Keyboard Shortcuts
ctrl + shift + ? :
Show all keyboard shortcuts
ctrl + g :
Navigate to a group
ctrl + shift + f :
Find
ctrl + / :
Quick actions
esc to dismiss
Likes
Search
TDS544A with strange display
#photo-notice
|
Hello gentlemen,
My first post here, but I've been looking for information on this site for a long while. Some 8 years ago I bought A TDS544A with normal caps issue; that was solved (removed all old caps, cleaned and replaced caps on all boards, repairing some traces). In the end the scope started to work fairly normal but some exceptions apply here. Whenever there is a white patch on the right side of the screen (after pressing Vertical Menu, for example), there is a row of horizontal... lets call that bands, starting on the left of the screen, diminishing toward the right (see attached pictures). These are not of constant length (visible from pictures); wild guess is that at the end of white patch (each line) there is a signal that changes (maybe from 1 to 0, or opposite) that triggers something like a sample and hold which allows dots to be displayed for a while during the next line (only on blue). No white patch at the right, no bands of blue at the left. VGA output at the back of the scope shows normal (no such bands of blue). Using a 2235 I tried to see some signals, mostly at J5, J62 (and synchro signals at J18), but that scope is not good enough to see specific lines while synchronized on frame signal. I've seen some posts that tell me there are people around that know much more than necessary to give me some guidance where to look (and what to look for, in fact). It maybe related, or not: it seems I cannot convince the scope to go to Auto (in Acquisition) after selecting Single Shot; the only way (i may be wrong in that is the only way) to see it running again is by selecting Autoset. I would expect a lot of questions, don't keep anything back... TT P.S. Not sure why there is no menu at the top of the post, where you can click to add a picture, or format text.. I put a new post to tek500 and it was no issue to upload. What did I do wrong here? |
|
Replaying to myself
I made an album with pi9ctures, same name as the post. Still not clear why I was able to directly upload pictures on Tek500 post and not here (maybe there you cannot make a picture folder; didn't check that yet) Forgot to mention that those blue annoyances disappear when Clear Menu button is pressed (that state is presented in the last - or first - picture). Wave-forms are from FG502, different points of triangle generator schematic - if interested about another device with abnormal behavior... there it is TT |
|
Hi Jeff,
This is in my initial message (too well hidden somewhere in the middle): "VGA output at the back of the scope shows normal (no such bands of blue)." That output shows clean, which is somehow normal; information is taken from the same place for both VGA and CRT display, but have to be differently timed (different frame and line frequencies). What is wrong is on the path from the memory keeping the information and the CRT display (probably around display-daculator or before that). Thanks for taking the time to think about this TT |
|
Okay, sorry I didn't read carefully enough.
From the documents on TekWiki I'm betting that the color is produced by the "shutter" assembly in front of the CRT, and that the CRT is monochrome. That would suggest a failure of the shutter driver circuitry, though I am at a loss to explain how that results in the specific symptoms you are seeing. I don't actually know much of anything about this scope, only what I've read on TekWiki, and what I've heard about Tek's use of filters to produce color CRT displays. -- Jeff Dutky |
|
Hello Jeff,
You are correct that the CRT is monochrome and the LCD shutter at the front is enabling the color; however, there must be some display on CRT, for the time a specific color layer of the shutter is enabled, to show on the screen. In this case the whole display (monochrome CRT + shutter) is displaying correctly the CH4 blue trace (white on CRT and blue LCD layer ON), but the CRT, during the same period of time, is displaying something unintended, due to some defective (or un-powered) part. I'll try to better understand the schematic (as of now I have a general idea how this works, but I don't know or understand all the details) and, in the meantime, I'll try to see if there isn't something that I missed at the right time, like a corroded trace in some hidden place or, plainly, a defective part. Hopefully other people might have encountered something similar and might be available to provide some insight... Thank you TT |
|
Hey TT,
My first thought was something on the LCD driver board too. My TDS784D allows me to set different color schemes, and maybe even to pick the colors for different screen elements (I don't remember that I ever played with this). If your scope has the same, maybe you can narrow this down by playing with the colors, see whether and how different color and brightness settings affect this display defect? Siggi On Sun, Mar 7, 2021 at 10:28 AM durechenew via groups.io <durechenew= [email protected]> wrote: Hello Jeff, |
|
Looking at the photos closer, it appears the blue trace is also compromised
- there look to be some dropouts on the vertical edges here and there? Did you re-cap the CRT driver board when you fixed up the leaked SMD caps on the other boards? If you squint at this just right, it looks like maybe you have decaying oscillation on blue after the last line displayed white at the end. On Sun, Mar 7, 2021 at 11:50 AM Siggi via groups.io <siggi= [email protected]> wrote: Hey TT, |
|
Hi Siggi,
