• Please review our updated Terms and Rules here
  • Exhibitor application for VCF West 2022 is now open! If you are interested in exhibiting, please fill out the form here.
  • Here are the results of the VCF East 2022 Post Event Survey: Survey Results

Possible Project: Simple Serial Board

glitch

Veteran Member
Joined
Feb 1, 2010
Messages
4,964
Location
Central VA
I've been using my Morrow SwitchBoard for some simple serial I/O tests lately, since its serial ports are WD 1602 UARTs and require no software initialization. After you set the appropriate switch settings, you can just push bytes into the correct I/O port and the UART will shift them out as serial data. This makes it really nice for debugging, as there isn't an init sequence as there is with the Intel 8251 or Zilog SCC. An added bonus, the SwitchBoard provides both current loop and RS-232 signaling!

I've located a possible source of cheap WD 1602 UARTs and I am considering building a board with a single UART, mainly for debugging purposes. Would anyone else be interested in seeing such a board get turned into a proper PCB layout? I'd probably include 20/60 mA current loop drivers and RS-232 drivers, as well as a FTDI serial-to-USB chip (these work at the TTL level with the UART, and are much more reliable than USB <> Serial converter cables). The board would include an onboard baud rate generator with a dedicated crystal (discrete TTL, no hard-to-find BRG chips) so that it doesn't have to rely on a possibly faulty S-100 bus clock. Looks like interrupt circuitry would be easy to provide, too.

If the UART circuitry turns out to be far too little to fill the board, I could also include simple I/O port driven LEDs/input switches (a la IMSAI front panel) or a DL-2416 ASCII character display. This would recreate a lot of the functionality of my S-100 debug board ( http://www.glitchwrks.com/vintage/s100/debugboard.html ).

Please provide input! Maybe this isn't something anyone else really wants, with the mix of vintage serial boards out there, and the excellent (albeit marginally harder to program) S100Computers/N8VEM Serial IO board.
 

Chuck(G)

25k Member
Joined
Jan 11, 2007
Messages
38,870
Location
Pacific Northwest, USA
GI and WD both had versions of the TR1602.

I was very happy to discover the Signetics 2650--a genuine programmable USART with integral baudrate generator. Far less quirky than the 8251A and in the same space.

No, I don't miss the 1602 at all.
 

NobodyIsHere

Veteran Member
Joined
Dec 21, 2006
Messages
2,394
Hi, There are a whole big stack of the S-100 Serial IO board PCBs left so if anyone is interested in those please let me know. True, the SCC does require some initialization but has its own clock source. Also the board can be dramatically simplified by only installing those component necessary for the dual serial port function. Just leave the USB and voice synthesis sections as "do not populate". It does make for a nice debugging board though. I accidentally ordered two batches of the S-100 Serial IO PCBs which is why there are so many left over (my fault). There are at least 30 PCBs remaining.

There are probably at least two dozen of the S-100 buffered prototyping boards and 8-10 S-100 EPROM boards left over. I am glad to offer the boards to the community but there has been a growing problem with people asking for boards, getting on the waiting list and then bailing out or disappearing once the boards arrive. The exceptional one or two I can handle but its become about half of the boards recently. Needless to say, I am reluctant to make/order more PCBs while there is this large stack of unclaimed PCBs. My goal is only to break even on the PCBs but even that is not working recently.

Thanks!

Andrew Lynch
 

glitch

Veteran Member
Joined
Feb 1, 2010
Messages
4,964
Location
Central VA
Yes, the Serial IO is pretty easy to initialize -- you can usually just pass it the same general init string used in John's test code. It does require a clock from the S-100 bus in its unaltered state though, which is why I've been using the SwitchBoard (no clock other than BRG required).

I'd been having the same problem with people saying they wanted something, then not following through after purchases had been made. That's why I've started requesting preorders on components I don't want to keep around for myself (like the HP hex display deal from last summer). I wish I didn't have to resort to it, but getting stuck with a bunch of inventory that you've paid out of pocket for sucks.

Andrew, if you're not opposed to it, perhaps I can assemble and test some of the Serial IO boards and sell them off as fully tested/working. That seemed to move XT-IDE boards pretty well. Even just providing parts kits might help move them. Selling the boards without the voice synth module or USB port would reduce the price considerably. I definitely don't want to see fewer N8VEM boards available in the future because people balked on their orders.
 

monahan_z

Experienced Member
Joined
Oct 19, 2008
Messages
260
Location
San Ramon, CA
Glitch, just want to mention that the clock signal going to pin 20 on the Zilog SCC can be more or less any value (so long as it less than the BRG clock rate). As I remember it the frequency is not critical (but check the docs). It only purpose is to "move things along" within the chip. I have two jumpers on the board going to pin 20. One the phi signal (S-100 pin 24), @2,4,8, or in my case 10Mhz for an 8086, the other the 2Mz signal on S-100 pin 49. All S-100 systems should have both signals and if you are using a phi <= 4MZ either can be used. The frequency going into pin 20 does not require any software changes.
 

glitch

Veteran Member
Joined
Feb 1, 2010
Messages
4,964
Location
Central VA
Glitch, just want to mention that the clock signal going to pin 20 on the Zilog SCC can be more or less any value (so long as it less than the BRG clock rate). As I remember it the frequency is not critical (but check the docs). It only purpose is to "move things along" within the chip. I have two jumpers on the board going to pin 20. One the phi signal (S-100 pin 24), @2,4,8, or in my case 10Mhz for an 8086, the other the 2Mz signal on S-100 pin 49. All S-100 systems should have both signals and if you are using a phi <= 4MZ either can be used. The frequency going into pin 20 does not require any software changes.

That's correct, it can be pretty much anything less than BRG clock. Still, if there isn't a local oscillator on-board, one can't always assume that the S-100 bus will be providing it if a board is faulty. Wouldn't be much trouble to tack one on, though!
 
Top