Stereo pre-amp controller project
awasson
Posts: 57
This thread is·intended to be about audio and digital amplifier controls. We'll see how it goes...
OVERVIEW
A year or so ago in the sandbox, there was a thread that discussed a·digitally controlled multi-channel audio system that looked pretty interesting however my project is a little simpler than that. I've got way too much audio stuff (in boxes)·and I want to consolidate my home entertainment system into a neat little all in one package that works off of one remote control. To keep it simple I'm only concentrating on the living room to start with.
The setup:
X-Box (1st Generation)
Motorolla DVR-Cable Box
Panasonic Plasma TV
Logitech Z2300 (2.1 Powered Speakers)
* I'll add a dedicated media PC to the mix later
This is a really great setup for a small living room plenty of sound and the Motorolla remote controls all of the components. The only hitch·is that the line out from the TV is not attenuated via the remote control and the mute button has no effect on the line out·either. My first project is to make a remote controlled line level attenuator with mute and make it respond to the remote control.
APPROACH
1) Audio Attenuation
It looks like a chip from TI called a PGA2311 will handle the attenuation of a pair of line level signals and also provides a muting control. It takes a·binary number from 1 to 255 which controls the level of output.
2) Digital Control
I should be able to use any of my Parallax BS2 or SX micros to receive a remote control's output and·produce a binary number to feed to the PGA2311.
So, it looks pretty simple but... I still have to figure out how to decode the signals and put it all together so any help would be great. Wish me luck and Happy New Year!
Andrew
OVERVIEW
A year or so ago in the sandbox, there was a thread that discussed a·digitally controlled multi-channel audio system that looked pretty interesting however my project is a little simpler than that. I've got way too much audio stuff (in boxes)·and I want to consolidate my home entertainment system into a neat little all in one package that works off of one remote control. To keep it simple I'm only concentrating on the living room to start with.
The setup:
X-Box (1st Generation)
Motorolla DVR-Cable Box
Panasonic Plasma TV
Logitech Z2300 (2.1 Powered Speakers)
* I'll add a dedicated media PC to the mix later
This is a really great setup for a small living room plenty of sound and the Motorolla remote controls all of the components. The only hitch·is that the line out from the TV is not attenuated via the remote control and the mute button has no effect on the line out·either. My first project is to make a remote controlled line level attenuator with mute and make it respond to the remote control.
APPROACH
1) Audio Attenuation
It looks like a chip from TI called a PGA2311 will handle the attenuation of a pair of line level signals and also provides a muting control. It takes a·binary number from 1 to 255 which controls the level of output.
2) Digital Control
I should be able to use any of my Parallax BS2 or SX micros to receive a remote control's output and·produce a binary number to feed to the PGA2311.
So, it looks pretty simple but... I still have to figure out how to decode the signals and put it all together so any help would be great. Wish me luck and Happy New Year!
Andrew
Comments
·
·· Yes, that thread (linked below), sadly, kind of died…It’s really funny that you should bring it up again since I am just now on eBay ordering an Amplifier case (see second link below) to get my amplifier finally into something instead of just sitting as a bunch of circuit boards and wires across the back side of my desk!· Anyway, I will attach the datasheet for the chip you refer to since, on Dave Paton’s advice I ordered some and happen to have it handy.· I will try very soon to continue on that project and share my progress.· Take care.
http://forums.parallax.com/showthread.php?p=527759
http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItem&rd=1&item=7624530988&ssPageName=STRK:MEWA:IT
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Chris Savage
Parallax Tech Support
The IR interface may be a little tougher. The original BS2 I still have in my preamp is a little underpowered, and my code is pretty linear, so it misses the rotary encoder commands sometimes ( I can't spin the wheel too fast). You'll want to pay close attention to the loop time when you're reading and decoding your IR commands to make sure that you don't miss any front panel button presses or anything.
Yup, that thread kinda died, and it was kinda my fault. I've been incredibly busy in the last year and a half, what with rehabbing a house, getting engaged, getting a new job, moving a few times, and getting a mountain of work. I'm still around though, and still play with audio toys from time to time
Chris-
Nice case! That guys also sells some very nice toroids. I'm using one in a client project here at work right now. eBay rules for one-off proto parts.
