Hi all,
I've just picked up a dragon sold as having no screen output. Power rails are all fine, and it always clicks the motor relay on power up. Interestingly I've managed to get occasional screen output from it, but when I do it'll run in normal 'green' text mode for a short period before spontaneously jumping to the alternate palette 'buff' mode (like SCREEN 0,1). Once this has happened Power cycling will then usually keep 'buff' mode but sometimes switch back to green.
I'm currently wondering if both the lack of display most of the time and this odd issue when the display does work are symptoms of a faulty VDG?
Stu
Dragon 32 jumping to buff mode spontaneously?
Re: Dragon 32 jumping to buff mode spontaneously?
Got same problem from both the RF and Monitor output?
Re: Dragon 32 jumping to buff mode spontaneously?
I've only tested composite monitor so far as I don't have a TV capable of tuning the RF to hand. I assumed the signal paths were basically the same apart from the modulator?
Re: Dragon 32 jumping to buff mode spontaneously?
Yes, it's the same signal from the mobo. If the case was that you only tested the RF.
Re: Dragon 32 jumping to buff mode spontaneously?
Thanks for the reply. I'd somewhat assumed this was not a problem at the analogue output end of things because the 'buff' palette matched buff mode so closely. I guess my main question is if it's worth getting hold of a new VDG and/or SAM and trying them? The SAM's socketted so is an easy swap, but the VDG's soldered in and has a patched in resistor soldered to pin 10 (IIRC) of the chip.
Stu
Stu
Re: Dragon 32 jumping to buff mode spontaneously?
Could you post an image to show how the screen looks like when you turn it on. As mentioned, if the colors are off it might be the chip if you got garbage on the screen then it most likely is the RAM issue. Or something else but an image would help out.
Re: Dragon 32 jumping to buff mode spontaneously?
Predictably thus far today it refuses to display anything, though I hear the tape relay clicking. MOTORON/MOTOROFF doesn't seem to be doing anything while it's not displaying.
I haven't seen random characters on the display that machines with bad RAM manifest, though the first time I got a picture out of the machine I did get a full page of random characters that cleared when I hit reset.
My troubleshooting is a little hampered by only having a multimeter, so I can't check for clocks and the like, I'm now resorting to doing DC resistance, voltage checks around the display output side, but I'm not really sure if this is useful in any way.
I haven't seen random characters on the display that machines with bad RAM manifest, though the first time I got a picture out of the machine I did get a full page of random characters that cleared when I hit reset.
My troubleshooting is a little hampered by only having a multimeter, so I can't check for clocks and the like, I'm now resorting to doing DC resistance, voltage checks around the display output side, but I'm not really sure if this is useful in any way.
Re: Dragon 32 jumping to buff mode spontaneously?
Leaving it sat, powered off on the bench for a bit has got it to life again! Pictures attached. The fuzziness is the composite to HDMI adapter I'm using I think, though I'm getting occasional flickers I don't get with other input devices. The attached images show the screen on power up and a quick bit of code I hacked out to get the colour palette on screen. This definitely looks off and I wonder if this issue is downstream of the VDG.
- Attachments
-
- IMG_20200221_142436.jpg (1.55 MiB) Viewed 7991 times
-
- IMG_20200221_140151.jpg (1.06 MiB) Viewed 7991 times
Re: Dragon 32 jumping to buff mode spontaneously?
One further development, holding reset for a few seconds then releasing has got it to show as normal for at least a little while, before flockering and jumping back to 'orange'.
Re: Dragon 32 jumping to buff mode spontaneously?
Having experimented a bit, I can't replicate by tying øA, øB or CHB to various values. While it looks like it's in the other colour set (CSS=1), if it is, that's not the whole story: even in that set, the block by 0 should remain green.
Maybe check around the LM1889? (Possibly IC9, can't really read the schematic).
Maybe check around the LM1889? (Possibly IC9, can't really read the schematic).