Building a VersaTerm - A Retro Computing Serial Terminal
Re: Building a VersaTerm - A Retro Computing Serial Terminal
I've also made this Versatherm. But I have a strange issue. If I connect RX and TX with a jumper, the Versatherm echoos my keys on the screen, so I suppose it works good.
If I connect a board, e.g. my Z80-retro-computer which works great on the ASCII VT100 video terminal of Geoff Graham, I don't see anything on my screen with the Versaterm-PCB.
I had the idea to invert the TTL-output and connect it to the RX of the Versaterm, and then I got some rubbish on the screen. Tried different baudrates, but no readable text when I pressed the reset button on the Z80-board.
No shortcircuits, no wrong resistors, diodes are okay, USB input works because I can go into the menu of the Versatherm.
Does anybody have an idea? I'm out of it at the moment.
If I connect a board, e.g. my Z80-retro-computer which works great on the ASCII VT100 video terminal of Geoff Graham, I don't see anything on my screen with the Versaterm-PCB.
I had the idea to invert the TTL-output and connect it to the RX of the Versaterm, and then I got some rubbish on the screen. Tried different baudrates, but no readable text when I pressed the reset button on the Z80-board.
No shortcircuits, no wrong resistors, diodes are okay, USB input works because I can go into the menu of the Versatherm.
Does anybody have an idea? I'm out of it at the moment.
Re: Building a VersaTerm - A Retro Computing Serial Terminal
Hi. It's hard to say without a closer look. I'm assuming you're just using the TTL serial connections, and have the voltage level switched to 5V (to match your likely Z80 ouput logic levels)?
Your external loopback was a good initial test. Do you have any other test equipment you can utilise? A scope would be very helpful here.
You shouldn't need to invert the output. You could test the non-transmission level easily with just a multimeter, to check you have the right signal polarity (but I'd say this is unlikely to be the problem).
The most common cause of serial issues, is either transposed Rx / Tx connection, baud-rate mismatch, or hardware handshake issues (but your external Rx <-> Tx link loopback test would seem to rule that one out).
Your external loopback was a good initial test. Do you have any other test equipment you can utilise? A scope would be very helpful here.
You shouldn't need to invert the output. You could test the non-transmission level easily with just a multimeter, to check you have the right signal polarity (but I'd say this is unlikely to be the problem).
The most common cause of serial issues, is either transposed Rx / Tx connection, baud-rate mismatch, or hardware handshake issues (but your external Rx <-> Tx link loopback test would seem to rule that one out).
Re: Building a VersaTerm - A Retro Computing Serial Terminal
Some terminals are fussy about the number of stop bits in use, this has caught me out once or twice. I always find it best to get comms working with a PC where it is easy to change setting and to then switch to a target system when you know things are working as expected. i.e. use a USB FTDI device to talk to the versaterm from a pc for starters. hth David.
Re: Building a VersaTerm - A Retro Computing Serial Terminal
Thanks all for the reply.
I think there is something wrong with the level of the Z80 board, despite working on the other ascii terminal.
I connected the output from the Z80 board directly onto the 9pins serial port and Versaterm shows something more, so I have to look for something else with a TTL-output and I will see what happens then.
This will take some time but I will continue with the terminal till it works
I think there is something wrong with the level of the Z80 board, despite working on the other ascii terminal.
I connected the output from the Z80 board directly onto the 9pins serial port and Versaterm shows something more, so I have to look for something else with a TTL-output and I will see what happens then.
This will take some time but I will continue with the terminal till it works

Re: Building a VersaTerm - A Retro Computing Serial Terminal
By the way: I have the Versaterm V1.2 PCB, while I noticed version 1.0 on github.
Does anybody know the difference between them? I've googled on the internet but I can't find anything about this.
Which PCB-version do you use?
Does anybody know the difference between them? I've googled on the internet but I can't find anything about this.
Which PCB-version do you use?
Re: Building a VersaTerm - A Retro Computing Serial Terminal
My VersaTerm PCB is also v1.2, as can be seen in my assembled PCB photo here (last photo): Building a VersaTerm – A Retro Computing Serial Terminal
I ordered my PCBs from the github published gerbers, so it must be all good.

Re: Building a VersaTerm - A Retro Computing Serial Terminal
This suggests that your Z80 board has RS232 serial drivers.PE1KEL wrote: ↑Thu Feb 20, 2025 9:05 pm I connected the output from the Z80 board directly onto the 9pins serial port and Versaterm shows something more, so I have to look for something else with a TTL-output and I will see what happens then.
This will take some time but I will continue with the terminal till it works![]()
If it's old-school, look for a MAX232 chip, or MC1488/MC1489 or SN75188/SN75189 RSR232 driver chips.
You definately don't want to connect RS232 levels to the TTL terminals. But you can use the 9-pin D connector instead, which carries the RS232 level serial interface.
Re: Building a VersaTerm - A Retro Computing Serial Terminal
I finally solved my problem.
Despite the Z80-board high level is 4.5 volts provided by an MC6850P ACIA (I can't measure the voltage in low-state) it didn't work.
For my radio-hobby I have a MFJ1270 packetmodem, and while it is called TTL-level, the high-level was 3.3 volt, and the low was about 0.6 volt.
And this MFJ-thing also didn't work with the Versaterm, but this was a bit understandable with these levels.
Some days ago I was looking for a TTL-IC with inverting buffers and a wide voltage range. I took the 74HC04.
