Author: Samuel M. Goldwasser
E-Mail: sam@stdavids.picker.com
Corrections/suggestions: [Feedback Form]
[mailto]
Copyright (c) 1994, 1995
All Rights Reserved
Reproduction of this document in whole or in part is permitted if both of the following conditions are satisfied:
You may also wish to see a typical SMPS found in a Panasonic VCR for a
general "How they do it". You can find it at URL:
http://www.paranoia.com/~filipg/HTML/FAQ/BODY/vcrps.ps (80KB).
It's a
Postscript file so deal with it accordingly. You may wish to read the
Postscript FAQ **FIRST** if you have no clue on how to proceed. It's at URL:
http://www.cis.ohio-state.edu/hypertext/faq/usenet/postscript-faq/top.html (SUBMENU).
After you read the FAQ, and your questions are still unanswered, you may choose to
ask your question on the group: comp.lang.postscript.
Note that the danger to you is not only in your body providing a conducting path, particularly through your heart. Any involuntary muscle contractions caused by a shock, while perhaps harmless in themselves, may cause collateral damage - there are many sharp edges inside this type of equipment as well as other electrically live parts you may contact accidentally.
The purpose of this set of guidelines is not to frighten you but rather to make you aware of the appropriate precautions. Repair of TVs, monitors, microwave ovens, and other consumer and industrial equipment can be both rewarding and economical. Just be sure that it is also safe!
Bad solder joints are very common in monitors due both to poor quality manufacturing as well as to deterioration of the solder bond after numerous thermal cycles and components running at high temperature. Without knowing anything about the circuitry, it is usually possible to cure these problems by locating all bad solder connections and cleaning and re-seating internal connectors. The term 'cold solder joint' strictly refers to a solder connection that was either not heated enough during manufacturing, was cooled too quickly, or where part pins were moved before the solder had a chance to solidify. A similar situation can develop over time with thermal cycling where parts are not properly fastened and are essentially being held in by the solder alone. Both situations are most common with the pins of large components like transformers, power transistors and power resistors, and large connectors. The pins of the components have a large thermal mass and may not get hot enough during manufacturing. Also, they are relatively massive and may flex the connection due to vibration or thermal expansion and contraction.
These problems are particularly common with TVs and monitors - especially cheaper monitors.
To locate cold solder joints, use a strong light and magnifier and examine the pins of large components for hairline cracks in the solder around the pin. Gently wiggle the component if possible (with the power off). Any detectable movement at the joint indicates a problem. With the power on, gently prod the circuit board and suspect components with an insulated tool to see if the problem can be effected.
When in doubt, resolder any suspicious connections. Some monitors may use double sided circuit boards which do not have plated through holes. In these cases, solder both top and bottom to be sure that the connections are solid. Use a large enough soldering iron to assure that your solder connection is solid. Put a bit of new solder with flux on every connection you touch up even if there was plenty of solder there before. However, remove any obvious excess. Inspect for solder bridges, sliver, splashes, etc. before applying power.
Alternatively, the male pins of the cable may not be making good contact with the female VGA socket. First try contact cleaner. If this does not work, gently squishing the male pins with a pair of needle-nose pliers may provide temporary or permanent relief if the pins are a tad too small. However, if you go too far, you can damage or break the pins or cause the female sockets to become enlarged and loose fitting for any other monitor you may use.
There may be cold solder joints on the VGA board itself at the VGA connector. These can be resoldered.
The video signals for red, green, and blue (or just a single signal for monochrome) are sent over cables which are generally 75 ohm transmission lines. These are coaxial cables that may be combined inside a single sheath for VGA, SVGA, MACs, and many workstations but may be separate coaxes with BNC (or other) connectors for other video applications.
Without going into transmission line theory, suffice it to say that to obtain good quality video, the following conditions must be met:
Failure to follow these rules will result in video ringing, ghosts, shadows, and other unsightly blemishes in the picture. It is often not possible to control all aspects of the video setup. The cable is often a part of the monitor and cannot easily be substituted for a better one. The monitor may not have properly designed circuitry such that it degrades the video regardless of the cable and display board quality. The display card itself may not have proper drivers or source termination.
Ironically, the better the video card, the more likely that there will be visible problems due to termination. This is due to the very high bandwidth and associated signal edge rates.
Some examples of common termination problems:
In most cases, the cover will need to be removed. The controls we are looking for may be located in various places. Rarely, there will be access holes on the back or side.
The controls may be located on the:
Set the user brightness control to its midpoint and the user contrast control as low as it will go - counterclockwise.
Let the monitor warm up for at least 15 minutes so that components can stabilize.
