CTX CVP-5468NI picture increased width then shrank
 

I've been making use of this VGA monitor with various computers since I got it in March, but its most recent application was as a monitor for my Linux box. That's when I started using the 1024x768 @ 60Hz (XGA) mode often.
Well, a few days ago I was in a white-on-black fullscreen terminal and suddenly the picture width increased drastically. I cranked up the vertical size. The horizontal geometry became distorted and the right edge was folding over. Then, I saw the picture shrink towards the center of the screen and I quickly hit the power button.
I turned the monitor back on a few hours later to be greeted with a dim, shrunken XGA image.
But, the picture is perfectly fine and bright in the 640x480 (VGA) and 800x600 (SVGA) modes.
Here are some pictures of the monitor with the back shell removed. It's fairly clean inside, no visible signs of leaking caps or other damage.
https://files.catbox.moe/smj7cv.JPG
https://files.catbox.moe/unga79.JPG
https://files.catbox.moe/5kx7at.JPG
https://files.catbox.moe/1v6po5.JPG
https://files.catbox.moe/ovw6qu.JPG
https://files.catbox.moe/a8oibs.JPG
https://files.catbox.moe/uf3tao.JPG
https://files.catbox.moe/ph2ygv.JPG
https://files.catbox.moe/7p1nga.JPG
A google search with "5468NI" in quotes returns a few pages mentioning this monitor and repairs.
https://crtdatabase.com/crts/ctx/ctx-cvp-5468ni
https://asavage.dyndns.org/Monitors/monitors.html (search in page for 5468)
https://pe2bz.philpem.me.uk/Html_faqs/Monitortips.htm (search in page for 5468)
I'm inclined to think that a component in one of the horizontal circuits has failed and is bringing down HV or B+, after all XGA is the highest mode supported and therefore highest stress. I'll have to look over the example multisync SVGA monitor block diagram and see how these things really work (coming from CRT TV repair where the frequencies are fixed).
Any feedback would be appreciated. I really want XGA mode to work again on this monitor. It looked really good especially when driven by modern video cards which output very crisp VGA signals.

With all TV's & monitors from anywhere look for cold joints.
ESPECIALY in the oz drive & output. Tapping & poking the chassis
often smoke out the problem.
BTW you can look up the FCC Id number or type acceptance ##
to find the real OEM model etc. Goto FCC.GOV

73 Zeno
LFOD !

Looks like this monitor is what they say it is... The model number is the same under FCC filings.
https://fcc.report/FCC-ID/DBLCVP-5468NI
Would be nice to have the service manual

I tried slapping the chassis and poking some parts near the flyback/HOT with a pencil while in 1024x768 mode, nothing happened.
I did notice that when the monitor is displaying 640x480 and 800x600, the HOT is warm. But with 1024x768 it gets burning hot, I can't leave my finger there.

I took pictures which show operation in 800x600 and 1024x768, but I'm hyperlinking them so they won't embed in the page. They're huge 10 megapixel shots just like the last ones.
800x600 (SVGA) mode.
1024x768 (XGA) mode. It's broken!

If the multisync selection is handled by discrete logic like it was in the early 90s, then you could have a bad chip or voltage in that area.

The way that they did that stuff before there were common ICs to do the resolution binning is that the incoming signal would hit a series of logic gates and comparators that would flip according to what frequency signal was there.

If anything is off in that part of the circuit, then the whole input section cascade fails.

By around 1994 or so, I think they had mostly transitioned to dedicated ICs that could do the resolution binning and tell the deflection circuitry how to behave.

It's really a big idk unless you have schematics though. Maybe someone else could help, but it's pretty hard with zero knowledge of the layout. That time period had a very, very large spread of potential circuit layouts.

The horizontal section has two ICs, an LA7850 for deflection and an HD74LS86P which is four XOR gates. It's mounted on a little card with a few passives and four wires going to the neckboard.
There's a cap (C427) which lives right next to the LA7850 and H-hold adjustment and it looks like the type that cracks and smokes. I totally forgot what it's called

This is definitely not using discrete logic for frequency binning. It's firmly after that (1996), but it could still be using some external components to aid in controlling the horizontal oscillator in the LA7850.

Without a schematic, I wouldn't be able to solve that puzzle. Too complicated for me.

.

Good Evening; Good monitor, well worth keeping... The second picture you posted
https://files.catbox.moe/yo5ysk.JPG Looks like you have horiz linearity problem,
as listed on the site https://asavage.dyndns.org/Monitors/monitors.html like
you said search cvp-5468. I noticed that in a lot of the symptoms & solutions
area there are little reference to what mode the monitor is working in when
the problem is observed, so the mode selection may be the wrong spot to
go looking for an answer.... + Your screen pic seems to show a properly
synced picture, just poor linearity - So check on the part they suggest
5468 Hori linearity problem C322 .39J 400v. ESR should be less than 10.
At least that is where I would start with the information available.

