120v LED bulbs -- innards

The persistence is not much more than about 100 lines of the field time, thus, about 5-10 ms. Try moving your finger rapidly across the CRT screen and observe the strobo effect. With LCD monitors, no such strobo effect will be visible.

The first TV that I built, (1949), used a surplus radar CRT. I remember watching the rodeo from Madison Square Garden. The cowboys stayed on the bulls for a long time!

I had heard that by leaving the CRT in the sun for awhile, the long persistence part of the phosphor would be de-activated. It didn't work!

The HV supply to the CRT was from a 60 cycle HV transformer. I was lucky not to have electrocuted myself.

Its real easy to stay on a bull.

Then some dammed fool opened the gate.

-- Many thanks,

Don Lancaster voice phone: (928)428-4073 Synergetics 3860 West First Street Box 809 Thatcher, AZ 85552 rss:

Please visit my GURU's LAIR web site at

You left out that they also pulled on the rope around the Bull's penis.

This worked well as the interlaced lines greatly reduced the apparent flicker.

If you ever displayed a pattern of all odd (or even) lines on an old computer, or tried to use a progressive scan monitor at below 60 Hz, the difference was VERY apparent.

Jon

Rather than use a switching supply to feed a series string of LED's, I wonder about an approach that just turns on ONE at a time at a very rapid rate? Like scanning an LED 7 segment display? The old fool the eye trick.

That way you really only need a power supply that will power ONE led.

Certainly not nearly as many lumens but if you just wanted a "broader light" it would work wouldn't it?

"mkr5000" wrote in message=20 news:22745262.737.1324589873362.JavaMail.geo-discussion-forums@yqjk1...

The problem is still the fact of having a high voltage (180 to 360V = peak)=20 which needs to be dropped to the forward voltage of the LED (3-4 volts = for=20 white). 30 to 40 LEDs in series comes close. And if you wanted to = multiplex=20 that many LEDs you would need essentially 30 times the normal current = for=20 equivalent power and brightness, and that is probably above the peak=20 allowable. The multiplex circuit would also add a lot of complexity.

I came up with a circuit that seems to be capable of 81 to 94% = efficiency at=20

more constant 2.2 to 3.3 watts over the same range at 70 to 95% = efficiency.=20 It uses a small inductor (10 mH) and a high voltage capacitor (10 uF at =

Paul

=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D LED_Lamp_120V_Inductor_10m.asc = =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D

