Contents:
Flameproof Resistor or Fusable Resistor are often designated by the symbol 'FR'. They are the same. You may see these in the switchmode power supplies used in TVs and monitors. They will look like power resistors but will be colored blue or gray, or may be rectangular ceramic blocks. They should only be replaced with flameproof resistors with identical ratings. They serve a very important safety function. These usually serve as fuses in addition to any other fuses that may be present (and in addition to their function as a resistor, though this isn't always needed). Since your FR has blown, you probably have shorted semiconductors that will need to be replaced as well. I would check all the transistors and diodes in the power supply with an ohmmeter. You may find that the main switch mode transistor has decided to turn into a blob of solder - dead short. Check everything out even if you find one bad part - many components can fail or cause other components to fail if you don't locate them all. Check resistors as well, even if they look ok. Since they function as fuses, flameproof resistors should not be replaced with higher wattage types unless specifically allowed by the manufacturer. These would not blow at the same level of overload possibly resulting in damage to other parts of the circuitry and increasing the risk of fire. Then, with a load on the output of the power supply use a Variac to bring up the voltage slowly and observe what happens. At 50 VAC or less, the switcher should kick in and produce some output though correct regulation may not occur until 80 VAC or more. The outputs voltages may even be greater than spec'd with a small load before regulation is correct.
Note: the following is just a brief introduction. For more detailed deflection
system theory of operationo and sample circuits, see the document: "TV and
Monitor Deflection Systems".
The electron beams in the CRT need to be scanned horizontally and vertically
in a very precise manner to produce a raster - and a picture.
For NTSC and PAL, the horizontal scan rates are 15,734 and 15,625 Hz
respectively, the vertical scan rates are 60 and 50 Hz (approximately)
respectively.
For PCs and workstation monitors, a wide range of scan rates are used.
For example:
Standard Horizontal, KHz Vertical, Hz
------------------------------------------------
MDA 18.43 50
CGA 15.75 60
EGA 15.75-21.85 60
VGA 31.4 60-70
SVGA (800x600) 35-40 50-75+
SVGA (1024x768) 43-52+ 43-75+
SVGA (1280x1024) 64-72+ 60-75+
Workstations 64-102+ 60-76+
Even in high resolution fixed frequency monitors, these high horizontal
(in particular) scan rates necessitate some fancy circuit design. All
components are running under stressful conditions and it is amazing that
failures are not more common.
With auto-scan monitors, the complexity of the circuits increases dramatically
to accommodate the wide range of horizontal scan rates. Relays or electronic
switches are used to select power supply voltages, tuning components, and
to make other alternations in the deflection circuits to handle DOS VGA
one minute and Autocad 1280x1024 the next. It comes as no surprise that
the most stressful time for a monitor is when switching scan rates.
Unfortunately, successfully diagnosing problems dealing with the scan
switching logic and circuitry is virtually impossible without a schematic.
The deflection yoke includes sets of coils for horizontal and vertical
scanning oriented at 90 degrees with respect to each other. Additional
coils are needed to correct for pincushion and other geometric defects.
The deflection circuits must be synchronized and phase locked to the
incoming video signal.
Therefore, we have the following functions:
1. Sync separator to obtain horizontal and vertical synchronization pulses
for monitors with composite video or sync inputs. Input sync detectors
and auto polarity switching circuits as needed for separate horizontal
and vertical sync inputs.
2. Horizontal oscillator which locks to horizontal sync pulses.
3. Horizontal drive followed by horizontal output which feeds deflection
yoke (and flyback for HV and other voltages), Yoke requires a sawtooth
current waveform for linear horizontal deflection. Horizontal output
in all but the smaller TVs or monitors is a large discrete power
transistor, most often an NPN bipolar type.
4. Vertical oscillator which locks to vertical sync pulses. Yoke requires
sawtooth waveform for linear vertical deflection.
5. Vertical drive/output which feeds vertical deflection yoke. Newer TVs
and monitors use ICs for vertical drive and output.
