Hugo Holden
Veteran Member
@nullvalue......What sort of VDU are you using and have you tried any others ?
- VCF Southeast - July 19 - 21, Atlanta, GA
- VCF West - Aug 2 - 3, Computer History Museum, Mountain View, CA
- VCF Midwest - Sept 7 - 8 2024, Schaumburg, IL
- VCF SoCal - Mid February 2025, Location TBD, Southern CA
- VCF East - April 2025, Infoage Museum, Wall NJ
-
Please review our updated Terms and Rules here
Cromemco dazzler replica project
- Thread starter Hugo Holden
- Start date
daver2
10k Member
Ok,
Colour mode - where have you gone to?
The interesting thing is that most of the documentation I have seen on the Dazzler ports (in particular port 0Fh) state that D4 is only used when in resolution x4 mode (D6=1). This is NOT true. It is used in ALL modes to differentiate between colour mode and grey scale mode. The KSCOPE program sets the 0Fh register to a value of 30h (0011 0000) which is: D7=0 (unused), D6=0 (normal resolution), D5=1 (picture in 2K RAM) and D4=1 (colour mode). D3 through D0 are not used and are set to 0000.
First of all run my little test program and set D5=1 and D4=1 with all the others =0. This is the same mode as KSCOPE uses.
; Static level checks - Monochrome DAC disabled and colour DAC enabled.
Check IC7/1 for HIGH (D4) and IC7/2 for LOW (/D4).
Check IC15 pins 13, 1, 4 and 10 for LOW (Monochrome DAC disabled).
Check IC17 pins 11, 3 and 2 for Pulsing HIGH. (Colour DAC enable).
Check IC25 pin 10 for pulsing HIGH.
Check IC25 pin 9 for either HIGH or pulsing HIGH (enable high brightness colour DAC).
Check IC16 pins 11, 3 and 2 for pulsing HIGH (enable high brightness colour DAC).
; Check for 3.58 MHz.
Check the following IC pins for a 3.58 MHz signal. We also need to make sure this signal is a TTL signal - so that the logic '0' is < 0.8 Volts and the logic '1' is > 2.0 Volts (preferably higher).
IC14 pin 10.
IC17 pins 1, 4 and 10.
IC16 pins 1, 5 and 9.
Check for pulses on the following pins. A logic HIGH should enable the colour gate, a logic LOW should disable the colour gate.
IC16 pins 13, 4 and 10.
IC17 pins 3, 5 and 9.
Check for pulses on IC14 pin 9.
Assuming the above is good, lets check the output of the gates. These should be bursts of 3.58 MHz when all of the inputs to the associated NAND gates are at a logic '1'.
IC14 pin 8.
IC17 pins 8, 6 and 12.
IC16 pins 8, 6 and 12.
Check the following gates for simple pulses.
Check IC14 pin 6.
Check IC25 pin 3.
Anything that looks strange, report back.
That should keep you busy and out of trouble...
Dave
Colour mode - where have you gone to?
The interesting thing is that most of the documentation I have seen on the Dazzler ports (in particular port 0Fh) state that D4 is only used when in resolution x4 mode (D6=1). This is NOT true. It is used in ALL modes to differentiate between colour mode and grey scale mode. The KSCOPE program sets the 0Fh register to a value of 30h (0011 0000) which is: D7=0 (unused), D6=0 (normal resolution), D5=1 (picture in 2K RAM) and D4=1 (colour mode). D3 through D0 are not used and are set to 0000.
First of all run my little test program and set D5=1 and D4=1 with all the others =0. This is the same mode as KSCOPE uses.
; Static level checks - Monochrome DAC disabled and colour DAC enabled.
Check IC7/1 for HIGH (D4) and IC7/2 for LOW (/D4).
Check IC15 pins 13, 1, 4 and 10 for LOW (Monochrome DAC disabled).
Check IC17 pins 11, 3 and 2 for Pulsing HIGH. (Colour DAC enable).
Check IC25 pin 10 for pulsing HIGH.
Check IC25 pin 9 for either HIGH or pulsing HIGH (enable high brightness colour DAC).
Check IC16 pins 11, 3 and 2 for pulsing HIGH (enable high brightness colour DAC).
; Check for 3.58 MHz.
Check the following IC pins for a 3.58 MHz signal. We also need to make sure this signal is a TTL signal - so that the logic '0' is < 0.8 Volts and the logic '1' is > 2.0 Volts (preferably higher).
IC14 pin 10.
IC17 pins 1, 4 and 10.
IC16 pins 1, 5 and 9.
Check for pulses on the following pins. A logic HIGH should enable the colour gate, a logic LOW should disable the colour gate.
IC16 pins 13, 4 and 10.
IC17 pins 3, 5 and 9.
Check for pulses on IC14 pin 9.
Assuming the above is good, lets check the output of the gates. These should be bursts of 3.58 MHz when all of the inputs to the associated NAND gates are at a logic '1'.
IC14 pin 8.
IC17 pins 8, 6 and 12.
IC16 pins 8, 6 and 12.
Check the following gates for simple pulses.
Check IC14 pin 6.
Check IC25 pin 3.
Anything that looks strange, report back.
That should keep you busy and out of trouble...
Dave
Last edited:
Gary C
Veteran Member
Yes, that's what I need.
I am currently using a Commodore 1701, and I was also using an AppleColor Composite monitor.. both are known/good color monitors. I may have some others around here I could try...
