PDA

View Full Version : OT: Anyone know anything about crt display repair?



macona
08-03-2011, 02:41 PM
I have a little 4.5" monochrome CRT Display in a deposition controller I have. It runs on 12v and uses NTSC. Pretty generic.

Its Dead.

I have high voltage, getting about 260v to the screen grid, filament is glowing. Scope shows activity on horizontal and vertical deflection. I can see the amplified video signal at the cathode on the tube.

What am I missing? Not enough HV? Poisoned cathode?

Checked the video output with another display and the video is good. If all else fails I will just run an external display.

-Jerry

Evan
08-03-2011, 04:03 PM
Check for failed electrolytic capacitors in the power supply. That can cause insufficient voltage to the flyback oscillator and reduce the HV enough that no beam current flows. The horizontal sweep osc usually supplies the drive to the flyback transformer. Other possibilities are bad flyback xformer or shorted high tension lead. Clean all and apply insulating laquer to the xformer and HT lead.

Try to avoid getting zapped. It's decidedly unpleasant.

arcs_n_sparks
08-03-2011, 04:08 PM
If you have an appropriate probe, check for high voltage on the CRT (sometimes you can sense it moving the back of your hand towards the CRT face and feel the electrostatic forces on your hairs). No high voltage is usually due to a failed switcher that drives the flyback.

macona
08-03-2011, 04:40 PM
There is high voltage. I get a spark when I discharge the tube.

I do have a HV probe, I just need to hook it up.

This is the board. Really compact so it will be hard to replace with something else.

http://farm7.static.flickr.com/6005/6005870423_fbe368b99f_b.jpg (http://www.flickr.com/photos/67292116@N00/6005870423/)
CRT board (http://www.flickr.com/photos/67292116@N00/6005870423/) by macona (http://www.flickr.com/people/67292116@N00/), on Flickr

MaxHeadRoom
08-03-2011, 05:18 PM
Very often with those small CRT's the filamament is fed from the H.O.T., if so and you have filament voltage it usually means the flyback is OK.
Maybe just very low emission, did it fail suddenly?
You could try placing a load resistor from cathode to grid to see if any kind of emission exists, this has the effect of reducing the grid bias voltage.
Max.

macona
08-03-2011, 06:43 PM
It was dead when I got it.

The filament is ran directly from the 12v power input. This circuit has the video signal to the cathode with the G1 at ground. I did try a resistor and nothing.

Found a datasheet on the controller, NEC uPC1379C. Gives me a bit of the circuit as they seem to have followed the app circuit pretty closely.

kc5ezc
08-03-2011, 09:53 PM
Sounds like no HV if I have understood your info.
Measure and then you will know. Or use back of arm (hairs) to see if you get a 'rise'.

macona
08-04-2011, 02:32 AM
Someone gave me an old Heathkit flyback tester a while back when I was working on the DRO with the bad display missing a HV probe for it. I found my HV probe and it looks like the one that is supposed to be for it. I hooked it to my 0-50KV adjustable power supply and the reading was about where it should be.

So measuring the HV I get about 5.5kv. Seems a little low to me but I can't find specs on this tube, and it is pretty small. Since I have that adjustable power supply I set it to positive out and hooked that up to the CRT and slowly raised it. Nothing. Until I got to about 15kv when it looks like it arced in the tube. When that happened I got a little bit of something.

Thinking the CRT is bad. Guess it wouldn't surprise me too much, it is 15 years old and heavily burned in. Might have been on that entire time depending on where it was used. When I had it hooked to an external monitor it showed the last time a coating run was ran was last october. The internal clock was only off by 30 minutes.

darryl
08-04-2011, 02:48 AM
Know anybody with a picture tube rejuvenator? We used to have one- hook up, wait for filament to heat, push button. Sparks fly internally, tube works again. Sometimes pretty good, often better but not really good, sometimes no change.

It could be gassy, in which case it's game over. Mind you- some of us have vacuum pumps and glass blowing equipment :)- you could try to re-evacuate the tube.

macona
08-04-2011, 02:59 AM
Hadn't thought about that. But then I would need to rig up something to refire the getter internally, either that or break the seal in an argon enclosure.

macona
08-19-2011, 12:21 PM
I picked up another bad unit from the same seller. This one would not start up but it turned out to have the cpu card come loose. Putting it back in it came up but all the video was compressed down to about 1/2" vertically. I tried the monitor board from the first one and it did the same thing, so the first CRT was bad after all.

Replaced a couple old electrolytics and the vertical deflection transistor and now I have a good image.

If anyone is wondering what this thing is, its a Film Thickness Controller for a vacuum system. This works with deposition systems like sputtering and evaporation to measure and control the rate of deposition. The controller connects to an oscillator that has a remote quartz crystal that is in the vacuum chamber that normally oscillates at 6MHz. As material builds up the frequency of the crystal changes and based on the density of the material. You can measure the film thickness down to the angstrom. These are also know as Quartz Crystal Microbalances. I can breathe on the sensor and it can measure the thickness of the water vapor film and watch it evaporate on the display!

http://farm7.static.flickr.com/6194/6058850661_60a7e596e2_b.jpg (http://www.flickr.com/photos/67292116@N00/6058850661/)
Sycon STC-201 (http://www.flickr.com/photos/67292116@N00/6058850661/) by macona (http://www.flickr.com/people/67292116@N00/), on Flickr

Evan
08-19-2011, 02:09 PM
BTW, DX has some nice colour LCD displays that would fit in the hole for as low as $35. Standard NTSC composite video input and excellent resolution.

macona
08-19-2011, 02:46 PM
BTW, DX has some nice colour LCD displays that would fit in the hole for as low as $35. Standard NTSC composite video input and excellent resolution.

