New run of orbshields

[vputz]

For all three of you watching...



I have a great team of beta testers working on the kits I sent out. There are Problems, though, and while I greatly appreciate their efforts and the feedback is fantastic, I think that kit is simply doomed to failure. I've been working on the new version and I think it's just about ready I'm having a few problems, though, since it's SMD and my production methods aren't exactly up to scratch yet. But today I plugged it in, it was recognized, and responded to the orb's button presses; unfortunately I chose the incorrect rs232 driver chip and it's not providing enough power for the orb's axes to be read, but I have ordered the replacement part and I may yet be able to test what remains of this board.



By "what remains" I mean the process of trying to solder surface-mount parts on a hot plate in the kitchen is somewhat dodgy, so the board is scorched, and even worse I think I connected an electrolytic capacitor backward and it turned into a smoking ember that I had to jumper with a wire. And since it's a prototype board and the connector wasn't at the edge, I had to drill a big stonking hole to plug the cable in. You can check out the fine, elegant craftsmanship below.


[attachment=0]130224131712451739093.jpg[/attachment]

...but it sorta works, first time out. The things I do for you people :) :) :)

[vputz]

Bad news for me, good news for orb stuff; the trip that was going to occupy the next three weeks was cancelled, so I'm back at work on this, and the parts for my "stabilize the hands for SMD production because darn, those things are small" platform arrived. I will be ordering another prototype for the "kit" version and if I can finish the controller for my SMD hot plate, we may actually be in business for the second (hopefully final) round of tests.

[A Future Pilot]

I'm sorry to hear that you can't go on your trip, but I'm not gonna lie I'm pretty excited that the new boards are coming along swimmingy ;) I'm looking forward to buying one when they're all finished! :)



How much will they cost?

[vputz]

Good question, pilot. I'll try and have a better idea after I get the "kit version" tested (the new chip requires SMD soldering, which most hobbyists wouldn't want to do, but I'm making an "all-SMD" board which I will assemble and test or a "just this chip is SMD, everything else is solderable" version which will cost a little less. For the kit version, I'm hoping for mid-20s; the fully-assembled a bit more as the SMD business is labor intensive at the moment, but don't hold me to that.



Timeline goes something like "new prototype order today or tomorrow, second internal testing next weekend, then boards to testers"... sorry it's taking so long--I'm learning as fast as I can!

[VDX]

... if you need a tester more with SMD soldering capabilities, I'll drop in :mrgreen:



Viktor

[vputz]

I'll keep that in mind :) Mostly the speed of all this is limited by my free time, the 5-day turnaround for prototype boards (my minimill can generally handle through-hole sized traces, but now we're in SMD land, I've got to subcontract), and shipping delays from suppliers. And to an extent my pocketbook; I like software where changing a design doesn't cost anything. I tried replacing the max3232 on my test board with a replacement and when removing the old chip lifted up some traces from the board that I don't think I can fix, so...delays: I have to order another prototype and just treat the components on that board (the processor etc) as mostly lost. That's fine--this is all about learning and it costs a lot less than a course at a university--but it does get tiring!



I did swap the max3232 for an older max chip, which [i]did[/i] increase the voltage out to the orb... so my gut feeling is that this will work and I'll order more boards for further abuse after this weekend (I'm trying to tack on a prototype for my hot plate controller so that I don't burn more boards). Hopefully that'll get the hardware done.



The new boards really are nice and tidy; 2 inches square for the "all-smd" version (the only through-hole components are the programming header and the DB9 socket) and 2.3 inches square for the "mostly through-hole" kit version.



This weekend:



"case" and maybe stencil for the boards

Build instructions sheet

Hot plate controller prototype

Prototype order

Labels for bags (you laugh; it's these little annoying details that get you)



My goal is to personally test the SMD and kits next weekend, order a round of boards (not sure if all SMD or all kits) after that. Everything takes longer than I want!

[vputz]

Good news: next prototypes are in production.



Bad news: they won't get there by this weekend, and my Windows partition died so I'll be busy anyway replacing the drive and fixing it (luckily Linux is still ticking).



So, sorry about the delays. But I have a lot of confidence in this latest iteration... I really think the last one would have worked fine if I'd gotten the parts, and this way I have the kit version as well which will be a little easier to debug. This has all been fun, but I'm rather eager to get the hardware part over with and move on to people actually using it!

[JimH]

Vic,

I'm sure I speak for most of us. we all appreciate your diligent efforts. Thank you for continuing the hard work.

[vputz]

Thanks, Jim. I just wish it wasn't taking so darn long; I'm used to the software world, where if I see a problem the only thing stopping me from fixing it is my own limitations--in other words no "well, I broke my only prototype board when the backwards capacitor burst into smoke, there's a week turnaround to order the next one" or "what do you mean I ran out of DB9 ports and have to order more?" If I could have boards and parts within an hour of needing them, this would have been done a while ago!



(I could just set up my minimill again and work through how to do double-sided boards, but I don't think it's good enough to do QFP pads for the chip anyway...)