Version 4 SHEET 1 1752 680 WIRE 432 -80 256 -80 WIRE 544 -80 432 -80 WIRE 480 16 96 16 WIRE 544 16 544 -80 WIRE 432 32 432 -80 WIRE -64 64 -112 64 WIRE -32 64 -64 64 WIRE 64 64 32 64 WIRE 96 64 96 16 WIRE 96 64 64 64 WIRE 256 64 256 -80 WIRE 64 96 64 64 WIRE 112 96 64 96 WIRE 112 112 112 96 WIRE 432 112 432 96 WIRE -32 144 -48 144 WIRE 64 144 64 96 WIRE 64 144 32 144 WIRE 160 144 144 144 WIRE -112 160 -112 64 WIRE 256 176 256 128 WIRE 256 176 208 176 WIRE 208 192 208 176 WIRE 432 208 432 176 WIRE 48 224 32 224 WIRE 112 224 112 192 WIRE 144 224 144 144 WIRE 144 224 112 224 WIRE 544 224 544 16 WIRE -64 256 -64 64 WIRE -32 256 -64 256 WIRE 32 256 32 224 WIRE 144 272 144 224 WIRE 160 272 144 272 WIRE 32 304 32 256 WIRE 48 304 32 304 WIRE 208 304 208 288 WIRE 208 304 128 304 WIRE 224 304 224 144 WIRE 224 304 208 304 WIRE -112 352 -112 240 WIRE -48 352 -48 144 WIRE -48 352 -112 352 WIRE -32 352 -48 352 WIRE 32 352 32 304 WIRE 144 352 144 272 WIRE 256 352 256 176 WIRE 288 352 256 352 WIRE 432 352 432 272 WIRE 432 352 368 352 WIRE 32 400 32 352 WIRE 64 400 64 352 WIRE 64 400 32 400 WIRE 208 400 64 400 WIRE 544 400 544 288 WIRE 544 400 208 400 WIRE 208 432 208 400 FLAG 208 432 0 SYMBOL LED 416 208 R0 WINDOW 3 -48 67 Left 2 SYMATTR InstName D1 SYMATTR Value NSSWS108T SYMATTR Description Diode SYMATTR Type diode SYMBOL voltage -112 144 R0 WINDOW 3 -53 -83 VRight 2 WINDOW 123 0 0 Left 2 WINDOW 39 0 0 Left 2 SYMATTR Value SINE(0 200 60 50m 0 0 60) SYMATTR InstName V1 SYMBOL nmos 160 192 R0 SYMATTR InstName M1 SYMATTR Value STP8NM60 SYMBOL res 144 288 R90 WINDOW 0 0 56 VBottom 2 WINDOW 3 32 56 VTop 2 SYMATTR InstName R1 SYMATTR Value 20 SYMBOL zener 48 240 R270 WINDOW 0 36 32 VTop 2 WINDOW 3 -4 32 VBottom 2 SYMATTR InstName D2 SYMATTR Value BZX84C8V2L SYMATTR Description Diode SYMATTR Type diode SYMBOL diode -32 80 R270 WINDOW 0 32 32 VTop 2 WINDOW 3 0 32 VBottom 2 SYMATTR InstName D3 SYMATTR Value UPSC600 SYMBOL diode -32 160 R270 WINDOW 0 32 32 VTop 2 WINDOW 3 0 32 VBottom 2 SYMATTR InstName D4 SYMATTR Value UPSC600 SYMBOL diode 32 272 M270 WINDOW 0 32 32 VTop 2 WINDOW 3 0 32 VBottom 2 SYMATTR InstName D5 SYMATTR Value UPSC600 SYMBOL diode 32 368 M270 WINDOW 0 32 32 VTop 2 WINDOW 3 0 32 VBottom 2 SYMATTR InstName D6 SYMATTR Value UPSC600 SYMBOL res 128 208 R180 WINDOW 0 36 76 Left 2 WINDOW 3 36 40 Left 2 SYMATTR InstName R2 SYMATTR Value 100k SYMBOL zener 448 96 R180 WINDOW 0 24 64 Left 2 WINDOW 3 24 0 Left 2 SYMATTR InstName D7 SYMATTR Value 1N5373B SYMATTR Description Diode SYMATTR Type diode SYMBOL zener 448 176 R180 WINDOW 0 24 64 Left 2 WINDOW 3 24 0 Left 2 SYMATTR InstName D8 SYMATTR Value 1N5373B SYMATTR Description Diode SYMATTR Type diode SYMBOL diode 272 128 R180 WINDOW 0 24 64 Left 2 WINDOW 3 39 43 Left 2 SYMATTR InstName D9 SYMATTR Value UPSC600 SYMBOL ind 272 368 R270 WINDOW 0 32 56 VTop 2 WINDOW 3 5 56 VBottom 2 SYMATTR InstName L1 SYMATTR Value 10m SYMBOL cap 528 224 R0 SYMATTR InstName C1 SYMATTR Value 10=B5 SYMBOL diode 480 32 R270 WINDOW 0 32 32 VTop 2 WINDOW 3 0 32 VBottom 2 SYMATTR InstName D10 SYMATTR Value UPSC600 SYMBOL res 48 368 R270 WINDOW 0 32 56 VTop 2 WINDOW 3 0 56 VBottom 2 SYMATTR InstName R7 SYMATTR Value 100k SYMBOL cap 224 128 R90 WINDOW 0 0 32 VBottom 2 WINDOW 3 32 32 VTop 2 SYMATTR InstName C2 SYMATTR Value 10n TEXT -184 400 Left 2 !.tran 1 startup TEXT 632 24 Left 2 ;C1=3D10u, R1=3D47, R2=3D470k, \nV1=3D150V P-P, Input =

I had a long persitance CRT in a video monitor. It was useless for aiming a C-band dish. ;-)

I saw it on lots of defective monitors, with vertical deflection problems.

If it only did 100 lines, the screen intensity would be uneven, and show hum bars.

Sure, if you film the TV screen with a high speed film camera (1000 frames/s) it indeed will show "hum bars". Apparently you have not seen such shots.

