I have some problems with my gigatron.

Information about getting support on Gigatron builds
denjhang
Posts: 23
Joined: 02 May 2021, 01:25

Re: I have some problems with my gigatron.

Post by denjhang »

I don't have an oscilloscope, I only have a multimeter. Can I find out the problem only by using a multimeter?
At present, I doubt whether it is a problem with my ROM programming.
I am willing to provide my address to receive the correct ROM.

In addition, I plan to buy an oscilloscope, please tell me how to use the oscilloscope to test gigatron. Compared to mailing ROM, perhaps using an oscilloscope will find the problem faster.
Last edited by denjhang on 23 May 2021, 09:25, edited 1 time in total.
bmwtcu
Posts: 54
Joined: 01 Nov 2018, 12:02

Re: I have some problems with my gigatron.

Post by bmwtcu »

Can you post a memory dump from your PROM programmer when you read a programmed part back into the device buffer? The goal being to confirm that you have the bytes programmed in the correct order and not swapped like I described earlier. Assuming you're using ROMv5a, you should be seeing the IR and D register values starting from line 498 onwards that corresponds to https://raw.githubusercontent.com/kervi ... ROMv5a.lst. The first instruction is all zeros and isn't very useful, but the 2nd instruction should have 0x18 going to IR and 0x80 going to D. The lower byte is the instruction and the higher byte is the data. I'm interpreting Marcel's comment in line 497 to mean he also used a voltmeter to check this. I assume he would have needed an Arduino or some other means to single step the clock feeding the PC.

If you have a way to erase and reprogram the 150ns EPROM (since your 45ns part is OTP), a quicker way would probably be to just bypass all that and reprogram it with the bytes swapped and try that again on your Gigatron.
rombyteorder.jpg
rombyteorder.jpg (125.2 KiB) Viewed 193 times
denjhang
Posts: 23
Joined: 02 May 2021, 01:25

Re: I have some problems with my gigatron.

Post by denjhang »

bmwtcu wrote: 23 May 2021, 07:58 Can you post a memory dump from your PROM programmer when you read a programmed part back into the device buffer? The goal being to confirm that you have the bytes programmed in the correct order and not swapped like I described earlier. Assuming you're using ROMv5a, you should be seeing the IR and D register values starting from line 498 onwards that corresponds to https://raw.githubusercontent.com/kervi ... ROMv5a.lst. The first instruction is all zeros and isn't very useful, but the 2nd instruction should have 0x18 going to IR and 0x80 going to D. The lower byte is the instruction and the higher byte is the data. I'm interpreting Marcel's comment in line 497 to mean he also used a voltmeter to check this. I assume he would have needed an Arduino or some other means to single step the clock feeding the PC.

If you have a way to erase and reprogram the 150ns EPROM (since your 45ns part is OTP), a quicker way would probably be to just bypass all that and reprogram it with the bytes swapped and try that again on your Gigatron.

rombyteorder.jpg
Do you mean to let me exchange the part of the red circle?
I plan to make an In Circuit EPROM emulator and debugger
(viewtopic.php?f=4&t=112) to replace ROM and perform debugging.
Attachments
rombyteorder.jpg
rombyteorder.jpg (125.87 KiB) Viewed 192 times
bmwtcu
Posts: 54
Joined: 01 Nov 2018, 12:02

Re: I have some problems with my gigatron.

Post by bmwtcu »

Read back one of your programmed PROM parts and confirm that the order of bytes is the same as the excerpt above. I was specifically talking about byte 0x2 and 0x3 above.
denjhang
Posts: 23
Joined: 02 May 2021, 01:25

Re: I have some problems with my gigatron.

Post by denjhang »

This is the data read from my MX27C1024DC-55.
Attachments
210524000959.png
210524000959.png (52.79 KiB) Viewed 173 times
bmwtcu
Posts: 54
Joined: 01 Nov 2018, 12:02

Re: I have some problems with my gigatron.