6. Various additional deflection signals to correct for the imperfections
in the geometry of large angle deflection CRTs. These may be fed into
the normal deflection coils and/or there may be separate coils mounted
on the neck of the CRT.
7. Auto-scan deflection control and selection circuitry (auto-scan monitors
only), probably controlled by a microprocessor which stores scan
parameters for each scan rate and automatically detects the appropriate
settings to use by analyzing the input video. For horizontal deflection,
the usual way of size constant regardless of scan rate is to scale the
B+ to the HOT with horizontal frequency. Thus, VGA resolution may us
60 V B+ while 1280x1024 at 75 Hz may require 150 V. Various other
components may need to be selected based on scan rate. Relays are often
used for this selection since they are easy to control and can handle the
voltages and currents in the various deflection circuits reliably.
These sorts of problems usually relate to the picture shifting when switching between applications or between DOS and Windows. A couple things to check if you have a setup program for your video card (1-3 are software adjustments in the DOS setup program): 1. Make sure you are running well withing the accepted scan rates for each resolution. 2. Toggle sync polarity and see if this makes any difference. 3. Adjust H position or phase and see what this does. 4. Make sure your cables are secure. While a bad connection would likely messed things up worse, it won't hurt to check. Your monitor may have a problem though it is not likely to be major (in a relative way). If you still like the monitor, repair may be worth the money. However, not doing anything now may lead to more serious (and costly monitor damage.
First, make sure you are not specifying incorrect scan rate for your monitor. Check your video card setup and/or monitor selection in Win95. Assuming you are not violating the scan rate specifications but have a picture that is twice the height of the screen and one half the width, for example, this could indicate a failure in the scan rate switching circuitry of an auto-scan monitor. Either the logic is faulty and ordering the wrong selections for power supply voltage and tuning components or the relays or the relevant parts are faulty. This could be due to bad connections as well - quite likely in fact. Also, try to reset the afflicted parameters using the digital controls (if relevant) and confirm that your video card is putting out the correct scan rate - try another monitor or examine the video signals with an oscilloscope. Try prodding the circuit boards with an insulated stick - this may identify bad connections or unstick a sticky relay. A schematics will likely be needed to proceed further with these sorts of problems.
Complaints about the picture not filling the screen with computer monitors are common but may not indicate problems (except with your expectations). Older monitors, in particular, often did not allow a full screen display at certain resolutions. There may be underscan modes/switches as well. Keep in mind that advertizing a large diagonal CRT does not necessarily imply that you can fill it! However, if this problem just happened with no changes to your computer system (video card, scan rates, O/S), then the following are possibilities: * The B+ to the horizontal output is lower than normal. The way width control functions is that as you increase the horizontal scan rate, the B+ to the HOT must increase to keep the width constant. It could be that yours is low to start with and not tracking scan rate changes either. * A bad capacitor might also result in reduced width but I would expect non-linearity as well. * As noted in the section: "Gross problems in size or position at certain scan rates", there could be problems in the scan rate switching circuitry selecting incorrect components for certain scan rates. * There might be a bad (low value or high ESR) decoupling capacitor. Scope the rail after the low-value decoupling R for H-rate stuff. There shouldn't be anything significant. If there is, the ESR of the decoupling capacitor is too high or its value is too low. Seen it often where it also cooks the decoupling R, because the efficiency of the H-out becomes poor. (gwoods@albany.net (Gary Woods)). * A more unlikely possibility is a open yoke winding. The horizontal deflection yoke consists of multiple windings in parallel so it is theoretically possible for one or more of these to open up. I don't know what effects the associated detuning of the horizontal output circuit would have in this case.
