PDP-8m Console Switch Problems - fixed!

[Allison]

>
>Subject: Re: PDP-8m Console Switch Problems - fixed!
>   From: Don North <ak6dn at mindspring.com>
>   Date: Thu, 07 Sep 2006 11:28:15 -0700
>     To: General Discussion: On-Topic and Off-Topic Posts <cctalk at classiccmp.org>
>     Cc: cctech at classiccmp.org
>
>I also find it somewhat suspicious in that the BOM for the PDP-8f/m LED 
>console panel calls out a plain 74175 device for these parts (rev F board
> etch). However in the three panels which I have, they are all BUILT with 
> SIG 74S175 devices
>of various date codes in 75 and 76. None of these parts looks to have been
>previously reworked (or else someone did a real nice job; hard to tell 
>sometimes).
>I find it somewhat fishy the BOM and the board have different parts; DEC was
>usually a little more reliable than this.

They have been reworked. In the 70s 74S cost enough more I doubt DEC would 
have used them in a function where speed wasn't reqired.  It's entirely 
possible someone trying to be true to date used "S" parts.  Myself I will
not put 74S or 74F in with slow 7400 and I'm wary of 74LS in the wrong places
as well.  I've been bit by those intermixes in the past, especilly when
the part was used as a oscillator, oneshot or other variations of normal 
logic.

In DEC the BOM was it and ECOs were required to deviate. 

>> Put all the parts on a header and plug it in rather than mess the board up.
>I plan to do that, just replacing the two S175 positions with sockets, 
>and I'll
>dead bug the resistors over the top of the S175s. That way, someday when 
>I come across some of these 'magic' 74x175s :-) that work without pullups
> I can replace them easily.

I find the _need_ for pullups worrying.  Despite the IC tester there is 
something odd going on.   It could be the parts pass dynamic but not static
parameters.


Allison