CurtPalme.com Home Theater Forum

gjsmo · Joined: 06 Jul 2012 Posts: 9

FYI, this is my first post on the forums here and my first experience, with CRT projectors, so I'm effectively a noob.

I have an NEC XG75A which I obtained as surplus electronics from the local university (the recently upgraded pretty much everything in their media department). It it currently setup running 1920x1080, 30Hz interlaced (according to AMD CCC), and connected to my computer with the default RGBHV BNC connectors.

Recently, I started having an issue with the horizontal picture size. Some time after turning the projector on, the picture would expand to the full width available, and randomly come back to the set size for about a second, then expand once again. More recently, it began to lose horizontal scan completely, with the picture collapsing to a vertical line, then expanding back out.

I would think this is a horizontal amplifier issue, but I do not know. I have some understanding of digital electronics, not too much of analog. I do have a 100MHz dual-trace oscilloscope to use, if this would help. Could I simply have blown some electrolytic capacitors? If so, I can replace these myself, I've fixed a few LCDs with power supply issues like this before.

The projector *is* in a relatively warm area. Here in Upstate New York, it's been rather warm recently, and since my computer and the projector are setup in my badly ventilated attic, it can get to 90F or above. Unfortunately, there is no better place right now for either - I don't particularly enjoy the heat. If necessary I could eventually move it to a lower floor. I do know heat can cause electrolytics to blow, and could this be the entire issue?

On a related note, what is the cheapest (though *good) way to get new or "refurbished" green and blue tubes for this projector? They both (particularly the blue) have rather noticeable burn-in, though you can't see any burn-in at all on the red tube. All tubes have slightly over 6300 hours on them.

Thanks for any help.

gjaky · Joined: 05 Jun 2010 Posts: 979 Location: Budapest, Hungary

Hello and wellcome to the forum.
First let's check some other resolutions like trusty old 800x600 or 1024x768, and tell us if the problem is still there?
I would reccommend to turn down the contrast to just see the actual picture, a bad deflection board can nuke your tubes (especially when the horizontal width is narrowing to a line).
Try to reconnect the connectors on the horizontal deflection board, maybe that helps.

first we make the pj running well, and then we might think about the new tubes Smile

gjsmo · Joined: 06 Jul 2012 Posts: 9

Perfect timing, apparently.

I've tried several different resolutions, including 1024x768 (currently the only other setup resolution - damn the lack of multi-scan!) and the problem still exists.
I'll look at the horizontal deflection boards, if I can find them. I presume at least one of the manuals shows their location?

gjaky · Joined: 05 Jun 2010 Posts: 979 Location: Budapest, Hungary

Here it is:
http://www.curtpalme.com/NECXG_Layout1.shtm

gjsmo · Joined: 06 Jul 2012 Posts: 9

That makes sense, I had to flip the deflection connectors because I don't have rear projection. Apparently I already knew where that board was.

Ok, so I tried to go to 1024x768@60Hz, and the projector shut off the picture (normal) during the mode switch, but didn't turn it back on. And then it shut off completely, I heard the power relay click, and I don't want to power it back on just yet.

I have the top cover off, no caps seem to be obviously blown, no obvious shorts of any kind. Now what?

gjaky · Joined: 05 Jun 2010 Posts: 979 Location: Budapest, Hungary

any error codes on the led display at the back?

gjsmo · Joined: 06 Jul 2012 Posts: 9

I didn't think to look - I'm somewhat (probably irrationally) afraid of HV electronics blowing up, so I flipped the main rocker switch off as soon as it turned off. Too much Hollywood, I'm guessing.

Should I leave it running until something happens, then? Would that damage something?
EDIT: It only takes a few minutes for these problems to show up, I have no problem leaving it running, as long as you can (truthfully) tell me I won't blow it up and I won't cause any damage by leaving it running.

gjaky · Joined: 05 Jun 2010 Posts: 979 Location: Budapest, Hungary

Well in theory the machine is full with protection circuits, so it is very unlikely to blow up, but as always there are accidents... Very Happy

When the pj shut down itself probably it was because the horizontal deflection problem triggered the protection, when the power is off then nothing to worry about, but the error codes gives us guidelines where to start.

I think there is nothing serious with your machine, maybe a cracked solder joint hiding somewhere, the trick is to find it...

