IOB6120 Page

As mentioned on the SBC6120 page, Jim Kearney did an expansion board for the SBC6120.

The original boards from Jim have a problem with the battery back-up circuit. Basically, the batteries end up trying to run a TTL device, which runs the battery down too quickly. In 2009, I decided to take the CAD drawings for Jim's board, rework them a bit, and fix the battery back-up issue. I called my version 'iob9c' because it ended up being my third version of it in 2009. Once I had done that, I wanted to own one! So I got together with Robert Ribbeck and Jack Lowry to set up a group buy in the Spare Time Gizmos Yahoo discussion group. In early January of 2010, I ordered 65 boards -- 48 for the group buy and another 17 for myself.

The way that board orders are priced, there's a $450 or so setup fee, then a small charge per board. I anticipated that there might be later interest in a smaller second run, so I took the money from the group buy for their 48, and chipped in the extra money to buy some extras.

Long story short, that group buy fell apart, in a spectacular way, that still hasn't really resolved itself. At least Jack and I had done our parts. As far as I know, Dave McGuire now has those boards, and still needs to assemble and deliver them.

That led to an odd situation -- the only "available" IOB boards were the ones I had ordered for "round two". In early 2010, I had acquired components for these boards, expecting the group buy to proceed, and eventually demand to appear for "round two". That left me with a lot of money expended on IOB boards, and without an obvious way to recoup. Finally, in 2011, I decided to give up on "round one" and to assemble and sell "round two".


The IOB bare boards.

The IOB parts box.

This was actually a big deal for me, as I had never done finely detailed SMT solder work of the sort required to assemble these boards. The solder connections for the flash chip, the FPGA, and the CF connector are about 0.5 mm apart!

Fine pitch soldering required!

The back is a lot easier.

I did manage to acquire the correct tooling, requisite skills, and courage to build one:

Fine pitch soldering accomplished!

The back was indeed much easier. Electrical tape covers the "+" terminals of the batteries.

If you look closely at the pictured example, you can see that it has an Intel E28F800 flash chip installed in the upper left corner. The E28F400 flash chip is what the Jim originally used in the IOB, but they are no longer manufactured or sold, and I was lucky to get a couple of these, which are software compatible and mostly pin compatible (you have to connect an extra address line with a solder bridge).

What I actually had for the other boards, however, were the newer 29F400 chips. These are pin compatible, but they are not software compatible. That is, they will solder to the board without board modifications or jumpers, but the programming sequences used to load their initial contents are rather different, and not nearly as compatible as I'd originally thought.

The flash contains the FPGA programming as well as extensions to the SBC6120 firmware, and initial programming of the flash is done with a SBC6120 firmware command, and a long slow download:

It takes hours to load flash this way.

Since this command was actually what I was debugging, It took quite a bit of time to get that working, so that the flash command could tell automatically which flash chip was installed, then modify itself to correctly program the part. Long after learning to hate the 29F parts, I got that done and sent the firmware modifications back to Bob Armstrong.

I had recently updated my Windows desktop to Windows 7 64-bit, and the old Willem programmer I'd used before didn't work with Windows 64-bit. While watching bytes load into these boards over and over (until my code worked), I ordered a replacement for my Willem programmer, an all USB version, which was compatible with 64-bit Windows. I also ordered the correct adapter for the flash chip along with it. In no time at all, I was able to program the rest of my 29F400 flash chips, and then was ready to assemble the rest of the IOB boards, test them, and rework them until they got the green sticker:

Ramdisk check out...batteries still good.

VT52 Emulation, TOD clock...

Built and ready to go.

The boards, finally built.

These boards have been mostly sold. The original asking price for the group buy was $215, which included shipping. I got my bare boards a little cheaper, but then again had to buy all the tooling and actually build the things up, so I sold mine for the same price.

As of 12/2011, I have a total of 4 IOB boards. One has the E28F800, one has a flash chip socket with a 29F400 in it, and two have the 29F400 soldered down, like the others pictured. (Since I have two SBC6120/FP6120, my plan is to retain at the very least the two oddballs for my own use.)

Note that so far, we've built some nice hardware and played with some firmware, but we haven't made it do much yet.




Last updated on 04/30/12 02:07

vrs