• Please review our updated Terms and Rules here

Weird characters in kaypro II boot

Ji, I've just burn a 28C16 but the kaypro only shows garbage in the screen. I've tried with a 81-159b, C and mon2k.bin created by durgadas but with the same result, garbage in the screen

Any idea?? It could be for the eprom type. I've erased the 2716 and now I can't program it..
 
So the debug ROM will leave garbage on the screen, as it does not initialize most of the hardware. You interact with the monitor using the serial port, typically connected to a PC (running a terminal program). I assume you meant you tried 81-149, since I've never heard of 81-159. With the monitor ROM, you should be seeing the first character of the screen toggle between 'A' and 'B'. I can't see any reason that a 28C16 would not work, I've been using a 28C64 to replace the 2764 in my Kaypro 2X. Looking at the pins, I see nothing incompatible with the Kaypro II and 28C16.
 
Hi folks, I've been testing several 28c16, with no success. Just garbage on the screen. Tried too the monitor rom, but no "A" or "B" in the first character at screen. I couldn't test the serial connection because I was thinking that the cable was a null modem one, but was a serial one, so I'm waiting for one.

I've tested the memory rom too, but no numbers from 1 to 9 nor E or R in the first character on the screen. There's no difference with no monitor eprom, eprom empty, eprom with 28c16 81-149c, eprom with monitor rom or with memory rom.

I've tested all the connections in the video section in the schematics pdf, and are Ok no cut leads.

Since I erased the MM2716Q-1 original EPROM never get the boot screen again. I'm just change the eprom and check continuity with the schema. I don't know why don't show the boot screen.

Do you have any idea??

Thank you!!
 
I can't think of any reason that the 28C16 would not work. But I'm assuming you have a first-generation mainboard. Maybe we need to see a photo of your mainboard to confirm what model it is.
 
Ouch... Never deprogram a working part if it's your only one... Well, I won't lecture on that anymore.
The difference between a 2716 and 2732 is A11. Which should be high for a 2716. Can you program a 2732 eprom? Just put the data into both halves. Also are you sure your binary file is correct?

A 28c16 is also just A11 different - ie, it's !WE on the 28c16.... So it should work. Check what pin 21 is doing on the main board - whether it's floating, tied to GND or VCC. If it's tied to GND, then take two sockets and connect all the pins except 21, which should be broken at the PCB and tied to Pin 24 (VCC). Then the 28c16 should work. Also, after you test it, stick it back in your burner and check that it still verifies to make sure you're not reprogramming it in-situ in the board.

You are going to need to check what PIn21 is doing - Though @durgadas311 has suggested that a photo may be enough for him to identify it. That's worth doing first.

David
 
That was my thought: Don't overwrite your original. But I too have not followed that advice occasionally. "I'll never need this again". DOH.
 
I know, I know... My fault, but I can't understand why the 28c16 doesn't work at all.
 

Attachments

  • 1700235450693632709335927210733.jpg
    1700235450693632709335927210733.jpg
    1.9 MB · Views: 8
  • 17002354871142370307791741284682.jpg
    17002354871142370307791741284682.jpg
    1.6 MB · Views: 7
The 21 pin is high, and the binaries, I've programmed it with the 149b,149c that I found on several sources and with a backup of the original (and now lost.. sigh) 2716
 
It does appear to be a first-gen mainboard, and while the photo of the back is difficult to read it does appear that pin 21 is connected to +5V as shown in the schematics. A couple ideas of what might have happened:

A) Something got damaged when removing the original ROM. Maybe carefully inspect the socket and surrounding pads/solder joints. If you have an oscilloscope, maybe probe key signals on the ROM chip (on the chip pin, not socket or PCB) to ensure all pins are getting proper connections. I've seen bent or broken "leaf" springs in a socket before, can be difficult to diagnose. Check for voltages, and if possible look for correct patterns on other pins.

B) The ROM programming is not doing what you expect. Maybe the ROMs are being programmed with garbage or nothing at all. Maybe perform a read to a file on your programmer, and post the file for independent verification.


Maybe there is someone on the same continent as you that can program a ROM and send it to you.
 
Hi, thanks durgadas, I'm afraid that the rom is well programmed I've checked it a lot of times, anyway I'll post a file to check. With respect of the socket I'll revise again, fingers crossed

Thank you so much
 
The mainboard looks clean and in good condition, and I don't see any obvious problem areas. I do notice that the CPU appears to have been replaced, at some point over the life of the system. One thing we used to encounter some times was that a pin would get bent over and make partial contact with the socket, only to lose contact after something as simple as flexing the PCB (like what happens when removing/inserting other parts). Maybe check the CPU (and other socketed components) to be sure that all pins are making full contact/insertion.

Back in the old days, I had access to a 40-channel logical analyzer that we could just clip on the CPU (or other key component) and see exactly what was happening. That's an invaluable tool, but even these days they can be expensive - especially for a full-40-channel one. At the very least, this may take dutifully probing with an oscilloscope and trying to piece together a picture of what is happening.
 
Hi durgadas, I've replaced the z80 a days ago. There were a lot of corrosion in the pins, so I changed. After that it works Ok, with the message bad, but working at all. I was testing all the socketed chips with the retro chip tester pro (a beautiful device by the way...) and all were fine.

The two PIO have a little corrosion too, but it seems that it wasn't the problem. I really think, as you say, that a bad contact elsewhere is doing the bad job.

Keep trying and thanks for your support
 
Hi, I'm completely stuck. Right now, each time I start up the computer, just fill the video area with garbage. I've checked the connections with the diagram and it seems all Ok. I think that it must be a chip failure. If I try Durgadas's monitor eprom there were no signs, even with just the memory eprom, in the first screen characters.

What do you think about the main chips responsible of the start up process?? Thank you for your help
 
The suggestions are correct. Right about now, you'd need enough knowledge to make a main board from scratch to troubleshoot the problem. The best bet is to either build some custom hardware or get a large logic analyzer to monitor what is happening.

You could do it with an 8 channel and build a composite picture of what is happening, but some serious hardware or skill is probably called for at this point. I am guessing you may be hoping for an easy fix, something like "It's probably a specific chip" but once you get this far into a problem and it gets worse while troubleshooting and then you can't get it back the way it was, you basically need to trace and confirm the operating of every single wire on the board as well as every single chip because something else has been damaged while troubleshooting. It's a fairly complex task only made easy by some pertinent decisions involved deep hardware knowledge.

There are some short cuts if you have the right equipment... If not, you're really down to checking every thing one pin at a time. It's slow and laborious. Fixing computers is challenging and time consuming.

Maybe write your own ROM and just get it to do something basic - eg, loop around in low memory around 0000... Then you can watch on an oscilloscope to see if it's executing just that. Once you confirm the ROM is working, next confirm that the entire ROM is working correctly... Then you start checking RAM... And you work your way out from there...

Last time I had a problem like this, I built a z80 with it's own display so it could show me what it was finding. Wiring up some dual-ported RAM that sits in the video memory space and something to send it to a screen would help you a lot.

The next step is not really determinable without some serious troubleshooting.

:( I really hope you can fix it.
 
jmellidg said:
Hi, I'm completely stuck. Right now, each time I start up the computer, just fill the video area with garbage. I've checked the connections with the diagram and it seems all Ok. I think that it must be a chip failure. If I try Durgadas's monitor eprom there were no signs, even with just the memory eprom, in the first screen characters.

What do you think about the main chips responsible of the start up process?? Thank you for your help
Click to expand...
That's starting to sound like some fundamental breakage, like the CPU is dead. If the memtest.bin ROM image does not do anything, that tends to point to something other than DRAM. Could be something like bus (address, data, control) driver chips, too, such that the CPU is unable to communication with the rest of the hardware. The garbage on the screen simply means that the CPU could not run the ROM code in order to clear the screen - the garbage is just the random power-on pattern for the video RAM chips. Note that the memtest and monitor ROM images do NOT clear the screen automatically, so you will still see garbage (until you clear it using monitor commands).
 
jmellidg said:
Hi, I'm completely stuck. Right now, each time I start up the computer, just fill the video area with garbage. I've checked the connections with the diagram and it seems all Ok. I think that it must be a chip failure. If I try Durgadas's monitor eprom there were no signs, even with just the memory eprom, in the first screen characters.

What do you think about the main chips responsible of the start up process?? Thank you for your help
Click to expand...
That's starting to sound like some fundamental breakage, like the CPU is dead. If the memtest.bin ROM image does not do anything, that tends to point to something other than DRAM. Could be something like bus (address, data, control) driver chips, too, such that the CPU is unable to communication with the rest of the hardware. The garbage on the screen simply means that the CPU could not run the ROM code in order to clear the screen - the garbage is just the random power-on pattern for the video RAM chips. Note that the memtest and monitor ROM images do NOT clear the screen automatically, so you will still see garbage (until you clear it using monitor commands).
 
durgadas said:
That's starting to sound like some fundamental breakage, like the CPU is dead. If the memtest.bin ROM image does not do anything, that tends to point to something other than DRAM.
Click to expand...

durgadas311 said:
That's starting to sound like some fundamental breakage, like the CPU is dead. If the memtest.bin ROM image does not do anything, that tends to point to something other than DRAM.
Click to expand...

I read an echo. Are you two twins? ;)
 
You must log in or register to reply here.
Back
Top