Good article, guys; well-written and explanatory. I've read a lot about how they work but never actually seen a proper disassembly before - very interesting stuff ;)
P.S: It's Pythagoras for the CRT pixel calculations and there's a typo in the Future section: "Arguably, there is not much of a future a all for CRT technology"
Quote:
Originally Posted by Fr4nk Nice article, very very helpful, now take a CRT apart
Not advisable unless you like the feeling of 30,000V across you ;)
Good read, nice to see another dead LCD in the world too :D
I am curious though, when are we likely to see some sort of LED or OLED panel catch up with a CRT in terms of overall image quality? This isn't me being fanboyish(oh ok, maybe a little, but humour me), I'm just curious when I'll actually be able to dump the trinitron in favour of something better. Will it happen with LCD, or, is it likely to happen only once one of the new tech's like SED or OLED gets fully developed as the primary display technology of the world?
Originally Posted by hitman012 Good article, guys; well-written and explanatory. I've read a lot about how they work but never actually seen a proper disassembly before - very interesting stuff ;)
P.S: It's Pythagoras for the CRT pixel calculations and there's a typo in the Future section: "Arguably, there is not much of a future a all for CRT technology"
Not advisable unless you like the feeling of 30,000V across you ;)
pssh, whatever, wussies. i cant count the number of crt monitors ive fixed. never been shocked once.
...i get the feeling im going to die at an early age
Personally, I couldn't notice any colour difference between good 6bit panels (TN panels) and 8bit panels.
What I did notice though, was the massive difference in response time between the monitors with 6bit and 8bit panels, when I was choosing a new monitor in February last year (when the first 8ms monitors came out).
Gaming on a monitor with ghosting was horrible, and I had to ditch the monitor because of it (1st-gen Dell 1905FP - 20ms, 8bit panel). I quickly replaced it with a Benq FP937S+ (8ms, DVI, 6bit TN panel) and I haven't looked back since! :)
Now that pretty much all panels are fast (i.e. under 12ms), ghosting is no longer an issue. :)
However, I'm still confused by the dot pitch conversions. For example, why a DP of 1 converts to 0.87 horizontal dot gap. Similarly, shouldn't the horizontal pixels be calculated by dividing, not multiplying, ie 401mm horizontal viewable screen area /divided/ by 0.228 is 1758?
Also, in the section starting with the heading LCD, the third paragraph introduces TFT with no explanation as to what TFTs are (Thin Film Transisters? But what are they?):
LCD
Flat panel monitors...
LCD panels are fairly simple to understand...
The actual image on a TFT is made up of a matrix of pixels. Unlike with CRTs...
Are LCD and TFT interchangeably useable with reference to displays? How about an article about Plasma Displays?
Originally Posted by qipod However, I'm still confused by the dot pitch conversions. For example, why a DP of 1 converts to 0.87 horizontal dot gap.
In the image you can see that, if you take any three red phosphors, you may form an equilateral triangle with them.
By definition, an equilateral triangle is composed of three 60° angles. If you assume that the dot pitch is 1, as Wil has, then the horizontal component of any of these three angles will be 1(sin 60), which comes in at around 0.867. Now we've got this ratio, you can multiply the actual dot pitch (0.26mm) by 0.87 to get the horizontal dot pitch. He could just as easily have done 0.26(sin 60°).
Quote:
Originally Posted by qipod Similarly, shouldn't the horizontal pixels be calculated by dividing, not multiplying, ie 401mm horizontal viewable screen area /divided/ by 0.228 is 1758?
I think that's a mistake - it'll be corrected soon :)
Quote:
Originally Posted by qipod Also, in the section starting with the heading LCD, the third paragraph introduces TFT with no explanation as to what TFTs are (Thin Film Transisters? But what are they?):
TFTs are a type of active matrix LCD. Unlike the LCD in, say, your calculator or digital watch, it does not have signal wires to apply a voltage across each individual cell (pixel element). That would require many millions of wires, which would be impractical.
