Okay - sounds like a plan is coming together! I'm going to order a few things I need tonite to be ready to get building soon.
Brian: I'd love to have a couple of those sockets. Also, couldn't the CPLD just be programmed before populating R3? Finally, I need to get a DOS loaded on my M102. I've never actually done this. I'm actually not 100% certain of the health of my 102 right now as it kept cold-starting just loading TEENY.CO last time, but that could have been operator error. I'll play with getting TS-DOS loaded tonite. -Josh On Tue, Sep 19, 2017 at 8:17 PM, Brian White <bw.al...@gmail.com> wrote: > Now what would be REALLY great, is tweaking the existing eagle files to > provide a more convenient way to pull PORT_EN high for xilinx programming. > > The simplest procedure I came up with is to get the software all ready to > go on the laptop so you just need to press enter, clip a jumper onto the > vcc pin to the programmer, and touch the other end to one side of one of > the resistors. It only requires one hand to hold the probe on the resistor > if the other end is clipped to the vcc pin, so you have a hand free to tap > the enter key. > > The spot that you have to touch is tiny, and pretty close to other spots > that you absolutely should not touch, since the result would be to short > out the on-board 5v-to-3.3v voltage regulator. Touching the wrong side of > that very resistor would do that, and that's easy to do. > > It would be great to have a dedicated pad to touch a probe to. Could be as > simple as simply extending the existing resitor pad a little on that side > to make a better target to touch. Or in a luxurious world, a set of pins to > stick a jumper onto like the write-enable jumper on figtronix, but there is > precious little real-estate available for that. > > Clipping the lead onto the vcc pin is a little finnicky too. There is > just barely enough pin length to contact the board, the clip, and still let > the programmer wire grip the pin. > > Maybe it would be possible to insert a small through-hole for both VCC and > PORT_EN, so that you could just stick a wire or pre-made jumper lead in > both holes. Maybe you'd have to pull sideways on the wire a little to > maintain solid contact while programming. One advantage to that way would > be you could put the wire in place while there was no power to the board, > so that there is no danger from touching the wrong parts on the way in. > > And it would be nice to have github hosted versions of the cad files so > that you could have separate versions for 100 and nec. That way the 100/102 > version could come with the half-holes already cut to the right size by the > router path instead of having to sand each board down by hand without > wrecking the board. > > -- > bkw > > > On Tue, Sep 19, 2017 at 7:40 PM, Brian White <bw.al...@gmail.com> wrote: > >> That would be awesome. >> >> One update to the pics and stuff I have so far, During the xilinx >> programming stage, I used a chip carrier socket in a breadboard, powered by >> a 5v wall wart, because I happen to have a bunch o those sockets. I'll send >> you a socket or a few if you want, but I would say, ditch that and just use >> a M100 itself to power the board for xilinx programming. You need a M100 to >> do the final flashing anyway, not to mention testing, so it's a no-brainer, >> just pop it in the m100 and let the m100 supply the 5v to the board. >> Perfectly fitting socket and 5v supply ready made. >> >> >> On Tue, Sep 19, 2017 at 4:48 PM, Josh Malone <josh.mal...@gmail.com> >> wrote: >> >>> I'm willing to take a stab at building a REX. I think the only tool I'll >>> need to acquire for assembly is the proper Hakko tip for drag-soldering. >>> However, I don't have the Xilinx programmer, so I'll have to get my hands >>> on one of those. They're <$30 shipped from China but that puts it in my >>> hands in around a month, so nothing will be quick. (Darn - would have made >>> a great RetroChallenge entry :) >>> >>> My disposable income these days is significantly less than my available >>> time, so I'll just throw this offer out there: >>> >>> If 1 or 2 persons want to help cover the initial costs of the parts and >>> tools, I'll build them a REX. >>> >>> A quick estimate says a run of 3 boards will cost me <=$140. I'm, of >>> course, willing to throw in some of my own money, so I'd be looking for $50 >>> per board from 2 contributors. >>> >>> I've done lots of soldering and even some SMT assembly (though not this >>> fine) before, so I'm pretty confident I can pull it off. I'm not sure how >>> long it would take to get everything together and then actually build it. I >>> think the risk of failure is pretty low, but you're not willing to see your >>> $50 disappear in the event that I completely screw this up, don't >>> volunteer. :) >>> >>> If you're still interested after reading this, please contact me >>> off-list. In the (possibly likely) event that I get >2 interests, I'll just >>> choose randomly from the pool; I'm not willing to commit to >3 units until >>> I've done this a time or two. I'll post on this thread when "orders" are >>> closed. >>> >>> Thanks, >>> >>> -Josh >>> >>> >>> On Tue, Sep 19, 2017 at 2:46 PM, John R. Hogerhuis <jho...@pobox.com> >>> wrote: >>> >>>> >>>> On Tue, Sep 19, 2017 at 11:42 AM James Zeun <james.z...@gmail.com> >>>> wrote: >>>> >>>>> I'll ask around here, I actually might know someone with experience in >>>>> this area lol >>>>> >>>>> My next questions might sound silly, but I'll ask anyway. >>>>> What components needs putting on the rex? >>>>> What excludes it from being a home built project? >>>>> >>>>> I'm asking so I can understand it properly in the event I need to >>>>> explain it :-) >>>>> >>>>> >>>> >>>> Brian documented it here: >>>> >>>> http://tandy.wiki/Building_a_REX >>>> >>>> >>>> >>> >> >