If there is a master brightness or background level control, use this to make the black areas of the picture just barely disappear. Them, increase it until the raster lines just appear. (They should be a neutral gray. If there is a color tint, then the individual color background controls will need to be adjusted to obtain a neutral gray.) If there is no such control, use the master screen control on the flyback. If it is unmarked, then try both of the controls on the flyback - one will be the screen control and the other will be focus - the effects will be obvious. If you did touch focus, set it for best overall focus and then get back to the section on focus once you are done here.
If there are individual controls for each color, you may use these but be careful as you will be effecting the color balance. Adjust so that the raster lines in a black area are just visible and dark neutral gray.
Now for the gain controls. On the little board on the neck of the CRT or on the video or main board there will be controls for R, G, and B gain or contrast (they are the same). If there are only two then the third color is fixed and if the color balance in the highlights of the picture was ok, then there is nothing more you can do here.
Set the user contrast control as high as it will go - clockwise.
Now adjust each internal gain/contrast control as high as you can without the that particular color 'blooming' at very bright vertical edges. Blooming means that the focus deteriorates for that color and you get a big blotch of color trailing off to the right of the edge. You may need to go back and forth among the 3 controls since the color that blooms first will limit the amount that you can increase the contrast settings. Set them so that you get the brightest neutral whites possible without any single color blooming.
Now check out the range of the user controls and adjust the appropriate internal controls where necessary. You may need to touch up the background levels or other settings. Check at the other resolutions and refresh rates that you normally use.
If none of this provides acceptable brightness, then either your CRT is in its twilight years or there is something actually broken in the monitor. If the decrease in brightness has been a gradual process over the course of years, then it is most likely the CRT. As a last resort (untested) you can try increasing the filament current to the CRT the way CRT boosters that used to be sold for TVs worked. Voltage for the CRT filament is usually obtained from a couple of turns on the flyback transformer. Adding an extra turn will increase the voltage and thus the current making the filament run hotter. This will also shorten the CRT life - perhaps rather drastically. However, if the monitor was headed for the dumpster anyhow, you have nothing to lose.
Don't expect to have perfect focus everywhere on the screen. Usually there will be some degradation in the corners. A compromise can generally be struck between perfect focus in the center and acceptable focus in the corners.
If the adjustments have no effect, then there is probably a fault in the focus power supply.
For most color TVs and monitors, the correct focus voltage will be in the 4-8 KDC range so you will need a meter that can go that high or some big resistors to extend its range or a HV probe. You must use a high impedance meter as the current availability from the focus power supply is very low.
The pots in the flyback are sometimes accessible by removing their cover, which may snap on. However, a typical focus circuit will have a large value resistor potted inside the flyback (like 200 Megohms).
Try to measure the focus in-circuit. If the value you read is very low (assuming your meter has a high enough impedance not to load the circuit appreciably), then disconnect the wire (from the PCB on the neck of the CRT or wherever) and measure again and observe any change in picture.
If still low, then almost certainly there is a problem with the pot or the flyback. See if you can open it enough to measure and/or disconnect the pot. If the problem is inside the potted part of the flyback, the only alternative is a new flyback or an external divider if you are so inclined. However, once the focus network goes bad inside the flyback, there is an increased chance other parts will fail at some point in the future.
If the voltages check out with the CRT disconnected, there is a chance of a bad CRT or of a shorted component on the PCB on the neck of the CRT. Look for shorted capacitors or burnt or damaged traces.
After unplugging the monitor and waiting a few minutes for the filter capacitors to discharge (check with a voltmeter but stay away from the CRT HV connector as it may retain a dangerous and painful charge for a long time), use an ohmmeter across the various diodes in the power supply. These appear commonly as black cylinders about 3/8" long by 1/4 diameter. (Kind of like 1N400Xs on steroids). The resistance of the diodes in at least one direction should be greater than 100 ohms. If it is much less (like 0 or 5 ohms), then the diode is probably bad. Unsolder and check again - it should test infinite (greater than 1M ohms) in one direction. If it now tests good, there may be something else that is shorted.
Replacements are available for about $.25 from places like MCM Electronics.
Check other power semiconductors as well, in particular, the horizontal output transistor.
Other possible causes: bad solder connections, other shorted components like capacitors, other problems in the power supply. Sometimes this is an indication of an over-voltage shutdown due to a faulty regulator or open load.
White acrid smoke may indicate a failed electrolytic capacitor in the power supply probably in conjunction with a shorted rectifier. Needless to say, pull the plug at once.