Instead of this method, the pixels were arranged in a grid; a cell is activated by applying a voltage to the appropriate row and column. This used to be done by applying a negative voltage to one and a positive voltage to the other, which resulted in the cross-point of the two having a high PD across it, meaning it switched.
However, technology has moved on now to using transistors positioned on glass (hence TFT), which is very complicated but allows for finer image quality and better response time.
Quote:
Originally Posted by qipod Are LCD and TFT interchangeably useable with reference to displays?
Pretty much - you'll find it difficult to locate an LCD monitor that isn't a TFT.
Originally Posted by hitman012 Not advisable unless you like the feeling of 30,000V across you ;)
Normally you would let the monitor sit for a few hours unplugged to let the high voltage side drain. And if you dont let it sit then you only use 1 hand to work on it, that way the 30k dosent go through your heart and kill you. :D
Originally Posted by qipod Similarly, shouldn't the horizontal pixels be calculated by dividing, not multiplying, ie 401mm horizontal viewable screen area /divided/ by 0.228 is 1758?
Great article, Wil - I had forgotten some of that stuff :)
Honestly, the only way to get good image quality is for each pixel to be its own light source; as such, CRTs for the win in that regard, but OLEDs for the best of both worlds. I wonder how well that phosphor CRT style tech works in practice though...
Fantastic read, And i might be willing to take my CRT apart in about a year. That or i go hunting for one out of a skip.
Didnt i see an artical on Bit-Tech about a LCD with lots of light sources? There was a test with a single white rectangle in the middle of the screen and it was part of a HDR Test/ bit-tech explination.
Originally Posted by LAGMonkey Didnt i see an artical on Bit-Tech about a LCD with lots of light sources? There was a test with a single white rectangle in the middle of the screen and it was part of a HDR Test/ bit-tech explination.
I just want smaller dot pitches which will allow for higher resolutions and sharper images, sadly there doesn't seem to be a shift towards this, just making larger pannels. :-(
This little snippet from page 3 seemed unintelligible to me:
Quote:
The two filters are exact opposites of each other. As the light from the light source behind the first filter comes in, the filter effectively whites it out - which means that if it was to pass through the liquid crystals with no interaction, the filter on the other side would polarise it back to black, leaving no colour being emitted. In fact, alternate current - leaving the crystals 'dead in the water' - is how black is created on a panel.
I don't pretend to be an expert on the subject by any means, but I usually see this being explained more along the lines of:
The two polarising filters are placed at right angles to each other. The light passes from the backlight, through the first polarising layer, then the LCD layer, and finally through the second polarising layer. The light is polarised by the first filter, so that only light travelling as a (let's say) vertically "aligned" wave can pass through. When there is no current flowing through the LCD layer, this light passes through it pretty much freely, and is then stopped by the second polarising filter, which only allows horizontally "aligned" lightwaves to pass. However, when a current is applied to the LCD layer, the suspended crystals form a spiral-like structure which "twists" the light, allowing it to pass through both filters. The more current applied, the more light is "twisted", hence making for brighter pixels.
i have a problem with an old 17" crt I like to use...it's an old Dell model, and from time to time the sides will suddenly pin cushion inward in an extreme way...but if i smach the monitor it goes back. i figures the gun was just not aligning properly...but is there some permanent way to fix this? i can never take this monitor to LANs because i would have to smack it constantly until it's been on for about a half hour.
Originally Posted by KypD i have a problem with an old 17" crt I like to use...it's an old Dell model, and from time to time the sides will suddenly pin cushion inward in an extreme way...but if i smach the monitor it goes back. i figures the gun was just not aligning properly...but is there some permanent way to fix this? i can never take this monitor to LANs because i would have to smack it constantly until it's been on for about a half hour.
Sounds like a dry solder joint somewhere in the coil control circuitry. If it is, it's a simple fix... AFTER you've tracked the little bugger down. Basically, I'd recommend taking it to a repair shop, if you can live with the cost. It shouldn't be TOO high, since you most likely don't need any new parts, but most of these guys still charge quite a bit by the hour.
