SC126 questions

[Mark Cohen]

Hi,

I built a couple of SC126s and they have issues.

On all boards I can only see mostly decent console output only at 38400 with ROMWBW 34/35.

So I am thinking I have some rebranded Z180 chips? How can I tell what they really are. I do not see any testers avaailable.

I did switch different crystals and it made no difference. My crystals are in sockets. Could that be interfering?

Also on 1 board I do not see the green leds going off in sequence? What chip should I check? The yellow LED is lit.

I have checked the solder joints and I do not see any issues although I may reflow them.

Thanks,

Mark

[MartinR]

Hi Mark - 

Sorry to learn you've been having issues with your SC126s. 

I have been using an SC126 I built from one of Steve Cousins' kits and it's been working fine for about 2.5 years with its 18.432MHz oscillator mounted in a socket, and it works fine at 115.2kbps into Tera Term. So, I don't think the oscillator being in a socket is likely to be your problem. If you have a spare oscillator you could swap them and see if it makes any difference.

For the board that is not showing any signs of working:

The yellow LED is the power indicator, so the fact that's lit is a positive sign.

Steve lists the purpose of each of the RomWBW diagnostic LEDs at: https://smallcomputercentral.com/romwbw/romwbw-getting-started/ This info can also be found in the RomWBW documentation. If no green LEDs come on then the SC126 is not really getting initialised. Is your 5V supply stable? Is the CPU being repeatedly triggered or held in a reset state by the DS1233 chip?

I think you need to start with the basics: is the board's construction is good? If you've not done it already, thoroughly check for bad solder joints or solder bridges; check for shorts on adjacent pins; using the circuit diagrams, thoroughly check for continuity; and then check for a good and stable 5V supply. 

You could email Steve and ask his advice. He has always been incredibly helpful in my experience.

For reference, the Z180 on my SC126 is a Z8S18033VSC and is rated at 33MHz, though I use an 18.432MHz oscillator. If you want to check the provenance of your Z180 then physically examine it for the top/bottom surfaces being different where the top has been remanufactured and a new logo applied. You may also be able to use some acetone to see if the logo and part number etc can be removed (it can take a little perseverance). Have a look on YouTube for videos about ICs that have been 'blacktopped'. Even if your CPU has been remanufactured ('blacktopped'), I wouldn't necessarily discount the part - I've used ICs (though not my Z180) that are obviously 'blacktopped' without issue.

Good luck!

MartinR

[MartinR]

Just a very quick thought... The SC126 has two ROMs. Have you set the jumpers correctly to boot from the ROM containing the code?

[Richard Deane]

I also have an sc126 with socketed 18mhz crystal , made shortly after sc126 release, and has been working fine ever since. Baud rates to 115200 without problem, typically with Tera Term.

Richard

--
You received this message because you are subscribed to the Google Groups "retro-comp" group.
To unsubscribe from this group and stop receiving emails from it, send an email to retro-comp+...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/retro-comp/381a1475-1999-46bb-910c-9bbed5e930a5n%40googlegroups.com.

[Steve Cousins]

Hi Mark

If the suggestions from Martin don't lead you to a solution, try following the detailed troubleshooting guide for SC126:

https://smallcomputercentral.com/rcbus/sc100-series/sc126-z180-motherboard-rc2014/sc126-v1-0-troubleshooting/

Where did you source the components?

Steve

On Tuesday, 20 May 2025 at 01:52:04 UTC+1 markc...@gmail.com wrote:

[Mark Cohen]

Hi,

Thank you for the quick replies.

RAM,ROM and CPU sourced from Aliexpress. CPUs might be from 1 vendor.

The rest Digikey or Mouser.

I suspect the CPUs are mislabled, but I don't really want to use acetone to clean the tops off in the hopes I can find what it was originally.

I will see about working through the diagnostics and reflowing the joints.

Thanks.

[David Reese]

Hi,

Steven's suggestion of checking the troubleshooting guide is best advice so far.  As for the green LEDs not firing, check the usual suspects:  polarity (ohmmeter leads forward/reversed bias) solder joint quality, and the pull-up resistors.  If those are okay, then definitely follow the troubleshooting guide in sequence, as it starts with the basics and progresses through the system in a very structured and organized manner.