No, the blue trace is not compromised, but your observation is correct; the scope is set to display vectors, not dots, therefore, from time to time a vector might be missing because there was not enough time to capture the needed points of start and end (that's a guess) - remember that at slow scan speed the number of samples are fairly low and increase with scan speed. I did not recap the video stage; as I remember, there is no SMD cap on any of those board (only through-hole components). Idea of oscillation came to my mind too but I had to abandon it. The blue layer of shutter is ON for a certain amount of time while everything that needs to be blue is displayed by CRT. So those vertical stripes on blue must be on screen; no connection to shutter functionality, otherwise everything that is on screen (including the trace) would be chopped. Plus is doesn't explain why those stripes are waving - like a flag (each horizontal line in that area has those oscillations occurring for a longer or shorter period of time during the frame). And it's not oscillation because those would not be synchronized with vertical raster lines (if you take a closer look, the second vertical line is perfectly aligned with one of those vertical stripes and so is every second vertical line; that tells me that those blue lines are synchronized with a clock, not random frequency. My scope has also that possibility to check the colors, but I didn't discover any possibility to change anything (maybe different scopes with the same basic design have that feature enabled). TT |
|
On Sun, Mar 7, 2021 at 4:02 PM durechenew via groups.io <durechenew=
[email protected]> wrote: No, the blue trace is not compromised, but your observation is correct;I beg to differ. Try to switch that particular signal to one of the other channels, my $0.02 says the trace dropouts will follow the trace color. Idea of oscillation came to my mind too but I had to abandon it. The blueThe length of the blue stripes clearly follows the amount of white content on the previous line and/or the line with the stripe. And it's not oscillation because those would not be synchronized withGood point, I can't explain that. I don't think there should be anything on the CRT board that follows the dot clock... I also don't think there's anything on the CRT board that's per color, other than on the LCD driver board. I don't know exactly how the raster scan occurs on these screens, but I'd suspect the full raster is scanned for each color, changing the LCD shutter between a full frame scan. ... time passes ... It's rather interesting that there are 12 lines from the LCD driver board to the LCD shutter. It would be unnecessarily complex and awkward to have an LCD that matches the CRT's dot pitch - what the hell is going on there? IDK - what is LCS_CLK - it's brought out to the LCD shutter logic. My scope has also that possibility to check the colors, but I didn't discover any possibility to change anything (maybe different scopes withTo bring up the Color menu: 1. Press DISPLAY to show the Display menu. 2. Press Settings in the main menu until you select Color from the pop-up menu (see Figure 3-4) Good luck, Siggi |
|
Siggi wrote:
I think that the per-color part of this is probably explained by the order of the color shutter actuation, but I can't quite make it make sense either. If we simply had something that was being saturated from the all-wite sections (it's pretty clear that the all white sections at the end of the scan are causative) then we would expect ANY solid color block at the end of a scan to result in bleeed-over on the start of the next scan line. Why we don't see that is hard to explain. I like the idea of swapping control lines on the shutter to see if we can make the bleed-over change colors, that would likely narrow down the region in which the malfunction lives. -- Jeff Dutky |
|
Here we go; first, I believe that it was my poor English and I wasn't properly understood when I tried to explain what I see and I apologize for that.
Second, Siggi, you may be right about an issue with blue trace; I still hold my position, but you may wish to look at the picture 4019-1 (this time with the scope well wormed up). Some vectors show in red (don't know what that means), but we've seen that on green too - it's only on fast changing portion of signal (rise, fall time), never on triangle. Third, I wanted to take the shutter out of the picture: I made a short clip of the screen with stripes on blue, then... I opened the scope and disconnected the shutter from the shutter driver board. Made a second clip, essentially in black and white or, if you wish, what CRT is displaying. I believe this completely takes out the shutter as reason for discussion if it has a role in this or not. Also clarifies (I hope) what I meant when saying the stripes wave like a flag. As I cannot upload the clips here (not supported), the two files were uploaded to my Dropbox, link below: I verified the link before posting, works for me; it would be nice if you could confirm it works for you too. TT |
|
On Mon, Mar 8, 2021 at 3:58 PM durechenew via groups.io <durechenew=
[email protected]> wrote: Here we go; first, I believe that it was my poor English and I wasn'tThere's nothing wrong with your English that isn't also wrong with mine :). Second, Siggi, you may be right about an issue with blue trace; I stillInteresting. It would help to understand how the different colors refresh. My guess is that the entire shutter switches R/G/B on successive frame refreshes, but I don't understand why the complicated shutter circuitry is necessary for that. I noticed that there are 2 lines (LCS_CSEL0/1) that presumably select the color to display. Third, I wanted to take the shutter out of the picture: I made a short It sure does - there's clearly intensity modulation going on there, and not (just) shutter weirdness. I initially thought the intensity input to the CRT would be a binary value, but on reflection that doesn't make sense. Any color that's a combination of two or more prime colors would then experience double (or triple for white) the exposure to the prime colors, and would thus be brighter. It's also possible that the rendering is done in four sweeps, like e.g. RGBW. I