Old PCs are nice, but guess what makes them even better.
Monitors.
Old monitors. Yeah, we’re going back in time again, this time to bring back the glory of a IBM P275, also known as 6652-T3N. This is my second P275, but it’s the least used one. My second, black one has ~9000 hours clocked in, but this one has only 4500! This display is almost new!
Now, for those that don’t know, the CRT, the display itself is a consumable item, with each second it’s emitting electrons at the phosphor it’s very slowly getting dimmer and dimmer. In general, an average tube lasts about 30000 hours, but I have a Samsung 959NF that has over that amount that isn’t dimm at all, especially after calibrating it, however, increasing the cathode drives made it a bit bloomy and soft-looking. It’s still perfectly usable, but it’s power-on time has taken a toll on the picture quality.
As You can see there are several reasons why we moved on from CRTs.
Back to the P275, I picked it up with a broken stand and a messed up color calibration.
Of course it lived in a shed for past how many years, so it’s very dirty.
Broken stand was very easy to fix, just glue it back together with a 2-part glue. No close-ups for this part 🙂
For calibration and more thorough cleaning I had to get inside. A quick warning before cracking the shell: inside, below the anode cap there might be upwards of 24kV, not leathal for most people, but touching that won’t be pleasant. Although this era CRTs pretty much always have a discharging resistor, there is no telling whether it hasn’t failed, so caution is advised.
Before proceeding, I have to say that I absolutely love the design of these older electronics. It was supposed to be slick and ‘modern’, but not infringing on repairablity or usability.
As CRT age they attract soot inside. On this particular one there hardly is any dust or soot build up.
To an average person this might look bad, but in fact it isn’t that dirty, on high hour count monitors soot creates these ‘hair’ around the cable and anode cap, here there is almost only dust.
As CRT ages, the heat from the filaments makes the wires go from metallish-bluish color to almost yellow. Again, the picture tube itself is in like-new condition. I’m very excited how it’s going to turn out in the end.
This monitor uses Sony’s Trinitron tube. It’s basically the best tech of these years, if you wanted a good display, you went for a Trinitron and you were happy. You also could have went for a Diamondtron, which was a Mitsubishis take on an aperture grille, which to me is indistinguishable from an ‘original’ Sony part.
After cleaning, I left some dust on the yoke, as the wires are very thin, and I was worried that they might snap. CRTs are an immensly complicated and precise piece of electronics, so there is a significant chance that such snapped wire could EOL this monitor.
Now, for something that I don’t like about it, for color and geometry calibration you must take off the shell and connect the UART bridge. There are no options in OSD, like in a Mitsubishi, nor a parallel port cable that uses the I2C (it connects to pin 12 and 15 on VGA connector) as it’s in LG and Samsung displays.
Calibration
For calibrating Sony and compatible displays I use “WinDAS”. It’s compatible with Windows 10, so at least juggling PCs isn’t a problem, but getting an UART bridge to work with it is a nightmare. Some work and some just don’t. After trying a few, including those based on FT232 and CH340 I finally found that the one based on CP2102 works the best. It still sometimes cuts out for no apparent reason, but experiences of people on the internet are similar to mine.
As for the calibration in itself I used the guide made by ‘spacediver‘. It has worked well for me in the past and this time was no different. Links obviously will be at the bottom.
The calibrator I use is a Colormunki Display, which essentially is a slower i1Display Pro. Glass filters, very good repeatability of measurements, for my needs absoulutely enough.
Before calibration the image is very red, and the G2 voltage is very high meaning that blacks are grayish.
Midway calibration, after the first round whites are no longer red tinted, but G2 is still too high.
End result is just stunning. Blacks are black, whites are white and grays are gray. Picture really doesn’t do this monitor justice.
Delta E <2 is considered to be near perfect. This monitor almost managed below 1. I’m sure that if I messed around with G2 more I could have gotten below 0.8 E, but that might have compromised the reliability of guns, so I decided to stop here. It’s good enough for me.
Thanks for reading!
Links
WinDAS: https://crtdatabase.com/faq/windas-installation-and-guide
Calibration guide: https://hardforum.com/threads/windas-white-point-balance-guide-for-sony-trinitron-crts.1830788/