Oh, and what about the internal testpatterns, are those varying as well?

**Curt Palme** · Posted: Fri Jul 06, 2012 9:01 pm Post subject:

I have had this on two sets. IN both of my cases, it was the H deflection board. I've got good working ones at $100 plus shipping. I spent hours working on one board, trying to find the source of the pulsing, but in the end scrapped it, as it was somewhere in the low level circuits.

The set is protected, but watch the spark gaps. Chances are one or more are arcing when the H width collapses, and eventually they will nuke themselves and blow up. I'd swap the board ASAP. With the scan collapsing, you also stand the chance of burning the tubes. Not a repair I'd risk killing other parts of the set over.

gjsmo · Joined: 06 Jul 2012 Posts: 9

@gjaky The entire raster is moving, including the actual image, the info text, menus, and test patterns.

@Curt Well that's convenient, and certainly affordable. Is it at all possible for me to attempt a repair? Or would there be too much pain and suffering?

I suppose the high humidity (not really fixable without installing house AC) could have to do with the arcing. I don't hear the sound of an arc (the set is literally directly above my head and behind me, I can hear it fairly well), but it can't hurt to be safe.

macgyver655 · Joined: 22 Aug 2007 Posts: 7059

I'm only speed reading today as I'm doing other stuff but, is the raster shrinking both horz and vert, or just horz when it's smaller?

**Curt Palme** · Posted: Fri Jul 06, 2012 9:15 pm Post subject:

I'm sure it CAN be fixed, but what's your time worth, and do you want to risk nuking the tubes?

That wasn't actually meant to be a snide comment either! Smile

I used to pride myself in thinking that I can fix any projector board, including ones that had been under water from a flood, etc, but I realized that as the prices of these drop, and seeing that I have about 10 H boards in stock, that there's no point in spending time repairing them if they can't be fixed in under 2 hours. Once in a while I'll catch myself spending 4-5 hours on a board, just for the challenge, but as I said, in this case you might just want to replace it with a known working one.

gjsmo · Joined: 06 Jul 2012 Posts: 9

@macgyver Only (oops EDIT) horizontally
@Curt I have plenty of free time. Think high school student on summer break before senior year - because that's EXACTLY what I am.
I suppose I don't really want to risk the tubes, but then again 1.) I got this projector for free 2.) I know how to solder and 3.) My typical attitude - What the hell, go for it.

I wouldn't say I'm the absolute most experienced electrical engineer - but I know enough to not mess something up that wasn't messed up before. What would be the procedure for checking out a board like this? I presume at some point there needs to be a powered test... But all I can think of is to check for cold/cracked solder joints, check the big caps, make sure nothing is blown up etc...

**Curt Palme** · Posted: Fri Jul 06, 2012 9:44 pm Post subject:

Well, that's exactly how I started! what I'd do then is pull the CRT sockets, so the tubes don't come on, or pull the HV leads out of the splitter. That way, you can turn the set on without the tubes coming on.

gjsmo · Joined: 06 Jul 2012 Posts: 9

... this might seem like a stupid question, but how will I know if the H deflection board is working if I can't see anything? oscilloscope? And then that would mean I would have to know what the waveform was *supposed* to be like, which I don't.

**Curt Palme** · Posted: Fri Jul 06, 2012 9:56 pm Post subject:

Yes, exactly, with a scope.

What you'll see though is the H output waveform collapse when the pix collapses, so that's your key right there.

The red tube is less prone to damage than the G or B, so leave it connected if you want. If you burn it, a lot of us have spares for cheap, but I'd say disconnect them all, and scope your way back from the H outputs to the drivers and see where you get a clean signal.

gjsmo · Joined: 06 Jul 2012 Posts: 9

Well, time to start take the board out I guess and look it over.
Mind if I PM you with progress (or lack thereof)?

**Curt Palme** · Posted: Fri Jul 06, 2012 10:21 pm Post subject:

No, go right ahead!

macgyver655 · Joined: 22 Aug 2007 Posts: 7059

I think you'd be better off shutting down those tubes and running it till it shuts down and see if it spits out a code. The code may tell you more.

gjaky · Joined: 05 Jun 2010 Posts: 979 Location: Budapest, Hungary