The other thing I would do is step through all the items in the menu,
be sure every item works as it should, geometry, pincushion, etc.

Also it may just be worth looking at, but in this pic:
https://files.catbox.moe/a8oibs.JPG Not quite center of pic there is a
cap that in just above a twisted Orange, Yellow, & Green wires that
looks a little big on top - Look at it closer..... Also lots of the
problems & solutions on that page reference bad solder
joints - Like Zeno said, so make that a priority.

Also, this was a High $$ Monitor so chances are
the parts were better than they use today, so
Don't go randomly changing caps just cause
you think you should, change only what's
necessary. Find the actual fault.

Good Luck.


.

If this is the case, then why would the other modes be working correctly? If there is a problem with the horizontal linearity, wouldn't there be an issue with all of the modes?

I'm not saying this is impossible, but could you suggest a situation where a component in the horizontal linearity circuit is bad, but it only manifests in a single resolution?

I know nothing about this monitor, but I can imagine components being switched in for a certain mode, and either the switch or the auxiliary component failing.

.

Well I don't know how this thing works, I did see the spec sheet for the deflection chip
you posted, and it seems to have a very wide frequency range that it will work under.
But still uses the same vert. & horiz output drivers, transformers, yoke etc. So that
means the output circuit needs to have a wide bandwidth with relatively linear
flat response. The Q&A Board with a symptom "5468 Hori linearity problem
C322 .39J 400v. ESR should be less than 10." Could lead to the problem,
or that part of the circuit where you need to test a lot of components blindly.
Without a schematic you have to start somewhere, it might as well be C322.

That cap they are saying something about it's ESR, so could that cap be in a
spot that might effect the frequency roll-off of the horiz. deflection? Don't know.

Without a service manual, which often may have a theory of operation section,
may explain what parts are dependent on mode, and which are not.

Someone here once had a tv they were fixing and no one could find a service
manual for, so he drew a circuit himself of the tube and attaching parts and
figured out what he had to do to get it fixed. If you don't find a service
manual, you may just have to track down this cap C322, and figure out
if it's important or not.... It's just my 2 cents as to where I would start.
Again with the information available, which ain't much, it's gunna be
hard to pick a starting point.

If there were multiple pictures side-by-side, or some non synced
picture, I might suspect the mode switching section, that's just
my thinking, it's somewhere else, I could be wrong. I have not
had this problem on a multi-synced monitor.



.

C322 as a suspect makes a lot of sense, especially if it's tied to horizontal linearity at higher scan frequencies. XGA would push the deflection system harder than VGA or SVGA, so a cap on the edge of spec might not show issues until then. ESR rising past 10 ohms could absolutely distort the waveform under load. Without schematics, starting with parts known to fail under similar symptoms is solid strategy.

Hi, this monitor does not have an OSD. Those adjustments are controlled by pots on the mainboard/neckboard.
My gut says you're right, given the failure mode I watched unfold.

I searched Jason Scott's Discmaster for "5468NI" and found old shareware versions of Anatek's "Resolve Monitor Repair Database" on some commercial CD-ROMs. They're DOS programs which list many monitors and repairs. Unfortunately these versions only have information on the CTX CVP-5468 which is very similar to mine except it doesn't support Non-Interlaced (NI) modes that the 5468NI can show.
I was able to run Resolve in DOSBOX and took some screenshots of listed repairs for the 5468.

List of repairs
List of repairs (continued)
SCREEN TO BRIGHT, RASTER BRIGHT, NO CONTRAST/BRIGHTNESS
NO COLORS
DEAD, NO RASTER, NO VIDEO, SMOKED, BLANK SCREEN, SHUTS DOWN #1
DEAD, NO RASTER, NO VIDEO, SMOKED, BLANK SCREEN, SHUTS DOWN #2
DEAD, NO RASTER, NO VIDEO, SMOKED, BLANK SCREEN, SHUTS DOWN #3
DEAD, NO RASTER, NO VIDEO, SMOKED, BLANK SCREEN, SHUTS DOWN #4
DEAD, NO RASTER, NO VIDEO, SMOKED, BLANK SCREEN, SHUTS DOWN #5
DEAD, NO RASTER, NO VIDEO, SMOKED, BLANK SCREEN, SHUTS DOWN #6
DIES, TRYS TO COME UP AND DIES, USUALLY JUST AFTER STARTUP
HORIZONTAL LINE, NO VERTICAL, VERTICAL COLLAPSE
NO HORIZONTAL SYNC, JITTERS, ROLLS
NO VIDEO, BLANK SCREEN, NO CHARACTERS, RASTER BUT NO VIDEO #1
NO VIDEO, BLANK SCREEN, NO CHARACTERS, RASTER BUT NO VIDEO #2
NO VIDEO, BLANK SCREEN, NO CHARACTERS, RASTER BUT NO VIDEO #3
VERTICAL LINE, NO HORIZONTAL, NO SCAN
VERTICAL SIZE IS WRONG, SCRUNCHED, VERTICAL COLLAPSE

However, it looks like Resolve 3.17 and above document the 5468NI.
http://web.archive.org/web/199610280...om/reslist.htm
It cost $249 back in the day..! The Internet Archive doesn't have anything newer than the freeware releases of Resolve 2. If anyone has Resolve 3.17 or greater, please let me know.