-dave
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
This is not a sig. This is a duck. Quack.
·
·· So I don’t get off-topic, please see the original thread for something I wanted to follow up with you on…
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Chris Savage
Parallax Tech Support
Thanks for all of the feedback. I wasn't sure that I would get any replies or this quick due to the time of the year and being that the sandbox is a little off topic from robots and the general uController stuff.
BTW: Nice case Chris. EBay is a great place to pick up bits and pieces. I did a little follow up on that original thread and that's where I found your references to the Crystal/Cirrus chip and the 2311. I'll have to get a handful of them. They do look incredibly handy to have around.
Here's the 2311 Datasheet: http://focus.ti.com/lit/ds/symlink/pga2311.pdf
I've been doing a little digging on the Parallax site and found some of Jon and Andy's articles on interfacing the PNA4601 with the BS2 so that's how I'll prototype it. I'm pretty sure I'll move it over to the SX one I get the gist of it working. Just off the top of my head I think I'll read the PNA4601 and then use a table to convert the command number into a command that I can feed to the 2311. For now I think I'll have a loop for up, another for down and 000000 for mute. (I think that's how it works but I haven't reviewed the datasheet in detail.)
I just picked up 500' of Interconnect and 20 RCA ends so that'll keep me busy for a little while. I'm moving all of the equipment to a cabinet and building an IR remote extender first. Then I'll move onto the preamp remote control.
Andrew
·
·· Dave is the one with the PGA2311 experience…I have some, but haven’t connected anything up pending getting the power supply built that I need for all the audio circuitry. ·It would need +/- 5V and +/- 15V.· In any event perhaps Dave (sorry Dave) can post an example line of code to set the volume and give you an idea of how easy it will be. ·Take care!
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Chris Savage
Parallax Tech Support
thatcorp.com/datashts/2181data.pdf
This is the link to the That corp 2181 VCA, which is a very nice sounding VCA that is controlled by a voltage you could produce by the DA on the Prop or other, maybe have to have an externally DA with the Stamp. Thee are some benefits to a VCA, like not having to deal with looking for zerocrossings or other effects produced by digital controlled leveleling(zipper effect for example should the 2311 produce such).
If you are running any distances, you my consider balancing the send and return with some balanced drivers and receivers like these are are great:
http://www.analog.com/en/prod/0,,765_1075_SSM2141,00.html
www.analog.com/en/prod/0,,765_1075_SSM2142,00.html
Chris for power supplies, Apex Surplus in Sun Valley Ca usually has some Lambda +-15volt supplies on hand for cheap, they are great little powe supplies for audio.
Anyways, just some thoughts.
Literally, that's all.
Chris-
I have some regulator boards I made a while back that would be perfect for you...+/- rails with 3x 3 terminal regulators for 3 separate bipolar regulated outputs. Once I get them slit and broken out I'll send you a couple, PM me your address.
-dave
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
This is not a sig. This is a duck. Quack.
Work is kind of getting in the way, but I'll keep you all posted on my progress.
Dave: Thanks for the code example.·I figured it would be that simple.
I'll have to play with the muting circuit. The 2311 has both h/w and s/w methods for muting. S/W just means you feed it a bunch of zeros but H/W is a pin. If the 2311 remembers the volume level and returns there after activating the pin, then·H/W might be a simpler system. If not, I'll have to create a pair of variables that remember the level before mute was hit and then use the S/W zeros method.
I'd also like to be able to turn it off and then back on having it recall the last volume which means that it needs to write the left and right values somewhere.
Anyway, that's it for now.
Andrew
I don’t know Dave’s take on this (he knows the chip better than we do), but many chips I have seen go right from one value to the other instantly. That is, when you press mute the volume is cut and when you restore it jumps right back to full volume. In many cases this can be undesirable. Here are a few points to consider in case this chip doesn’t handle this for you…
Controlling the muting via the hardware will require an extra output from the BASIC Stamp. This may not be a big deal. Doing it from software though will allow you to ramp at your own rate between the current value and mute, then ramp back to the original value. It will require more variable/code space to do this.