Correct
I don't have any modifications on this IC? should I?
Correct
I think that's all
Thanks haha! should have time this afternoon.
Gary C
Veteran Member
From post #255
Pin 10 of IC 23 should be connected to the adjacent rail
Pin 10 of IC 23 should be connected to the adjacent rail
daver2
10k Member
It wouldn't...
Don't forget, a not connected pin on a TTL chip generally floats high. It is just not good practice to leave any input floating. They should always be pulled high (preferably by pull-up resistors) to avoid electromagnetic interference (EMI) from causing a malfunction.
Dave
Don't forget, a not connected pin on a TTL chip generally floats high. It is just not good practice to leave any input floating. They should always be pulled high (preferably by pull-up resistors) to avoid electromagnetic interference (EMI) from causing a malfunction.
Dave
Ok FYI here's what my screen looks like in that mode:
Correct
Correct
Pulsing high, correct
Pulsing high, correct
Yep, pulsing high
Pulsing high, correct
I am measuring the right frequency, but the waveforms look different for these so I'm going to post pictures. High/Low voltages is measured in the scope pictures.
IC14/10:
IC17/1 (voltage a little lower here but still in spec)
IC17/4:
IC17/10:
IC16/1: same as 17/1
IC16/5: same as 17/4
IC16/9: same as 17/10
Will get to the rest shortly..
daver2
10k Member
They are OK.
There is a reason for the look of some of the 3.58 MHz signals being as they are. I will explain why later.
Dave
There is a reason for the look of some of the 3.58 MHz signals being as they are. I will explain why later.
Dave
These are all pulsing. They all look generally the same, except for IC17/3 which had a slower pulse. Let me know if you want any pictures.
stable ~16khz clock here
Yes, I am seeing intermittent pulses @ 3.58MHz
yep ~16khz with a narrow +width
same, ~16khz but with a wider +width
daver2
10k Member
I will take a more thorough look tomorrow.
However, I am not seeing anything jumping out at me.
Next is to fill the screen with a single primary colour and check the output from the DAC. I will post how to do this tomorrow.
The 'strange' 3.58 MHz clock is a result of the phase shift introduced by the inductor, capacitor and variable resistor (LCR) circuit to produce the red and green chroma signals (from the blue). They can be varied by adjusting the two potentiometers.
We are then into the analogue part of the mixer/video amplifier and potentiometer adjustments.
There perhaps is nothing wrong at all now, it just requires adjustment in a controlled manner.
Dave
However, I am not seeing anything jumping out at me.
Next is to fill the screen with a single primary colour and check the output from the DAC. I will post how to do this tomorrow.
The 'strange' 3.58 MHz clock is a result of the phase shift introduced by the inductor, capacitor and variable resistor (LCR) circuit to produce the red and green chroma signals (from the blue). They can be varied by adjusting the two potentiometers.
We are then into the analogue part of the mixer/video amplifier and potentiometer adjustments.
There perhaps is nothing wrong at all now, it just requires adjustment in a controlled manner.
Dave
daver2
10k Member
Yep, there is a procedure for doing that!
I am on my phone, so it is not easy to do a copy and paste on a section of the Dazzler manual describing the procedure.
I have found one Dazzler manual to be better than another for this...
Dave
I am on my phone, so it is not easy to do a copy and paste on a section of the Dazzler manual describing the procedure.
I have found one Dazzler manual to be better than another for this...
Dave
daver2
10k Member
I think we will be done by then!
Dave
Dave
Hugo Holden
Veteran Member
Of that list, I am responsible for the missing link to IC23 pin 10 and IC33 pin 10. At least though the cards still work without them, but better for noise immunity to connect them. I will on my boards.
Gary C
Veteran Member
IC 43 pins 10,13 & 14 - connect to adjacent rail - fixed
IC 39 pin 14 - connect to adjacent rail - fixed
IC 23 pin 10 - connect to adjacent rail - fixed
IC 33 pin 10 - connect to adjacent rail - fixed
Cosmetics
R11 should read 30K - fixed
Two R23's ! - fixed
IC 44 should be a 7474 - fixed
IC52 should be a 7403 - fixed
PCB's uploaded to github
IC 39 pin 14 - connect to adjacent rail - fixed
IC 23 pin 10 - connect to adjacent rail - fixed
IC 33 pin 10 - connect to adjacent rail - fixed
Cosmetics
R11 should read 30K - fixed
Two R23's ! - fixed
IC 44 should be a 7474 - fixed
IC52 should be a 7403 - fixed
PCB's uploaded to github
Hugo Holden
Veteran Member
Possibly so.
The only defect that was measured before, what that there appeared to be no color carrier signal on the active video, aside from the burst, on the output video signal, even though they were there on the outputs of the three gates of IC17 & 16.
But there is no issue with the other gate that supplies just the burst, mixing that signal into the virtual earth point, so it makes the idea of a defect in the mixing circuit and output amplifier improbable, or there would be no burst.
It is as though the connections from those color gate outputs via the resistors are not connected to the virtual earth mixing point, because if the burst is and it makes to the video output, so should the color signals. As noted it is not the usual suppressed carrier system of NTSC, and on a white signal there will (should) be R,G & B color carrier signals easy to see on the output video signal.
We need another recording of the output video signal. If the color carriers are there now, then the most likely thing would be the 3.58 MHz signal is a little too far off for the set to lock to and the color killer is on. But I think, probably, if the previous recording was correct , there are no color signals on the video.