(Extra) Buffers/Transceivers for PL6 ❓

emulators, hardware and classic software for atom + system machines
Post Reply
User avatar
Louis.m
Posts: 37
Joined: Fri Jun 07, 2013 12:36 am
Location: the Netherlands
Contact:

(Extra) Buffers/Transceivers for PL6 ❓

Post by Louis.m »

Hello 10 all,

For safety while experimenting, I like to have answers on some questions about (extra) PL6 Buffers/Transceivers on the Atom.

First here's a listing of the layout of the original PL6 connector on the mainboard:
PL6 list.png
(Click on image to view better detailed one)

Address and Data buses are buffered on the Atom mainboard, making connections to experimental circuitry relatively safe.

But what to do with the control bus lines, can these be buffered too, or will timing (or something else) become an issue than, and what about the lines to and from the VIA (6522), will buffering those cause problems ?

Does anyone have experience buffering these other lines ? If you have any answers or suggestions, please post a reply [-o< !

Gr :D :D tings, Louis
User avatar
roland
Posts: 4486
Joined: Thu Aug 29, 2013 9:29 pm
Location: Born (NL)
Contact:

Re: (Extra) Buffers/Transceivers for PL6 ❓

Post by roland »

From the control bus only phi2 and r/w are used in most of the expansions. Phi2 has some kind of buffer by a double inverter. So only r/w is completely unbuffered. Afaik this is not a real problem. The load on the address bus is quite high as almost every memory ic is connected to (at least apart of) the address bus so I can imagine that expansion boards can cause to much load for it.

However, I know that bitd some people just place wires in the sockets of ic 2, 3 and 4 to avoid difficult stuf with enabling the data bus driver.

Buffering the 6502 is not practical because those lines can be used as inpu5 or output so buffering must match each I/O pin.
FPGAtom: 512 KB RAM, Real Time Clock and 64 colours
MAN WOMAN :shock:
User avatar
Louis.m
Posts: 37
Joined: Fri Jun 07, 2013 12:36 am
Location: the Netherlands
Contact:

Re: (Extra) Buffers/Transceivers for PL6 ❓

Post by Louis.m »

Hello 10 all,
roland wrote:
Thu Oct 07, 2021 10:56 pm
From the control bus only phi2 and r/w are used in most of the expansions. Phi2 has some kind of buffer by a double inverter. So only r/w is completely unbuffered. Afaik this is not a real problem. The load on the address bus is quite high as almost every memory ic is connected to (at least apart of) the address bus so I can imagine that expansion boards can cause to much load for it.
OK, the I/O lines of the VIA (6522) will be difficult (if not impossible) to buffer.

But for safety reasons, while experimenting, it would be nice to have output lines (φ2, NBlkB, R/NW, NWDS, NRDS, NRST, SYNC) and input lines (IRQ, NMI, RDY and SO) separated from any experimental circuit on a breadboard, to keep the inner Atom (not only µP but also ROM, RAM, etc.) unharmed in case of a mistake or malfunction.
I am also thinking about a second set of buffers/transceivers on address and data lines specially for a breadboard.

I'm still wondering if anyone has any experience with buffering like this and/or can tell something more about this subject.

Gr :D :D tings,Louis
User avatar
lovebug
Posts: 814
Joined: Sun Jan 31, 2021 5:07 pm
Location: Magrathea
Contact:

Re: (Extra) Buffers/Transceivers for PL6 ❓

Post by lovebug »

i wouldnt bother connecting the 6502's SO input pin to anything unless you like the overflow flag being set in the status register :lol:

for the port you could do something like this protect the pins from over/under voltage, any excess voltage above 5v or below gnd will be clamped by the diodes

pins will still function as inputs or outputs
 
protection.png
Image Image Image
User avatar
Louis.m
Posts: 37
Joined: Fri Jun 07, 2013 12:36 am
Location: the Netherlands
Contact:

Re: (Extra) Buffers/Transceivers for PL6 ❓

Post by Louis.m »

Hello 10 all,
lovebug wrote:
Sat Oct 09, 2021 12:25 am
i wouldnt bother connecting the 6502's SO input pin to anything unless you like the overflow flag being set in the status register :lol:

for the port you could do something like this protect the pins from over/under voltage, any excess voltage above 5v or below gnd will be clamped by the diodes

pins will still function as inputs or outputs
and
Hello Lovebug,

Well, as I mentioned earlier, it's all for experimental purposes.