My solution (specific to the amplifier chip I am using, LM3886T) comes from a discussion with an Engineer from National Semiconductor in which it was pointed out that if I used an RC circuit on the muting pin, my muting would have a nice curve to it. So in my audio system the muting is done at the amplifier, not in the controller where the line level signals are. I simply make an I/O pin LOW to mute and HIGH to return to normal. Another advantage of this is it helps curb turn-on thump at the output of the amplifier since the relay of OFF at power-up. Take care.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Chris Savage
Parallax Tech Support
* I'm planning on using it with my logitech setup for the time being but in the long run it will be providing volume control for those tube amps I mentioned a while back. I got my dad interested in them again after he retired and now his basement looks like an amp manufacturing outfit.
Based on that thought, I think creating a muting routine in the software with instant mute and soft return to the last value is going to be my best approach.
Thanks,
Andrew
Yeah, definitely bad for the speakers, but before you decide, see what the chip can do or wait for a reply from Dave Paton.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Chris Savage
Parallax Tech Support
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Chris Savage
Parallax Tech Support
I do software musting in my design with the CS3310. I store the last value and the mute value in 2 separate values and load them directly, in response to a flag in the software (flag_mutestate). It's actually a pretty easy thing to work with...I have the mute button change the flag, and as the loop rolls, it either mutes or not, each time through. If I used up and down buttons instead of a quadrature encoder for my volume control, it'd be crazy fast, but because of the quadrature stuff, things are a little slow.
Hardware muting puts more stuff in the way of the audio, which to me is bad. I want a straight wire with gain. Muting is a sore subject with me, so I use the tools I have available, in this case, a really really nice attenuator that can go from 0 to hero (unity gain) in a single shiftout(). I can't think of a reason to do muting any other way, unless you have a ton of analog stuff in the box and need to use relays to disconnect your outputs while it powers up and stabilizes.
As far as separate right and left volume, I don't do that. In the volume control routine, I add the volume level (global) to a trim offset for each channel (my preamp does analog 5.1 volume control) right before I do the shiftout()s. The example I posted was from when I was doing straight stereo with no balance adjustment (the only code I have here at work...the rest is at home). The current code looks a little more like this (from memory, might be off a little):
The trims can run from 0-255, just like the actual volume number, and I do some tricks with swapping trims to mute the surround channels with trim offsets in stereo mode, but that's mainly because the surround stuff was hacked in later, and the code is in a truly scary state right now.
-dave
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
This is not a sig. This is a duck. Quack.
-dave
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
This is not a sig. This is a duck. Quack.
I knew you would have your own methodology and explanation of the chip functions. In case I confused anyone about my setup, the “relay” isn’t interrupting the analog signal at all…The LM3886T has a muting pin right on it. By default the amplifier is already muted at power up. Until you draw current from the pin the amplifier chip remains muted. So my relay controls the voltage on that pin and the RC circuit controls the curve. The muting sounds so much cleaner than any hardware or software muting I ever used. But everyone has different tastes and I always encourage people (Andrew, in this case) to get more opinions than mine before doing something. Very nice though!
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Chris Savage
Parallax Tech Support
·· You just reminded me of something I forgot to mention on the ramping thing…I have a routine which I refer to as a running average loop.· When I make a sudden change to my volume each pass through the loop the new value is averaged with the old value and then assigned to the old value.· The effect is kind of logarithmic and would work perfect for muting.· I wonder if anyone would be interested if I demonstrated this piece of code using the current test hardware, which would only require an ADC0831, POT and an LCD (or you could use DEBUG)?
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Chris Savage
Parallax Tech Support
Your muting scheme with the mute pin on the 3886Ts makes perfect sense, and when I eventually do build my integrated amp (with LM3886TFs actually), I'll use something similar. I was speaking specifically to the case of a preamp only device, which wouldn't have the (very well thought out) 3886T mute feature. I was mainly trying to make the point that muting with FETs or transistors is a messy process, especially when dealing with oddball topologies like I do (fully balanced and servo'd). The only muting schemes I consider 'audiophile' are done with in the 3310/2311/whatever, done at the amp like with the LM3886 mute pin, or done with a DPDT relay that disconnects an audio out and ties the output jack to ground through a resistor (1k usually). Anything else will molest the signal to a degree I find unacceptable, and I've tried a lot. My very opinionated $0.02.
