PGA2310 Digital Volume Control

[alaskanaudio]

I got a update on new products from Texas instruments by Email today discribing the new Burr Brown digital volume control. The link does not seem to be working, but you may be interested in looking at this device by searching their web site.

Here is a partial copy of the TI news letter.

Texas Instruments Unveils Industry's Highest Performance Audio Volume Control IC for Professional Audio Systems -- PGA2310
TUCSON, Ariz. (Oct. 19, 2001) -- Designed for a wide
variety of professional recording and playback
applications, Texas Instruments (TI) (NYSE: TXN)
introduced the industry's highest performance audio volume
control IC (Integrated Circuit) from the company's Burr-
Brown products division. Featuring 120dB dynamic range,
0.0004 percent distortion (THD+N), -126dB interchannel
crosstalk, and +/-15V analog power supplies, the PGA2310
improves system performance in professional audio
applications such as digital mixing consoles, multi-track
recorders, broadcast studio equipment, high-end A/V
receivers, effects processors, musical instruments, and
high-end car audio systems. See
http://www.ti.com/sc/aap6908u.

I copied this from one of my previous entries in order to start a new thread on this subject.

John Fassotte
Alaskan Audio
http://www.audioamps.com

[alaskanaudio]

In reference to my previous post for the new TI/Burr Brown PGA3210 digital volume control IC.

I have some interest in testing this device in the future. However I have no time to devote to such a project at the present time since I’ am working hard on a all discrete version of my MOSFET class a power amp.

Since I have very little time to attack new projects right now my contribution would be limited to helping others to develop a test bed for this device when my time allows.

If anyone is interested in my help for testing this device and also has programming experiance with suitable controllers please let me know.

[paulb]

Hey, John, thanks for starting this thread as I suggested, I thought it deserved better visibility.
I'm interested in this chip, perhaps as an upgrade to my preamp. I'm currently using an SSM2018 voltage-controlled amplifier and am very disappointed (it seems to amplify noise on its supply rails). The micro part doesn't bother me (my background is in firmware), but I don't have a development system at the moment. I'm in the middle of another amp at the moment, so it would be a few months before I could start on it as well. Whatever I build, I'd like it to be accessible to DIYers who don't want to spend money on a development system, so I think I'll start looking for a reasonable platform - something like the Basic Stamp comes to mind.

[alaskanaudio]

PaulB,

This sounds good to me. I will likely order some of these parts to have some on hand for testing later on. Your time frame for working on this appear to match mine fairly well at the present time.

[Janne Nivala]

A newbie-question about the PGA2310's "Recommended Connection Diagram"/"Typical PCB Layout Floor Plan" found in the PGA2310's data sheet:

The digital and analog planes are supposed to be connected to each other at a single point. Why do they have to be connected? Why aren't the digital and analog grounds left unconnected with each other, since the chip has a digital ground pin and ground pins for both left and right analog channels? And, in the datasheet connection there are separate digital and analog power supplies.

Janne

[alaskanaudio]

Janne,

Typically the analog ground and the digital ground should be connected at one point on the printed circuit board using the shortest possible connection between the two. Any traces than may need to route from the analog side to the digital side should be run directly beneath this ground bridge. This is pretty standard practice to minimize the antenna effect that lenghty connections can cause.

I have not looked at this failrly new device carefully and perhaps it is possible to clock data into it using logic and a optical shaft encoder. If the processor/logic could be stopped when no volume changes are made it may make PCB design simpler. A digital readout is also required to know what volume level is programmed in.

Not to many years back I was at a Stereophile show where a digital volume control creaped up to full volume on its own. Luckely this was noticed before it caused damge to equipment and ears.

Some times errors are made on manufactors data sheets that do cause confusion.

[alaskanaudio]

I have added a link to the data sheet for this device. It looks extremely atractive and will be quite easy to use. I expect that many high quality audio manufacturers will be using it in the future. The high signal levels it can handle should gice plenty of headroom.

Perhaps the link below will save you some time looking for the data sheet.

The link is:

http://www.audioamps.com/diyparts/ics/ics.htm

I have also uploaded the latest diagram of one of my RIAA equalized preamplifiers and a picture of the bare circuit board.

These can be found at:

http://www.audioamps.com/diyparts/pcb/pcb.htm

John Fassotte

[mlloyd1]

I have had some CS3310 devices for a while and had no time to play :-( The specs on the Burr Brown PGA3210 "upgrade" sure look impressive; Crystal's CS3310 was no slouch itself.

I was roaming around and found this link. It may be useful to folks looking to prototype quickly. I wanted to build something like this for my prototyping but using an ATMEL AVR device as the micro instead of the PIC.

http://digilander.iol.it/paeng/CS331...e_control_.htm

So, has anybody tried the Burr Brown part yet? We're waiting....

Michael

[paulb]

Michael,
Yes, the Atmel looks like a good choice, a friend is using it and seems quite happy with it.
Maybe I'll get him to write the firmware...
What's the price on the Crystal part? Have you compared the specs? Yes, I could do it myself but thought you might have already done so.

[Pete Fleming]

Janne, digital and analogue grounds are kept separate to reduce the contamination of the analogue system with noise from the digital domain, which tends to be quite noisy. Connecting at one point in a “star” arrangement minimises contamination while providing a common reference to the two separate, but interconnected, domains.

Hope this helps.

Cheers,

Pete