exxos's Atari, Amiga & retro forum

Serial (RS232) port on ST is done by the MFP, not the ACIA. As said before, serial buffers were replaced with MAX version.. no more -12v needed.

Another thing ... RP900 (10K pull-up on CD[0..7]) ... "Helps fix unstable floppy drive"... but not present on the original ST ... could it be creating issues there ?

rpineau wrote: 12 Oct 2018 23:05 Another thing ... RP900 (10K pull-up on CD[0..7]) ... "Helps fix unstable floppy drive"... but not present on the original ST ... could it be creating issues there ?

He tried without it a while ago I think. I have the same pack on my V7 floppy mod pcb in my test STFM and not had any issues yet. Think its a 10k one.

Ok.
I'll keep digging.. there are not a lot of component in the path so not sure what else to look at.

rpineau wrote: 12 Oct 2018 23:17 Ok.
I'll keep digging.. there are not a lot of component in the path so not sure what else to look at.

I think its a "design bug" with the circuit itself... pretty sure needs STE buffers to solve 1772 conflict.. but that should be all it really needs... But hard to prove at the moment..

exxos wrote: 12 Oct 2018 23:15

rpineau wrote: 12 Oct 2018 23:05 Another thing ... RP900 (10K pull-up on CD[0..7]) ... "Helps fix unstable floppy drive"... but not present on the original ST ... could it be creating issues there ?

He tried without it a while ago I think. I have the same pack on my V7 floppy mod pcb in my test STFM and not had any issues yet. Think its a 10k one.

Every iteration I have tried with and without R900

rpineau wrote: 12 Oct 2018 22:29 Serial (RS232) port on ST is done by the MFP, not the ACIA. As said before, serial buffers were replaced with MAX version.. no more -12v needed.

Understood but there was an issue n the board where we had to cut a track as Pin 7 of the MAX238 was routed to pin 3 (RD) of the serial connector which in addition is routed to pin 19 (RD) of the 1772. If the WD1772 was already having issue may be the tipped it over the edge.

I can not test that WD1772 is bad as I can not plug into anything i have easily as all my existing systems WD1772 are soldered not socketed.

Waiting for new WD1772 which may help in diagnosis

We don't care where the serial comes from for the purposes of the discussion. On a standard ST TTL serial was converted bidirectionally to rs232 levels via a 1489 and a 1488 which have been replaced by a MAX238 which is powered by 5Vs only. RS232 didn't have the spec changed.. the MAX238 generates the the rail voltages +10 to -10.. They are still there, didn't go away. Still need them.

The error was that the 1772 RD pin (19) was tied to the the RS232 input pin on the MAX238 which comes from my MAC's RS232 output. That level was approximately -8 volts which showed up on the 1772 RD pin which clamped it to -1 to -2 volts. The discussion was concerning possible damage to the 1772 having the negative voltage applied to its 1772 RD pin. This discussion really has absolutely nothing to do with which chip created the ttl serial. In the the future just for you I will make it my priority to say MFP.. :) :) :)

Icky wrote: 13 Oct 2018 01:02

rpineau wrote: 12 Oct 2018 22:29 Serial (RS232) port on ST is done by the MFP, not the ACIA. As said before, serial buffers were replaced with MAX version.. no more -12v needed.

Understood but there was an issue n the board where we had to cut a track as Pin 7 of the MAX238 was routed to pin 3 (RD) of the serial connector which in addition is routed to pin 19 (RD) of the 1772. If the WD1772 was already having issue may be the tipped it over the edge.

I can not test that WD1772 is bad as I can not plug into anything i have easily as all my existing systems WD1772 are soldered not socketed.

Waiting for new WD1772 which may help in diagnosis

Yes I forgot that it was the RD pin which comes from another device and can definitely swing between -12 and +12v as the serial port was used at some point (the MX238 will not generate +/- 10v on this pin as it's an input, but the connected computer serial port could).
So this could totally have damaged the 1772... My mistake.

I spent a lot of time on serial port and after the fix the 1772 passes all the drag cart floppy tests. I think the saving grace is the -12 from the MAC didn't have very much drive capability so the 1772 was able to clamp it without hurting its self. Hopefully Ickys didn't get hurt either.