Lab V2.0 is underway. Getting all fancy and modern now with the addition of a 22" flat screen connected to my TDS544A digical Tek scope !
When I am old at least I will be able to see what's on the scope, eh ?!
The other day a friend gave me not one but two old 22" monitors of his, that decided to pack up recently in unison. Both of them are from 2007 so almost 15 years of age.
He bought brand new monitors to replace them, so he gave me the old ones. I could potentially make use of them, given that the last of my 3 beloved 17" CRT Samsung monitors died very recently, plus I intend to add a dedicated computer to lab V2.0 in the making. So I was kinda motivated to try to fix them. Both of them have a black screen, and no backlight, and "some" activity from the power button LED. I mean I can't make sense of it but at least I know the CPU is alive enough to play with the LED, so that's good I guess...
Given their age, both have fluorescent tubes (4 of them) as a backlight, no LED yet.
I started with the Samsung. Throwing a torch at the screen, I could see there was a picture there, CPU was displaying a moving " No Signal check cable " message. So no backlight it is.
Checked power board, low voltage going to the CPU board was there (13.5V), but not a sausage coming out of the inverter feeding the backlight. So at least I knew the primary side of the PSU was working, and problem was limited to the HV inverter. It runs from the 13.5V. Then 2 transformers each with two outputs, power the 4 fluo tubes. The primary winding of the two transformers are connected in //, and are switched between ground an 13.5V using a couple dual complementary MOSFET 8 pin chips. Zero sign of activity or voltage anywhere in the circuit... was at a loss. So tried my chance : replaced the two electrolytic caps that looked like they were at the heart of the oscillator. They looked fine visually, but well no schematic and circuit way too complex to improvise, so replacing these two guys was about the only thing I could do, short of spending my life troubleshooting it. Checked them with my new Chinese gizmo, that dirt cheap AVR based component tester everyone gets. Glad I bought it, now I can check ESR on old caps. Was not that bad actually, single digit ohms, a bit high of course but not delirious. Caps are identical, small ones, 820uF at 25V. However capacitance was way down at 150uF or so IIRC. So clearly not healthy, so I had "some" hope, it might make at least some difference. Searched my box of 40+ year old salvaged caps, found a couple 1000uF... about OK. ESR checked great and capacitance was spot on 1000uF, so I had some degree of confidence they would be better than the old ones I just pulled... And indeed it did the job, backlight/monitor now works !
Well the backlight flickers badly for a minute or two following power up, then stabilizes / works fine. So either the fluo tubes are tired (?) and/or it's my old caps that are "reforming" at power up. 30 years+ sitting in a box, I could understand that... might throw brand new caps in there to see if it fixes it.
The only video source I had in the lab was my Tek scope, so well I gave that a try ! Works fine... picture looks blurry and washed out, but in the flesh not so at all, it's just fine. My old camera might need replacing, it's getting worse and worse. Only problem is that being an LCD screen ie crap, any resolution that's not native, poses some sort of problem. So feeding VGA / 640x480 into a 1680x1050 screen just can't go smoothly. Both screens have the same problem : up close, up close only, doesn't show in the pics of course, you can see subtle vertical artifacts. The background is not perfectly / solid black. Instead you get a pattern, a succession of thin (5mm or so wide) vertical stripes, alternating between black and very dark purple/violet or dark grey depending on the screen.
So that was one of the two 22" monitors.
The other one same age same fluo backlight, same symptom (no picture), a Chinese brand I can't even pronounce "Chimei" or something. Looks less crappy, material and build quality wise, and my friend said it had a better picture than the bottom of the barrel Samsung.
That one I could not see a picture with the torch light sadly... and the power button LED was also showing some life... so that was looking bad.. pointing to a defective digital board not sending data to the screen. Hopes of fixing a digital board is next to nil, so I was feeling down.
Checked power board, a couple low voltage rails. They looked OK... didn't know what to do. However the 4 electrolytics on the LV secondary side were visually kaput, swelled. So I thought OK let's at least try to replace those, see if somehow it makes a difference. It did, screen works now !
But the voltages were fine so I don't know... maybe them being shot, made their performance at power up give bad transients. The CPU board maybe required a very accurate power up sequence and rise time, and the dodgy caps would not let that happen anymore ? I don't know...
Anyway, fixed two 22" monitors in record time for zero money, right when my last CRT monitor died, so pretty pleased, I am covered now : I am using the Chimei right now on my computer to write this, and the crappy Samsung is still on the bench displaying the output of the Tek scope, and will later use it probably to hook to a future second PC I will set up just for the lab / bench. So less money to spend !
So that was my X-mas gift(s) I guess...
Home
Help
Search
Login
Register
