Does anyone know anything about "Modulation Depth"? (Philips PM5680)

[inaxeon]

Hello

I'm currently trying to throw together a setup to create an analogue telly Broadcast [by that I mean signal - not actual on-air transmission] (for fun). It's mostly based on Philips PTV gear (I have spammed these forums with links to my teardowns of them in the past).

The latest piece I'm trying to grapple with is a Philips PM5680 modulator ( https://www.radiomuseum.org/r/ptvphi...5680_all.html# ). Usual problem - no manual available.

I have one which is PAL-I capable but am a little confused about the modulation depth adjustment.

I think I understand with the vision carrier, it's the ratio between the amplitude of the carrier tone and the video signal and I can see this clearly on my spectrum analyser. Some reading suggests 90% "modulation depth" is a good starting point.

I'm a bit more confused about the audio side of things. It measures it in KHz instead of %. Basically I have no clue what this measurement actually is.

Anyone have any experience of this?

[TonyDuell]

Originally Posted by inaxeon

I'm a bit more confused about the audio side of things. It measures it in KHz instead of %. Basically I have no clue what this measurement actually is.

Anyone have any experience of this?

Isn't the audio signal frequency-modulated? This might be the deviation or something related to it.

[Radio Wrangler]

PAL-I has FM sound, so the amount of FM is specified as the frequency deviation, usually in kHz.

Vision is amplitude modulated with the sync pulses being the greatest amplitude. With a colour signal, there has to be allowance in carving-up the amplitude cake to allow for the maximum chrominance voltage at 4.43361874MHz superimposed on the luminance ( DC component).

Video isn't a symmetrical easy waveform like a sine or AC-coupled AM audio, It will have to undergo DC restoration at the receiver. So it's best to not think of an average carrier level. Think of the RF voltage at peak sync and the RF level at the black reference in the porches and scale everything from these.

David

[inaxeon]

Originally Posted by TonyDuell

Isn't the audio signal frequency-modulated? This might be the deviation or something related to it.

Ah yes that sounds more like it. So if adjusting the modulation depth, what is that actually doing in practice then?

I would have thought deviation from the carrier would be determined by the actual signal, not an adjustment on the modulator?

[inaxeon]

Originally Posted by Radio Wrangler

Video isn't a symmetrical easy waveform like a sine or AC-coupled AM audio, It will have to undergo DC restoration at the receiver. So it's best to not think of an average carrier level. Think of the RF voltage at peak sync and the RF level at the black reference in the porches and scale everything from these.

David

OK. I'm gonna put the IF on my scope and see if I can get my head around what you are saying.

[lesmw0sec]

Originally Posted by inaxeon

Ah yes that sounds more like it. So if adjusting the modulation depth, what is that actually doing in practice then?

I would have thought deviation from the carrier would be determined by the actual signal, not an adjustment on the modulator?

It is both. In practical terms, your modulation (or deviation) setting is probably nothing more than a pot. to set the deviation from a standard audio test signal.

[wd40addict]

FM audio deviation is normally 75kHz, at least on the 88-108MHz band.

[marceljack]

Originally Posted by wd40addict

FM audio deviation is normally 75kHz, at least on the 88-108MHz band.

For 625 lines standards (B/G, I, D/K) the max deviation is +/- 50 kHz.
For 525 lines standard M (NTSC) it is +/- 25 kHz (sound carrier at + 4,5 MHz).

[inaxeon]

Originally Posted by marceljack

For 625 lines standards (B/G, I, D/K) the max deviation is +/- 50 kHz.
For 525 lines standard M (NTSC) it is +/- 25 kHz (sound carrier at + 4,5 MHz).

Would it be fair to say then, that the "pot" on the front panel is setting the frequency deviation of the maximum representable AF amplitude (as you state 50KHz in this case)?

[marceljack]

Originally Posted by inaxeon

Would it be fair to say then, that the "pot" on the front panel is setting the frequency deviation of the maximum representable AF amplitude (as you state 50KHz in this case)?

If the pot in question has graduations in kHz, I would say yes.
Can you post a photograph of this part of the front pannel ?

[FERNSEH]

Inexeon wrote:
"I think I understand with the vision carrier, it's the ratio between the amplitude of the carrier tone and the video signal and I can see this clearly on my spectrum analyser. Some reading suggests 90% "modulation depth" is a good starting point."
Is 90 per cent video modulation depth too much even for negative modulation TV systems?