The idea was to reach a real TTL-level and not depending on the input level, and take advantage these inverters will buffer too.
So the RX input onto pin 1, the output of the first inverter to pin 3, the input of the next inverter, and the output to the RX of the Versaterm.
Power is supplied from the same TTL-connector of the Versaterm.
I use 2 buffers in serial, so the data will not be inverted and see: the Z80 and the packetmodem worked on the ttl-input of my Versaterm.
That means that I have to use 2 buffers in series for the CTS-line as well.
I wanted you to let it know and thanks for all the help.
Maybe it can help someone else if one finds the same problems.
Despite the Z80-board high level is 4.5 volts provided by an MC6850P ACIA (I can't measure the voltage in low-state) it didn't work.
For my radio-hobby I have a MFJ1270 packetmodem, and while it is called TTL-level, the high-level was 3.3 volt, and the low was about 0.6 volt.
And this MFJ-thing also didn't work with the Versaterm, but this was a bit understandable with these levels.
Some days ago I was looking for a TTL-IC with inverting buffers and a wide voltage range. I took the 74HC04.
The idea was to reach a real TTL-level and not depending on the input level, and take advantage these inverters will buffer too.
So the RX input onto pin 1, the output of the first inverter to pin 3, the input of the next inverter, and the output to the RX of the Versaterm.
Power is supplied from the same TTL-connector of the Versaterm.
I use 2 buffers in serial, so the data will not be inverted and see: the Z80 and the packetmodem worked on the ttl-input of my Versaterm.
That means that I have to use 2 buffers in series for the CTS-line as well.
I wanted you to let it know and thanks for all the help.
Maybe it can help someone else if one finds the same problems.
Re: Building a VersaTerm - A Retro Computing Serial Terminal
Great to hear you have it up and running!
Please note that standard TTL levels are generally: logic low is < 0.8V, and logic high is > 2.4VPE1KEL wrote: ↑Sun Mar 09, 2025 12:38 pm Despite the Z80-board high level is 4.5 volts provided by an MC6850P ACIA (I can't measure the voltage in low-state) it didn't work.
For my radio-hobby I have a MFJ1270 packetmodem, and while it is called TTL-level, the high-level was 3.3 volt, and the low was about 0.6 volt.
And this MFJ-thing also didn't work with the Versaterm, but this was a bit understandable with these levels.
On this basis, a level of either 3.3V or 4.5V would certainly be compliant with a high TTL logic level.
Also, for reference, I regularly use my VersaTerm TTL inputs directly with a MC6850 ACIA's TTL level Tx / Rx pins. As I expect, this works perfectly.
As noted above, you already have appropriate TTL levels, so I believe there is something else going on here. Using a couple of TTL invertors, to buffer the signal, should not be needed.PE1KEL wrote: ↑Sun Mar 09, 2025 12:38 pm Some days ago I was looking for a TTL-IC with inverting buffers and a wide voltage range. I took the 74HC04.
The idea was to reach a real TTL-level and not depending on the input level, and take advantage these inverters will buffer too.
So the RX input onto pin 1, the output of the first inverter to pin 3, the input of the next inverter, and the output to the RX of the Versaterm.
Power is supplied from the same TTL-connector of the Versaterm.
I use 2 buffers in serial, so the data will not be inverted and see: the Z80 and the packetmodem worked on the ttl-input of my Versaterm.
That means that I have to use 2 buffers in series for the CTS-line as well.
It would be nice to get to the true root-cause of your problem.
A couple of things that I would check first:
a). Have you carefully checked the assembly of the VersaTerm around the TTL input circuitry. In particular the R25 - R27 and R29, R30 resistor values, D3 - D6 1N4148 diodes, and the Q4 - Q7 2N7000 mosfets?
b). You mention 2 buffers on the CTS line. Have you tried simply grounding the CTS input and using only Rx / Tx lines (no hardware handshaking)? An error I've made more than once over the years (when using a MC6850 ACIAs), is forgetting to ground the CTS input when using a basic Rx / Tx serial interface. With CTS floating, or held high, the serial interface does not work!
Hopefully, you can find the time to investigate further, as I don't believe your need for a TTL buffer is not the root-cause solution (for the reasons noted above). I'd love to hear that you eventually find the underlying issue.
Re: Building a VersaTerm - A Retro Computing Serial Terminal
The 2 buffers for the CTS line is for the future, because on the RX line the 2 buffers worked with me.
Of course I have checked the resistors, the value, the diodes and the mosfets, measured them and it is okay. I did that many many times. I even checked the PCB for cracks in the cupper lines. But with the loopback jumper everything worked like it should.
My setup was only RX and GND, because my Z80-board and packetcontroller gives a messages during startup.
Grounding the CTS input I've never heard of. I only know that in the settings of the terminal you can change the settings of the RTS and CTS line.
And I played with that too.
But next week I will try your proposal to ground the CTS input and I will let you know !
Best regards.
Of course I have checked the resistors, the value, the diodes and the mosfets, measured them and it is okay. I did that many many times. I even checked the PCB for cracks in the cupper lines. But with the loopback jumper everything worked like it should.
My setup was only RX and GND, because my Z80-board and packetcontroller gives a messages during startup.
Grounding the CTS input I've never heard of. I only know that in the settings of the terminal you can change the settings of the RTS and CTS line.
And I played with that too.
But next week I will try your proposal to ground the CTS input and I will let you know !
Best regards.