A visual inspection should be able to easily confirm the bad capacitor as it will probably be bulging and have condensed residue nearby. Check the rectifier diodes or bridge rectifier with an ohmmeter. Resistance across any pair of leads should be more than a few ohms in at least one direction. Remove from the circuit to confirm. Both the faulty diode(s) and capacitor should be replaced (though the capacitor may work well enough to test with new diode(s).
If a visual inspection fails to identify the smoking part, you can probably plug the monitor in for a few seconds until the source of the smoke is obvious but be prepared to pull the plug in a real hurry.
That is the same diagnosis a friend of mine got for her monitor with that identical problem. Replacing the capacitor did fix the problem.
That 'capacitor' is a big red thing is a Sony part which includes some kind of low voltage connection as well. The guy at the place where she got it repaired said that the capacitor is one of the most common problems with those monitors. $70 for the part + $50 for labor, ouch! Only, apparently available from Sony. Why can't Sony design monitors like everyone else? Sure, I know, theirs are better (well, except for the unsightly stabilizing wires on Trinitrons!).
The shadow mask or aperture grill of the CRT - the fine mesh just behind the phosphor screen - is normally made of a material (steel or InVar) which is easily magnetized. This can happen just by rotating the monitor on its swivel, by moving it from one place to another, by switching on or off some piece of electronic equipment near the monitor, even by a local lightning strike.
Since any stray magnetism effects the color purity and convergence, it is important that the CRT is demagnetized before use.
Next time you scrap a computer monitor (or a color tv), save the degaussing coil (coil of wire, usually wrapped in black tap or plastic) mounted around the front of the tube. To adapt it for degaussing sets, wrap it into a smaller coil, maybe 4"-6". To limit the current to something reasonable, put it in series with a light bulb (60-100W). You need AC current to degauss, so just put the bulb in series with the coil and use the your local 120V outlet. BE VERY CAREFUL that you actually wired it in series, and that everything is properly insulated before you plug it in (A fuse would be a real good idea too!!)
A few circles over the affected area will usually do it. Note that it will also make your screen go crazy for a little bit, but this will fade out within a minute or so.
Disclaimer: This has worked consistently for me in the past, but I make absolutely no warranty as to the safety or effectiveness of this procedure in your situation. But if you try it and like it, feel free to let me know!
Just a couple of points for emphasis:
If one or two activations of the degauss button do not clear up the color problems, manual degaussing using an external coil may be needed or the monitor may need internal purity/color adjustments. Or, you may have just installed your megawatt stereo speakers next to the monitor!
You should only need to degauss if you see color purity problems on your CRT. Otherwise it is unnecessary. The reasons it only works the first time is that the degauss timing is controlled by a thermsistor which heats up and cuts off the current. If you push the button twice in a row, that thermister is still hot and so little happens.
One word of clarification: In order for the degauss operation to be effective, the AC current in the coil must approach zero before the circuit cuts out. The circuit to accomplish this often involves a thermister to gradually decrease the current (over a matter of several seconds), and in better monitors, a relay to totally cut off the current after a certain delay. If the current was turned off suddenly, you would likely be left with a more magnetized CRT. There are time delay elements involved which prevent multiple degauss operations in succession. Whether this is by design or accident, it does prevent the degauss coil - which is usually grossly undersized for continuous operation - to cool.
Assuming that the focus has just been gradually getting worse over time, tweaking the internal focus control may be all that is needed.
On most monitors, the flyback transformer includes two control - FOCUS and SCREEN. The one you want is, of course, FOCUS.
Safety: As long as you do not go near anything else inside the monitor while it is on AND keep one hand in you pocket, you should be able to do this without a shocking experience.
Plug it in, turn it on and let it warm up for a half hour or so. Set your PC to display in the resolution you use most often. First turn the user brightness and contrast fully counterclockwise. Turn brightness up until the raster lines in a totally black area appear, then back a hair until they disappear. Then, turn the contrast control up until you get a fairly bright picture. Fully clockwise is probably ok. Adjust FOCUS for generally best focus. You will not be able to get it razor sharp all over the screen - start at the center and then try to get the edges and corners as good as you can without messing up the center too much. Double check that the focus is OK at your normal settings of brightness and contrast.
The SCREEN control adjusts background brightness. If the two controls are not marked, you will not do any damage by turning the wrong one - it will be immediately obvious as the brightness will change rather than focus and you can then return it to its original position (or refer to the section on brightness adjustments to optimize its setting).
Other than recommending moving the monitors, there is no easy solution. They can be shielded with Mu Metal but that is expensive. Or you could run all displays at a 60 Hz vertical rate (or 50 Hz depending on where you live). However, this is inconvenient and will never be quite perfect.