The running average loop example sounds really cool, I'd like to see that. It's been a while since I had my fingers deep into PBASIC (C has been the rule of the day at work for a while, and I haven't been keeping up with SX/B like I wanted) so it'd be a good refresher for me too
-dave
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
This is not a sig. This is a duck. Quack.
I just sent myself an e-mail reminder and I will definitely post that code this weekend since I will be working on the amplifier anyway. Yeah, I was going to use an encoder, as you are, however I wasn’t happy with the performance for certain things such as volume control, so I instead implemented a standard POT connected to an ADC0831 to get a direct digital value to feed to the audio control chip. The problem is that sometimes I get jitter in the readings.
I was going to filter them at first using a bump technique where you specify a dead band in which the value doesn’t change until you exceed that. But then I thought of averaging the values out and subtle changes wouldn’t make that big of a difference. I have proven that the subtle changes won’t affect the audio output though so I put it on the back burner.
And interesting effect of this technique is that when I very quickly go from one extreme to another the value starts a slow ramp which increases almost exponentially until it reaches the new value. The effect should be quite interesting for audio and should be more pleasing than a sudden change. Yet it happens so quickly there is very little lag. Obviously as more routines are added to the main loop this could change, but in the current test code it looks good.
As for the LM3886T muting pin…As I said, since I use it anyway for startup, it works well for muting. With the RC circuit you get a smooth fast change. The engineer from National suggested this which is often used on other integrated amplifiers…When the power button is pressed on my system an I/O pin goes HIGH turning on the main Power Relay in the Power Amp box. That energizes the main Power Transformer. About 2 seconds later a small PCB Relay energizes connecting the muting pin with the VEE line turning off muting (enabling the outputs). When I press the mute button it simply turns that relay off, which is just removing current from the muting pin on the LM3886T. It works great. When I click the power button to shut the system down it de-energizes the Muting Relay first and then the Power Relay 1 second later eliminating shut down transients. Works very nicely.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Chris Savage
Parallax Tech Support
I was trying to find a straight forward way of demonstrating my cumulative average technique but creating a real time loop with nothing to do was making it difficult. So what I have done is created a small program that demonstrates the concept by choosing a pseudo random value then averaging it in a loop with the current position. It does this for 20 iterations just to simulate the point.
After that it pauses for one second so you can see the resultant value then runs another random value through the 20 iterations. Bear in mind that within a real system this formula would be running in the main loop continuously and receiving real-time changes. I will come up with something once I perfect the routine without adding a lot of computational overhead. As it stands now the target doesn’t always match due to the way the average is calculated, but you get the idea of what I mean. I am open to questions and/or comments. Take care.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Chris Savage
Parallax Tech Support
-dave
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
This is not a sig. This is a duck. Quack.
That looks interesting. I'll have to wait to play with it though, work is insane today, and my personal life is a bit full for the next day or two.
Thanks!
-dave
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
This is not a sig. This is a duck. Quack.
No problem…The information is there…As I said, it’s not even perfected at this point. But the concept is what I am testing…I never did get anything done with my new case this weekend…
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Chris Savage
Parallax Tech Support
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Chris Savage
Parallax Tech Support
I hear you!
I prefer to avoid shunting of any kind. Given the niceness of the available parts in the preamp in question (CS3310/PGA2311), I can't figure a reason for any other kind of muting in the box, short of a hard bypass or disconnect with relays (something I can't really rationalize for myself, but I'm sure someone else could).
Specifically, as Chris noted, shunting the preamp output/amp input to ground dumps the preamp output to GND directly, which is a nono to me. If the preamp output was disconnected, that puts a switch inline with the audio, which is also something I like to avoid. Short of a big isolation resistor inline with the output of the preamp, but before the shunt (100k or so) and a buffer after the shunt (to provide the required low-z output to the amplifier) I'm not sure how to do it. Of course, adding those extra parts violates my straight wire with gain principle, so I choose to stick with the active volume control unit for muting.
This is a really interesting discussion...nothing personal here, just differing engineering opinions. