John Curl's Blowtorch preamplifier part II

Status
Not open for further replies.
Member
Joined 2014
Paid Member
TPA6120A2 - is a cfa specialty,
i think i am going to play with this one a bit, they have a complete ref design available, i am looking over the gerber data now. a nice 4-layer design.
High Fidelity Audio Headphone Playback Reference Design for Portable and Smartphone Applications - TIDA-00385 - TI Tool Folder

Many years ago was given a couple of TI ADSL dev boards using the THS6002 ADSL chipset, and was told that it should make a good headphone driver. My understanding is that at least one of the high end headphone drivers TI sells is basically that chip without the Rx part in.

And why not. I believe (and sure Scott will correct me if wrong) but the Walt Jung composite line stage from 1996 also used an ADSL part.
 
diyAudio Member RIP
Joined 2005
Many years ago was given a couple of TI ADSL dev boards using the THS6002 ADSL chipset, and was told that it should make a good headphone driver. My understanding is that at least one of the high end headphone drivers TI sells is basically that chip without the Rx part in.

And why not. I believe (and sure Scott will correct me if wrong) but the Walt Jung composite line stage from 1996 also used an ADSL part.

Yes, but a more conventional brute force part. ADSL drivers migrated to the geometric mean or Montichelli biasing.
 
normal listening levels are particularly important with the perceptually weighted noise shaping - the better curves are only valid at near human audibility threshold - turning up the volume to listen to the dither noise isn't "fair" as Loudness effects change our hearing frequency sensitivity shape

few home systems can do clean 120 dB SPL at the listening position - good system gain structure would put digital source 0 dB full scale not too much higher
 
Disabled Account
Joined 2012
normal listening levels are particularly important with the perceptually weighted noise shaping - the better curves are only valid at near human audibility threshold - turning up the volume to listen to the dither noise isn't "fair" as Loudness effects change our hearing frequency sensitivity shape

few home systems can do clean 120 dB SPL at the listening position - good system gain structure would put digital source 0 dB full scale not too much higher

Why are there more than one dithering noise shape?

Which is best one and/or when to use that one vs another?


THx-RNMarsh
 
Last edited:
Richard,
I just wonder if the FCC will like they have with broadcast TV outlaw analog FM so they can steal some of the bandwidth of the FM band for other uses. There is no more analog TV signal so why wouldn't they do the same with radio?

I am still trying to understand what you are going to do with the dither information such as which is the best method of dither usage? Even if you know and understand this what as a consumer of music and not a producer of music can you possibly hope to use this information for?
 
Last edited:
...each time i was○able to replace original OPAs by CFAs, there, the improvement was obvious, and sound back to life again (when lot of tracks active in a mixing desk );-)
that makes sense, to me.
It is because, you know one of the advantages of current feedback amplifiers: their bandwidths don't vary with gain, contrary to VFAs, where it decrease as their gain increase.
In blue: VFA 1 channel in
in green: VFA 32 channels in
in red and yellow: CFA, 1 & 32 Channels in
That is what noticed my wife from the kitchen which is built in the middle of all the recording studios, as everybody knows.
 

Attachments

  • bdwcvsv.jpg
    bdwcvsv.jpg
    157.9 KB · Views: 186
Last edited:
Disabled Account
Joined 2012
Richard

I am still trying to understand what you are going to do with the dither information such as which is the best method of dither usage? Even if you know and understand this what as a consumer of music and not a producer of music can you possibly hope to use this information for?

Why wouldnt I want to know how to make the best digital recordings? What some of the pit-falls may be? Best practices?

Really?!

We just went thru a few pages hitting on how LP are made right and wrong ways and best practices.

And, who says I wouldnt again record and not just play music?


-RNM
 
Last edited:
Not so much as we suppose, in fact. We have a very little area witch is sharp and detailed in the center of our vision. (Just try to read a text while you focus on the same surface same distance) a little aside. It is our eyes witch constantly move to bring our center of interest in the center, and our brain witch correlate the successive images.
True and that's not bad for us...I'll have to find the exact number,
we can perceive a significantly higher resolution than our computers
screens provide, I want to say something in the order of 11,000 or 17,000
lines per inch.

Thinking the audio perception range is 2 - 4 time our hearing bandwidth.
Stewart Hegeman and others have described the research. I don't think
I've ever read the original or primary research papers only secondary
if someone has a primary source and care to share I would appreciate it.
 
AX tech editor
Joined 2002
Paid Member
You can easily make your calculations based on your component choice: it depends on the skew of the F-F outputs and the difference between H-L and L-H transitions and delays of the buffer.
Let's say you use AC logic, raw (no tuning or compensation), that could amount to 500ps + 1.5ns, 2ns total for a period of 1ms if you work at 1Khz, that would be 50.0002% d. This is an almost worst case, unadjusted, for quasi-obsolete logic.
I'll let you compute yourself the H2 content of such a waveform. If it isn't satisfactory enough, you can use faster logic, compensate it, and even use additional closed-loop tricks to gain a 1K factor if you think you really need it

Thanks Elvee and Brad, that's instructive. Seems a little more easy than building a 0.0001% analog sine generator :cool:

Jan
 
Okay Richard,
If you actually intend to do some recording I can understand that. But for the consumer I don't see where dither information would ever be available or necessary. One question would be do the engineers doing final mix down even have a clue what the engineers before them have done, is there even any type of documentation in the least that keeps track of what has happened at each level or each track?
 
AX tech editor
Joined 2002
Paid Member
Guys, the Holman test is obsolete, the TIM30 test replaces it.

You mean the DIM30 test? Is that more sensitive than Holman's square wave test? Both exercise the amp dynamically. The DIM30 (or DIM100) I think looks at intermodulation; Holman's test looks to me more directly looking at transfer curve non-linearity. But in practise yes DIM30 has eclipse the Holman test.

Jan
 
AX tech editor
Joined 2002
Paid Member
It is because, you know one of the advantages of current feedback amplifiers: their bandwidths don't vary with gain, contrary to VFAs, where it decrease as their gain increase.
[/I]

But why is that in practical amps a disadvantage? You design your amp with a specific gain and bandwidth, and that's that, right?

I can imagine that it is of importance in say switchable gain instrumentation amps, but in audio amps?

I know that there can be gain/bandwidth modulation in less well designed amps but that's minimal, if it is there at all.

Jan
 
I would think you would have to take a cd and the high rez file and compare them bit for bit. I highly doubt that the producer is creating two different versions, so the question would be where would any changes come into play and why would the two media have any differences. We surely aren't talking the same thing as comparing the CD or High Rex files to a down sampled low bit rate MP3, then I would expect to see very different information if you compared files.

So we will have to wait and see if you can find and measure any difference between the Hi-Rez and CD files, I would think the dither again would be identical between files.
 
But why is that in practical amps a disadvantage? You design your amp with a specific gain and bandwidth, and that's that, right?
We were talking of mixing desks bus preamps. They work in inverting mode. Each time you plug-in a new track in the mix, the gain of the OPA increase automatically.
I noticed that, when you listen to a single track, if you plugged many other muted tracks, on most of the mixing desks (VFA), its sound turned dull, compared to the same all alone.
According to my wife from the kitchen, CFAs cured the issue, to the great delight of our friend Waly. ;-)
 
Last edited:
Status
Not open for further replies.