The easiest way to confirm that interference is your problem is to move the monitor or suspect equipment to a different location. The only real solution is to separate the monitor and interfering device.
One cause of these lines is Moire (interference patterns) between the raster and the dot structure of the CRT. Ironically, the better the focus on the tube, the worse this is likely to be. Trinitrons, which do not have a vertical dot structure should be immune to interference of this sort from the raster lines (but not from the horizontal pixel structure).
You can test for Moire by slowly adjusting the vertical size. If it is Moire, you should see the pattern change in location and spatial frequency as slight changes are made to size. Changes to vertical position will move the patterns without altering their structure - but they will not remain locked to the moving image.
The patterns will remain essentially fixed in position on the face of the CRT for horizontal size and position adjustments - the patterns will remain fixed under the changing image.
How to eliminate it? If Moire is your problem, then there may be no easy answer. For a given resolution and size, it will either be a problem or not. You can try changing size and resolution - Moire is a function of geometry. Ironically, I have a monitor which is nicer in this respect at 1024x768 interlaced than at 800x600 non-interlaced.
Another cause of similar problems is bad video cable termination creating reflections and ghosting which under certain conditions can be so severe as to mimic Moire effects. This is unlikely to occur in all colors with a VGA display since the termination is internal to the monitor.
A good quality multisync monitor should not mind switching screen resolutions frequently (though doing it every few seconds continuously may stretch this a bit).
Newer multisync monitors should also be smart enough not to blow up if you feed then a scan rate which exceeds their capabilities. However, there are a lot of poorly designed monitors out there.
If it is supposed to run SVGA, use it at SVGA. If it blows up, switch to a different brand. There are a lot of crappy monitors being sold on their own and bundled with PCs.
The only "tuneup" that a monitor should need, exclusive of adjustments needed following replacement of a failed component, would be video amplifier and/or CRT biasing adjustments to compensate for the aging of the tube. These are usually done only if you're using the thing in an application where exact color/brightness matching is important. Regular degaussing of the unit may be needed, of course, but I'm not considering that a "tuneup" or adjustment. (Bob Myers | myers@fc.hp.com)
Some of the newest ('green') monitors have energy conserving capabilities. However, it is necessary for the software to trigger these power reduction or power down modes. Few monitors in actual use and fewer workstations or PCs are set up to support these features. If you have such a monitor and computer to support it, by all means set up the necessary power off/power down timers. However, using the power saving modes of a 'green' PC with an older monitor can potentially cause damage since some of the modes disable the sync signals. A 'green' monitor which can detect a blank screen and and use this as a trigger can easily be used with a screen saver which can be set to display a blank screen - on any PC or workstation.
Please make it a habit to turn your monitors off at night. This will extend the life of the monitor (and your investment) and is good for the environment as well. For workstations, there are good reasons to leave the system unit on all the time. However, the monitor should be turned off using its power switch. For PCs, my recommendation is that the entire unit be turned off at night since the boot process is very quick and PCs are generally not required to be accessible over a network 24 hours a day.
How much is 'ever so slightly'? There are a fair number of components whose values could alter the position/size of a monitor image. I do not find it at all surprising that there should be a small shift due to heat. It really depends on many factors including the basic design, quality of components, ventilation/cooling, etc. Of course, it is possible to have a monitor that has a component that is worse with respect to temperature. Could also be related to line voltage depending on the regulation of your monitor's power supplies.
In general, my feeling is that if it is not objectionable (a 1/2" shift would be objectionable) AND it's severity is not changing with time, you can ignore it.
Many monitors do this. TVs do this but you are not aware of it since they are already 5-10% overscanned for just this reason, as well as compensating for component aging and line voltage fluctuations.
Probably the most common is deflection derived - the voltages are obtained from auxiliary windings on the horizontal flyback (HV) transformer. A number of components can be at fault and produce a 'dead' monitor. The switching supply could also be independent of the HV supply, but this is less common, especially on inexpensive monitors.
Some possibilities:
Horizontal output transistor (usually a TO3 metal or TOP3 plastic case shorts out. This will usually blow a fuse or fusible resistor as well.
Horizontal drive chain - horizontal oscillator, driver, or driver transformer. Newer monitors may use an IC for the oscillator and this can fail.
Startup - There may be some kind of startup circuit which gets the whole thing going until the auxiliary voltages are available. This could be as simple as a multivibrator or transistor regulator to provide initial voltage to the horizontal oscillator chip or circuit.
Output rectifier diodes can fail shorted and load down the outputs to the point of shutting down.