The short answer is - quite possibly. Don't push your luck. Mostly, there are problems at scan rates which exceed the monitor's specifications. However, some poorly designed monitors or just a particular combination of events can blow a monitor with too low a scan rate or an absent or corrupted signal input. There was one case where a very expensive high performance monitor would consistently blow its horizontal deflection circuits when driven by a particular ATI video card. It turned out that during the power-on self test of the ATI BIOS, just the wrong video timing was being generated for a fraction of a second - but that was enough. As far as scan rate limits, there is no way of knowing - it really all depends on the quality of the design of your monitor. Some will happily run continuously at 25% above specifications. Other will blow out totally at the first excuse. The specification that is likely to be more critical is the horizontal rate as it probably puts more stress on the components than the vertical rate. I have found that as you approach the upper limits, there is a good chance that the geometric accuracy of the raster near the top of the screen may start to deteriorate due to lock in problems as well. However, it would be foolhardy to depend on this sort of behavior as an indication of going over the edge. It will be much too late when you find out. If the manual says 75 Hz V and 64 KHz H, stay below **both** of these. If you exceed the safe ratings and the design isn't really good, there is the possibility of blowing components in the horizontal deflection and high voltage sections which will result in expensive repair bills. You will likely get no warning of impending failure. In addition, even if the monitor does not immediately turn into a pile of smoking silicon and plastic, components may be under more stress and running at higher levels of power dissipation. Total failure may be just around the corner. More subtle degradation in performance may occur over time as well. You won't see the difference anyhow beyond 75 Hz and your programs may run slightly faster at lower refresh rates since the video is not using as much bandwidth (however, the difference here may be very slight or non-existent depending on your board, computer, applications, etc..
You were happily playing 'Doom' when the sides of the picture squeezed in two inches or so when the entire monitor went dead - has remained like this since. Sound is fine, but no activity at all from the tube. Has it died? How much time, effort, and expense to fix? No, it's not dead, at least it certainly is not the picture tube. Your probably shot the monitor instead of the bad guys! Is there any indication of light on the screen? Any indication of the horizontal deflection running at all as evidenced by static on the screen? In any case, there is a problem in the horizontal deflection and you probably have no high voltage as well assuming no light on the screen. The fact that it squeezed in first indicates that a partial short or other fault may have developed in the horizontal deflection circuits - possibly the deflection yoke or flyback transformer. It could also have been a bad connection letting loose. Once it failed completely, the horizontal output transistor may have bought the farm or blown a fuse.
Confirm that the horizontal deflection is shutting down (along with the high voltage since it is derived from horizontal deflection: listen for the high pitched deflection whine (NTSC/PAL/CGA), test for static on the screen, see if the CRT filaments are lit, turn up the brightness and/or screen control to see if you can get a raster) and then why: 1) Power is failing to the horizontal output transistor - this could be due to a low voltage power supply problem, bad connection, etc. 2) Base drive to the horizontal output transistor is failing - could be a fault in the horizontal oscillator or bad connection. 3) Problem with the flyback transformer or its secondary loads (flyback may provide other power voltages). 4. X-ray protection is activating - either due to excess HV or due to a fault in the X-ray protection circuitry. If the problem comes and goes erratically it sounds like a bad connection, especially if whacking has an effect. If it comes and goes periodically, then a component could be heating up and failing, then cooling, etc.
A very narrow picture may indicate problems with the power supply to the horizontal deflection circuits, incorrect scan rate selection or defective components, faulty deflection yoke, or bad connections. If the size is erratic and/or gently whacking the monitor makes the width change, bad connections are likely. See the section: "Monitor manufacturing quality and cold solder joints". Confirm that your video card is running at the proper scan rate - particularly that it is not violating the monitor's specifications. An excessive horizontal scan rate is a common cause of a reduced width raster. Try its software setup adjustments as these may have been lost. Beyond this, a schematic will probably be needed to isolate the fault.
Most modern monitors are nearly perfect with respect to non-linearity. There almost never any user adjustments and there may not even be an internal adjustments. See the section: "Position, size, and linearity adjustment". A sudden change in linearity or a monitor that requires a warmup period before linearity becomes acceptable may have a bad component - probably a capacitor in the horizontal deflection circuits. For the latter, try some cold spray or a heatgun to see if you can locate the bad part. (From: helio (mmccann@usa.pipeline.com)). You should likely begin in the area immediately around the HOT, perhaps there might be a high frequency NP (non polarized) electrolytic just starting to go. Some larger monochrome monitors actually have working H-lin adjustment coils (believe it or not) especially if they are older ones. But most are glued/potted down or fixed value. If you locate it (the coil) the problem should be nearby.
