• Please review our updated Terms and Rules here

Commodore 3016 scrambled screen

Hello All,

I am back again with new 3016 project ;)
CPU is OK. (Did NOP test)
replaced 2114 ic's <-- one of them was defect.
I have checked all the 4116 ic's and all seems ok. Tested with RetroChipTester.
Where should i look first now?
Pettester now give this output:
 

Attachments

  • IMG_7232.jpg
    IMG_7232.jpg
    3 MB · Views: 11
Last edited:
The 4116 DRAM in the lower bank is returning a hexadecimal value of FE. This is wrong.

When you performed the NOP test, did you look at the activity on the DRAM control signals and address decoding?

Dave
 
Sorry, travelling today...

There is no point in retesting the DRAMs if you have already tested them.

I assume (now) they are in sockets?

If it is simple to retest them, who knows.

Check the /RAS and /CAS pins of each DRAM.

You should have /RAS and /CAS action on all DRAM devices.

The R/W pin should always be HIGH (reading).

You should have a 'block' of /RAS and /CAS action on the lower 16K followed by a block of /RAS and /CAS action on the upper block of 16K.

However you should always see action on the /RAS line as the refresh operates.

When the DRAM is being accessed, you should observe one memory access followed by one refresh cycle. When the DRAM is not being accessed, you should only observe a refresh cycle (/RAS only).

Dave
 
Last edited:
Hello,

I have new project. A CBM4016
I think with 2 problems.
1. There is no screen on the cbm monitor.
2. I have problem with mainboard.
I now have connected the cbm with a RGB2HDMI converter.
First i will focus on the mainboard.
I now have this output (and after a while like second screendump)

What should i check first? (Cpu is already checked and Ok)
 

Attachments

  • IMG_7682.jpg
    IMG_7682.jpg
    854.5 KB · Views: 7
  • IMG_7684.jpg
    IMG_7684.jpg
    1.5 MB · Views: 7
First off, check the power supply rails - always.

for the 8032080 assembly, the technical data is here: https://www.zimmers.net/anonftp/pub/cbm/schematics/computers/pet/univ/index.html

I am expecting (from the display) something wrong with the video RAM in UC4 and/or UC5.

Have you installed my PETTESTER yet?

Until we start testing it with known data patterns, we won't know for sure which one(s)...

EDIT: Based on the various characters I am seeing in the first image, I am going to guess UC5 is the faulty video RAM. However, it could also be a faulty data buffer (UB5) or the latch UB3. The more likely device would be UC5 though...

Dave
 
Last edited:
Indeed UC5 is defect(replaced). I already figured that out ;) There is always at least on faulty ram in a Pet.
I installed the pettester eprom in UD7 (The only one with a socket)
And i see this:
 

Attachments

  • IMG_7685.jpg
    IMG_7685.jpg
    1.8 MB · Views: 6
Page 0 and page 1 RAM is dead then (or the associated support circuitry).

The bulk of the RAM reads seem to be returning a data value of $02.

Can you double-check the voltage supplies on the pins of the DRAM please. They are 4116 devices, so they should have +12V, +5V and -5V. Any voltage rail not present will cause the RAM to not work...

Dave
 
So, to the DRAM address multiplexers (UE8, UE9 and UE10)...

Check is the RAn signals connected to the pins of UE8 (74LS244) for RA1 = 500 kHz, RA2 = 250 kHz, etc. up to (and including) RA7. Each being half the frequency of the previous RAn signal.

Then check CLK 1B on UE8 pins 1 and 19 for a 1 MHz clock. All using the oscilloscope.

Just thinking - @Hugo Holden has already documented all of this... It is for a 4008 machine, so slightly different in terms of the IC references, but the signals should be very similar...

Dave
 
Check the BAn signals on the next two ICs (UE9 and UE10). Also check for activity on pins 1 and 19 of each IC for activity.

Dave
 
UE9 is getting hot.
1Mhz signal on 1 and 19 but no correct signals on BAn signals (jumps from khz to Mhz)
UE10 has 500Khz signal on 1 and 19 and BAn signals also jumps from khz to mhz.
 
If the BAn signals are jumping all over the place, pop the NOP generator in. They should be nice and stable then!

Dave
 
You must log in or register to reply here.
Back
Top