However, the eye is quite slow, especially in middle of the field of view, particularly at the relative low light intensities available from a CRT. The eye or the neural system will handle the integration and no hum bar will be observed.

If you try to display a picture with details only in odd (or even) lines on an interlaced monitor, you will get 25 (or 30 Hz) flicker, which is quite annoying.

For this reason, the vertical resolution must be reduced, if intended to be displayed on an interlaced CRT display. In the old days, the scanning electronic beam in the camera was so wide that it took some electron charge from the line above and below (belonging to the other field), thus automatically reducing the vertical resolution.

With CCDs, some active means are require to reduce the vertical resolution to avoid the 25/30 Hz flicker on interlaced CRTs. For instance scan rows 1+2, 3+4, 5+6 etc. could be added in one field, while 2+3, 4+5 etc. could be added in the other field.

Are the 1080i HDTV transmissions still vertically filtered at the transmitter end ? This would explain why the 1080i format only get similar visual ratings as 720p transmissions, even for stationary scenes.

However, with the small number of HDTV CRTs used today, it would make much more sense to do the vertical filtering digitally in the set top box, and then only, if driving an interlaced CRT. As far as I understand, most of the US 1080i CRT TVs are actually 540p displays, in which no attempt is made to do proper interlace, but simply write the odd and even lines on top of each other and for most tubes, the shadow mask pitch would not allow much above 540p anyway.

I was a TV broadcast engineer, and I despised Kinescope.

'High speed film' also has very short, non syncronized shutters that add to the problem. WTF would you want to shoot 1000 farames per second of something with a 50 or 60 Hz field rate? Are you really that stupid?

Sigh. The human eye's 'Persistance of Vision' is taken into account when chosing the phosper. If the phospor was too slow, the additional 'Persistance of Vision' will cause smearing of any motion.

The processing is done in the Camera, and everything is digital past that point. Are you suggesting that the video is processed once again at the transmitter? That would also require the audio be re-processed, to match the additional delay.

While the non-synchronous film 24 frames/s vs 50/60 fields indeed is causing problems, running the film at something like 1000 frames/s is severely oversampling the TV screen, readily satisfying the Nyquist criterion.

For the exact reason of showing the CRT scanning mechanism and the persistence. The shot I have seen a few decades ago, showed a CRT displaying a test card and the filming sequence lasted for a few TV-fields and of cause, when the film was replayed at normal speed, the CRT screen update could be seen in slow motion.

I believe in what I see, but not necessary in all kinds of urban legends.

The MPEG compression will add a huge amount of delay, so the sound has to be delayed anyway.

What I was thinking about was capturing a scene with proper 1080p60 cameras, simply throwing out every other line to make a 1080i60 for transmission. Since many customers have 1920x1080p60 capable displays, it would be possible to deinterlace such transmitted signals and restore the 1080p60 resolution at the screen at least for stationary areas in the picture.

However, if the image is vertically filtered, either directly in the CCD by summing charges from different rows or later on digitally, there is no way that the original 1080 horizontal line resolution could be reconstructed on the display. In practice, the effective resolution would be something like 1920x540p60.

to get a definitive answer on what is REALLY going on inside a typical =

of leds? (and 60hz flicker isn't noticed?)

above?=20

manufacturers. Certainly there aren't sophisticated drivers in a lamp = (maybe just some higher priced ones?)

Lazy cheap s*it. DPA a few and know for your self.

?-)

Ah but it does somewhat due to some flicker in the backlight. Small but there, see for your self.

?-)

Paul -

I disagree with your 200V results. I am wondering how you measure these.

Try this: copy and paste the following as a Spice Directive into your schematic.

.meas tran pd1 avg V(A,C)*I(D1) trig V(N002,N004) val =.1m rise=10 targ V(N002,N004) val=.1m rise=11

.meas tran pv1 avg V(N002,N004)*-I(V1) trig V(N002,N004) val =.1m rise=10 targ V(N002,N004) val=.1m rise=11

Run the simulation. View/Spice Error Log

It will show the power in the LED (D1) as 151mW and the power delivered by the source as 5.43W integrated over a full period (it omits fractional periods by using trigger and target statements).

If you are using some other output/input power conditions, let me know and I will change the Measure statements to comply.

If you can show that my method is not correct, I will also revise my method.

Thanks and Merry Christmas, John S