Help with getting an original Altai running: 8080A clock circuitry and MITS88 2-SIO board

drlegendre . drlegendre at gmail.com
Sun Feb 7 20:05:10 CST 2016


Pardon this, it was meant to be part of the previous message..

Kip, a modern switching supply won't really win you anything, and IMO, kind
of wrecks the look (and the vintage fun)..

Again, the only regulated supplies produced in the chassis are the +/-18V
and they are limited to 1A. I believe the original design calls for zener
regs controlling pass transistors, but if you're looking for an upgrade,
it's quite reasonable to replace them with the 7818 (+18) and 7918 (-18)
three-pin regulator packs. No muss, no fuss.. built in current and thermal
limiting, and cheap to boot.

And the ~8V supply is a simple one-stage design. Xfmr secondary -> 30A 50V
bridge -> 95,000uF 3" computer-grade bulk capacitor filter.. that's it. At
the most, you'll need a new cap and a $5 bridge pack. Regulating it gets
you nothing, as each card has at least one independent 5V regulator. For
example, my 16K SRAM card actually has 4x 7805 5V regs, one for each of the
four 4K banks.

On Sun, Feb 7, 2016 at 7:51 PM, drlegendre . <drlegendre at gmail.com> wrote:

> Kip,
>
> Assuming the initial visual inspection goes well, the PSU is +always+ the
> first thing to check.. and that goes for everything from digital computers
> to vacuum-tube radios.
>
> Do you have a general background in electrical / electronic
> troubleshooting? Do you have a Variac on-hand, or at least a
> current-limited AC supply - such as the old light bulb socket in-series, as
> was popular in the days before Variacs became affordable to the common man?
>
> On Sun, Feb 7, 2016 at 6:15 PM, Robo58 <robo58 at optonline.net> wrote:
>
>> Hi Brent, Drlegendre and Eric
>>
>> Brent:  I took your advice and did a little program to output "A"'s and
>> monitoring Transmitter empty before looping again.  That works fine and it
>> will run for hours.
>>
>> Drlegendre:  Thanks for the excellent document link.  My buddy has the
>> MITS
>> 1K static ram and the MITS 4K dynamic ram cards.  I've been doing my
>> testing
>> with the 1K static ram to limit potential problems.
>>
>> Eric:  This Altair has the smaller PS with the smaller Electrolytic caps.
>> They were all replaced and the PS brought up without cards with a Variac.
>> My buddy bought Kemet Capacitors on the basis of very good ripple current
>> rating and low ESR.
>>
>>
>> Thanks Ron
>>
>>
>>
>> -----Original Message-----
>> From: cctalk [mailto:cctalk-bounces at classiccmp.org] On Behalf Of Brent
>> Hilpert
>> Sent: Sunday, February 7, 2016 4:14 PM
>> To: General Discussion: On-Topic and Off-Topic Posts <
>> cctalk at classiccmp.org>
>> Subject: Re: Help with getting an original Altai running: 8080A clock
>> circuitry and MITS88 2-SIO board
>>
>> On 2016-Feb-07, at 6:17 AM, Robo58 wrote:
>>
>> > c). I hand assembled some code to exercise the 6850.  I know it has a
>> > software reset and then you set attributes.  I tried my buddies echo
>> > program.  It looks for a received character and then echo's it.  I'm
>> > using a laptop and PuTTy along with an RS-232 breakout box.  I can see
>> > characters going in but nothing coming out.  The 6850 appears to drive
>> > the bus for a finite amount of time and then turn off its drivers.  I
>> > don't have any experience with the part and I don't know if that's the
>> > way it's supposed to work or that the Altair has a bus timing issue.
>>
>> How about doing a program that simply repeatedly outputs a character
>> first,
>> getting that going before trying to deal with input - cuts the problem
>> space
>> in half.
>> The repeat loop should give you some consistent bus activity and
>> observable
>> waveforms on a scope if you have one, rather than trying to catch events
>> initiated from external input activity.
>>
>>
>>
>>
>>
>


More information about the cctech mailing list