On a sunny day (Wed, 02 Jul 2014 21:26:44 +0100) it happened Mike Perkins wrote in :
Right, the circuit is a DC resorter, but not a black level stabilizer.
A black level stabilizer will also restore DC level. The black level stabilizer is *required* to keep the black level in the display device constant, and usually all the way at the end of the 'chain' of amplifiers . Sync amplitude changes, and cannot be used for that, and should not be dependent upon for display.
 It is done by clamping at the black level, just after the sync, and needs a special correctly timed pulse for that, as there is also the color burst there in a composite PAL or NTSC signal, so it should be done where there is no burst, just before the CRT or whatever, and that is where you want the black level constant anyway.
You can make a good DC restorer by clamping on the sync tip to keep your signal in the linear amplifier range, but that is ALL it is good for. Not much accuracy is needed in such a case, as usually the amplifier range is much bigger than the composite signal, so a simple diode will do in most cases.
I see you have now removed the offending circuit from your website.
You "had" a 100k resistor down to ground to bias the circuit. The current flowing through this, and the clamp will be dependent on video level. The clamp will have an impedance and so there will be a small variation of clamp level with video level.
Ordinarily you would need a buffer after this to ensure the clamp is not affected by anything downstream.
Agreed, but why would your customer specifically introduce AC coupling?
Jim isn't doing a back porch clamp. It appears to grab sync tips and hold t hem at 0 Volts and as long as the sync amplitude is consistent the black le vel will be too. This isn't a broadcast rig. You see this sort of thing in consumer gear but not this slick.
Simple diode clamps are often used for 'crude' sync detectors to drive mono stables to sample/hold on sync tip and back porch to guarantee 50% slice po int for a precision sync detector. You find those in composite analog Time Base Correctors genlock circuits where you absolutely need nanosecond preci sion to maintain SCH phase.
Ah, that makes sense for old surveillance cameras.
Gently pushing the signal to 0V doesn't work for OTA video because the H-sync becomes a total wreck. That's where using the sync detectors for restoration does a much better job. (The sync detector gets help from a PLL in OTA decoders.)
Thankfully, surveillance cameras are going digital too. I just bought a nice one that has the entire surveillance control system running on a tiny Linux board inside it. I think it has more computing power than my old Mac Pro.
Not sure exactly what you want, but a PNG is just an image file... it can be any size you like.
I just recently scanned a 2-page article, saving each page as a separate PNG. Then fired up GIMP (Windows Paint would work as well) opened Page 2, selected All, copied to clipboard, opened Page 1, doubled the height of the work area, and pasted Page 2 at the bottom. Worked great, and was about 1/4 the file size of a PDF (but that may have been due to the cheesy free PDF converter I tested against).
You could paste the second page next to the first instead of below it, if you needed that for some reason.
Bob Masta DAQARTA v7.60 Data AcQuisition And Real-Time Analysis
Scope, Spectrum, Spectrogram, Sound Level Meter Frequency Counter, Pitch Track, Pitch-to-MIDI FREE Signal Generator, DaqMusiq generator Science with your sound card!
Old TV circuits are art and magic - getting the most done with the fewest parts. I have no idea where to find a good schematic that's legible.
Clean H and V signals were absolutely critical. Not only were they needed to produce a steady picture on a weak signal, but it kept the TV from burning out or, in late models, triggering the X-ray protect circuit. 99% of the TV was powered off inductors driven by sync signals.
I did it in a very crude circuit when I was about 18 years old but I can't remember exactly how it worked. I think a 555 chip generated pulses slightly narrower than sync and used that as a mask on a filtered video signal that went back into 555 pins 2/6. HSYNC and VSYNC were pretty much the same except for polarity and frequency. HSYNC clamped the video going to the gun drive for DC restoration. I was using a CRT module that, thankfully, had X-ray protection.
On a sunny day (Sat, 05 Jul 2014 00:56:46 -0700) it happened Kevin McMurtrie wrote in :
There are some basic things about TV receivers (as of old analog). I have designed several ones, camera too. And that was in the late sixties. Nothing really changed, except more semiconductors.