Some load could be shorted or a capacitor could be shorted leading to overload and shutdown.
Flyback transformer can have shorted windings which load down the output. These (primary shorts in particular) may cause the horizontal output transistor to fail as well. Common problem with older MacIntosh computers and video terminals.
Also, look for cold solder joints - monitors tend to have these as a result of temperature cycling and bad manufacturing. (Is this sounding repetitive yet?)
Sometimes there is a series regulator after the filter cap and this could be bad as well.
I guess I would attempt to trace the circuit from the main filter cap assuming that has the proper (approx. 150-160 VDC usually) voltage.
If there is no voltage at the main capacitor, then there is probably a blown fuse or bad connection somewhere. However, the fuse may have blown due to a fault in the switcher.
If you can locate the horizontal output transistor, see if there is voltage on its collector, should be the same. If there is, then there is probably a drive problem. If you have an ECG or similar semi cross reference, that will help you identify the ICs and transistors and locate the relevant portions of the circuitry.
Check for bad solder connections between the main board and the deflection yoke. Could also be a bad horizontal coil in the yoke, linearity coil, etc. There is not that much to go bad based on your symptoms assuming the high voltage and the horizontal deflection use the same flyback. It is almost certainly not an IC or transistor that is bad.
This could be due to:
This sounds like the filament to the CRT is losing power. Since the picture is otherwise unchanged, look really carefully for cold solder joints particularly on the little board on the back of the CRT if there is one and/or on any connectors associated with the CRT socket. There isn't much to the filament power supply - it is usually a couple of turns of wire on the flyback. If all colors are more or less equally effected, the it is not the CRT itself as the 3 filaments are welded in parallel to the socket pins.
"HI-RES Computer Display Systems"
part # ST1496-1093LE/KGPGC
I am sure this can be ordered from Philips Service Co. P.O. Box 555, Jefferson City, TN 37760 phone 615-475-0044
This book does an excellent job of explaining how these monitors work. Most is about Philips monitors but the material is applicable to most manufacturers. This course and reading this text has help me a lot with my monitor repair efforts.
Skip, e-mail skipperm@mtc2.mid.tec.sc.us
(4) (and sometimes (3) as well) will drastically lower the Q and increase the load the flyback puts on its driving source with no outputs connected. A Q meter or transformer tester can be used, but since many people do not have such equipment, here is a circuit that should work. The only caution is that this tester probably does not put enough stress on the flyback to find an intermittent that fails only under operating conditions.
It is just a 12 V chopper feeding the salvaged core from a flyback. The secondary of this core is a 10 turn coil. You will need to remove the suspect flyback from the TV or monitor. A second 10 turn coil is wound on the suspect flyback anywhere it will fit. Connect one end of each coil securely. Provide an easy way of connecting the other ends momentarily - a push button comes in handy.
If the flyback is good, then with the coils connected there will be several KV at its output - enough to create a small arc (1/8" typical). The load imposed on the oscillator will be modest (the frequency increases in response to load). If there are any shorted windings, then there will be no significant HV output and the load on the oscillator will increase dramatically.
The circuit is shown below. None of the component values are critical.
+12 Q1 +-----------------C || o | C || | B |/ C C || <-- Flyback Under Test ---> | +------| 2N3055 C || | | |\ E 5T C || C-------|>|----------o +HV | | | C || C HV Diode, usually | | -_- C || C built in | | C || _____ C +--|--------------------------C ||C C||C | | Q2 _-_ C ||C 10T C||C | | | C ||C ea. C||C | | B |/ E 5T C ||C C||C | | ----| 2N3055 C ||C__/ _C||C | | | |\ C C || C | | | | C || C | | | +-----------------C || C | | | || C | | ------------------------C || C-------------------o -HV | | 2T C || | | +----------C || (Numerous other windings not shown) | | | 2T C || | +--------------------------C || | | | R1 | R2 +----------/\/\/\--+--/\/\/\---+ 110 27 _|_ 2W 5W -
An easy and flexible solution for both these needs is to construct an electrical box with a duplex outlet where the individual outlets have been separated and wired in series - line-1->hot 1, neutral-1->hot-2, neutral-2->line 2. Clearly mark on the box as to how it is wired.
Now, you can plug your monitor into one outlet and a suitable lamp into the other. For a TV or monitor, use a 100-150 W bulb. For testing a VCR power supply, use a 24-40 W bulb. You could use a lamp with a 2-way switch as well. For your degaussing coil, as large a load as your wire can stand can be plugged into the other outlet. This is safer and easier than using alligator clip leads.