Help with Tynemouth 68B50 serial port v1.1

[Christer Karlsson]

I have some problem, I have a small system made from duplicates.
RC2014 Backplane 5
RC2014 z80 CPU v2.1
SC602 memory v1.1.0 (SCM-S7, RomWBW 3.5.1)
RC2014 IDE HardDrive Module v1.0
and
Tynemouth 68B50 Serial Port v1.1

This system is all of a sudden not communicating with TeraTerm over FTDI cable.

I have so far on the card checked:

1. I have tried different settings for flow control in TeraTerm. 
2. I have tried swapping positions of the card on the backplane.
   
3. All connections from the 40- and 6 pin connectors have continuity.
4. The clock produces 7.3728 MHz (not perfect square-wave but in my mind good enough)
5. The clock signal are visible on pin  21 on the 40-pin connector, and on pin 3 and 4 on the 68B50.

I have a another system as well (RC2014 Pro set-up for just CP/M 2.2). By swapping card between the systems I have narrowed the problem to the Tynemouth card.  The CPU card works, the RAM/ROM card works, and the IDE card works, but the Tynemouth does not work in any configuration.

So now I don't know what to do, so I need some ideas.  My uneducated guess would be that it is the 65B50 that is damaged, short of buying a new one, is there anyway I can confirm this?

/Christer

[7alken]

hi, not having this card (waiting for some parts for my wild system still, I plan to use cmos hd63b50 in sc139), I peeked on that card and ... did you tried both clock from system bus and the internal card oscillator? there are jumpers (from theirs build instructions) - may be even pulling out that 7404 may help to isolate something? they drive also 2 LEDs and support IRQ, well;
BUT, peeking into schema of sc139, it has also some 100k pullup/down on RX and CTS, not sure how important there, interesting ... I have also ebay links for those hitachi ACIAs, both DIP and SOIC packages, ... cant wait to test it all;

the tynemouth schema
https://45ab1267a0f78266cbcc.b-cdn.net/wp-content/uploads/2023/11/Tynemouth-68B50-Serial-with-clock-for-RC2014-V1.1.pdf

the sc139 schema
https://drive.google.com/file/d/1ynwaNImr62LZED6-mIoRNQO-ScKGtn6y/view

here I found also some crazy history and development around that ACIA card ... he went wild later ))
http://blog.tynemouthsoftware.co.uk/2025/11/a-new-serial-card-for-minstrel-4th-and-rc2014.html

Petr

[7alken]

.. and that nice/sexy layout of parts for xosc on tynemouth look ... suboptimal ))
P.

[7alken]

and ... wow, it has no decoder ... is this okay in your system??

[Michael Druckenmiller Sr]

First thing I see is a 7Mhz clk for TX, I don't see any divider and the spec for a 68B50 is 2 Mhz.

The decoding he is doing will work, but may conflict with other ports. I'd map out what ports are actually affected and double check what other stuff is on the bus. I wrote my own driver for mine.

But, with the RC2014 accepting so many different software packages...

I've attached a blurb for the 68B50 I created for my TS2068 which after I corrected all my memory induced assumptions works quite well.

Also attached is the schematic for the TS2068 IO which may give some insight. It's all Z80, right?

I've really got to get my bench cleared off and finish my "Pro".

I assume that Tynemouth included driver data and addressing information? And, how to use it with CP/M or the other RC2014 software?

Mike Sr aka Micro Mikie

--
You received this message because you are subscribed to the Google Groups "RC2014-Z80" group.
To unsubscribe from this group and stop receiving emails from it, send an email to rc2014-z80+...@googlegroups.com.
To view this discussion, visit https://groups.google.com/d/msgid/rc2014-z80/6b0a7cc3-c958-44be-bce3-b9732faa97b8n%40googlegroups.com.

[Alan Cox]

The Tynemouth card I think follows the original original 68B50 design so occupied 25% of the I/O space rather than having a tight decode. Fine for the original setup but not for much else.

The 2MHz thing commented on is I think a misunderstanding. It is rated for a 2MHz E clock on a 6800 bus, and as the Z80 bus works differently is in timing spec 

--

You received this message because you are subscribed to the Google Groups "RC2014-Z80" group.
To unsubscribe from this group and stop receiving emails from it, send an email to rc2014-z80+...@googlegroups.com.

To view this discussion, visit https://groups.google.com/d/msgid/rc2014-z80/fe540a35-803f-43fe-b803-6297739d32dan%40googlegroups.com.

[Spencer]

Hi Christer,

You say "This system is all of a sudden not communicating..." which implies that it had been working. The obvious question is what has changed?

As this module has two functions (clock and serial) then maybe it is worth testing those individually.  If you remove the 68B50 then it's just a clock module, so you should be able to run that with your SIO/2 module from the RC2014 Pro. As you have already checked for a square-ish signal, then I suspect that this will be fine.

If you set the jumpers so that the system clock is driving the 68B50 instead of the one on that module, then you can use the clock module from your RC2014 Pro to see if that makes a difference.

Note that the 68B50 chips went End Of Life many years ago, so all of these chips are used and reclaimd. (New Old Stock chips are available from China. These are also used and reclaimed, but also cleaned and remarked!).  I do a pretty thorough test on all the 68B50 chips I supply, so I know that it would have been working in 2024 when it was supplied. I cannot guarantee that it is still working now though.

Spencer

--

[Christer Karlsson]

Thank you all!

"All of a sudden" = I had mostly played with my RC2014 pro and a system built around an SC722.

Spencer, I will do as suggested and try to use it first as only a clock and see if that is my problem, then do the opposite, and take the clock from the bus. Yep, that 68B50 worked when it was delivered. I got this card because the SIO was in short supply around then, and also it reduced the number of cards by one as it combines the IO and clock on one card.  It worked great for a long time, so there is no problem with your products Spencer. When I expanded it to the 'Blue-box', The SIO was back in stock and I had room for a dual clock module as well, so I removed this card, and put it together with other spare parts. I know I ran it at least once, and the ROM is also RomWBW. I know I must have had it up at least last year.   

Mike, thanks, it is a good idea to try to see if I have a port conflict.

So I have a plan for the moment.

Isolate the two parts on the card and see what happens (I should be able to scope the CPU and ROM without having IO). Make sure there is no port conflict.  I can also look into SC610, a similar idea but it allows you to address the ports. This would also allow me to recycle most of the components on this card.

/C

To view this discussion, visit https://groups.google.com/d/msgid/rc2014-z80/8sMJS8krad9JxbIr1J5RLO3L6zrnbEXCkQ-Hq-wDLH_F7GipI0xr7g9wgU072rkMlzuWT2tBqpUoI8Gh7tPSqlRfSdhJty-KJcKyhD2OnG4%3D%40z80kits.com.

[Christer Karlsson]

All well, all is working again.

Do not know why, or what went wrong.  I was measuring continuity and noticed that slot 5, where I had the ROM/RAM module signal A15 didn't reach. So, I changed the order of the cards. The only limitation I had was that the Hard Drive module had to be the first (or the 44-pin IDE to CF adapter could potentially shortcut to the previous card), and I put the Tynemouth 68B50in slot 5, and behold now it worked!

Why? Well, Tynemouth 68B50 does not use A!5 (as pointed out earlier it only uses A7 and A6. The chip is active when A7 & ~A6 (address 0x80 - 0XBF)), and for some reason pin 1 on slot 5 seems to be unconnected on my Backplane-5, there is no trace between pin 1 of slot 4 and pin 1 on slot 5 on solder side. I had never noticed that before.

Thanks everyone for the help and support, many great ideas, great community!

/C