Yeah old computing technology always amuses me and I think I am going to gradually work my way through everything and possibly even design a little plug-in expansion card. No idea what function it will perform. Possibly just to let the smoke out of the motherboard.
There’s definitely lots more retro and discrete design coming up. I’m finally getting around to cleaning out the basement and seeing all the things I never remembered I had down there and putting it all to use. We’re talking over 20 years of obsolete technology! I might even get into some vacuum tube stuff.
I got my hands on an old hitachi analog scope, one of the later ones with a digital mode that can be toggled and a data save function. To transmit this data to an HP plotter, it has an RS232 D25 port that spits out ASCII HPGL language. So I set to building a simple little RS232 datalogger with an arduino and SD-card module, and a MAX232 from alibay. It worked just fine to keep one of the lines (RTS I think) at a constant value to tell it to keep sending data, but when my code toggled this pin twice per byte (in order to synchronise each byte) or when my testing button bounced on this pin, the MAX232’s +/-10V rails would collapse to both be about +2V and neither of them would return to their proper values until I cycled power to the entire circuit. Not sure if this was typical behaviour of a MAX232 or not (I know I had the recommended capacitances on it) but I ditched it and replaced my receiving circuit with an NPN inverter with an inverted diode from base to collector (plus base resistor) in order to get a 0V-5V signal, which worked fine. What surprised me more is that simply sending a 0V-5V signal up the RTS and into the oscope worked absolutely fine. Reading the datasheet of the scope’s RS232 transceiver it had a 2V threshold, so I guess it’s more important to be within spec for your ICs than for the standard itself. A $8 USB logic analyser proved vital for debugging the thing, since I couldn’t use my oscilloscope while it was transmitting data. Then to ditch the clumsy USB lead powering the thing I attempted to run a rectifier+filter cap+7805 off the unused CTS pin to power the arduino off that, but as it turns out my nano’s quiescent current was a few mA thanks to the power LED, and the 7805’s minimum current was also suboptimal, and the circuit lost power each time I tried to power RTS. So I ordered some better linregs and am attempting to get a stand-alone ATMega working for this, the caveat being they come with their fuses set to their internal RC clocks, not for the 16M crystal that I planned on slapping on there. Quite the rabbit hole for a free oscilloscope. I look forward to what you do with this project as it might help my own escapades.
Even at this initial stage, I also already went down rabbit holes just to get this much working and there’s a whole big unknown territory coming up with what I’m trying to do next. Based on GnuReligion’s comment that many RS 232 interfaces might except ground as the negative voltage, and if the positive voltage has to be at least 3 V, then 0 V to 5 V may very well work. I have at least two working analog Tektronics scopes down in the dungeon and one of them has some sort of digital on screen Readout and storage I think, but I’m not sure what kind of ports and features are on the hardware interface. I might dig those out and check out if there’s anything to hack.
Hahah I remember waaay back when.. those mice with the rubber balls were so annoying, cause after a while, the rollers or sensors would get dust and crap on them.. messing it up and making it not work right.. so you had to take it apart and clean it out.. nice!!!!
My worst memory of those mice would be how the cover that holds the ball might get loose over time so I would epoxy it shut and then I can’t easily open it to clean the dust out.
I can remember the rs232 cable cook books, where there was about 100 or more ways of connecting the PC to the other device. before the PC our company computer used VT100 terminals. 100'sof rs232 cables running around the lab. the hard drive was about the size of a washing machine. RS232 is still a pain to code for, without handshaking it's a bugger to know when the data has stopped being sent.
Well now you are just encouraging me to go get out an old microcontroller and try doing some RS 232 code! In my earliest jobs sometimes there were still 12 inch monochrome monitors in use as dumb terminals with keyboards hooked up somehow to a big tangled mess of technology. And I think I had more cases of RS 422/RS 485 long cables running all over the place. Connecting one machine to another or connecting serial debug interfaces to machines etc. Lots of damaged and severed cables when rolling big machines around the lab for various testing purposes and in and out of hot and cold environment chamber rooms. Those were the days.
@@GadgetReboot we had a VAX -11 something, made up of a CPU, tape drive, hard disk with a top loading removable single disk. it had it's own air condition room the only one in the whole building. we would take it turns on a hot day to sit at the small desk in there. poor old thing got sold as scrap as it was replaced by a micro-Vax about the size of a PC. we had to keep it running because of the data stored had to be kept for 10 years or more. Happy days.
