Re: 7712 info

From: Clockmeister <clockmeister_at_internode.on.net>
Date: Fri, 19 Jun 2020 20:36:53 +0800
Message-ID: <cb64ec4a-67bb-e59b-b11f-0a7d5951f0f3_at_internode.on.net>
On 19/06/2020 5:48 pm, Gerrit Heitsch wrote:
> On 6/19/20 11:23 AM, Clockmeister wrote:
>>
>>
>> On 19/06/2020 4:00 am, Francesco Messineo wrote:
>>> On Thu, Jun 18, 2020 at 9:49 PM Gerrit Heitsch
>>> <gerrit_at_laosinh.s.bawue.de> wrote:
>>>> Yes, but often, if you have a dead C64 with a lot of 77xx chips on it
>>>> it's one of those that is the problem (unless it's the PLA, but 
>>>> that is
>>>> also known as '7700-xx' or '8700-xx').
>>> I've repaired dozens of C64 :)
>>> Once I've found a C64 that had almost all 77xx logic chips and one by
>>> one I had to change them all to revive it.
>>> I don't desolder parts before making sure they are really bad usually.
>>> I've never said they're reliable, but they aren't either all failed by
>>> now for sure.
>>>
>>> What is really "disturbing" in the MOS reliability issue is that it
>>> appears to  depend on the part number too.
>>> Either some parts were made in different farms, so the less reliable
>>> farms were producing the less reliable part numbers
>>> or it was heat-related maybe (like the hotter parts failing more 
>>> often)...
>>> I tend to believe it was a "farm" issue.
>>>
>>> Frank
>>>
>>
>> Sorry Frank my previous replies were meant for the list not just you 
>> by email, apologies for that.
>>
>> C64 DRAM has a high failure rate but that's true of DRAM generally in 
>> all my vintage equipment these days it seems (yes some brands fare 
>> much worse like mT).  I rarely had to replace was the 77xx logic on 
>> any Commodore computers though, I would say about the same rate as 
>> for other systems that used 74XX logic.
>
> I had a lot more dead 7708 (74LS257), 7709 (74LS258) and 7707 (7406) 
> than their 74xx equivalent.
>
>

There you go. Not my experience but clearly a pattern there in that event.

>> It is however my opinion that power supply ripple, not only over 
>> voltage is quite capable of damaging IC's in the long term including 
>> DRAM and C= supplies were notoriously rubbish back in the day and 
>> haven't gotten any better with age.
>
> A 78S05 (as used in the C= supplies) doesn't produce much ripple 
> though, at least as long as the large capacitor in front of it is 
> still within spec. You get a lot more from any switching regulator.
>
>  Gerrit
>
>

It's the capacitor that is the issue. On the potted supplies there isn't 
much option to fix that issue which is why they are only suitable for 
the bin.
Received on 2020-06-19 15:02:06

Archive generated by hypermail 2.3.0.