Comments 1 to 23 of 23
-Fr4nk
because if it was a perfectly good 17" monitor it would have been a waste, a broken 15" or a broken 17" would seem much less "wasteful"
anyways, I hope it was a broken 17" LCD
P.S: It's Pythagoras for the CRT pixel calculations and there's a typo in the Future section: "Arguably, there is not much of a future a all for CRT technology"
I am curious though, when are we likely to see some sort of LED or OLED panel catch up with a CRT in terms of overall image quality? This isn't me being fanboyish(oh ok, maybe a little, but humour me), I'm just curious when I'll actually be able to dump the trinitron in favour of something better. Will it happen with LCD, or, is it likely to happen only once one of the new tech's like SED or OLED gets fully developed as the primary display technology of the world?
...i get the feeling im going to die at an early age
Personally, I couldn't notice any colour difference between good 6bit panels (TN panels) and 8bit panels.
What I did notice though, was the massive difference in response time between the monitors with 6bit and 8bit panels, when I was choosing a new monitor in February last year (when the first 8ms monitors came out).
Gaming on a monitor with ghosting was horrible, and I had to ditch the monitor because of it (1st-gen Dell 1905FP - 20ms, 8bit panel). I quickly replaced it with a Benq FP937S+ (8ms, DVI, 6bit TN panel) and I haven't looked back since! :)
Now that pretty much all panels are fast (i.e. under 12ms), ghosting is no longer an issue. :)
Don't worry - it was kindly donated for the purposes of this article by our friends at ViewSonic: http://www.bit-tech.net/hardware/2006/03/20/how_crt_and_lcd_monitors_work/5.html
However, I'm still confused by the dot pitch conversions. For example, why a DP of 1 converts to 0.87 horizontal dot gap. Similarly, shouldn't the horizontal pixels be calculated by dividing, not multiplying, ie 401mm horizontal viewable screen area /divided/ by 0.228 is 1758?
Also, in the section starting with the heading LCD, the third paragraph introduces TFT with no explanation as to what TFTs are (Thin Film Transisters? But what are they?):
LCD
Flat panel monitors...
LCD panels are fairly simple to understand...
The actual image on a TFT is made up of a matrix of pixels. Unlike with CRTs...
Are LCD and TFT interchangeably useable with reference to displays? How about an article about Plasma Displays?
By definition, an equilateral triangle is composed of three 60° angles. If you assume that the dot pitch is 1, as Wil has, then the horizontal component of any of these three angles will be 1(sin 60), which comes in at around 0.867. Now we've got this ratio, you can multiply the actual dot pitch (0.26mm) by 0.87 to get the horizontal dot pitch. He could just as easily have done 0.26(sin 60°).
Instead of this method, the pixels were arranged in a grid; a cell is activated by applying a voltage to the appropriate row and column. This used to be done by applying a negative voltage to one and a positive voltage to the other, which resulted in the cross-point of the two having a high PD across it, meaning it switched.
However, technology has moved on now to using transistors positioned on glass (hence TFT), which is very complicated but allows for finer image quality and better response time.
Hope this clears it up for you :)
Good article, Im still not giving up my CRT yet.
:o
FIXD.
Honestly, the only way to get good image quality is for each pixel to be its own light source; as such, CRTs for the win in that regard, but OLEDs for the best of both worlds. I wonder how well that phosphor CRT style tech works in practice though...
Didnt i see an artical on Bit-Tech about a LCD with lots of light sources? There was a test with a single white rectangle in the middle of the screen and it was part of a HDR Test/ bit-tech explination.
well not any time soon any way.
The two polarising filters are placed at right angles to each other. The light passes from the backlight, through the first polarising layer, then the LCD layer, and finally through the second polarising layer. The light is polarised by the first filter, so that only light travelling as a (let's say) vertically "aligned" wave can pass through. When there is no current flowing through the LCD layer, this light passes through it pretty much freely, and is then stopped by the second polarising filter, which only allows horizontally "aligned" lightwaves to pass. However, when a current is applied to the LCD layer, the suspended crystals form a spiral-like structure which "twists" the light, allowing it to pass through both filters. The more current applied, the more light is "twisted", hence making for brighter pixels.
Nezuji :)
Nezuji :)