8051-class "learner" board - is it worth making one?

This is a crosspost from comp.arch.embedded, with some enhancement.

The story: I semi-accidentally acquired a fairly large quantity of surplus ROMless

8051-class parts in DIP-40 packages. These include Intel and Signetics 8031, 8032, and lots of Dallas DS80C310 and DS80C320. Along with them came a bunch of 6264 and 62256 SRAMs.

Somewhere in my archives I have a layout for a board that takes a DIP-40 8031 and has 32K of program flash and either 2K, 8K or 32K of RAM. It also has a level-shifted serial port, and some miscellaneous headers for GPIOs and such.

In this day and age, is it worth my while to do a production run of these boards and offer them for sale to get rid of the surplus chips? I would write a small bootloader and preload that into flash so you wouldn't need additional hardware to load code onto the board.

I figure I could [afford to] sell an assembled board for ~USD35.

When I posted the above in c.a.e I got some suggestions, including to ask the question again here.

Is that old design of mine even useful as it is, or would I have to build something more exotic to make it interesting?

Reply to
zwsdotcom
Loading thread data ...

enhancement.

ROMless

Signetics

miscellaneous

chips? I

including to

Personally, I would offer the chips for sale on eBay or something like that to see what you could unload. You could add a note at the bottom of the auction ad asking people to email you with any interest in the production run board.

On the other hand, if you just want to create a product that you think is useful and better than others, I say go for it! Kind of a personal decision ...

JCD

Reply to
pogo

Hi.

I have a shelf full of 8051 boards and chips, which is probably why I am the only one in my club using 25-year-old technology in a robot. I wanted a board with a serial boot loader, where I did not have to re-vector the interrupts from the boot loader. I designed a small board that ANDed PSEN and RD, and used a TTL signal to swap between the boot loader (EPROM) and RAM at location

0x0000. That way I can load code, toggle the signal, reset and run the code from RAM.

I did finally find a handful of boards I could modify to work this way, and used them instead of my own design. It was also nice that these boards had a hook for a memory mapped device (8255, quadrature decoder, etc.), and could support a few different RAM and EPROM sizes.

I am not sure how much of a market there is for "vintage" 8051 boards. I am in the process of porting some of my stuff to ARM, and have started to really appreciate the newer 8051-core offerings from companies like Atmel and Analog Devices. The extra peripherals are great, and it is nice not to have to use I/O ports for external memory. I might still be a customer if you go ahead with the project, and your board supports code developed with SDCC. I would probably be more inclined to buy bare boards, however, and then probably only two or three.

Jeff.

Reply to
Jeff Shirley

I purposely bought 30K of them in PLCC to offer a cheap 8051 board. That was like 10 years ago. They're mostly still here.

We sold/sell them at $39. We've offered specials at $29. They don't move very well. Probably our best documented boards, with many options for language.

So my opinion is the only way I'm going to move this inventory of 8051's is to design some little comsumer product that doesn't require much smarts, and sell them embedded in it. Even that is iffy, because the glue logic to go with the IC and larger board to hold it all costs as much as an ARM now.

Reply to
Randy M. Dumse

This is what I was afraid of. Thanks for the response.

Reply to
zwsdotcom

PolyTech Forum website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.