It'll get there. The latest prototype board has a bunch of tiled layouts on it: the SMD board (which should work--everything worked last time but the wrong serial chip, and I have those now), the "mostly-kit" version (which is the same thing electrically so it SHOULD work and will be easier to toy with if it doesn't) and very importantly initial designs for my hot plate controller so I don't scorch the SMD boards in production (I already had a PID controller design that was automating my smoker and cooks a mean custard in an electric kettle, so modifying it for a hot plate seemed logical, but the only one I had with me was a 110v version and it needed a redesign anyway)

[VDX]

... on my Isel-mill I'm outline-milling with 0.025mm accuracy and 0.2mm wide grooves, what's precise enough for SMD/QFP - modified onother Isel some years ago for 5 micron precision, but hadn't done this for my old gantry at home --- but have plans and mechanics, motors and controllers for a sub-micron XYZ-stage lying around, waiting :roll:



But I can give you some IR-laserdiodes with 5Watts or 9Watts of power for engraving black paint on copper-PCB's, so you can etch even with 0.1mm wide isolation grooves, if you fix the head to your mill and 'draw' the outlines :wink:



Viktor

[vputz]

Ha! Nice. My little minimill is just a beginner's kit, if you will... it does pretty well, but it's got just enough play that I can't quite get it that precise (one problem is depth; I even have it "touch probe" the PCB with conduction to get a heigh map that interpolates, but it still varies depth too much; the Z-stage apparently has a little backlash). So I have to dig a little too deep to assure breaking the copper layer, which means QFP precision is tricky.



Instead, I'm working with Spirit Circuits here in the UK, which will get me a very good quality prototype (no solder mask, and not routed out so the edges are "inside" the board) within 5 days for free, so long as I place a paid order every 3 boards. This actually works out really well, because to a point they will let me "tile" gerbers; my last board had the SMD and kit versions of the Orbotron as well as the power and logic board for my temperature controller. I can't complain; they do great work and have been good to work with.



I like the laser idea! But honestly even if it was good enough I don't have room in this place for the mill to go up. As it is I have to unpack my soldering station and set it up in my tiny living room every time I do work, then pack it back up. I love the place, but it's basically a large 3-room apartment. I love living here, but sometimes I do miss the space to have a workshop.

[VDX]

... what's with using a flatbed-protter for laser-engraving coated PCB's?



I've got a stable flatbed-plotter with vacuum-table for 50€ and have plans to equip it with different laserheads for engraving or SLS.



Attached an image of the plotter and a small laserdiode-head+driver with a (blue) 445nm-diode emitting 1Watt of power I'll fix to the moving head.



And here i have some other diode types with wavelengths in the UV, visible and IR and powers up to 9Watts: http://reprap.org/wiki/Laser_Cutter



Viktor

[A Future Pilot]

So how goes the prototyping these days? :)

[rayz007]

Hello and thank you for your efforts. I'm in the states and recently dug up my SpaceOrb 360 and would love to get it working in Win 7 64 bit. I was an electronics

technician in the Navy, and not too scared of soldering, though I have never soldered SMC parts. I would prefer a completed solution and don't mind paying a premium

for it. Please let me know what I can do to obtain a working prototype. I have a Keyspan USA-19HS Serial to USB adapter, but of course the SpaceOrb 360 didn't work

as expected. I hooked it up via the Virtual Windows XP Mode in Win 7 and installed the HIDsporb drivers and Win XP recognized the SpaceORB, but only the buttons

worked, no axis commands. This wouldn't be nearly as cool as native support in Win 7 64bit, so I am looking forward to the completed prototype. Again, thank you so

much for your efforts. Looking forward to seeing updates.



rayz007

[vputz]

Sorry for the delays! It's going Very Well, folks. Some serious delays in getting parts put things on hold, but the parts came in and I finally had a weekend. My solder stencil order did NOT work so I still had to do this SMD reflow "by hand" (tweaking the on/off on a hot plate!), but was able to do it without scorching the board (and my hot plate controller is coming along too!)



The bad news is I missed a couple of necessary traces so I'll need another prototype which I'll order tonight or tomorrow (the blue and green wires in the pic). The good news is that as of today the "smd+kit" version works! Capabilities-wise it's where I want it to be and it's nice and small. There will be a few TINY tweaks, but this is just about it. I also found a place with affordable solder stencils, and it's worth my time to order a sheet of those even with the full-smd still in prototype.



Rayz: Ha, awesome that the hidsporb driver even worked in "Virtual XP Mode"! The reason you got buttons and no axes is because the keyspan adapter doesn't provide enough voltage to the orb (that's what forced the last prototype change here, actually). As for the new version, I will be able to do a full-SMD version and provide a fully-assembled board, but it'll take a bit longer... I actually have more confidence in that board, but I won't be able to test it until the solder stencil comes in (fixing solder bridges on one single component is annoying, but on lots it's just painful--the stencil will make it much more accurate)



So, sorry for the delays, but I'm back in action, I think I have all the parts for the next couple prototypes. One more run of protos and then its beta time!


[attachment=0]2013-03-24 17.50.09.jpg[/attachment]