How about the flip-side to this idea: sending power to the rs232 to recharge a 6v internal battery? How would I hook up the red and black wires coming from an adapter and into the rs232 pins? 7 of the pins are used on my device.
great com port rs232 video... i was looking to see if the old pc port was male or female? then found this vid and it is what i am working on, rs232 signal power, perfect. thanks a lot...:)
Unsure of where you are going with this, but if you want to go to Serial TTL without a MAX232, there are a number of conversion hacks. Some even steal negative voltage. It seems this is usually unnecessary. 0V is undefined, but most PCs will politely interpret it as low. www.kswichit.com/ap275/ap275.htm
Yeah that sort of transistor interface with diode clamps to tame the levels is where I’m headed, although I still have some design requirements to figure out for example I may end up needing to not just level translate but multiplex two drivers down to one receiver, and I haven’t decided if it’s OK to OR them together sort of like open collector first come first serve with the intention that only one should ever be really using it at a time so it’s safe, or if I want to completely cut one off while the other is talking and therefore I need to do something more involved. It’s good to know we might be able to get away with ground as a valid low-voltage. Given the technology in question, I probably don’t need it to be totally compliant, just something that works with the hardware I’ve got on hand so I will breadboard test. Although if I end up making a PCB layout, it wouldn’t hurt to throw a few extra components on there just in case. And since I like to put jumpers all over the place maybe that could be one of the options, switching between a negative supply or just ground as the low-voltage.
@@GadgetReboot Not sure that I understand, unless you mean you mean to put TX+RX one one wire. Have seen that project online too, but never built it up. Will look out for your next video though ... am sure you are doing something cool! A thing I have wanted to do is connect a 10-pin IDT fe to a mobo's UART header, and another cable to floppy power, merge the two behind a bay cover with a MAX232, then come outside the case with 5 wires to control my homemade Arduino-like boards. It would be 5 wires because DTR is needed to pull the reset low. But the truth is ... I like the buffer that a USB hub offers, insulating my PC from evil MCU experiments. The MAX232, at $0.15/ea, calls to me though: www.aliexpress.com/item/32944645394.html
I should stock up on some Max232s at the rate I’m going lately. They might come in useful. I’m going to need to do some buffering and isolating of both power and data for the stuff I’m going to be doing so I think I have a big stack of optocouplers down in the dungeon that I have to locate and if I do have them, those are going to be put to use to give me peace of mind so I don’t blow up the rare motherboards. With the multiplexing or channel switching, what I would be looking to do is have something like a modem transmitting to the computer as well as my own circuit transmitting to the same receive port on the computer, only one at a time but being able to electronically control which one is allowed to talk.
That was great! I never knew that was possible, and that rs-232 didn't have a dedicated VCC pin. Mind blown!
Yeah old computing technology always amuses me and I think I am going to gradually work my way through everything and possibly even design a little plug-in expansion card. No idea what function it will perform. Possibly just to let the smoke out of the motherboard.
Please keep doing these videos, your content keeps getting cooler!
There’s definitely lots more retro and discrete design coming up. I’m finally getting around to cleaning out the basement and seeing all the things I never remembered I had down there and putting it all to use. We’re talking over 20 years of obsolete technology! I might even get into some vacuum tube stuff.
I got my hands on an old hitachi analog scope, one of the later ones with a digital mode that can be toggled and a data save function. To transmit this data to an HP plotter, it has an RS232 D25 port that spits out ASCII HPGL language. So I set to building a simple little RS232 datalogger with an arduino and SD-card module, and a MAX232 from alibay. It worked just fine to keep one of the lines (RTS I think) at a constant value to tell it to keep sending data, but when my code toggled this pin twice per byte (in order to synchronise each byte) or when my testing button bounced on this pin, the MAX232’s +/-10V rails would collapse to both be about +2V and neither of them would return to their proper values until I cycled power to the entire circuit. Not sure if this was typical behaviour of a MAX232 or not (I know I had the recommended capacitances on it) but I ditched it and replaced my receiving circuit with an NPN inverter with an inverted diode from base to collector (plus base resistor) in order to get a 0V-5V signal, which worked fine. What surprised me more is that simply sending a 0V-5V signal up the RTS and into the oscope worked absolutely fine. Reading the datasheet of the scope’s RS232 transceiver it had a 2V threshold, so I guess it’s more important to be within spec for your ICs than for the standard itself. A $8 USB logic analyser proved vital for debugging the thing, since I couldn’t use my oscilloscope while it was transmitting data.
Then to ditch the clumsy USB lead powering the thing I attempted to run a rectifier+filter cap+7805 off the unused CTS pin to power the arduino off that, but as it turns out my nano’s quiescent current was a few mA thanks to the power LED, and the 7805’s minimum current was also suboptimal, and the circuit lost power each time I tried to power RTS. So I ordered some better linregs and am attempting to get a stand-alone ATMega working for this, the caveat being they come with their fuses set to their internal RC clocks, not for the 16M crystal that I planned on slapping on there.
Quite the rabbit hole for a free oscilloscope. I look forward to what you do with this project as it might help my own escapades.
Even at this initial stage, I also already went down rabbit holes just to get this much working and there’s a whole big unknown territory coming up with what I’m trying to do next.
