Hello =) 1) I started to read the i8275 (D8) datasheet. It seems to be, that HRTC and VRTC active from power-up even if the controller is not initialised. (Page 8-239) 2) i8275 is working in pair with DMA i8257 (D2). DMA and CRT controller also are used for DRAM refresh in the 86РК. Them must be intialised on startup as soon as possible. Probably, DRQ/DAC signals and HOLD/HLDA signals can be checked by oscilloscope for any activity. 3) There are startup ROM mapping circuit in the 86РК. On power-up or reset it will map the ROM to te beggining of the adress space. At the beggining of rom command C3 36 F8 = JMP F836. The ROM is mapped to the end of adress space on that command. D13.1, D10.1, D11.2, ... I think, if them don't working properly, it can cause RAM to be mapped to the beginning of the address space on the startup and CPU executes randomness. 4) Also, modern 86RK clone is beeping on keypresses.
Thanks again for your excellent research! Yeah, I am concerned that the machine isn't actually working (no keyboard response) so that is good information about HRTC & VRTC signals. I was hoping to get a stable display (just B&W screen) and then move to troubleshooting the machine, but it sounds like it's better to check the other parts first.
Thanks - I should be getting back to this fellow later this week - I have some test PCBs for the video interface arriving, so will have a proper look at the rest of the system
It looks like for whatever reason both HRTC and VRTC frequencies are 2 times lower than they should be. Proper horizontal sync is ~15kHz and ~50Hz for vertical.
Sergij (@sergijtolmeros) checked the 8275 datasheet and apparently it outputs these signal regardless if the CPU is actually working ... so our suspicion at the moment is that the machine isn't actually booting and the 8275 is just outputting at a default frequency. So I'm going to check the rest of the machine. Maybe the frequencies will be correct when the rest of the machine is working (I've not had much success with these, i.e. Альфа-БК) ... but am going to try!
Is the original SECAM? That would explain the lack of sync lock on the monitor. Compare crystal frequencies between the original and Sergei's. When you were testing the output of the shift register the trigger level was set at or above the high level voltage, resulting in triggering on noise or no triggering. Try it again with the trigger set between high and low. If there is indeed no bit stream from the shift register the video output could be white instead of black screen. Better to leave the output circuit until you have a good bit stream and maybe save yourself some work.
@@Brfff Horizontal and vertical scan frequencies are different also, not just color encoding. That might be why you saw the different frequencies on the scope, but oscilloscopes are notoriously bad frequency counters.
If you do get PCBway to make that board, I would add the character generator and taps to drive either the original video circuit with your generator, or use the computer's generator to drive your video circuit on the PCBway board. And I spent the evening like a kid in a candy shop in an electronics design/build shop with a friend. The owner is looking for a design engineer, and I might be the guy. All sorts of toys, and gear from the 80's to early 2000's Including at least one Tandy CoCo (that was used for driving a label printer_ and a couple of Mac (IIe or something like that), a pristine IBM CRT monitor, and lord knows what else.
Oh baby! Sounds like fun! I'll put the Kicad files for the 86RK video output board on my Patreon so you can see what I've done so far ... I figure it might be useful for all the 86RK based computers with RF video output I'm trying to fix ... the Alpha BK, the Partner, this UMPK-R ...
@@Brfff Seems like it's a reoccurring theme, so you might as well make a Swiss Army knife for video troubleshooting. PAL, NTSC, SECAM, and whatever else would be cool.
![🇺🇦 УМПК-Р (UMPK-R): Part 1 (First Look) [TCE #0544]](http://i.ytimg.com/vi/Kr2Yl7z__qQ/mqdefault.jpg)
![🇺🇦 УМПК-Р (UMPK-R): Part 2 (Trying to get video output) [TCE #0545]](/img/n.gif)
Hello =)
1) I started to read the i8275 (D8) datasheet. It seems to be, that HRTC and VRTC active from power-up even if the controller is not initialised. (Page 8-239)
2) i8275 is working in pair with DMA i8257 (D2). DMA and CRT controller also are used for DRAM refresh in the 86РК. Them must be intialised on startup as soon as possible.
Probably, DRQ/DAC signals and HOLD/HLDA signals can be checked by oscilloscope for any activity.
3) There are startup ROM mapping circuit in the 86РК. On power-up or reset it will map the ROM to te beggining of the adress space. At the beggining of rom command C3 36 F8 = JMP F836. The ROM is mapped to the end of adress space on that command.
D13.1, D10.1, D11.2, ...
I think, if them don't working properly, it can cause RAM to be mapped to the beginning of the address space on the startup and CPU executes randomness.
4) Also, modern 86RK clone is beeping on keypresses.
The same screen pattern apeared on another computer th-cam.com/video/x_RBBbvt5lU/w-d-xo.html
Thanks again for your excellent research! Yeah, I am concerned that the machine isn't actually working (no keyboard response) so that is good information about HRTC & VRTC signals. I was hoping to get a stable display (just B&W screen) and then move to troubleshooting the machine, but it sounds like it's better to check the other parts first.
The picture youv'e got at 17:00 is typical for a Radio-86RK sans reset. If the reset does not change it - might be the CPU or the ГФ24.
Thanks - I should be getting back to this fellow later this week - I have some test PCBs for the video interface arriving, so will have a proper look at the rest of the system
It looks like for whatever reason both HRTC and VRTC frequencies are 2 times lower than they should be.
Proper horizontal sync is ~15kHz and ~50Hz for vertical.
Sergij (@sergijtolmeros) checked the 8275 datasheet and apparently it outputs these signal regardless if the CPU is actually working ... so our suspicion at the moment is that the machine isn't actually booting and the 8275 is just outputting at a default frequency.
So I'm going to check the rest of the machine. Maybe the frequencies will be correct when the rest of the machine is working (I've not had much success with these, i.e. Альфа-БК) ... but am going to try!
Is the original SECAM? That would explain the lack of sync lock on the monitor. Compare crystal frequencies between the original and Sergei's.
When you were testing the output of the shift register the trigger level was set at or above the high level voltage, resulting in triggering on noise or no triggering. Try it again with the trigger set between high and low.
If there is indeed no bit stream from the shift register the video output could be white instead of black screen. Better to leave the output circuit until you have a good bit stream and maybe save yourself some work.
It’s B&W output only - no colour, so not SECAM/PAL/NTSC. OK, I’ll have a better look at the character data
@@Brfff Horizontal and vertical scan frequencies are different also, not just color encoding. That might be why you saw the different frequencies on the scope, but oscilloscopes are notoriously bad frequency counters.
If you do get PCBway to make that board, I would add the character generator and taps to drive either the original video circuit with your generator, or use the computer's generator to drive your video circuit on the PCBway board.
And I spent the evening like a kid in a candy shop in an electronics design/build shop with a friend. The owner is looking for a design engineer, and I might be the guy. All sorts of toys, and gear from the 80's to early 2000's Including at least one Tandy CoCo (that was used for driving a label printer_ and a couple of Mac (IIe or something like that), a pristine IBM CRT monitor, and lord knows what else.
Oh baby! Sounds like fun! I'll put the Kicad files for the 86RK video output board on my Patreon so you can see what I've done so far ... I figure it might be useful for all the 86RK based computers with RF video output I'm trying to fix ... the Alpha BK, the Partner, this UMPK-R ...
www.patreon.com/posts/video-output-for-116498402
@@Brfff Seems like it's a reoccurring theme, so you might as well make a Swiss Army knife for video troubleshooting. PAL, NTSC, SECAM, and whatever else would be cool.