1) gated AGC. When sync is maximum amplitude in the AM modulated video (the French did it in reverse to the rest of the world), you would, _after_ H lock by a PLL, only look at the signal amplitude during the H pulse for AGC control.
2) sync detection Impulse detection on peak of signal (that now is at a stable level due to AGC).
3) H sync separation CR differentiator
4) V sync separation RC integrator, although later models counted H pulses and what not.
5) H oscillator normally a PLL
6) V oscillator normally was an oscillator running slightly slower than the transmitter, and then hard synched. Later sets used PLL assistance for V too.
7) H deflection Almost always combined with HV generation, flyback, parabolic pulse into magnetic deflection coil. For shadow mask type color sets DYNAMIC geometry correction circuits were needed (east west, cushion, etc), sometimes used transductors to modulate H amplitude.
8) V deflection In the early sets all sorts of correction circuits were added to make a linear ramp in the 44 mH or so V coils. later (I think Philips started with that) a simple linear ramp was generated and the current in the deflection coils measured with a resistor in series with the coils, and compared to that ramp, no more adjustments. That was one of the biggest improvements ever.
9) Audio Old sets were mono, there were basically 2 systems: 1) separate IF for sound. 2) intercarrier sound here the video IF bandwidth was a bit bigger and the 5.5 MHz (Europe) FM component was taken from the video detector, so AFTER AM demodulation. then amplified and limited in a separate simpler 5.5 MHz IF, ratio detector for demodulation. When stereo came a second slightly higher in frequency FM carrier carried L-R, processed in the same way. There also was a sound in sync system, where the sound was in digital packets in the H sync, no experience with that.
Video in general: You cannot depend on the sync amplitude being constant EVER. Even in a TV studio the sync is regenerated many times, new sync replacing old. All that counts is the BLACK LEVEL, and white level. Imagine an analog video mixer, getting input from different (synchronized) sources. Apart from the PAL or NTCS color phase requirements, the black levels must be the same. Now if you mastered all that, there are a zillion ways to design electronics that does that, some very simple and some very complicated, the simple often working just as well as the complicated ones, in NORMAL signal conditions.
But remember Einstein said: "Things should be as simple as possible, but not simpler than that." He was not always right (almost never actually), but probably right about that,
So after this long course in TEEFEE in now present to you the SCOPE TV
I did not bother about black level at all, because the scope has a capacitor in the input AFAIK, would have to modify it to clamp at the CRT grid, anyways that was just for fun.
The sync separator is the simplest possible AFAIK. See how it works?
Note the the linear V ramp generator, not even a V oscillator needed (scope wont burn in like a TV CRT).
If you want to be more precise:
This uses a PIC to grab black level during V sync, stores it in the 100 nF cap, and then, during interfering pulses, replaces the interfering pulses by that black level. The H and V decoding and front porch black level timing (sample pulse) generation is done in software. A simple diode clamp clamps bottom sync to make sure the video is in the PIC analog hardware comparator range, there are 2 diodes, running at different currents to make 2 slice levels of the sync pulse. Note the 680 Ohm 330 pF burst filter. An analog switch switches between signal and stored black level.
In the old TVs, the PLL consisted of a monostable multivibrator set to run sightly lower than the expected line frequency. A signal edge assumed to be the H-sync pulse would retrigger the multivibrator a bit earlier before reaching the end of period. The voltage range was quite limited and so was the frequency locking so I do not see how that would burn the TV due to false sync pulses. The PAL, SECAM and NTSC line frequencies are quite close together, so you could get good H-sync, while the 50/60 Hz difference was often too much for V-sync.
Anyway, the H-sync pulse has an other important duty in OTA reception, since it is also used for generating the a.g.c. signal for the front end RF and IF stages at least in countries, with negative video modulation. The amplitude of the synch pulse is sampled and that value is then used for the _whole_ horizontal line. You can't use the average amplitude of the whole line as a.g.c., since the IF gain and hence average picture brightness would vary depending on picture content.
I have no idea, how they handled the a.g.c. in systems with positive video, such as the 405 line system in UK or 819 line system in France.
That was about 1980, such as the Salora/Nokia IPSALO