Post by bmwtcu »

OK, that rules the byte swap out. You've already ruled out bad/fake ICs, so unless there's some other BOM error, it's more likely a soldering issue at this point. If you have a scope, start from the clock signal quality, but otherwise even with a voltmeter, you should be able to work your way through verifying PC->EPROM->IR/D->CU functionality if you can single step instructions with your debugger.
Last edited by bmwtcu on 23 May 2021, 17:03, edited 1 time in total.
denjhang
Posts: 23
Joined: 02 May 2021, 01:25

Re: I have some problems with my gigatron.

Post by denjhang »

at67 wrote: 22 May 2021, 19:02 The second video without the crystal looks like the MCP100 is continually causing your Gigatron to reset 3 times a second in a loop, (entering a random Blinkenled state). I'm not sure why it would do that without a crystal, probing crucial signals with an Oscilloscope would probably find an explanation.

It's starting to look more and more like a ROM issue, as @bmwtcu hinted at; see this thread, (if you haven't already), for a comprehensive debugging session with Marcel that turned out to be a dud ROM: viewtopic.php?f=4&t=137

P.S. If you private message me your address I can send you a spare working ROM that I have, I have no idea how long you would take to receive it, as I am located in Australia.
After re-checking the solder joints and the ic socket, I found that U14 was not plugged in tightly. After dealing with this problem, I added some suspicious points by the way.
At present, all blinkenlights are all on without inserting the crystal, and the previous blinking situation will not appear.
But when the crystal oscillator is inserted, it still cannot work normally.
.............
Well, after I replaced a rom, the blinkenlights flickering phenomenon came back without inserting the crystal.
It seems that I need to use an oscilloscope to find the problem.
bmwtcu
Posts: 54
Joined: 01 Nov 2018, 12:02

Re: I have some problems with my gigatron.

Post by bmwtcu »

Did you already confirm that you're getting enough voltage at the MCP100 supervisor chip? Basically it will reset the PC if voltage droops under 4.5V. The quality of your USB cable makes a difference to how much voltage drop you're going to get at the end of the cable. It might also be a sign of a resistive short on your board.
denjhang
Posts: 23
Joined: 02 May 2021, 01:25

Re: I have some problems with my gigatron.

Post by denjhang »

After I changed a better power supply, the problem of not inserting the crystal and power ok flickering has been solved.
But when I inserted the crystal, blinkenlights still behaved the same as I described at the beginning (first light up the first one, and then quickly switch to full light), and the program cannot run normally, and there is no display or sound output.

At present, I have placed an order to buy an OSC482 oscilloscope, which I think is enough for Gigatron.
jbailey
Posts: 29
Joined: 20 Nov 2018, 08:54

Re: I have some problems with my gigatron.

Post by jbailey »

One other valuable troubleshooting solution is using the diode function on a multimeter, and connecting your positive lead to ground of the PCB and probing every single point on the board and noting the voltage drop at that point (without the board connected to power). The creators of the Gigatron created a Excel file which list the expected voltage drop at every point, you may have a slight verance from their documentation (I did) but it should be small and somewhat consistent.

On my third Gigatron a installed machined sockets, and it wouldn't start. I found that D7 on the ALU was shorted, no amount of looking at the board with magnification would allow me to find this issue as the short must have been under a socket. Once I identified that D7 was the issue, I pulled every "D7" leg up on any chip that was part of that "bus". Then using some test clips, I tied all of the D7's together outside of the sockets and every function of the Gigatron worked perfectly.

That confirmed I had a short. In my case I unsoldered each D7 connection and its adjacent two connections and more carefully resoldered them, which solved my issue.

Even if you have a single bad component this test may yield a useful hint.

The document can be find from the main Gigatron.IO web page under MANUALS AND REFERENCES --> TROUBLESHOOTING. about half way down the webpage you will find the link to the troubleshooting.xls document.
Post Reply