Well, that's embarrassing

[Adam Thornton]

I decided to update 2.11BSD (after totally destroying mvsevm.fsf.net doing that, because I destroyed my SD card with too many writes) on the pidp-11, and one thing led to another, and I ended up getting the new version of the software and the new images...

...and in so doing, determined that all the lights are working fine, but switch 3 isn't, and neither are switches 11-21, which indicates row0 isn't correctly connected, I think.

So I guess I know what I'm doing with my weekend.

Adam

[Flavio Villanustre]

I empathize with you! I've been there! Unplug your raspberry pi and check continuity between the GPIO connector and the input diode for that row; sometimes overheating destroys a trace and you can bridge it with just a piece of wire. Also check that diode (a cold solder can have the same effect).

Yeah, it kinda sucks, but it's an stimulating time-sink 😀 Let us know what you find!

On the SD card dieing, I use a USB drive on my PiDP-11 for that reason: it's more robust, faster and can be replaced without opening the back. On the PiDP-8 and 10, because the SD slot is externally accesible, it doesn't matter as much, IMO.

Flavio

--
You received this message because you are subscribed to the Google Groups "[PiDP-11]" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pidp-11+u...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/pidp-11/CAP2nic0ARwAzu3COYpHAoHug9b1ghB9Whrr4HkXp0%3DmMCgaPww%40mail.gmail.com.

[terry-...@glaver.org]

On Friday, December 13, 2024 at 9:36:08 PM UTC-5 atho...@gmail.com wrote:

I decided to update 2.11BSD (after totally destroying mvsevm.fsf.net doing that, because I destroyed my SD card with too many writes) on the pidp-11

There are many different grading systems for SD cards. In fact, there are at least 4 that I know of, none of which address robustness, just speed. I've been using Samsung Pro Endurance cards which are designed for video recording. Samsung claims 140,000+ hours of continuous recording. These cards allegedly implement wear leveling and over-provisioning. A run-of-the-mill SD card is likely going to wear out from lots of writes to the same "disk" blocks (the PDP-11 operating systems weren't SD-aware, and so tended to re-write the same disk blocks over and over again.

[Mike Katz]

I've worn out several micro SD cards so I've been using small 2.5" SSD drives.  It's overkill for space but they are designed for considerably more read and write cycles than memory cards or USB drives.  There are USB 3.0 to SATA/NVME converters for the Pi.  If I recall correctly the Pi5 has a PCI external port on it for even faster access if you need it.

To view this discussion visit https://groups.google.com/d/msgid/pidp-11/CAEUQ4QFyfmj%3D6scH48nZJy7tgYbftjMsfKoD9ORNLDaEYoQQ5Q%40mail.gmail.com.

[timr...@gmail.com]

I don't really use my pidp-11 these days.  I stole it's Rasp-Pi for my pidp-8i and have never replaced it.  I know it has a few LEDs that don't work, but not any full rows.

I think there are 2 or 3 that don't work.  I never did track down what connections had to be bodge wired to make them work.  I am sure no one would even notice

if the system was running.  Some day I may get back to this and fix it, but having too much fun with the pidp-10.  It could have been worse and be more switches or

something else harder to fix.  Maybe it's just a cold solder joint and that would be easy to fix.

[Adam Thornton]

Yeah, I'm not expecting it to be a complex fix.  I did the PiDP-11 years ago, and I'm not sure I ever even fully tested the switches.  I did the -8 as a warmup for the -10, and I think I remember that it had a nice test routine as part of the assembly process; I know the -10 did and I had a nontrivial amount of rework to get everything shipshape, so I wouldn't be surprised if some of my -11 work was subpar.

My guess is bad joints on one switch and somewhere on the row0 line, but we'll see.

[Tim Radde]

Same here.  Did the 11 long ago.  I got the 8 later as I like it's front panel and the 8i was the first computer I used.  Learned computers on one.  When the 10 was announced I was not sure right off if I wanted one.  Yes, I had used a KA-10 and it's a beautiful machine.  So I held off from the first batch.  Seems I still got one though as I have no SN.  The 8 worked well right off and I am sure there was a test to check the lights and switches.  Not sure if the 11 had that or not.  I only checked the LEDs on the 10 as there were so many.  Didn't have any LED failures at all.

--
You received this message because you are subscribed to a topic in the Google Groups "[PiDP-11]" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/pidp-11/cuycb5bh7fY/unsubscribe.
To unsubscribe from this group and all its topics, send an email to pidp-11+u...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/pidp-11/CAP2nic1SyByUcwhFfOCRedgh-hHhnm35E%3DQRL1yKRxqTE8_AOQ%40mail.gmail.com.

[Curtis Dean Smith (史國興)]

This is why I installed SD card extenders on both my 8 and my 11.  It is now simple to switch out SD cards at will.  I am also planning on running off of HDDs installed in USB cases, but only when I have the time to more than the occasional messing around.  For now, I enjoy the blinkenlights and the idea that I can use one of these systems whenever I am in need of a DEC.

[Socratease]

If you have a switch register switch with a bad contact, you can desolder it and flip it over to use the other contact, but you also have to pry out and flip the plastic handle.  Had to do that with mine just the other day.

[oscarv]

On Sunday, December 15, 2024 at 5:02:17 AM UTC+1 Socratease wrote:

If you have a switch register switch with a bad contact, you can desolder it and flip it over to use the other contact, but you also have to pry out and flip the plastic handle.  Had to do that with mine just the other day.

Or ask Jose for a replacement one: j.l...@ceds.dev!

Kind regards,

Oscar.

[oscarv]

On Saturday, December 14, 2024 at 3:36:08 AM UTC+1 atho...@gmail.com wrote:

...and in so doing, determined that all the lights are working fine, but switch 3 isn't, and neither are switches 11-21, which indicates row0 isn't correctly connected, I think.

For switches 11-21, check along the highlighted trace:

 

And also,

Reflow the GPIO pins around the highlighted one, in case of a bad contact. It must be that this connection, somehow, is not made between GPIO and the center pin of the row of switches.

For a dead SW3, the following traces:

and

and

It has been forgotten, I think, but the Kicad project file for the PiDP-11 can be downloaded here, in case you want to inspect more. Right-clicking on any solder point, selecting Net Inspection, gets you these pictures. Download link: https://drive.google.com/open?id=1SPvF5xFyG-lbpwkwjp5FTWIWAg1EhxTR

It's not likely that the switch #3 itself is faulty. But test by shorting middle and bottom pin.

90% of the time it is forgotten or bad solder points. 8% of the time, reflowing the GPIO pins solves a problem even though a multimeter shows no issues - the hypothesis is, small current leaks to neighbouring GPIO pins through excess flux. I did not believe it at first, but came to believe it over time. For the rest: 1% of the time, broken trace. 1% of the time, a Pi with bad GPIO (the Pi works fine, but one or two GPIOs do not behave anymore).

Kind regards,

Oscar.

[Adam Thornton]

Thank you!  I had entirely forgotten about the KiCAD file, or at least I guess I'd figured, "I don't know CAD programs, what could could that possibly be to me?"  This will give me enough targeted debugging to get the PiDP-11 off my table before Monday when friends need it for boardgames.