"I'm trying to repair a Target DN-1564 monitor with a problem in the horizontal deflection: on both the left and right side of the screen the picture gets squeezed together, regardless of H-width and other settings. I've checked most semiconductors in this part, but I can't find anything wrong there." This sounds like an S-correction capacitor may have too small a value or failed open. Check the capacitors in the vicinity of the deflection yoke connector and HOT. It could be due to bad connections as well. S-correction is needed to linearize the horizontal scan (and vertical as well scan but that is a separate circuit). Without S-correction, the scan current would be nearly linear. This would result in greater coverage in a given time near the edges of high deflection angle CRTs. The picture would appear stretched near the edges In this case, the correction appears excessive. (From: David Henniker (david.henniker@cableinet.co.uk)). I had a similar problem with a monitor (here in Edinburgh Scotland). The S-correction cap was open-circuit altogether. Other caps in parallel allowed the distorted scan. If it had been a TV there wouldn't have been other caps in parallel and the result would have been no line scan, maybe a vertical line (line collapse) or nothing at all.
This means the vertical size is reduced with or without distortion. Before attacking the circuitry, make sure your vertical scan rate is within the monitor's capabilities and that the user vertical size control is adjusted properly. If there is no distortion, this is likely as many (but not all) circuit problems would result in non-linearity or cutoff of the top or bottom portions of the picture. All you may need to do is change your computer's video settings! Swap the monitor or computer to be sure it is not a problem with the video card. However, if failure happened suddenly and the vertical is squashed at all scan rates, this is likely a vertical deflection problem - possibly a bad capacitor, bad connection, bad flyback/pumpup diode, or other component. None of these should be very expensive (in a relative sort of way). If the symptoms change - particularly if they become less severe - as the unit warms up, a dried up electrolytic capacitor is most likely. If they get worse, it could be a bad semiconductor. Freeze spray or a heat gun may be useful in identifying the defective component. It is often easiest to substitute a good capacitor for each electrolytic in the vertical output circuit. Look for bad connections (particularly to the deflection yoke), then consider replacing the vertical output IC or transistor(s). A defective deflection yoke is also possible or in rare cases, a bad yoke damping resistor (e.g., 500 ohms, may be mounted on the yoke assembly itself). The following are NOT possible: CRT or flyback. I am just trying to think of really expensive parts that cannot possibly be at fault :-).
This means that the size of the picture is not constant from top to bottom (width changes) or left to right (height changes). Note that some slight amount of this is probably just within the manufacturing tolerance of the deflection yoke and factory setup (geometry magnet placement, if any). With a monitor, such defects are more noticeable than with a TV since much of the display is of rectangular boxes - i.e., windows, lines of text, graphics, etc. Furthermore, the monitor is usually run just barely underscanned to maximize the viewing area without cutting anything off. Any deviations from perfection show up in relation to the CRT bezel. However, a sudden increase may indicate a problem with the deflection yoke. An open or short in a winding (or any associated components mounted on the yoke assembly) will result in the beam being deflected less strongly on the side where that winding is located. However, with a high scan rate monitor, there may be many individual windings connected in parallel in the yoke so the effect of only one opening up may not be as dramatic as with a TV where there may only be a single pair of windings for the horizontal and another for the vertical. A simple test of the yoke in this case can be performed by simply swapping the connections to the yoke for the affected direction (i.e., if the width changes from top to bottom, interchange the connections to the vertical windings). * If the keystone shape remains the same (but of course the picture flips), it is likely the yoke. * If the keystone shape flips, it is a circuit problem (see below). See the section: "Deflection yoke testing". If the monitor has been dropped off a 20 story building, the yoke may have shifted its position on the neck, of the CRT resulting in all sorts of geometry and convergence problems (at the very least). (From: James Poore (aw133@lafn.org)). I have seen the 'reverse keystoning' in several monitors and the fix is usually the same. In the horizontal leg of the pincushion transformer are 1 or more electrolytics to ground. The caps have + going to transformer and - to ground. Anyway when they start loosing capacitance and/or become leaky the reverse keystoning effects become more pronounced.
