few things that could get in the way of that.
Also got a local source that pre cuts the mdf for the uni cutter.

cheers
lilafisch

[Richard Barlow]

On 15 Feb 2015 22:15, "Peter Law" <peter...@gmail.com> wrote:
> Either way, as I understand it we don't even know if the designs
> themselves work (ie, is there a tab or hole missing somewhere). While
> I have confidence in Rich's design abilities, actually making a
> prototype seems the next step.
>
> Rich: did you end up getting a prototype from either of the companies
> you got quotes from?

I ordered one copy of the design from RazorLab last week. It should be delivered today.

Rich

[Richard Barlow]

On 16 Feb 2015 01:29, "Samson Danziger" <samson....@gmail.com> wrote:
> At the moment they have said no to cutting MDF, but they are going to do some experiments and get back to me. So the only available material is acrylic.

You need 'laser' MDF, otherwise the glue will release chlorine when it's cut. This is bad for both the machine and your lungs.

> I think Freddie was going to, or has purchased some acrylic for a test flag. 

If you want to properly assemble this you'll need to get hold of some acrylic cement. It's pretty nasty stuff, so be careful.

Rich

[Lila Fisch]

Hi,

On 16 Feb 2015 01:29, "Samson Danziger" <samson....@gmail.com> wrote:
> At the moment they have said no to cutting MDF, but they are going to do some experiments and get back to me. So the only available material is acrylic.

You need 'laser' MDF, otherwise the glue will release chlorine when it's cut. This is bad for both the machine and your lungs.

I have the source of MDF that is used for the 600mmx400mm lasercutter in EDMC, where I can get them pre cut for the laser cutter.

I talked to Freddie, we will go with that cutter and the mdf.

Rich: since you'll hopefully have a prototype by today, let me know if the design works. I'd then try to spread flags to TechDays from here.

cheers
lilafisch

 

[Richard Barlow]

On 16/02/15 08:52, Lila Fisch wrote:
> Rich: since you'll hopefully have a prototype by today, let me know if
> the design works. I'd then try to spread flags to TechDays from here.

https://www.flickr.com/photos/richardpbarlow/sets/72157648545964963/

I've built a prototype and it's good. The only thing is assembling the
inner walls is fiddly and requires some bending/persuasion. However, it
is possible and I don't want to change the design now that this one has
been proven.

It weighs 1.225kg without ballast, has 11.25mm of ground clearance
(specified to be 10+-5mm) and the side length is exactly 250mm
(specified to be 250+-10mm).

It took about an hour to assemble.

Please note that we don't currently have a budget line for these. I have
paid for this prototype out of my own pocket and do not expect
reimbursement.

Rich

[Lila Fisch]

Hi,

Please note that we don't currently have a budget line for these. I have paid for this prototype out of my own pocket and do not expect reimbursement.

What are you trying to say?

Should we not aim to build flags for the upcoming Kickstarts and wait until the budget line exists if we expect to get reimbursement?

cheers
lilafisch

[Richard Barlow]

Since we're not under extreme time pressure (i.e. the competition isn't
tomorrow and everyone is currently not on fire) there is no reason for
us to not be following our normal purchasing procedure. The procedure is
to get a purchasing ticket ACK'd by two treasurers and a purchasing
ticket needs a budget line to be valid.

We can aim to get the flags built for the upcoming Tech Days (which I
presume you meant by Kickstarts), but we should get a budget line in
place for them. From a very quick look at budget.git, we do have a small
amount of money that can be allocated which will probably cover this.
Please can you submit a budget mod for the costs associated with your
proposal.

The only reason I paid for the prototype out of my own pocket was to
avoid the extra paperwork and delays. That was my own choice and I don't
expect anyone else to do the same.

Rich

[Peter Law]

Rich wrote:
> https://www.flickr.com/photos/richardpbarlow/sets/72157648545964963/
>
> I've built a prototype and it's good.

Looks great, thanks for doing this.

Peter

[Tom Leese]

Hi,

I've added a page on Trac with information about the flags we created in Southampton: https://www.studentrobotics.org/trac/wiki/2015/Flags

More photos will be added by Andy Barrett-Sprot shortly.

Tom

[Peter Law]

Hi,

Tom wrote:
> I've added a page on Trac with information about the flags we created in
> Southampton: https://www.studentrobotics.org/trac/wiki/2015/Flags

Thanks, it looks good.

I'm sure it's just my eyes, but the casters look further in from the
edges than I was expecting -- what's the distance from the edge to the
centre of the ball?

Thanks,
Peter

> --
> You received this message because you are subscribed to the Google Groups
> "Student Robotics Development" group.
> To unsubscribe from this group and stop receiving emails from it, send an

> email to srobo-devel...@googlegroups.com.