Accommodating varied input signal levels?

Forum for discussion of narrow-bandwidth mechanical television

Moderators: Steve Anderson, Dave Moll, Andrew Davie

Accommodating varied input signal levels?

Postby ac7zl » Wed May 16, 2007 4:49 am

The club standard provides for a one volt video signal. That said, I don't think a robust input circuit for an experimental televisor should bank on that level of standardization.

Question (a) What range of amplitudes might the televisor have to accommodate?

For example, from what I've read, consumer audio gear "line outs" are set up for -10 dBV. I compute 0.316 VRMS. Assuming a sine wave (admittedly not the case with video) I compute a pk to pk value of 0.894 volts.

It follws, then, that the televisor front-end must provide a gain factor of 1.12.

Professional audio gear has +4 dBu outputs. I compute 3.47 volts pk to pk, which means the televisor front end would have to attenuate the input signal by a factor of 0.288.

If the televisor is being fed from a CD player or other media device via its headphone jack, the input signal could be almost anything.

It's easy enough to add a level pot to the input of a televisor, but it would helpful to be able to scale the circuit appropriately. Making the gain/attenuator circuit broader than necessary only serves to make the proper adjustments more "touchy."

b) Should the televisor's input circuit be designed for a certain input impedance? I think consumer "line input" impedance is on the order of 1000 ohms, pro-gear impedance is figured at 600.

It has been my past experience that the headphone outputs of some equipment won't deliver signals properly unless the jack is driving an appropriate load. This implies to me that a well-designed televisor input circuit would feature a load switch that would cut in a low-value resistor, perhaps 47 ohms, to emulate headphones and act as a suitable load for the media device providing our video signal.

Your thoughts?

User avatar
Mad Scientist
Posts: 62
Joined: Thu Feb 01, 2007 5:04 am
Location: Tucson, Arizona, U.S.A.

Standards, are there any?

Postby Steve Anderson » Wed May 16, 2007 2:16 pm

Amongst the professional analogue video world there is one standard that is adhered to, that is that the video circuit has an amplitude of 1V at a circuit impedance of 75 ohms for a monochrome signal from bottom of sync to peak white. There it ends.

Throw in different lines, 405, 525, 625, 819 (not to mention HDTV), frame/(field) rates 25Hz(50Hz) and 30Hz(60Hz). Add in colour, NTSC, PAL and SECAM. Even here there are differences of sub-carrier frequency, 3.58MHz and 4.43MHz. And any permutation you can think of, take a look at some of the South American 'standards'.

Even when it comes to power the world is divided between 120V/60Hz and 220/240V/50Hz. That is not the end of it! Japan uses both in varying areas and the Philippines uses 220V/60Hz!

When it comes to analogue audio the situation is even worse. The professional 'reference' level is a 1kHz sinewave tone of 0.775V RMS. This was derived from the UK post office telephone network where at an impedance of 600 ohms, 0.775V RMS results in 1mW of dissipation in the load.

However, when that voltage is fed into a professional VU meter, what does it read? -4VU. Into a BBC spec PPM, 4. The use of 600 ohm circuits in Europe and most parts of the world vanished years ago, it's now a low impedance output (less than 50 ohms) and a high impedance input (greater than 10k). You do find 600 ohms sometimes today in the US, which is why on a Sony DVW VTR you'll find a switch to select between 600 or 10k.

But when it comes to domestic gear, there is no defined standard. 'Hi-Fi' gear can have a 'reference' level from 100mV to 3V! There is also no defined circuit impedance. But generally follows the concept above. There is also the issue of signal polarity as mentioned in a different thread.

It simply is a mess. But you might think that with digital audio all that will have gone away. Sorry.

Different sample rates, different bit-depths and different compression systems used. Surely the professionals sorted it out? No. AES/EBU audio is (thankfully) 'usually' at a sample rate of 48kHz, but at varying bit-depths from 16 to 24 bits. AND 4 different reference levels!

Also there is two major 'standards', 1V/75 ohms unbalanced on BNCs (Sony), almost identical to analogue video, and 2-8V balanced using 3-pin XLR connectors (almost everyone else).

Then you have optical/fibre, networks...and AES/EBU sampled at never ends....

I gave up despairing a long time ago.

Steve A.
User avatar
Steve Anderson
"Fester! Don't do that to 'Thing'"
Posts: 4211
Joined: Fri Mar 30, 2007 10:54 pm
Location: Bangkok, Thailand

Postby AncientBrit » Thu May 17, 2007 11:22 pm


For "receivers" I tend to design my NBTV circuits with a nominal input level of 1 volt pk/pk, but with 6db gain in hand.

So I preface all my circuits with a simple 10k linear pot at the input and ensure that I have around 2x gain in the main chain.

Normal operation then is with the pot at 50%, but there is still enough in hand to cope with a smaller signal.

On "transmitters" I always send at 1volt pk/pk.


Green padded cells are quite homely.
Posts: 858
Joined: Mon Mar 26, 2007 10:15 pm
Location: Billericay, UK

Return to Mechanical NBTV

Who is online

Users browsing this forum: No registered users and 9 guests