That was the first thing I tried. I also tried that with the DRO that I had that had a bad CRT. Two problems with them, first the resolution is pretty low on the standard NTSC input displays. Text is rather hard to read as the dot pattern is different from regular LCD panels. NTSC is 480 lines vertical and most of the NTSC LCDs are only about 240 pixels high.

Second problem is they overscan and most equipment that uses a crt for data underscans so there is clipping at the edges.

I did try one little data input LCD we have here at work and it did work fine, its just the little display was near $300.

Evan
08-19-2011, 03:18 PM
Yeah, that is true of most of the small LCDs, but not all. They have one that I bought that is very high res and cost $35.

This is from an NTSC camera in my video microscope.

http://ixian.ca/pics9/lcd1.jpg

http://ixian.ca/pics7/vmscope1.jpg

macona
08-19-2011, 03:51 PM
Do you have the model of that?

Evan
08-19-2011, 04:04 PM
http://www.dealextreme.com/p/et-350-3-5-tft-lcd-digital-monitor-for-vehicles-reverse-camera-ntsc-pal-12v-dc-39044

It says 960 by 468 res but I think that is incorrect. It's closer to 720 x 512 or so. Regardless, the resolution is excellent. Also, hidden inside the right bezel are 4 push buttons that they didn't bother to bring out side. I easily added some external buttons to actuate them. That gives complete access to a wide range of the usual functions including brightness, contrast, Video 1 and 2 priority (it has two inputs) sharpness etc.

macona
08-19-2011, 04:08 PM
Probably still overscans but at that price its worth picking one up for other projects. Thanks.

Paul Alciatore
08-19-2011, 04:09 PM
OK, fillament glows so one common problem is eliminated.

I assume from your comments that you do not see ANY illumination on the CRT's face. If there is even a single spot then HV is probably OK, but you seem to have none so HV is the next suspect. Often only a 10 or 15% decrease in HV can cause a CRT to go dark so the fact that you get a discharge is of no significance. You could look up that tube (internet) and see what the HV should be. You have a probe so use it and check the actual HV value with the tube hooked up.

Electrolytic capacitors are a frequent problem. I have a TV in the next room that failed. I replaced all the electrolytics and instant new life. Been running four or five years now.

You can look for fried components. Replace any that are discolored, providing more air circulation if possible (leave the leads longer to raise them above the circuit board).

Beyond that it is probably not economically repairable. If you absolutely must fix this very one, then you will probably need the service literature - schematics, parts list, layout drawings, Voltage readings, waveforms, etc. If your time is worth anything, your $300 replacement quickly looks more reasonable.

Paul Alciatore
08-19-2011, 04:12 PM
Probably still overscans but at that price its worth picking one up for other projects. Thanks.

It sounds strange that a digital display would overscan. There is no reason for it.

Are you sure that the actual display does the overscanning or perhaps only the bezel of the mount is blocking some of the pixels. If that is the case, you could file away some of it to show the whole display.

Evan
08-19-2011, 04:25 PM
I don't think it overscans much since it is also compatible with PAL. Paul, it's using an analog signal as input so it doesn't have precise pixel timing that it can use. It's pretty standard for an LCD display intended for analog composite input to implement overscan. The NTSC television standard allows for that so the consumer doesn't think that they are losing valuable screen space to black edges. The latest generation of chips can compute the timing from the video signal and get very close to zero overscan but they still overscan a bit to avoid ragged edges.

macona
08-19-2011, 04:27 PM
Even with generic NTSC there is stuff outside of the normal visible field like the encoding for subtitles. You dont want to see that on a LCD either so they overscan on there as well. They do sell small NTSC LCD's for video work that do under scan but they start around $800. Even my 1080p projector at home has an underscan/overscan option in the menu.

With digital encoding there is none of this any more but trying to retrofit the old stuff can be a pain. One other issue is a CRT can have a display almost full diagonal of the screen and have a very small area in the around the viewing ares. Most LCDs that I could replace them with have a decent size frame around the glass so you have to put a smaller LCD in place.

MaxHeadRoom
08-19-2011, 05:17 PM
Have you checked out EarthLCD? I have used a few of their products from time to time, they make their own interfaces including 'Intelligent Programmable'
http://store.earthlcd.com/LCD-Products/ezLCD
Max.

flylo
08-19-2011, 05:31 PM
Great site,Thanks! Now I can see the walkway in my shop shrinking fast with more projects I'm sure I'll get to soon!

macona
08-19-2011, 05:56 PM
Have you checked out EarthLCD? I have used a few of their products from time to time, they make their own interfaces including 'Intelligent Programmable'
http://store.earthlcd.com/LCD-Products/ezLCD
Max.

Yeah, I have known about them for a long time. Their prices are sky high.

darryl
08-20-2011, 12:49 AM
Back to the other monitor you got- most of the time when there's vertical collapse (narrow band of light horizontally) the problem is in electrolytic capacitors in the vertical deflection circuit. Often, these caps stand close to hot parts and they dry out. Many times I've gone into crt displays with no schematic or info, and just looked for an area that says 'vertical deflection', then just replaced about three caps that were closest to hot parts- and got it working again.

macona
08-20-2011, 01:48 AM
I did change caps in the vertical circuit. But that was not it. The image was not fully collapsed, it was actually full size in the middle of the display and the parts of the image above and below the center of the CRT were squashed. the transistor that controls the vertical sweep from the vertical deflection IC had issues.