Power supply issues can be eliminated by using a battery pack that delivers 5 VDC, but most commercial models these days will "sleep" after a little while because the SC126 doesn't draw enough current to keep them alive.  If you can find a 10,000 ma/h pack that will stay alive for it, your SC126 will run on it for about 100 hours.  (There are also "keep alive" circuits you can build that will help with this)

Some "wall wart" Class 2 types are quite noisy, and could inject ripple into your power feed at 50/60 Hz, depending on where you live.

What sort of terminal are you using for your console output?  I see you're using RomWBW 34/35.  I can't right now, but when I get some time, I'll fire mine up and see what I have.  I know I flashed mine a few years back and haven't had any problems running at 115,200 baud using minicom for a terminal. 

One thing you might try:  replace your FTDI to USB board if you have a spare one.  I bought (I think) three of them with my kit so I would have spares.

The SC126 is a great little board.  Mine still starts every time I turn it on, and runs like greased lightning compared to any other CP/M-based system I've used over the years.  Stay with it, man, and you'll have a couple of great machines.

Regards,
David

--

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

To view this discussion visit https://groups.google.com/d/msgid/retro-comp/541a9708-ba08-4d09-bc40-da0ad22f8ddcn%40googlegroups.com.

[Mark Cohen]

I actually have a working sc126 so I have something to compare. I have many FTDI boards. That is not the issue.

I soldered the 18.x crystals in directly. These are from mouser.

Now I can see that scm works. The leds work with scm.

But the baud rate is off with romWBW so I do think I have mislabeled CPUS, maybe 25 of them! The vendor is no longer on AliExpress. Wonder why?

Getting some new ones this week via ebay.

Thanks,

Mark

[Wayne Warthen]

Yes, I suspect you are right Mark.

In the RomWBW boot messages, the variant of Z180 is identified.  It will say "Z180", "Z180-K", or "Z180-N".  Z180 is the older variant and will only run at 1/2 oscillator speed.

Thanks, Wayne

[Mark Cohen]

great!

I will take a look and catch that.

I recall there is a spot to change the speed like half it and recompile?

I have to look that up again

Thanks,

Mark

To view this discussion visit https://groups.google.com/d/msgid/retro-comp/fbb758c6-e14b-4b3a-bf4a-22115056aa8en%40googlegroups.com.

[Wayne Warthen]

Yes, you can change the ROM to start in half-speed mode.  However, it won't really do anything.  If RomWBW detects an older Z180 CPU, it automatically realizes it is running at half-speed.  The CPU speed in the boot messages should reflect this (9.216 MHz).  The CPUSPD command should also indicate that the CPU is running at half speed.

The real problem is that a Z180 running at 9.216 MHz (as yours is) has no baud rate divisor that will result in 115,200 baud operation.  See https://github.com/wwarthen/RomWBW/blob/master/Doc/Z180%20ASCI%20Baud%20Rate%20Options.pdf.

You could change the baud rate specified in the build config to a baud rate that is available at 9.216 MHz.

Thanks, Wayne

[Mark Cohen]

I can see the boot message.

RomWBW HBIOS v3.4.0, 2024-11-10

Small Computer SC700 [SCZ180_sc700] Z8S180-K @ 18.312MHz IO=0xC0
0 MEM W/S, 2 I/O W/S, INT MODE 2, Z180 MMU
512KB ROM, 512KB RAM
ROM VERIFY: 00 00 00 00 PASS

AY: MODE=RCZ180 IO=0x68 NOT PRESENT
ASCI0: IO=0xC0 ASCI MODE=115200,8,N,1
ASCI1: IO=0xC1 ASCI MODE=115200,8,N,1
DSRTC: MODE=STD IO=0x0C Sat 2000-01-01 00:00:24 CHARGE=OFF
MD: UNITS=2 ROMDISK=384KB RAMDISK=352KB
FD: MODE=RCWDC IO=0x50 NOT PRESENT
IDE: IO=0x10 MODE=RC
IDE0: NO MEDIA
IDE1: NO MEDIA
PPIDE: IO=0x20 PPI NOT PRESENT
SD: MODE=SC OPR=0x0C CNTR=0xCA TRDR=0xCB DEVICES=1
SD0: NO MEDIA
CH0: IO=0x3E NOT PRESENT
CH1: IO=0x3C NOT PRESENT

Unit        Device      Type              Capacity/Mode
----------  ----------  ----------------  --------------------
Char 0      ASCI0:      RS-232            115200,8,N,1
Char 1      ASCI1:      RS-232            115200,8,N,1
Disk 0      MD0:        RAM Disk          352KB,LBA
Disk 1      MD1:        ROM Disk          384KB,LBA
Disk 2      IDE0:       Hard Disk         --
Disk 3      IDE1:       Hard Disk         --
Disk 4      SD0:        SD Card           --

To view this discussion visit https://groups.google.com/d/msgid/retro-comp/da0b72e4-58e6-4225-9b08-29fd39ef6d3cn%40googlegroups.com.

