• Please review our updated Terms and Rules here

Poly88 is now running

Dwight Elvey

Veteran Member
Joined
Jun 21, 2003
Messages
4,991
Location
Santa Cruz
I've been bring my old Poly88 back to life. I'm making slow progress but mostly going slow.
The keyboard I have has an interesting modification. I'm not sure why it was there. It had a switch on the side that connected two of the data out pins together of the KR2376. The funny thing is it didn't even wire these pins out to the data pins of the Poly88. It looked like work I'd done years ago but who knows. I removed the switch and ran the wires that looked to be grounded to the Poly88 and now it see numbers as well as other characters. Anyway, the switch was old and didn't work. I'd like to find a way to make a shift lock that didn't shift the numbers but may have to live with it as it is. I wish I knew what that switch did?
Ahh, the smell of a large tantalum going up in smoke. It was a 100uf. There were two of them in parallel right next to each other so if figured it would work fine with just one. It really burned good. There was a nice flame I had to blow out and black smudges all along the board. What a mess. This was on the CMOS memory board I was talking about. I cleaned up the soot and the board seems to work fine.
I'm about ready to see if some old tapes I have have got any information left on them. Some are tapes I did and some are others work. For those using old machine with cassettes, a word to the wise. Don't store cassette tape wound with fast forward of fast reverse. Always store them packed slowly and smoothly with the play. Only use the fast while you are actively going to run that part again in a few minutes. The loose packing leaves wrinkles at the edges and some time wrinkles across the tape, that cause dropouts.
Hopefully, I can repack them and get them to read after sitting some. This usually works but these have been sitting for years.
The Poly88 did two speed modes for recording. One was the typical KC byte format at slow speed. The other was a high speed 1200 baud rate. It required a higher quality tape recorder. I've been bring an old Superscope Marrantz back to life. These old Marrantz were built like tanks. I've got belts on order but made some temporary belts from rubber bands and super glue. So far they seem to work OK. The deck has issues with the auto shut off and no fast reverse. I need to look into the fast reverse but I can live with the auto shut off for a while. It stops the play by releasing the pinch roller but sometimes the motor keeps running. It looks like some old grease keeping one of the levers from moving, freely.
So far all the old memory seems to be working. The memory on the video board looks good, the memory on the CPU board looks good and the ones on the CMOS board look good ( although, the ones at the edge have smoke and fire damage ). I'm quite surprised because the last time I pulled this stuff out, it required some chips to be replaced ( about 10-15 years back ). Maybe, I got all the infant mortality out and the rest will last for another 100 years.
It is interesting that I have the issue with the vertical pulse being too long as mentioned in one of the earlier articles. It is causing the first few lines to drag to the left some as the horizontal resynchs. I need to look at the schematic for the video to see if there is a convenient modification that can be done.
Enough fun stories, back to hacking it some more.
Dwight
 
It is interesting that I have the issue with the vertical pulse being too long as mentioned in one of the earlier articles. It is causing the first few lines to drag to the left some as the horizontal resynchs. I need to look at the schematic for the video to see if there is a convenient modification that can be done.
Enough fun stories, back to hacking it some more.
Dwight

The problem basically relates to the time constant of the AC coupling into the transistor(or tube) that is the sync separator device in the set. If it is too short with respect to the total length of the vertical sync pulse, it causes an amplitude drop in the H syncs that immediately follow the V sync. I have attached a remark on it from Grob's television book (it relates to tubes, but the principles are still the same for transistors). I found also that in transistor separators, the transistor's bias resistor also affects this as with a little more base current the transistor will stay in saturation longer, even if lowering the resistor value shortens the time constant a little more.

Looking at a lot of different sync separator circuits (in transistor TV sets and monitors) the value of the bias resistor runs in the range of 270k to 1.5M, and the coupling capacitor in the range of 0.1uF to 0.33uF. So its just pot luck really if the V sync is too long, whether the particular monitor will object to that or not.

The graph attached is for a sync signal with equalizing pulses, to aid interlaced scanning, where there is a 1/2 line delay between consecutive V sync pulses, but you can ignore those, as they don't relate to the problem of what happens when the V sync pulse is longer than the time constant of the sync separator stage in the TV monitor was designed for.
 

Attachments

  • sync3b.jpg
    sync3b.jpg
    103.5 KB · Views: 1
I suspect something can be done but for now, I'll live with a bent first line. The Polymorphic uses a counter to create the V-Synch so, I suspect more clever use of the counter will shorten it up a little more. I have to start looking for a spare gate or do some piggy back mounting.
I've got other issues to fix before next weekend so I'll have to ignore the tilted letter.
Dwight
 
I stared at the VTI schematic for a while a week or two ago while we were discussing the vertical sync issue and didn’t see an easy mod to shorten vsync significantly. It can be shortened from 8h to 7h with a jumper setting on the board, but we probably need more. A two input OR gate used as an active low AND could be used to easily halve vsync, but no 7432’s are on the board, so I know there are no spare OR gates. I haven’t done a study to see what spare gates - if any - are available.

Mike
 
I stared at the VTI schematic for a while a week or two ago while we were discussing the vertical sync issue and didn’t see an easy mod to shorten vsync significantly. It can be shortened from 8h to 7h with a jumper setting on the board, but we probably need more. A two input OR gate used as an active low AND could be used to easily halve vsync, but no 7432’s are on the board, so I know there are no spare OR gates. I haven’t done a study to see what spare gates - if any - are available.

Mike

I don't know the Poly schematic, but sometimes I have been able to get around this sort of problem with a couple of diodes and a resistor to make an OR gate function without having to add any extra IC's or transistors. But, probably there will be spare gates and enough logic signals about in the divider chain where the sync pulses are generated to figure out a mod. It is likely just a matter of using the V reset pulse to gate through a 3H width pulse, which can also be done if there is a spare flip flop about.

I don't know if this could be of help but have a look (on page 25 of this document) at the way Atari generate a /Vsync pulse that is 4 horizontal periods long (254uS) using what amounts to a S-R flip flop and a three input gate using signal VBLNK and 4V and 8V, but there would be many other ways to do it, scroll to page 25:

http://worldphaco.com/uploads/LAWN_TENNIS.pdf
 
Last edited:
Using diodes may be the way to go. I've got to spend a little time examining the display schematic. I have Tiny BASIC on EPROM so I loaded StarTrek from cassette tape. I only have one good working cassette machine out of 4. The Marrantz is going to need at pinch roller replaced as well. It looks to be skewing the tapes ( Baaa ). It need the belts replaced but I have those on order. I have a GE machine that I used to use but its belt is broken. I've got a little Sanyo ( same problem ). I have a Optimus that is working well enough to easily read my PolyPhase tapes ( requires good fidelity ). The belts should have been in by Friday but still not in. I'll most likely not have time before Friday to do the Marrantz SuperScope. It is the best quality drive I have but will still need some repairs. I have some tapes from someone else and I was tiring to read them. They are in poor shape. There was some interesting stuff to look at but I may not be able to do much with them. One cassette had the pressure pad come off and jam the tape. I will be more careful with these tapes. Most are on 60 or 90 minute tapes. Some of my own old 120 minute tapes are not doing well in either machine. The 60 and 90 minute are working well. I don't know why I got those 120 minute tapes. When using the 60 and the high speed PolyPhase, I can probably get more than 150K on a single side. My only excuse was that I was young at the time and didn't think about 45 to 50 year later, I might want to read them and why wouldn't everyone still be using cassette tapes with their computers?
Dwight
 
Back
Top