Mechanical Affinity Battery Backed Clock Board

[bwinkel67]

So I recently did a QL upgrade video:

https://www.youtube.com/watch?v=GLc_H-7hgIE

I had posted the video on the Facebook QL page and Johan Engdahl wanted to know some details of the board as he wants to recreate it (in a previous post I had already posted the images but no specifics on the parts). The board is pretty simple so I thought I'd post it on the forum for future reference (in case someone here also wants to create one).

I realize if you have a GoldCard there is no need but still nice to capture old hardware. The board does work through resets and power offs, and seems pretty stable. Once, after a QL crash, the clock did reset itself so it's not 100%. But for day-to-day use it works pretty well.

First the pictures (note battery holder removed as I borrowed it for another project, but it's just a button cell one -- I've since put it back as evident in video above):

Top of board with all the diodes/capacitors needed:

Bottom of board:

Parts list as best I can determine:

• ceramic cap is 103
• diode to the right is DT K102
• diode next to blue cap is 1NZ 001 752 (first part could also be 1N with a small 2)
• blue cap says 1047 but seems partially rubbed off
• no marking on diode furthest left

[Cristian]

Very interesting

[javier2112]

So I recently did a QL upgrade video:

https://www.youtube.com/watch?v=GLc_H-7hgIE

I had posted the video on the Facebook QL page and Johan Engdahl wanted to know some details of the board as he wants to recreate it (in a previous post I had already posted the images but no specifics on the parts). The board is pretty simple so I thought I'd post it on the forum for future reference (in case someone here also wants to create one).

I realize if you have a GoldCard there is no need but still nice to capture old hardware. The board does work through resets and power offs, and seems pretty stable. Once, after a QL crash, the clock did reset itself so it's not 100%. But for day-to-day use it works pretty well.

First the pictures (note battery holder removed as I borrowed it for another project, but it's just a button cell one -- I've since put it back as evident in video above):

Top of board with all the diodes/capacitors needed:

IMG_1433.JPG

Bottom of board:

IMG_1434.JPG

Parts list as best I can determine:

• ceramic cap is 103
• diode to the right is DT K102
• diode next to blue cap is 1NZ 001 752 (first part could also be 1N with a small 2)
• blue cap says 1047 but seems partially rubbed off
• no marking on diode furthest left

Great.

What schematic have you used for the circuit?

[bwinkel67]

What schematic have you used for the circuit?

None, I bought this in the early 90's and just dug it out recently and stuck it back into my QL. Someone else wants to recreate it and I just posted the info I gave them here in case anyone else also wants to recreate it.

[Dave]

I have one of these. It is problematic, because of the design of the 8302. The battery backup regularly results in clock corruption during power-off. Nasta seemed to understand the problem well when I spoke with him about it many years ago. I am sure he will magically appear and give the lowdown.

[bwinkel67]

I don't recall having that issue much when I used it back in the 90's. It does seem to reset sometimes but mostly it keeps the time between power offs.

[Pr0f]

I think the problem is the clock register could get written randomly at reset / power cycle - and the simplicity of the circuit can't prevent that. Some additional 'helper' logic is required - such as is found on the QIMI interface, to prevent any writes occurring until the voltage is stable. The 8 pin DS1210 for instance will gate both VCC voltage and the CE line for RAM's - and can make any CMOS RAM non volatile - you could use that chip and gate DSL on the ZX8302 for instance - it doesn't have to be CE line

[Pr0f]

This old post covers it I think :

viewtopic.php?f=2&t=2072&p=18650&hilit= ... ery#p18650

[bwinkel67]

Yeah, there was another thread where Martin Head suggested that when you power off and held the reset switch it would mostly fix the issue and that's what that PDF in your link tries to do.

[bwinkel67]

So perhaps not as useful but I just don't recall it being much of a problem back in the 90's. Maybe if you don't power on/off quickly...but from what I read it supposed to be pretty random.