Daniel's Work Thread: Toward a Hackerspace Scanner

[daniel_reetz]

So, it's been incredibly difficult to get the Hackerspace Scanner off the ground, even with all the help here. And because I have a demanding (but awesome) day job, it's been hard for me to get in the right mindset every night when I get home.

This thread is where I will document what I'm doing and what problems I'm facing getting my CNC Router into production.

My biggest issue at this point is regenerating all the artwork in a way that makes sense. It's one thing to offer a package of Open Hardware artwork suitable for one person to copy. It's another thing to go into production yourself - the quantities change everything.

My task today/tonight is the same it's been for the last three weekends: hack the artwork into shape. Right now, the art is 1 sheet per scanner. However, due to the fact that most of these parts are small and irregularly shaped, that is unlikely to be the most efficient way to make them. Most likely, it will be best to tile them and cut, say, twenty parts per sheet rather than one scanner per sheet. So by tomorrow I will report back on my tilings. I think I'm going to shoot for 10 scanner kits per week; 10x quantities are high enough that I should be able to optimize things in an interesting way, and 10 sheets of ply are about as many as I can fit/carry in my truck before I start hurting it and myself.

[dpc]

Most likely, it will be best to tile them and cut, say, twenty parts per sheet rather than one scanner per sheet.

What does this mean for the fixture area of the sheet where the backside counter bores are made?

[daniel_reetz]

What it probably means (I spent quite a bit of time on this yesterday) is that I'm going to make a special fixture sheet that fits only those parts and I'll make them in quantities of 50 or so at a time. My idea right now is to get them cut out and then have a sheet which has fixtures that fit in the bearing pockets instead of pockets for the outside of each part - so I can flip them and do them en masse.

[dpc]

There's an advantage of cutting the pocket (whether it's for inserting the entire arm, or just a hole for a c'bore "stub") on the sheet at the same time that you're going to insert the part and cut the backside counter bore - registration.

How difficult is it to put a sheet down and get the router to register the exact position of the center of all of the "stubs" you'll be using to position the parts for cutting the backside counter bores? If you can't get the sheet with the stubs positioned and secured easily, i'm wondering if it may actually be quicker to have a drill press fixture to make the backside counter bore one at a time (provided you can get a bit that is the correct diameter - 22mm forstner?).

[daniel_reetz]

That's true. However my current plan is to set up my machine with optical limit switches so that I get a repeatable home position every time. From that, I'll be able to drill some peg holes into the bed and make future sheets register relative to the home position. It's work, but so far everything about this has been slog work. The drill press approach has no appeal at all for me except for one-off scanners; if I can't make these things en masse with the machines, knowledge and effort I have in this already, it's just not worth it. I might change my mind about that later, but right now I just want to automate everything. Part of my internal thinking about this is that I want time to work on other projects - I have only a few short hours per night and I'd like to at least have one of those hours available for non-bookscanner activities.

That said it may be a great approach for others. Also, when I was in Portland I tried CFMorrill's method (well, a modified version) with the drill press/Forstner and though it oversized the hole slightly it worked well. So the total diameter was not that critical.

[jck57]

That's true. However my current plan is to set up my machine with optical limit switches so that I get a repeatable home position every time. From that, I'll be able to drill some peg holes into the bed and make future sheets register relative to the home position. It's work, but so far everything about this has been slog work. The drill press approach has no appeal at all for me except for one-off scanners; if I can't make these things en masse with the machines, knowledge and effort I have in this already, it's just not worth it. I might change my mind about that later, but right now I just want to automate everything. Part of my internal thinking about this is that I want time to work on other projects - I have only a few short hours per night and I'd like to at least have one of those hours available for non-bookscanner activities.

That said it may be a great approach for others. Also, when I was in Portland I tried CFMorrill's method (well, a modified version) with the drill press/Forstner and though it oversized the hole slightly it worked well. So the total diameter was not that critical.

IMO you guys would be just as happy with one bearing per joint and save a bunch of extra work. Here's my idea for a single bearing pocket with retainer. Two flat washers fit into shallow pockets and overlap the outer edge of the bearing.

[dpc]

Dan, if you can register the sheet that accurately, then you might be able to get away with cutting the backside counterbores on the sheet, flipping it, then cutting the front side counter bores and individual pieces from the sheet. No fixture tom-jiggery involved.

[dpc]

jck47,

I must be missing something. Won't the shoulder bolt that goes through the center keep the bearing in each arm from popping out?
[edit] I can see why you'd need this. If the bearing isn't a tight fit in the arm, the arm could move off the bearing and rub against the other arm.

I think you'll also need a spacer behind the bearing, that protrudes slightly from the backside of the arm so that it would keep the arms from pinching against each other as the bolt was tightened.

[jck57]

jck47,

I must be missing something. Won't the shoulder bolt that goes through the center keep the bearing in each arm from popping out?

I'm not following what you're saying. Can you elaborate?

I think you'll also need a spacer behind the bearing, that protrudes slightly from the backside of the arm so that it would keep the arms from pinching against each other as the bolt was tightened.

Yeah, the bearing will end up almost flush with the plywood on the retainer side. So a couple washers between that side and the other arm.

[Fab52]

Hi Daniel,

Here's the solution so you will not need two bearings in the arms and have to flip the sheet to do the bearing trust (pocket)....

I'll try to explain this in my best Franco-English....

It's simple has 1, 2, 3, just a different way of thinking....

You put one bearing in the arm and one bearing into the support... Is it simple enough ????

Hey don't hit your head on the desk for that...))

Sometime when we work to hard on something we don't see it anymore...

I'll do mine this way since the guy who cut it does not flip the sheet.

I hope this help you!!!

See you

Fab

P.S. English is a second language for me, you may laugh at me or help me improved....)))