looked back at the schematic, and the VIDEO_OUT_LCS signal comes out of the IOR pin of a TLC34075. This is an analog current-mode output, supposed to be terminated with 37.5Ohm - 75Ohm on either end. It's capable of at least 256 different intensity levels, plus sync. I'd assume the intensity modulation across the color fields would be implemented by color palette switching on VBLANK. For something silly to do, try and jiggle the contrast pot on the CRT board - see whether this is as simple as a marginal pot? The video amplifier uses several supplies including +5, +21, +75 and they're all generated on the CRT board. Checking out the power supplies on the CRT board would make sense to me. Otherwise I can't really see anything there but perhaps the TLC34075 or the data lines that feed it, that might cause this display weirdness... Also clarifies (I hope) what I meant when saying the stripes wave like a flag.It works fine for me. |
|
Today I'll re-read the manual, more attention to details
"My guess is that the entire shutter switches R/G/B on successive frame refreshes, but I don't understand why the complicated shutter circuitry is necessary for that. I noticed that there are 2 lines (LCS_CSEL0/1) that presumably select the color to display." While I don't have a proof yet, I agree with you about successive frames. But I believe control commands of the shutter are serialized (only one or two wires, out of which one is already available - frame blanking) and then decoded in the (complex) shutter control board. "Any color that's a combination of two or more prime colors would then experience double (or triple for white) the exposure to the prime colors, and would thus be brighter. It's also possible that the rendering is done in four sweeps, like e.g. RGBW." I think the complex colors would be diminished if seen through two or more layers of colors, more likely the same frame is successively exposed to different colors and the human eye makes the integration (I noticed that the working frequency of CRT is much higher than that for standard VGA, available at the back connector) And for white would make no sense to have all fundamental colors ON; you just make all layers of color transparent. I already tried Contrast pot a few weeks back, when I had all the video board out, same for Brightness and Screen, nothing wrong there. Idea of checking voltages on the video board is worth considering, but it's painful to do it (should have checked that when i had it open) I believe it's good, but without proof... Ones again, those stripes synchronized to a clock signal... tells me the issue is before or at TLC34075. |
|
On Tue, Mar 9, 2021 at 11:31 AM durechenew via groups.io <durechenew=
[email protected]> wrote: Idea of checking voltages on the video board is worth considering, butIf you have the board out again, maybe you can just replace the power supply caps preemptively - or you can tack new caps over the suspect ones. See this video for some concerns when running the CRT outside the scope case. |
|
I think that the way that the shutter works is to apply one of three filters (R/G/B) to the white CRT image. To get a white image you have three frames of the same illuminated image with all three filtered colors, and the human eye integrates the three colors together, just like a bicolor, red-green led that illuminates red with positive current flow, and green with negative current flow, appears yellow with fast alternating current.
This would suggest (but I am not able to make this explanation work now, in light of your further investigation) that something is saturating in response to the white region (which will be one for all three frames) at the end of the scan line, and it staying saturated for a portion of the next scan during the blue filter period (which I am assuming is the "first" shutter setting in each cycle: B/G/R or B/R/G). The colors produced by subtractive filters (where you would have two filters active at the same time) would be different than the colors you are seeing. They would be the colors you expect from kindergarten paint mixing (Red/Yellow/Blue -> R+Y = Orange, R+B = Purple, Y+B = Green, and R+Y+B = Brown/Black). The colors you are seeing on your display indicate an additive color scheme (also called RGB, what you are familiar with from color CRTs: Red/Green/Blue -> R+G = Yellow, R+B = Magenta, G+B = Cyan, R+G+B = White). I am at a complete loss to explain the symptom you are seeing by any obvious physical model. Even though the white region at the end of the scan line would be on for all three cycles, the electron beam would be off for most of each scan line, so it's hard to understand what would be getting saturated and holding that saturated state across the horizontal retrace to the next scan line. It's also hard to explain why this only happens when the white block is at the end of the scan: you don't seem to be seeing a bleed-over effect after a white block in the middle of the scan. The symptoms are so unusual and specific that they MUST be indicating a very specific part of the display circuitry, and you have clearly demonstrated, I think, that the malfunction is not in the filter system. Neither is the malfunction in the early portion of the display system, because we don't see the malfunction in the VGA output. That leaves the display driver for the internal CRT, but I don't understand that nearly well enough to make a specific diagnosis. -- Jeff Dutky |
|
"To get a white image you have three frames of the same illuminated image with all three filtered colors, and the human eye integrates the three colors together" in my answer to Siggi, I asked, why would you put three filters for white when you can leave all the color layers transparent...
"it's hard to understand what would be getting saturated and holding that saturated state across the horizontal retrace to the next scan line."; yeah, essentially that's my problem |
to navigate to use esc to dismiss