Posted on

Parts Availability

One of the difficult parts about building accessories and reproductions of vintage machines is that some of the parts are getting more difficult to find.  Over the years I did few prototypes using modern CPLDs and FPGAs but always opted to stick with 74xx series parts because then anyone could understand my designs and fix them if something went bad.  It was also nice to use parts that were available in that time frame.

More and more of the parts I use are getting difficult to find in the surplus market, some of them are going up in price, and others aren’t available in small quantities at reasonable prices.  This is forcing me to reconsider going to a modern CPLD for some designs.  A modern CPLD is in the $1.50 range, which can replace a few parts that cost $5 or more, and are readily available from many distributors.

During a recent session of building SS-50 boards, my supply of female Molex connectors ran low.  That’s usually not a problem because several vendors normally keep them in stock, but when I went to order more, all but one of my supplies said they will no longer stock that part and had none in stocks.  The one remaining vendor had none in their inventory but will be getting more in late February, so I ordered more than 300 of those connectors.  This is a concern because there are no second sources of this very old connector, so if the manufacturer stops making them, the available supply will disappear.

I normally have a large stock of assembled products ready to go, but until the new connectors arrive in about six weeks, some SS-50 products won’t available once the existing stock is sold.  When the connectors appear, I’ll start building boards and increase the inventory of assembled products.

Posted on

Workbench Notes

My full time job is keeping me very busy, but there is still some time during the week to squeeze in some Corsham Tech fun.  This post isn’t about any one thing, so bear with me.

Over the years I’ve accumulated a number of oscilloscopes, always used, usually from hamfests, computer shows or friends’ basements.  Yes, I do like the old Tektronix scopes, but after using modern scopes at work for many years, I finally decided to break down and buy a new Rigol scope.  I’m 55 years old and finally bought a new scope!  It was very handy the last few days debugging a new version of the SS-50 motherboard (more on this later).  Frequency counter, I2C decoder, and four channels… I should have done this a long time ago!

For those of you who with our SS-50 systems, we’ve been through a lot of revisions of the motherboard.  Version 1, 2, 3, 4, then 4A.  As each was finished it seemed like a perfect product until either a small bug was found or a new feature was needed, like when it was modified to support both 6800 and 6809 processors.  Rev 5 is the latest and fixes all known problems and does everything it needs to do.  It’s still going through additional testing but does support 6800 and 6809 CPUs.  The only big remaining test involves NitrOS/9 interrupt support.  This was the problem with the rev 4 motherboard.

During testing, it worked fine for the 6800 but gave very unpredictable results for the 6809.  Hours were spent reviewing the schematic but nothing seemed wrong.  Fortunately I had a new scope to work with, so after some debugging it came down to one IC not being quite fast enough for some of the decoding.  The scope showed it was plenty fast for the 1 MHz 6800 but a bit slow for the 2 MHz 6809 and the tighter timing.  One chip was tossed into the trash, a new one taken from tube, and the motherboard sprang to like and has been running memory tests non-stop for over 50 hours.

Most of my software work recently has been on the 6502, including writing a Tiny BASIC from scratch, and bringing up fig-FORTH on the KIM Clone.  I also want a better 6502 debugger and have been spending a bit of time on that.  I am technically a software engineer and only dabble in hardware so I have something to write code on.

Oh well, the soldering iron is hot, so it’s time to get back to the workbench…