If the picture area is expanding or contracting without any changes to your video card settings or other software. then there is a problem with the power supplies in the monitor. This would be confirmed if the change is (1) gradual over the course of say, an hour, and/or (2) gently whacking the monitor has some effect indicating bad internal connections. Software problems would not result in either of these characteristics. Note that if the change is very small - say, less than 1 or 2%, then it may simply be normal for your monitor due to poor design or the use of inferior components - some parts associated with power supply regulation may be changing value as the monitors warms up. A way to confirm that something is drifting due to thermal problems would be the monitor from another computer and see if the same thing happens. Just powering the monitor by itself (but not in any power saving mode) might also work for this test. One possible cause could be that the high voltage is drifting gradually due to a faulty component - increasing and making the beam 'stiffer' or vice-versa. If this is the case there might also be a gradual change in brightness as well (decreasing image size -> increase in brightness). Alternatively, the HV may be stable but the power to both H and V deflection is gradually changing. Excess high voltage can increase the X-ray emissions and any kind of power supply problems may ultimately result in total failure and an expensive repair. Therefore, these symptoms should not be ignored. See the sections on low voltage and high voltage power supply problems.
For monitors using BNC cables, first make sure that the cable connections are correct - interchange of H and V sync or G with one of the other video signals (sync-on-gree setups) can result in all kinds of weird sync problems. There are a wide variety of causes for a monitor that will not display a stable or properly configured image. Among the symptoms are: * Lack of sync horizontal - drifts smoothly horizontally. This may mean that the horizontal sync signal is missing due to a bent, pushed in, or broken connector pin (pin 13) or other bad connection or a fault in the sync processing circuitry. * Incorrect lock horizontal - torn picture (like a TV with the horizontal hold control misadjusted - if you remember these). This means that the sync signal is reaching the monitor but that it is having problem locking to it. Check the rate specifications - you may be exceeding them. * Lack of sync vertical - rolls smoothly vertically. This may mean that the vertical sync signal is missing due to a bent, pushed in, or broken connector pin (pin 14) or other bad connection or a fault in the sync processing circuitry. * Lock not stable vertical - jumps or vibrates vertically. This may be due to scan rate problems or a fault in the vertical sync circuitry of the monitor. * Multiple or repeated images horizontally or vertically. There may be multiple images side-by-side, on top of each other, or interleaved. Most likely cause is driving the monitor with an incorrect scan rate. However, faulty circuitry could also be to blame. Additional comments on some of these problems follow in the next few sections.
A monitor which loses horizontal lock when changing resolutions, momentarily losing the signal, or switching inputs may have a horizontal oscillator that is way out of adjustment or has drifted in frequency due to aging components. Alternatively, you may be running at scan rates that are not supported by your monitor. Check its user manual (yeh, right, like you have it!). Use the setup program that came with your video card to adjust the default scan rates to match the monitor. Not only will it lock better, you are less likely to damage the monitor by feeding it improper scan rates. Note that the characteristics of this are distinctly different than for total loss of sync. In the latter case, the picture will drift sideways and/or up and down while with an off frequency oscillator, the torn up picture will try at least to remain stationary. Assuming you are have your video card set up properly - double check anyhow - this could be a capacitor or other similar part. Or, the oscillator frequency may just need to be tweaked (particularly with older monitors). There may be an internal horizontal frequency adjustment - either a pot or a coil - which may need a slight tweak. If a coil, use a plastic alignment tool, not metal to avoid cracking the fragile core. There may be several adjustments for auto-scan monitors - one for each major scan range. A schematic will be useful to locate the adjustment if any or to identify possible defective parts. If it is a heat related problem try cold spray or a heat gun in an effort localize the offending part.
