Monday, April 12, 2010

TEasy Mod Project, Part 4, the balding continues...

Hooray, got a few things in the mail today, one of which was unexpected.





The board came out well, and as a total plus, the drill holes are pretty spot on..almost a little big. Although one thing I noticed was that I never adjusted the end header annulus size for the through holes. Take a look, it's like soldering vias:


Got a little nervous seeing that, so the first thing I did was sit back down at my computer and fuck around with the size of those pads. They're usable, but it's be less of a hassle if I had a bit more room to work with. I ended up with this:


If you look at those in comparison to the original, you'll see they're much bigger. Obviously this won't pass a DRC, but for now I don't give a shit, I just wanted to fix that up before I forgot about it. There's a few other things on the obvious to do list, one of which is to add silk screen to the USB section of pins, somehow I missed that one.. oh well, can't catch everything.



The board put together, but again I missed something. As you can see I needed to cut the shrouding on the header to accommodate for the stand offs. I ended up just taking down that entire front wall to be sure I didn't run into any weird spacing issues. I'll be sure to lengthen the board just a bit, or just hang the header off the side more to give myself some breathing room on the side of the TE PCB.


Put together with the shrouding cut, fits like a glove that's a size too small. And what do we have here? Those are flanged spacers that I looked for for a very long time, turns out the same place that carries the standoffs has these on hand...and for a reasonable price too :).

They are put into the case before you install the board so that it's raised up a little bit. In my first post, I mentioned the length of the pins and what effect they'd have on this whole ordeal. Well, it just so happens that I needed to make some modifications to the pins themselves, as well as raising the board another 1.5mm or so.



The first picture shows the spacers installed, they're a #10 screw inner diameter (.2") to fit around the little lip that the plastic stand offs built into the TE have, FYI that lip is roughly 1.5mm tall. The second picture shows my modified receptacles, they have the crimp tail cut off since 1. we don't need that, I'm soldering them in place, and 2. without doing that this just won't fit. There's only about 9mm of clearance under the board and the overall length of the receptacles before cutting is around 11mm.

This is part of the reason why I haven't locked a price yet, because I'm going to try to get the factory to custom make me these receptacles so that I don't have to dremel thousands of little pins. The last picture shows it installed, as you can see I've designed it so that you can use the stock screws that are already in there, allowing you to install this with only a screw driver... awesome.

That center'ish hole is for a PCB support I have yet to decide on, you'll see why I've allowed for it as a just in case.


After tinkering around with it, and finally getting it clipped in I found that the board still has the bending issue as I saw before. Thus we need a board support, I think I know how to make that work...but it may require going to a company whom I believe charges more than most do for simple shit like male/female hex standoffs. At least that's what a co-worker tells me, we'll see.

I'm not terribly happy on how it went together, it looks okay in the final picture, but there's some weird alignment issues going on that will probably take me the most time to fix. I'm hoping it's a simple fix, but honestly I have no idea how deep that hole will go. It could be a measurement issue, it could even be the soldering of the receptacles. Since they're not super tight in the holes they have some room to move, which causes inaccuracy on the other end. Again, this is all part of the R&D process, but man is it a pain in the ass.

After checking continuity, I decided it was time to try my hand at crimping ribbon cable! I didn't do a terrible job for my first time, it was actually kind of neat.


I matched up everything, and made sure that Pin 1 on the shrouded header was the marked side (it's usually on the side with the little cutout for keying). I really wanted to make sure my pinout was correct, so I crimped the other end with what is probably the most exciting part of this whole thing:


Yeah, that's a direct ribbon cable connection with the Dual Strike! I made sure that the keying was correct on his shrouded header, as it turns out, he's got it turned backward. There's a little arrow pointing to position 1 of the header, and that's supposed to be the front left pin (VCC), but instead it's on the other side. No big deal, I just flipped the connector, crimped it in place, and voila...we have a direct connection.

If I wanted to, I could have actually made a USB B end to the stock cable, plugged it in, and would've been done right then and there with a dual mod on this stick. However, as exciting as that is, this is still all testing..so I didn't want to get ahead of myself. Also, I found that with the length of the stock cable, pointing the dual strike the other way would probably work out best in the TE. However, I didn't leave myself enough ribbon cable to do that, so I ended up with a pseudo-install, lol.


You get the idea though, you'd run the USB cable under and around the TE PCB, and the length of the cable would be perfect for plugging into the front of the Dual Strike.

Alright, I'm tired, and for now, I'm done. More testing will be done this week to try and hammer out how to deal with all these minor details. I believe I'll do one more round of prototyping just to be sure, since I'd rather make sure and spend less than 1/3 of what I would if I went full production and found something still wrong with the design.

The problem is that I've got 4 more boards that have these minor issues, but they're still technically usable, but they take a lot more tinkering than they should. I also don't have any more of those edge holding standoffs, I'm still waiting on the rest of my samples. I hope they get here soon so I can send Jochen (the nice guy that sent me the Dual Strike for testing) a couple of my prototypes so he can take a look and see if he spots anything I missed.

3 comments:

  1. > I needed to cut the shrouding on the header to accommodate for the stand offs

    LOL, really great: Had exactly the same issue on first version of Dual Strike.

    > Again, this is all part of the R&D process, but man is it a pain in the ass.

    Time to make a break and get the head clear again :)

    > There's a little arrow pointing to position 1 of the header, and that's supposed to be the front left pin (VCC), but instead it's on the other side.


    Jep, and this was second fault in dual strike v1 ;) The eagle lib part is messed up here. When I`ve time I correct the part labels.

    > I hope they get here soon so I can send Jochen (the nice guy that sent me the Dual Strike for testing) a couple of my prototypes so he can take a look and see if he spots anything I missed.

    Would be great, but don`t stress you :)

    Ehm, does it work? I think I`ve updated Dual Strike to V1.6.

    Really great work. That`s really awesome stuff!

    ReplyDelete
  2. Phreak, sorry for being nagging but you said your baby wouldn't be out before end of june/debut of july ? Thats a bit late for me :D

    Is there any hope you manage to release it earlier :D (by the end of the month) ?

    ReplyDelete
  3. Well, as appreciated as the enthusiasm is, I'm pretty positive I wouldn't have a production run by the end of the month.

    However, with any luck, I'll have a second round of prototypes figured out by next week. I'm taking a little break at the moment to let my brain digest the information I've gathered so far. If I do, then I could potentially have working prototypes early next month.

    The biggest concern right now is making sure that my edits to the board are as 100% spot on as possible. Another order will set me back another 90 bucks for boards alone, so accuracy is a must. I figured these first prototypes would be off a little bit, but I've realized quite a few mistakes that basically make this batch a throw away, I don't want to make that mistake again. Luckily I can use these boards to make sure of that ;)

    ReplyDelete