[Mark Cohen]

The other 1.

RomWBW HBIOS v3.4.0, 2024-11-10

Small Computer SC700 [SCZ180_sc700] Z8S180-N @ 18.288MHz IO=0xC0

0 MEM W/S, 2 I/O W/S, INT MODE 2, Z180 MMU
512KB ROM, 512KB RAM
ROM VERIFY: 00 00 00 00 PASS

AY: MODE=RCZ180 IO=0x68 NOT PRESENT

ASCI0: IO=0xC0 ASCI W/BRG MODE=115200,8,N,1
ASCI1: IO=0xC1 ASCI W/BRG MODE=115200,8,N,1
DSRTC: MODE=STD IO=0x0C Sat 2000-01-01 00:00:17 CHARGE=OFF

MD: UNITS=2 ROMDISK=384KB RAMDISK=352KB
FD: MODE=RCWDC IO=0x50 NOT PRESENT
IDE: IO=0x10 MODE=RC
IDE0: NO MEDIA
IDE1: NO MEDIA
PPIDE: IO=0x20 PPI NOT PRESENT
SD: MODE=SC OPR=0x0C CNTR=0xCA TRDR=0xCB DEVICES=1
SD0: NO MEDIA
CH0: IO=0x3E NOT PRESENT
CH1: IO=0x3C NOT PRESENT

[Wayne Warthen]

Hmmm... The -K and the -N should both run at full speed and 115200 baud just fine.

I may have misunderstood your problem.  One of your posts above says that the baud rate is "off" when running RomWBW.  Can you expand on that?  What exactly are you seeing?

Thanks, Wayne

[Mark Cohen]

38400 is the only baud rate that works somewhat  reliably with these boards in teraterm.

Other boards run at 115200.

No leds except power.

Small computer monitor seems to always work at 115200 including LEDs.

To view this discussion visit https://groups.google.com/d/msgid/retro-comp/e57f360e-e423-42aa-8f92-83594cdc8c62n%40googlegroups.com.

[Wayne Warthen]

Wow, that is odd.  Especially the LEDs.

I am a little confused by something.  Your first post said the problem boards are SC126, but the ROM builds are for SC7xx series boards.  Can you clarify what exactly the system is made up of?

Thank you, Wayne

[Mark Cohen]

Yes sc126.

I'll take a look at that build.

That's the one I had on hand as I originally thought the downloaded 3.5 was the problem.

That could be the led issue.

To view this discussion visit https://groups.google.com/d/msgid/retro-comp/90bac64a-61e1-461a-869e-1a6132b2fe93n%40googlegroups.com.

[Mark Cohen]

With the correct version of the firmware the LEDs work correctly.

But the baud rate issue remains.

New (to me) CPUs arrive this week.

TY!

[Sam Sawyer]

This problem with wrong baud rate with only 38400 baud working sounds like something I encountered when I used an an 18.000 MHz oscillator instead of the specified 18.432 MHz oscillator. Double check the oscillator and replace it with correct one if its is the more common 18.000 MHz version.

[Ronny Ribeiro]

Yes, one boot message is showing 18.312 MHz and the other 18.288 a MHz, I think Sam spotted something here...

To view this discussion visit https://groups.google.com/d/msgid/retro-comp/574328bd-f21b-4d1e-8886-033ca94c6a40n%40googlegroups.com.

[Mark Cohen]

I actually used 3 different brands. I finally settled on the mouser ones. But perhaps it's something related. Isn't there a capacitor or something that goes with the crystal?

To view this discussion visit https://groups.google.com/d/msgid/retro-comp/CAEg%2B7vBAW5HPe_B09HDKeaoNC61ZEXBnNguRCPEqgq_NH%3D7GEg%40mail.gmail.com.

[Ronny Ribeiro]

I don't think there must be capacitors for a canned resonator to resonate. Crystal oscillators do need a pair of capacitors, but canned resonators have them inside. Can you confirm your resonators are not 18.432 MHz? This is certainly going to cause issues by generating baud rates slightly off. And for larger rates, larger deviations, which would explain your system not being able to communicate at higher rates...

