July 1st, 2013, I finished installation of Harald's USB board into HP30b. It was quite a challenge for me, but now it seems to be over.
I started with crystal modification so I can have something I can try and test quickly. I must confirm, those capacitors are really tiny indeed. I was gently holding one of the capacitors with a toothpick by pushing it towards the board when toothpick slid off and sent it flying into unknown direction. It took me about hour and a half to install capacitors and crystal. I reassembled case, installed batteries and enabled crystal with key combination.
Couple of days later I continued with USB board installation. I followed directions in the manual. There are a few notes I would like to make.
First, there is not nearly enough emphasis on attaching the board to the case. It is, in my opinion, the most critical part -- if something goes wrong it is difficult or nearly impossible to correct. Case in point: to attach the board I used polyurethane glue (aka gorilla glue). I smeared thin layer of it on the bottom of the board, pressed it to the case with couple of clothpins and left it to dry for a day. Next day I attached LED and made wire connections to the board. It took me another couple of hours. When I tried to put the case (with new holes in it) back on I found out that that extra layer of glue between case and board, though very thin, is still enough to raise connector beyond the flat ledge of gray plastic case, where it would interfere with shiny black battery cover.
I decided to ignore it for now and try to connect the usb cable without the battery cover. Here is where second surprise came in: the overhanging "cliff" of the gray case gets in the way of micro USB connector housing. Maybe there are slimmer micro usb connector housings out there, but the ones I have are all rather thick. I shaved off some of the "cliff" to let the connectors reach each other, but it still did not feel like it went all the way in.
That is when the scariest thing happened. As i was pushing on the connector, trying to connect them, the board inside the calc came loose. It was still sitting on a central screw post, but it was freely moving around it, no more adhesion between board and case. To fix it I had to remove the glue first, then apply new glue in a different way. There were almost no glue remains on the case, I guess the plastic it is made of is not good for gorilla glue. But there was a lot of glue on the bottom of the board and it did not want to come off easily. I ended up scratching it off with a small utility knife, tiny bit by tiny bit. Very long and tedious process. With the wires still attached to the board and the board being pretty small and difficult to hold, it was an experience I would like not to repeat. I probably should have removed wires first. Putting them back should not be too difficult as soldering pads are pretty big compared to tiny spots on the main board.
So, I cleaned as much glue as I could, got some clear plastic glue (E6000 if it means anything to anybody) from the hobby store, squeezed some of it out to see what it looks like when it hardens and next day (which was yestareday) I glued it again. This time I tried to put it as flat on the case as I could. I removed all remains of post supports that could get in the way, cleaned off all the plastic shavings, pressed the board flat to the case and poured glue on the board around the screwpost. It was pretty viscous and did not want to flow well, I can only hope that it will hold. (BTW, quite a few of the plasic shaving get underneath the LCD. I opened it and cleaned it up before gluing the board. After the board is glued and wires are connected, there is no way to get under the LCD.)
After that I put everything together, plugged in USB cable, communicated with emulator first (RECV/SENDA). It worked, I celebrated, then I flashed IR software, it went in, I celebrated again and finally, tried to print rX with [f] [EXIT] and the screechy sound of moving printer head was as sweet as I've ever heard it.