• Please review our updated Terms and Rules here

Poly 88/8813 keyboard compatibility

falter

falter

Veteran Member
Joined
Jan 22, 2011
Messages
6,580
Location
Vancouver, BC
Just got an 8813 lid and keyboard from deramp (many thanks sir!). I'm just curious.. trying to find the answer in the manuals here -- but am I correct to assume the pinout for the Poly keyboard matches what both the 8813 and 88 expect on their keyboard headers?
 
falter said:
Just got an 8813 lid and keyboard from deramp (many thanks sir!). I'm just curious.. trying to find the answer in the manuals here -- but am I correct to assume the pinout for the Poly keyboard matches what both the 8813 and 88 expect on their keyboard headers?
Click to expand...

They should both work with the keyboard. You just need to make sure the jumpers on the video board is setup correctly. The keyboard I got from Mike uses the unregulated 8V only ( it has the 5V regulator on board ). It doesn't use any negative voltage. Some of the keyboards use the regulated 5V so there may be a cut and jumper to deal with. You may need to fiddle with the strobe polarity jumper as well. It may work in either position, as it used to for a previous keyboard I had.
Dwight
 
Yeah this one appears to have a regulator on it... I'm guessing since it is Poly's own design it probably uses the standard they developed.

Right now I'm fighting with those 'sliders'.. the ones the keytops plug into that split apart and then bind in the keyshaft. I don't have any spares now so I'm not sure how I'll deal with this. Was hoping to figure out a way to pull the sides tight and glue again.
 
According to the VTI manual, the VTI should already be configured for a 'Polymorphic' keyboard. I confirmed mine has none of the additional voltage regulator or associated parts installed on the VTI that you would need for keyboards that use negative voltages, etc. I've got 5 volts at pin 14 of the Keyboard connector on the VTI.

The thing that is throwing me with the Poly keyboard I got is that it has a voltage regulator on it. I wouldn't think you'd need that if you were looking for 5V. So I'm wondering if they did something on the Poly 8813 VTI boards to give 8V or something.
 
So I found a few bits of keyboard manual and yeah, definitely looks like this keyboard was set up to receive anywhere from +7.5VDC to +10VDC. There is an NEC 14305 +5V regulator on the keyboard PCB itself, and according to my DMM the input leg is connected directly to pin 14 of the VTI, which is +5V in its normal configuration but can be altered to provide +8V.

The mystery to this is why Poly would set up the VTI to provide +5V by default, and then set up the keyboard to receive a higher voltage only to step it back down to +5V. And why in the instructions it says if you have the Poly keyboard, you don't need to do any mods to the VTI. So far it seems like the other pins line up properly with the keyboard.. so yeah. That's a weird one. I'm wondering if maybe those late revision VTI boards, like the Rev F board I had, had +8V out the gate (I seem to recall testing it and finding +8V there at pin 14, but the memory is dim), and Poly is assuming a totally different keyboard design in the original VTI manual. Wasn't really keen on cutting into my VTI board more, but if it has to happen.. so be it. Although I wonder if it would make sense to just yank the voltage regulator on the keyboard and patch the +8V in to the +5V out pads.

Side note - remember that Scientific Devices KBMO keyboard I bought a while back? They use the exact same keyswitch 'matrix' or system that the Poly keyboard does. SD used smaller shift keys but the positions of all keyswitches are exactly the same as they are on the Poly. Too bad the Poly uses that 'box' style plunger that breaks over time and binds.

20190426_213116.jpg

20190426_213148.jpg
 
There is a place, as I recall, were you cut a trace and jumper the raw 8V. Just don't cut the trace to the rest of the board from the regulator.
I have an idea for the split plunger. I'll try to get to it tomorrow.
Dwight
 
The keyboard connector on the VTI board in my 88 is left at +5v. However, the VTI boards I have from multiple 8813’s are all strapped for +8v. My guess is that Poly found the 5v was too noisy at the keyboard after running across a long ribbon cable, so for the “professional” target market of the 8813, they went with local regulation at the keyboard.

Mike
 
I'm not clear on what was the proper 'Polymorphic' keyboard for the 88. I've seen pictures with everything ranging from SWTPC KBDs to ones that look like the 8813 keyboard.
 
Okay.. so I modded the VTI, triple checked, fired up.. and was able (just barely, because of the key binding problem) to get to the front panel screen. However, the output was a bit unexpected:

20190427_100647.jpg

Not really sure why it's coming up backwards and upside down??
 
Man this is a weird one.

I'm looking at the screen and some characters *are* oriented the right way. And if I'm typing along the bottom of the screen, most come out the right way also. It's like something is tripping out and flipping every so many characters. Very odd.

EDIT: Nevermind... I just used photo editing software to flip and rotate 180 degrees, and it all looks normal. I guess some characters look right no matter which way they are pointed.
 
I see only a vertical flip. Sounds like your character ROM is an MCM6571 instead an MCM6571A.

Mike
 
Last edited:
Yup. Thanks for that. I happened to have a 6571a handy from another project. Plugged it in and everything is correct direction now.

I guess there's no way to modify things to flip characters generated from a plain 6571 around?
 
Almost wishing I hadn't touched on this topic. It reminded me that I have a 6575L somewhere that I bought for my prototype Sol project. Spent half the afternoon tearing apart the office looking for it. Hoping I didnt toss, they're easy to come by!
 
Thanks! For some reason the original Sol terminal board used a 6571l (the prototype board, that is). Not sure if this will work but I'll check it out. I'm sure the one I bought is around somewhere. I had it in a little red case. Probably fell somewhere.
 
The 6571A (typical Poly VTI ROM) and the 6574 (typical Sol-20) ROM are interchangeable with each other and differ primarily in the symbols that are generated for the control character range (00h-1Fh).

However, if you want to play most of the coolest Sol-20 games, you'll want the 6574 ROM because the symbols used for aliens, spacecraft, etc., are unique to the 6574 PROM.

Mike
 
To my knowledge, there is no fault in the video board that can do this??
It is possible with the wrong part installed but not possible for a part to fail and cause this. This can't be caused by the display's yoke being upside down. I assume you've verified the video display and it is not the yoke being upside down. Then I thought about this and the text was on the correct side of the screen. Even reversing the vertical coils couldn't put the registers at the top and the memory window at the bottom. The characters are in the correct location on the screen.
It had to be something was inverting the data from the character ROM or the character ROM was the wrong one. The counter chip IC15 that is a 74393 could be wires wrong ( 4 wires ). I don't know if there is a replacement for the 74393 that just reorders the Q bits of one half of the IC. The similar 74390 has the clocks on different pins. Do look around the 74393 chip for jumpers??
That leaves the character ROM being incorrect.
Dwight
 
You can make it work but you'll need to invert the 4 signals going to IC37's pins 32, 22, 23 and 24. You could mount a 7404 on top of another IC and then cut the wires to IC37, and tack in the inversion. You need to check that you don't cut the wires going the the two muxes for the graphic functions. They might pass across IC37.
Dwight
 
Thanks Dwight.

It was definitely the wrong chip.. 6571 instead of 6571a. I cannot explain why the 6571 was there in the first place though. As I recall the board did have some jumpering on the back which I removed to go to stock... I only retained wires that were required due to artwork errors. Maybe that's part of what it was doing.
 
You must log in or register to reply here.
Back
Top