Apple 1 terminal section crashes soon after boot

7 posts / 0 new
Last post
Offline
Last seen: 10 months 1 week ago
Joined: Oct 9 2023 - 14:36
Posts: 12
Apple 1 terminal section crashes soon after boot

Hello,I'm building an apple 1 replica. I got the terminal section to output video (yay!). The output however does not seem entirely correct.For one the screen is all @'s. Am I correct in my understanding that it's supposed to be all "@_"? Or is it just completely uninitialized?Second problem is, the screen will sometimes blink a a few seconds and then freeze, or just entirely freeze on boot. The times that it does blink it seems I can clear the screen. This gives me the impression of either a component warming up and then failing, or some kind of weird latch up problem. (Or a counter that reaches a state where a pin isn't working?)I figured I'd ask for pointers before I dive into the schematics all by myself head first in the weekend.Does anyone have any clues where I should start looking? I do actually have a pretty reasonably well equipped hobby electronics lab so I can take proper measurements.Kind regards,Jero32

 

edit: One thing I forgot to mention. The picture seems to increase in size whenever it "freezes". I can take a video if anyone wants to see tommorow.

Macintosh_nik's picture
Offline
Last seen: 12 hours 6 min ago
Joined: Jan 8 2021 - 05:18
Posts: 465
Hi Jero32!

A video would be very helpful. Also take a good photo of your board so you can read the markings on each chip.

Offline
Last seen: 10 months 1 week ago
Joined: Oct 9 2023 - 14:36
Posts: 12
Uploaded a video at https:/

Uploaded a video at https://www.youtube.com/watch?v=JW7FzN91xkw I've been doing some work with the oscilloscope. And unfortunately I think the issue lies with the 2519.I can see clock signals on pin 6 "CL" And the blinking cursor signal seems to be present on pin 3 "I6". (assuming that I understood how it works correctly.)When I boot I can briefly see signals coming out of pin 7 "O6" but this quickly dies down. So what I'm guessing is happening is the chip briefly works, something (temperature, capacitive fault inside the chip?) makes it fail and stop outputting a signal.This is further compounded by the fact that the blinking usually works briefly only when the computer has been off for a while, whereas if I repower immediately after it stops it often wont work.I'd upload pictures but I don't have a good camera right now so that'll have to wait. 

Macintosh_nik's picture
Offline
Last seen: 12 hours 6 min ago
Joined: Jan 8 2021 - 05:18
Posts: 465
Hi Jero32!

It seems that this problem is really related to the 2519. At least the picture in your video is very similar to the photo of the symptoms of a non-working 2519 shown by Uncle Bernie.

 

https://www.applefritter.com/content/horror-room-symptoms-bad-cis

Offline
Last seen: 10 months 1 week ago
Joined: Oct 9 2023 - 14:36
Posts: 12
Well, I have ordered another

Well, I have ordered another 2519. I'll post an update when it arrives. 

Offline
Last seen: 1 week 3 days ago
Joined: Apr 1 2020 - 16:46
Posts: 996
It would be helpful to know from which source / batch the ...

... allegedly bad 2519 is. In my kits I mostly had 2519B with a date code 76xx (xx=week, varies) and only ~1% were bad, and did not make it through my screening / burn-in process.

 

In my earlier kits I had 2519N with date codes from 7734 to 7743 of which ~25-33% were bad. I have two 2519N with 7803 and 7814 date code, also bad.

 

So it seems that Signetics had an issue in their wafer fab beginning around Fall 1977 and these 2519N tend to "die" during the first few days of burn-in, if they are not bad out of the tube. Remember, I did a 4 week 24/7 burn-in of all ICs in my famous kits (now sold out). And from the pile in the "graveyard" of bad ICs I was able to eliminate by the process I can say that bad ICs happened often enough that selling untested IC in Apple-1 kits would have caused so much trouble that it can't be done (unless the seller is a pervert who enjoys having angry customers complaining about bad merchandize, and has time to waste on that. I preferred to ship 100% tested ICs, because my time is too precious).

 

However, be aware all '@' also can be caused by missing or bad clocks PHI3, PHI4 on the 2504 shift registers. But since you have used your oscilloscope, you sure did check that, too.

 

- Uncle Bernie

Offline
Last seen: 10 months 1 week ago
Joined: Oct 9 2023 - 14:36
Posts: 12
I'm happy to report that a

I'm happy to report that a new 2519 (79 date code) fixed the issue.I still have to run a more extensive test but the demo code from the manual seems to run. Unfortunately I've caught quite a nasty stomatch bug so that will have to wait.

 

edit: video https://www.youtube.com/watch?v=SR1lo21GAhk

Log in or register to post comments