DFWB.

[inaxeon]

Originally Posted by FERNSEH

Inexeon wrote:
"I think I understand with the vision carrier, it's the ratio between the amplitude of the carrier tone and the video signal and I can see this clearly on my spectrum analyser. Some reading suggests 90% "modulation depth" is a good starting point."
Is 90 per cent video modulation depth too much even for negative modulation TV systems?

DFWB.

At this point in time I don't even know what 90% is a measure of. I saw a mention of it somewhere (on these forums I think), also the summary datasheet from Philips pictures it set at 90%

[FERNSEH]

"At this point in time I don't even know what 90% is a measure of. I saw a mention of it somewhere (on these forums I think), also the summary datasheet from Philips pictures it set at 90%"
Peak white relative to zero modulation. Sync tips to maximum carrier level.

DFWB.

[Restoration73]

This might explain things;

To avoid transmit overmodulation with chroma present, the modulation level is
reduced. In the receiver the chroma gain can be increased in the RGB
video stages.

[jhalphen]

Hi to all,

To illustrate this thread, here are photos (web search) of the Philips PM5680 TV modulator.
From looking at the I/O connectors, the unit requires an RF TV channel input frequency and only accomplishes A/V modulation.
The photos show a (possibly) multistandard unit with NICAM audio or German stereo/dual sound option.

Best Regards
jhalphen
Paris/France

[inaxeon]

I'll probably end up doing a full teardown of it at some point. Here's a photo inside the front panel (Also attached):

https://xpander.mattmillman.com/file...ges/pm5680.jpg

Regarding the Zweiton version. I've searched long and hard and cannot find a single documented example of this actually being fitted (Even to the multi standard units). I would guess the Netherlands is the only place such a thing could be found because if it were Germany it'd be Rhode & Schwarz :P

In that photo the Zweiton audio (second channel) is done by the missing Unit 8.

There is no option of NICAM. The "BW NICAM" Switch on the front (mine has this fitted) changes the composite video input filter to allow the NICAM carrier to be inserted through this input rather than through the IF. NICAM modulation is done by the PM5687 (which I also have. It is stacked on top it).

Another point of interest is that Unit 10 can either be a fixed channel up-converter or a linearity corrector. I believe mine has the linearity corrector although as of yet I have no clue what this actually does.

For the up-converting I'm planning to use this beast: https://www.mattmillman.com/a-quick-...-up-converter/.

[inaxeon]

Originally Posted by Restoration73

This might explain things;

To avoid transmit overmodulation with chroma present, the modulation level is
reduced. In the receiver the chroma gain can be increased in the RGB
video stages.

That is a really good diagram. Gives the clear explanation I was looking for.

Couple of points / questions:

A) Looks like it's probably quite handy to have a colour bar generator to calibrate a modulator.

B) The "weighting" that is described there. That I'd guess has to be done by the video source?

[rambo1152]

The weighting is not applied to the chrominance signal as a whole, it is applied to its components. The weighting factors are
0.877 applied to (R' - Y')
and
0.493 applied to (B' - Y')
these weighted signals are then referred to as V & U respectively.

As to where these weightings would be applied, I am not sure, as I was only a humble TV repair person, and not a broadcast engineer, but I would suspect it was done inside your PM5680.

[Radio Wrangler]

What you have there is the modulator rack from a full broadcast TV transmitter.

It takes baseband video and baseband sound signals and creates an IF signal having both video and sound modulated carriers.

From this unit, the IF signal goes off to be mixed with a local oscillator to create a signal on the intended channel. Filtering is needed to remove the image product. and the rest is power amplification with harmonic filtering.

Everything after this unit has to be carefully kept linear.

The spare slot is either for a second analogue sound modulator for analogue stereo (or dual language) service. It could also be for a NICAM modulator for digital sound.

The VSB filter means vestigial sideband. Broadcast TV has the sideband energy around the vision carrier trimmed asymmetrically so one sideband is more extensive than the other.

So, what you have is fixed frequency and not on any used TV channel frequency. You need some more hardware to get there.

David

[inaxeon]

Originally Posted by Radio Wrangler

The spare slot is either for a second analogue sound modulator for analogue stereo (or dual language) service. It could also be for a NICAM modulator for digital sound.

...

So, what you have is fixed frequency and not on any used TV channel frequency. You need some more hardware to get there.

Did you read what I said?