No DA is perfectly real. Bob Pease showed how to model a cap as a ladder of simple RC circuits down to an almost arbitrary accuracy. Back then they were trying to make extremely accurate integrating A/D's with them. Substituting that model you can get all of the results of the null test in SPICE. Depending on how much time you want to waste even down to uV levels.
As we have agreed that DA is a linear phenomenon it is hardly surprising that it does not explain capacitor non-linearity!ThorstenL said:
As I said, is it the LF phase shift caused by DA which is audible, or is DA a marker for other (non-linear) dielectric problems?
Could the LF phase shift be audible in stereo because it is unlikely that the responsible capacitor will be identical in both channels? Given that DA is essentially an error (albeit a linear one) it is unlikely that dielectric manufacturers will control it precisely.
It is not totally unreasonable to postulate that a really bad dielectric might also have non-linear mechanisms at a higher level.
Hi,
Yup, I grew up in a country that had Secam colour TV... 🙂
I made money from modding these TV;s to take a PAL colour decoder and to switch depending on the signal. The decoders where mostly equipment pulls from West Germany, though towards the end we had our own fully discrete design.
RC (well) above cutoff = constant groupdelay, last time I looked.
Ciao T
Yup, I grew up in a country that had Secam colour TV... 🙂
I made money from modding these TV;s to take a PAL colour decoder and to switch depending on the signal. The decoders where mostly equipment pulls from West Germany, though towards the end we had our own fully discrete design.
An externally hosted image should be here but it was not working when we last tested it.
RC (well) above cutoff = constant groupdelay, last time I looked.
Ciao T
John,
DA has impact, but there is so much variabiliaty... If you run appreciable DC across the Cap (if you don't, why have the Cap?) microphonics tend to play the largest role.
The AMR Phono was originally designed with a special cap for the LF EQ.
This was a wound part, long discontinued, east german mil spec and normally exceedingly good as coupling cap (plus I have huge stocks).
In the EQ position with DC across it it was so microphonic (and that with only 26dB gain after it to line level) tapping it with a pen made the bass drivers bottom out. I (with sadness) changed the design to monolithic caps (literally, they are "stone" [lithos] cap's)...
No more microphonics...
Ciao T
DA has impact, but there is so much variabiliaty... If you run appreciable DC across the Cap (if you don't, why have the Cap?) microphonics tend to play the largest role.
The AMR Phono was originally designed with a special cap for the LF EQ.
This was a wound part, long discontinued, east german mil spec and normally exceedingly good as coupling cap (plus I have huge stocks).
In the EQ position with DC across it it was so microphonic (and that with only 26dB gain after it to line level) tapping it with a pen made the bass drivers bottom out. I (with sadness) changed the design to monolithic caps (literally, they are "stone" [lithos] cap's)...
No more microphonics...
Ciao T
I see what you mean but not the same thing. A true delayed signal would propagate as a wave. An infinite RC propagates as a diffusion like heat, add enough L and you can get a wave solution. The glass delay lines have mechanical L and C.
What a kluge SECAM was, with that slab of quartz cut at several angles to get enough length. Trimmed with pencil lines too.
A kluge that worked, tho. Over the air or on VHS it always looked better than PAL (NTSC not even close).
Sorry for the OT.
Sorry for the OT.
sCOTT,
I must have missed the part in your post where you set out the limits...
Well, I know people that use RC based delay lines to compensate mechanical alignments, have done so since the 1980's and the measured impulse and amplitude response support that contention.
Yes, they do.
Sure, but at least it delivered the same colour every time, unlike NTSC (aka "Never The Same Colour")
Ciao T
I must have missed the part in your post where you set out the limits...
Well, I know people that use RC based delay lines to compensate mechanical alignments, have done so since the 1980's and the measured impulse and amplitude response support that contention.
Yes, they do.
Sure, but at least it delivered the same colour every time, unlike NTSC (aka "Never The Same Colour")
Ciao T
The 50Hz refresh is a non-starter for me, I have a sensitivity in my peripheral vision.
Scott,
I take it that you cannot watch movies in the cinema at all then, or Movies in Cinema mode (24Hz) on your TV? Poor you.
Ciao T
PS, PAL and Secam had their shortcomings, but next to NTSC they where solid gold. It's all history now.
I take it that you cannot watch movies in the cinema at all then, or Movies in Cinema mode (24Hz) on your TV? Poor you.
Ciao T
PS, PAL and Secam had their shortcomings, but next to NTSC they where solid gold. It's all history now.
Indeed NTSC's nickname 'Never The Same Colour' was earned, it wasn't gratuitous.
Uhhhh...... you may want to rethink that one, Thorsten. What's the real refresh rate there?
The 50Hz took some getting used to for sure. That's why the 100Hz refresh rates came out some time ago in Europe. Yes, all history now. Glad we've moved on.
The gate flashes typically 2 to 3x the film speed and the there a numerous issues related to the actual on/off ratio of light not to mention a CRT and film projector are quite different. The old TV's (long retrace) in the UK didn't cut it, and yes a CRT monitor on an old PC running 60Hz annoys me too.
So you're telling me what I see now? If you think a film on an LCD panel is actually turning off and on at 24Hz, you need to do a little research. 24 FPS gives a good margin on the minimum rate to percieve continuous motion. The flicker is related to light on/off time and frequency, people have different sensitivities.
Hi,
Film in the Cinema? 24 Frames per second.
Projector and/or TV running at cinema mode, 24 Frames per second.
The difference is that each frame is full frame with modern systems (LCD, DLP, Plasma) while with CRT TV's the picture was drawn line by line and the "afterglow" of the phosphor was not long enough to avoid flickering.
But that is a problem of CRT technology, not of Pal/Secam/NTSC in principle.
If you run a full picture with 24 (or 30) Frames per second but change from one full frame to the next without switching off the picture in-between, there is no flicker...
I grew up with 50Hz TV's, so I am used to it. But I found the colours on US TV's a little strange, to be polite, when I spend time there in the 90's (never mind the programming).
Ciao T
Film in the Cinema? 24 Frames per second.
Projector and/or TV running at cinema mode, 24 Frames per second.
The difference is that each frame is full frame with modern systems (LCD, DLP, Plasma) while with CRT TV's the picture was drawn line by line and the "afterglow" of the phosphor was not long enough to avoid flickering.
But that is a problem of CRT technology, not of Pal/Secam/NTSC in principle.
If you run a full picture with 24 (or 30) Frames per second but change from one full frame to the next without switching off the picture in-between, there is no flicker...
I grew up with 50Hz TV's, so I am used to it. But I found the colours on US TV's a little strange, to be polite, when I spend time there in the 90's (never mind the programming).
Ciao T
Again we're talking about two different things. If film were actually projected with a single flash per frame and sufficient dead time indeed I would find it unwatchable as it is I can see the gate on the edges of a bright scene.
Now on "Sprockets" it's time to dance.
- Status
- Not open for further replies.