Re: Kits received -> HW question

From: Jim Brain (
Date: 2008-01-21 07:16:14

Mike Naberezny wrote:
> Jim Brain wrote:
>> The parts kit has 2 20x1 header pins and 2 40 pin sockets.  The idea 
>> was:
>> Solder the 40 pin socket on the board.
>> Solder the 20 pin headers to the bottom side of the board to the pins.
>> Plug the second socket into the bottom of the pins (or solder it there.
>> Plug the lower socket into the 1541
> I received my board on Friday and assembled it today.  Thanks to all 
> who have helped realize this project, particularly Ruud, Jim, and Al.
> The holes for the 40-pin socket appeared to be a bit smaller than the 
> others.  I could not get the 40-pin socket included with the kit to 
> fit in these holes easily.  I enlarged the holes slightly and then a 
> socket with machine-tooled pins just fit.
> Soldering the headers directly below the socket worked for me as you 
> described but I think some people with less experience soldering will 
> find it challenging.  If there is a future revision of this board, 
> putting parallel rows of pins just for the headers would make it a 
> little easier.

I assembled my board this weekend, and did some testing against the 64.

First, a found a bug with the board.  Looking at the schematic, note 
that the 74LS245 G line is connected to  /(SEL * PHI2).  Well, for the 8 
bit users, that is fine, but a 16 bit user will find that the 74LS245 is 
active for all addresses in the address range.  In fact, it should only 
be active for addresses where A4 is low.  The solution is to gate the G 
line with /A4. but that requires an OR gate (when they are both low, G 
should be low).  I temporarily solved the issue by using the spare 
74LS244 and a pullup resistor on the G line:

   1. Cut power trace to pin 1 of the '244
   2. Cut trace from pin 1 of '244 to the resistor in front of it.
   3. cut trace from pin 15 of '139 to pin 19 of '245 (the best place is
      between the '244 and the '139, it's the trace on the top right
      next to the caps by the edge of the board)
   4. wire '139 pin 13 (A4) to '244 pin 1
   5. wire '139 pin 15 (gated SEL) to '244 pin 6
   6. wire '244 pin 14 to '245 pin 19
   7. attach 10K pullup on '245 pin 19. (it seems to work fine without,
      during initial testing, but YMMV)

Now, that all said, I am noticing that the /R strobe is being sent too 
often to the board from the 64.  This may not be a problem with the 
1541, I don't know.  I am getting back a sector 0 from the board, but I 
am missing bytes.  The AA55 signature (which is supposed to be at 
0x01fe) shows up around $140 or so. 

I am going to add Ruud's '74 PHI2 gate idea and see if that helps.


       Message was sent through the cbm-hackers mailing list

Archive generated by hypermail pre-2.1.8.