The hardware manual says a negative transition on the SO pin sets the overflow bit (V) in the status code register, and the signal is sampled on the rising edge of PHI2. This was intended for fast input recognition.

Who knows what fun one can have with that ? 😎

Now to the main point of my reply:
Thank you for the ✨excellent suggestion✨, it's such a simple and elegant discrete solution❗ =D>

I'm ashamed I didn't came up with it myself. :oops:
Forgotten because these days I'm no longer used to think in Discrete Parts but instead in Integrated Circuits. ](*,) #-o

Together with some buffers and transceivers this makes an almost perfectly safe connection, for all lines of PL6 between an Atom and a breadboard, possible. :D


Gr :D :D tings, Louis
Btw, is that Magrathea Brunel ?
User avatar
lovebug
Posts: 814
Joined: Sun Jan 31, 2021 5:07 pm
Location: Magrathea
Contact:

Re: (Extra) Buffers/Transceivers for PL6 ❓

Post by lovebug »

I know how it feels to forget the simple things, back in the day I used to design a lot with simple logic gates and discrete parts and later I discovered the 6502 and did a lot of the logic in code and later moved on to pic micro controllers and by this time I had started to forget the old ways :lol:

then a project came up that needed fast vector drawing for controlling 10KW white lasers, I started with a 6809 cpu but after a few experiments I realised that the only thing fast enough was logic and some analogue drivers so I had to use binary adders calculating real numbers for the x and y velocities to move the beam from its current position to the new x y position in a way that the speed was constant regardless of angle so that all lines had the same brightness, there were also some 64K eproms thrown in there for fast sine/cosine vector rotation tables. the output from this mess went into 12 bit DtoA's and analogue drivers to drive the laser galvanometers

that got my brain working again :lol: :lol:
Image Image Image
User avatar
lovebug
Posts: 814
Joined: Sun Jan 31, 2021 5:07 pm
Location: Magrathea
Contact:

Re: (Extra) Buffers/Transceivers for PL6 ❓

Post by lovebug »

sometimes I rant on, im sorry. sometimes things pop into my head and my brain gets excited, im a bit impulsive at times :roll:
Image Image Image
User avatar
Louis.m
Posts: 37
Joined: Fri Jun 07, 2013 12:36 am
Location: the Netherlands
Contact:

Re: (Extra) Buffers/Transceivers for PL6 ❓

Post by Louis.m »

lovebug wrote:
Sat Oct 09, 2021 2:47 pm
sometimes I rant on, im sorry. sometimes things pop into my head and my brain gets excited, im a bit impulsive at times :roll:

Hi Lovebug,

Sorry ❓
Please don't ❗
Your simple but luminous :idea: idea for the 6522 I/O lines really helped me out, my thinking about a solution for that problem got stuck in a Cule de Sac, that's why I wrote "difficult (if not impossible)", and your story about the lasers is quite interesting.
So please be anything but sorry, 🖖😎👍 ❗

Gr :D :D tings, Louis
User avatar
lovebug
Posts: 814
Joined: Sun Jan 31, 2021 5:07 pm
Location: Magrathea
Contact:

Re: (Extra) Buffers/Transceivers for PL6 ❓

Post by lovebug »

if you want to be super safe you could use opto isolators too but then a line could be either input or output but not both

you could however build a breakout board with jumper links to connect either an input or output opto isolator to each port line
Image Image Image
User avatar
Louis.m
Posts: 37
Joined: Fri Jun 07, 2013 12:36 am
Location: the Netherlands
Contact:

Re: (Extra) Buffers/Transceivers for PL6 ❓

Post by Louis.m »

Hi !

Noh, I'l stick with your first suggestion (with some additions).
viewtopic.php?p=337170#p337170
It seems perfect to me.

Gr :D :D tings, Louis
Post Reply

Return to “acorn atom and acorn system series”