[Ronny Ribeiro]

PS: Are you using crystals to build the oscillators?

[Wayne Warthen]

On Tuesday, May 27, 2025 at 9:10:08 AM UTC-7 ronny...@gmail.com wrote:

Yes, one boot message is showing 18.312 MHz and the other 18.288 a MHz, I think Sam spotted something here...

Don't put too much confidence in the CPU speed shown in the boot messages.  It is not terribly accurate.  The measurement depends on the accuracy of the DS1302 RTC, so if it is running fast or slow, the CPU speed will be off.  Even if the DS1302 is running perfectly, there is a fair amount of slop in the measurement technique.  The measurements I see are well within the tolerance of the measurement technique if the CPU is running exactly at 18.432 MHz.

With that said, if the CPU oscillator is truly off, it will certainly cause problems at higher baud rates.

Thanks, Wayne 

[Ronny Ribeiro]

Thanks for the clarification, Wayne.

--

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

To view this discussion visit https://groups.google.com/d/msgid/retro-comp/13615824-9413-48da-848a-fd17a818cb63n%40googlegroups.com.

[Mark Cohen]

Changing the CPU made no difference.

romWBW is now 3.6

The SD card works fine.

Could the X2 crystal interfere with the baud rate if it is not correct?

All the resistors and capacitors were tested prior to use. all the 74HCT* chips were also tested. 

Its probably something real simple.

I will also start doing some testing.

To view this discussion visit https://groups.google.com/d/msgid/retro-comp/CAEg%2B7vCtFAfHVgZordR4%2BixwJ7fZGWtzJ0Tca4y_TFkBvjMkZA%40mail.gmail.com.

[Wayne Warthen]

On Tuesday, May 27, 2025 at 5:20:47 PM UTC-7 markc...@gmail.com wrote:

Changing the CPU made no difference.

romWBW is now 3.6

The SD card works fine.

Could the X2 crystal interfere with the baud rate if it is not correct?

All the resistors and capacitors were tested prior to use. all the 74HCT* chips were also tested. 

Its probably something real simple.

I will also start doing some testing.

OK, your question about the X2 crystal made me think of something.  When the baud rate divisor is computed by RomWBW, it does so based on the CPU frequency.  If the measured CPU frequency is off enough, then the divisor programmed will be wrong.  That may be what is happening here.

As a way to test this, you could disable the DSRTC device in your config file.  That will also disable the dynamic CPU speed measurement and the CPU speed will just default to 18.432 MHz.  To do this, you need to add the following line to your config file and build a custom ROM.

DSRTCENABLE .SET FALSE ; DSRTC: ENABLE DS-1302 CLOCK DRIVER (DSRTC.ASM)

Thanks, Wayne

[Mark Cohen]

Hi Wayne!

Yes that was it!

I guess I need to order a couple of parts for the RTC now.

A replacement crystal and a DS1302+. Although I thought my DS1302+ would be okay as I just got them from mouser.

So probably just the crystal then.

Thanks all for the debugging help.

Regards,

Mark

--

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

To view this discussion visit https://groups.google.com/d/msgid/retro-comp/619650d1-c386-43f1-ab3a-2a902d122a54n%40googlegroups.com.

[Steve Cousins]

Mark

It is important the 32.768 kHz crystal has 6pF load capacitance rating. eg:
https://www.mouser.co.uk/ProductDetail/ABRACON/AB26T-32.768KHZ-6?qs=4ZuRpliPd6xfrWwj%252BakZdw%3D%3D

Steve

[Mark Cohen]

Yes. That was probably my issue.

Thanks.

To view this discussion visit https://groups.google.com/d/msgid/retro-comp/468819eb-cfa8-457d-aa09-62e571011959n%40googlegroups.com.

[Wayne Warthen]

Glad to hear that resolved it.

RomWBW does some rounding tricks to try and accommodate minor variances in the CPU speed measurement when calculating the baud rate divisor.  It appears that your measured CPU speed was just a little too far off for this.

Sorry I didn't think of this earlier.

Thanks, Wayne

[Mark Cohen]

It all worked out. Ty

To view this discussion visit https://groups.google.com/d/msgid/retro-comp/d7cb21b8-7e2c-478d-806c-fc936e5261f0n%40googlegroups.com.