So, it seems it is now working as expected, although I'm not counting my chickens yet. In the SPI routine from the Teensy to the DAC, I changed the gain for the DAC from two to one, loaded the code, edited the code back to a gain of two in the DAC, loaded it again and voila'. Odd!
It's working as expected. Although I don't understand why! Il giorno domenica 23 giugno 2024 alle 02:46:09 UTC+1 gregebert ha scritto: > So another possibility is that one of the amplifiers is clipping, usually > because the input signals are trying to drive the output beyond the supply > rails. BTW, most op-amps are not rail-to-rail, but they can get pretty > close. It will show up quickly on a scope. > > On Saturday, June 22, 2024 at 3:35:15 PM UTC-7 Dekatron42 wrote: > >> So what happens if you put this Teensy in the other working board, does >> the error follow the Teensy? >> >> If so, have you tried to reprogram the Teensy or even erase it to set it >> to its original setup condition? >> >> /Martin >> >> On Saturday 22 June 2024 at 22:48:49 UTC+2 Max DN wrote: >> >>> @Martin, if I try to bypass the DAC using POTs for NEG5 and POS5 it all >>> seems fine (the dot goes across the whole X axis and Y axis). >>> @gregebert, no Z-axis. >>> >>> I am probing X and Y just out of the DAC and not connected to anything >>> else. I can see the image on the oscilloscope as a full round circle, but >>> as I try to center it to what would be the center of the CRT then thr >>> circle flatlines. >>> >>> It feels as though past a certain point on the vertical axis, the dots >>> are just collapsing into the same line and cannot go past (on one end of >>> the Y vertical axis, X axis is fine). >>> >>> Very odd as I have already used same code and same pcb for another scope >>> clock. After 10 hours on it today, I'm going to let go until I have any >>> other ideas. I have tried replacing the processor, the DAC (3 times), >>> checked all resistors and bypass caps and triple checked all soldering >>> points. Maybe it's time to bin this pcb, but I do like a challenge! >>> >>> >>> Il giorno sabato 22 giugno 2024 alle 16:38:19 UTC+1 gregebert ha scritto: >>> >>>> Is there a Z-axis ??? That would be my first suspect; if the trace is >>>> still visible, then the appropriate grid needs to be driven a bit more >>>> negative. >>>> >>>> If the CRT uses rapid movement for dark (technically dim) regions, and >>>> slower movement for bright regions, then my best guess is some kind of >>>> software tweak would be needed to get the beam to move faster, or off the >>>> phosphor-coated area (or both). >>>> >>>> On Saturday, June 22, 2024 at 3:49:04 AM UTC-7 Dekatron42 wrote: >>>> >>>>> Can you remove the processor and feed the amplifiers directly from a >>>>> potentiometer connected to NEG5 and POS5 and see how the dot moves on the >>>>> screen? You need one potentiometer for X & Y so that the dot will appear >>>>> on >>>>> screen. This way you will test out the amplifier chains without the DAC >>>>> involved. >>>>> >>>>> /Martin >>>>> >>>>> On Saturday 22 June 2024 at 12:10:54 UTC+2 Max DN wrote: >>>>> >>>>>> Martin, that's what I am thinking... however I don't think it's the >>>>>> amplifier chain as I get the same issue if I just display X and Y >>>>>> signals >>>>>> straight out of the DAC on the oscilloscope. That brings it down to a >>>>>> few >>>>>> components only: 2 potentiometers for centerting X and Y, their two >>>>>> bypass >>>>>> capacitors, the DAC and its 1 bypass capacitors, the +3.3V Linear >>>>>> regulator, the +5V supply to the DAC. I'm going to try and replace >>>>>> everything once more by one by one... Not the smartest way of doing it >>>>>> but >>>>>> let's see what it brings... >>>>>> >>>>>> Il giorno sabato 22 giugno 2024 alle 11:01:12 UTC+1 Dekatron42 ha >>>>>> scritto: >>>>>> >>>>>>> Could it be the that any of the power lines sags as the signal tries >>>>>>> to increase to the right (in the picture right where the line is) or >>>>>>> that >>>>>>> any of the resistors in the amplifier chain has an incorrect value? >>>>>>> >>>>>>> /Martin >>>>>>> >>>>>>> On Saturday 22 June 2024 at 11:32:21 UTC+2 Max DN wrote: >>>>>>> >>>>>>>> Tried to replace a few more parts, including potentiometers for >>>>>>>> Focus and Intensity, made no difference. So, back to basics: using an >>>>>>>> oscilloscope I see that the X and Y signals straight out of the DAC >>>>>>>> look >>>>>>>> exactly the same as in the picture I had attached. Maybe a faulty DAC? >>>>>>>> Tried to replace the DAC twice with new ones and made no difference. >>>>>>>> Looks >>>>>>>> like I'm hitting some sort of physical limits... More debugging to >>>>>>>> do... >>>>>>>> >>>>>>>> Il giorno venerdì 21 giugno 2024 alle 18:31:51 UTC+1 Max DN ha >>>>>>>> scritto: >>>>>>>> >>>>>>>>> Hello, >>>>>>>>> >>>>>>>>> I'm building a scope clock, same board as another one I built >>>>>>>>> before with no issues (after some help I got on this group, thank >>>>>>>>> you!) and >>>>>>>>> I'm getting a flat line on the display. >>>>>>>>> >>>>>>>>> I have tried replacing the CRT, the DAC, the Op Amps and the >>>>>>>>> MPSW42 deflection amplifiers but made no difference. CRT is 3RP1A. >>>>>>>>> Code and >>>>>>>>> circuitry have been fully tested on another working board. >>>>>>>>> >>>>>>>>> I do recall having a similar issue before with a DH3/91 CRT and I >>>>>>>>> seem to recall that I just needed to adjust the pot for Astigmatism >>>>>>>>> that >>>>>>>>> controls the voltage on A2. Schematics as in here: >>>>>>>>> http://www.cathodecorner.com/sctv/sctv-schem-2.png >>>>>>>>> >>>>>>>>> I'm a bit at a loss here, although I'm sure it's something simple >>>>>>>>> that I'm missing. >>>>>>>>> >>>>>>>>> Pic attached. >>>>>>>>> >>>>>>>>> Any suggestions welcome, >>>>>>>>> >>>>>>>>> Max >>>>>>>>> >>>>>>>> -- You received this message because you are subscribed to the Google Groups "neonixie-l" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To view this discussion on the web, visit https://groups.google.com/d/msgid/neonixie-l/7660e525-baef-41d4-8315-831565f5018cn%40googlegroups.com.