If there are hum bars or wiggles in the picture, see the section: "Reduced width picture and/or hum bars in picture". If both width and height are affected, the cause is likely something common: low, low voltage power supply voltages or excessive high voltage (resulting in a 'stiffer' beam). (From: Jerry G. (jerryg@total.net)). Lack of width is usually caused by defective power supply, low horizontal drive to the yoke and flyback, defective circuits in the pincushioning amplifier section, excessive high-voltage caused by defective voltage regulation, and or excessive loading on the secondary side of the flyback.
The problem lies either in the horizontal oscillator or in the sync system. If it really is a problem with sync pulses not reaching the oscillator, the picture will move around horizontally and can be brought to hold momentarily with the hold control. If the picture breaks up into strips, there is a problem in the horizontal oscillator. If there is an accessible hold control try rotating it: if the frequency is too far off, the picture will not settle into place at any adjustment of the hold control. Look around the horizontal oscillator circuit: all of the oscillator parts will be right there, or check on the horizontal oscillator module. If only one resolution on a auto-scan monitor is affected, the there could be a separate oscillator circuit for each range.
Multiple images on the screen horizontally or vertically indicate that the scan rate is way off (by a factor equal to the number of complete pictures.) This could be a fault in the monitor or you could be running way outside of the monitor's specifications. Even slightly exceeding these for the horizontal or vertical may confuse the scan rate selection logic and result in the monitor setting itself with incorrect scan rate settings. A situation where successive sweeps alternate position slightly resulting in double or triple images may be caused by a incorrect or out of range video timing, a bad component, or improper sync signals. Check the settings of the video card and any sync termination or selection on the monitor. Beyond this, a schematic will be required.
The following applies if the part of the picture is missing but not otherwise squashed or distorted. For example, 85% is missing but the portion still visible is normal size. Wow! That's an interesting one, more so than the typical run-of-the-mill "my TV just up and died on me". Or, "my pet orangutan just put a hole in the CRT, what should I do"? With a monitor, this is more likely than a TV. But the cause is probably not in the monitor (though not impossible). Check that your video parameters are set up correctly (particularly if you have full control of them as with Linux). You may have set the active too short or blanking too long. If your video is confirmed to be ok (looking at it with an oscilloscope would be best), then with the size of the picture fragment correct but 85% missing, check waveforms going into the vertical output stage. The supply voltage is probably correct since that often determines the size. It almost sounds like the waveform rather than being mostly on (active video) and off for the short blanking period is somehow only on during the last part of the active video thus giving you just the bottom of the picture. If there is a vertical output IC, it may be defective or the blanking input to it may be corrupted. The problem may be as far back as the sync separator. Then again who knows, schematics would be really handy.
These may be sharp-edged or blurry. The latter could result when a portion
of the active video is unblanked during retrace.
* Where the entire picture is present, the problem is one of the video
blanking not occurring properly beyond the picture boundary.
* Where part of the picture is cut off with a bright horizontal or vertical
line at that point, it is either a video timing problem or a fault in the
deflection circuitry preventing the beam from being where it is supposed to
scan in enough time.
You may be seeing part of the active video during retrace or as the beam
reverses direction at the start or end of retrace. Horizontal timing
problems would produce vertical bars on the right or left edge; vertical
timing problems would produce horizontal bars at the top or bottom edge.
* If your video card permits control of video timing parameters, try reducing
the relevant active time relative to the blanking period. The relevant
software settings might be horizontal position, phase, size, and sync
polarity. If this does not work, your video card may be incompatible with
the monitor.
* If the problem just happened without any changes to the video source, the
monitor may have a problem:
- Deflection circuits - coil or capacitor, a power supply fault, position or
size settings or control, or deflection yoke.
- Video amplifier or drive (CRT neck board), or blanking circuits - chip
decoupling capacitors or filter capacitors in scan derived power supplies.
If the bars are significantly colored - not just shades of gray - then
a video problem is likely.
An oscilloscope would help greatly in identifying the source of the problem.
Go to [Next] segment
Go to [Table 'O Contents]