As for CTX, their service manuals were only $15 direct from them.
http://www.repairfaq.org/REPAIR/F_mo...ml#MONFAQI_003
So I'm sure some old repair shops or people still have the schematics for this monitor. They just haven't been archived yet.

BTW: this monitor have S-cap switching for each resolution? And have variable HOT supply (one voltage for each resolution, probably)?

To be sure if for eg. for a higher resolution the supply will need to change (for measuring and avoiding damage). If remains untouched, the PSU have stuck with one single resolution setup (naturally the possibly culprit will be the resolution freq. select already cited here), and image or are shrink with one resolution, or too big with another.

I've had luck in the past doing a saved search for a repair manual I couldn't find and waiting until it came up on ebay. In some case, it took like 10 years for that to happen, but others more like 1-3 years. I know that doesn't sound ideal, but if you can just stick it in a closet and use something else for now, it could be a future project. That's more or less how I do things when I get stuck.

Of course, there's been some good suggestions too, you could try checking the ESR on that cap and see if it's okay.

Also, reflowing the solder on the PCB will fix a surprising number of faults; old solder fails.

This. It looks like a mini RIFA to me...
https://files.catbox.moe/woa2em.JPG

Pull it and check the ESR. Did you pull C322 and check it?

I have to buy a component tester.

I would recommend it. You can either buy one of those cheapo testers they have on ebay or if you want something a little more trusted, the Peak ESR meters have a decent reputation.

Personally, I have one of those cheap ones, but I also have a real LCR meter. Between the two of those things (and my DMM) I'm able to test most anything I need to.

Honestly, I've seen you around here enough that at this point you probably want to have some tools to make your life a little easier. You have to consider how much your time is worth, stabbing around in the dark and replacing component hoping it will help.

I pulled C322 and C427 today but the cheap LCR-T4 I bought only measures their capacitance, not ESR.
C322 rang up as 50pF on both the LCR and my multimeter!
It should be 220nF. The code on the package is 224J which means 22 x 10^4 = 220000pF with 5% tolerance. The second line says MP400 which I assume means rated 400V DC.
https://files.catbox.moe/7prbnh.jpeg
I’ll find a replacement mylar cap on DigiKey but let me know if there are any special considerations to be made.

Here is C427. Measured capacitance was close enough to 330pF.
https://files.catbox.moe/w1rpj1.jpeg
Should I replace it? I still don’t know its ESR.

Are you sure? I have a similar cheapo tester that has ESR on the screen when I test for capacitance.

yep it's bad.

I can't see your picture. It's a bad link for me.

I believe that 330J is a 33pF code (33 for the digits and "0" for the number of places). What does it measure?

Also, the cheapo testers sometimes don't go down very low, and they definitely can't go up very high. So it might be outside of it's limits.

I put a 2200uF electrolytic cap in my tester and it measured the ESR of that. I don't think the ESR test works for caps in the pF range.
Works fine here... It's just a picture of a film cap that says 330J and S50F.
317pF. Again, the formula is [first two digits] multiplied by 10 to the power of [third digit]. 10 to the power of 0 is just 10 so 33 x 10 = 330 (which is the full three digits anyways). If it was 33pF, why wouldn't they just save ink and write only "33"?

Okay, nevermind. I got myself confused about it because we were talking about pF. 330J is 33nF, which is the same thing as 330pF.

Here’s the C322 replacement I picked.
https://www.digikey.com/short/5c3crh37
Hopefully it’s appropriate.

Looks good to me. Low loss, 450V. Should be better than what came out of there.

After replacing C322 the XGA mode works again. Here's a picture.
The brand new 220nF cap measured 30Ω ESR on the cheap LCR

Awesome! Glad to hear you got it rolling.

That is definitely not right. Film caps are generally very low ESR. I suppose there could be something about the way the cheap tester does the ESR measurement that isn't good for filim caps, or it's possible that it has a narrower range for ESR.

Also possible that it's just not super reliable. I generally use my cheap tester as go/no-go tester and I don't rely on it for precision.

If you decide to get something that is specifically more reliable for ESR, people seem to like the PEAK devices.

I decided to get a DE-5000 LCR meter after watching a bunch of reviews, and I'm happy with it. It is not quite as convenient to use as the PEAK meter, but it's got a wider range of functions and seems to be pretty accurate according to the hobbyist community.

Perhaps a not so good ESR meter can measure with low frequency so low value caps, not being electros of high µF, will end measuring reactance and not ESR.