Based on GnuReligion’s comment that many RS 232 interfaces might except ground as the negative voltage, and if the positive voltage has to be at least 3 V, then 0 V to 5 V may very well work.
I have at least two working analog Tektronics scopes down in the dungeon and one of them has some sort of digital on screen Readout and storage I think, but I’m not sure what kind of ports and features are on the hardware interface. I might dig those out and check out if there’s anything to hack.
Hahah I remember waaay back when.. those mice with the rubber balls were so annoying, cause after a while, the rollers or sensors would get dust and crap on them.. messing it up and making it not work right.. so you had to take it apart and clean it out.. nice!!!!
My worst memory of those mice would be how the cover that holds the ball might get loose over time so I would epoxy it shut and then I can’t easily open it to clean the dust out.
I can remember the rs232 cable cook books, where there was about 100 or more ways of connecting the PC to the other device. before the PC our company computer used VT100 terminals. 100'sof rs232 cables running around the lab. the hard drive was about the size of a washing machine. RS232 is still a pain to code for, without handshaking it's a bugger to know when the data has stopped being sent.
Well now you are just encouraging me to go get out an old microcontroller and try doing some RS 232 code!
In my earliest jobs sometimes there were still 12 inch monochrome monitors in use as dumb terminals with keyboards hooked up somehow to a big tangled mess of technology. And I think I had more cases of RS 422/RS 485 long cables running all over the place. Connecting one machine to another or connecting serial debug interfaces to machines etc.
Lots of damaged and severed cables when rolling big machines around the lab for various testing purposes and in and out of hot and cold environment chamber rooms. Those were the days.
@@GadgetReboot we had a VAX -11 something, made up of a CPU, tape drive, hard disk with a top loading removable single disk. it had it's own air condition room the only one in the whole building. we would take it turns on a hot day to sit at the small desk in there. poor old thing got sold as scrap as it was replaced by a micro-Vax about the size of a PC. we had to keep it running because of the data stored had to be kept for 10 years or more. Happy days.
Awsome & inovative video, thanks for sharing.
Useful video 👍
How about the flip-side to this idea: sending power to the rs232 to recharge a 6v internal battery? How would I hook up the red and black wires coming from an adapter and into the rs232 pins? 7 of the pins are used on my device.
great com port rs232 video... i was looking to see if the old pc port was male or female? then found this vid and it is what i am working on, rs232 signal power, perfect. thanks a lot...:)
Just curious how much current each LED consumes in your schematic?
Unsure of where you are going with this, but if you want to go to Serial TTL without a MAX232, there are a number of conversion hacks. Some even steal negative voltage. It seems this is usually unnecessary. 0V is undefined, but most PCs will politely interpret it as low.
www.kswichit.com/ap275/ap275.htm
Yeah that sort of transistor interface with diode clamps to tame the levels is where I’m headed, although I still have some design requirements to figure out for example I may end up needing to not just level translate but multiplex two drivers down to one receiver, and I haven’t decided if it’s OK to OR them together sort of like open collector first come first serve with the intention that only one should ever be really using it at a time so it’s safe, or if I want to completely cut one off while the other is talking and therefore I need to do something more involved.
It’s good to know we might be able to get away with ground as a valid low-voltage. Given the technology in question, I probably don’t need it to be totally compliant, just something that works with the hardware I’ve got on hand so I will breadboard test.
Although if I end up making a PCB layout, it wouldn’t hurt to throw a few extra components on there just in case. And since I like to put jumpers all over the place maybe that could be one of the options, switching between a negative supply or just ground as the low-voltage.
@@GadgetReboot Not sure that I understand, unless you mean you mean to put TX+RX one one wire. Have seen that project online too, but never built it up. Will look out for your next video though ... am sure you are doing something cool!
A thing I have wanted to do is connect a 10-pin IDT fe to a mobo's UART header, and another cable to floppy power, merge the two behind a bay cover with a MAX232, then come outside the case with 5 wires to control my homemade Arduino-like boards. It would be 5 wires because DTR is needed to pull the reset low. But the truth is ... I like the buffer that a USB hub offers, insulating my PC from evil MCU experiments. The MAX232, at $0.15/ea, calls to me though: www.aliexpress.com/item/32944645394.html
@@GadgetReboot I have one of these gizmo's, but that would be cheating. Too expensive! www.aliexpress.com/item/32722395554.html
I should stock up on some Max232s at the rate I’m going lately. They might come in useful. I’m going to need to do some buffering and isolating of both power and data for the stuff I’m going to be doing so I think I have a big stack of optocouplers down in the dungeon that I have to locate and if I do have them, those are going to be put to use to give me peace of mind so I don’t blow up the rare motherboards.
With the multiplexing or channel switching, what I would be looking to do is have something like a modem transmitting to the computer as well as my own circuit transmitting to the same receive port on the computer, only one at a time but being able to electronically control which one is allowed to talk.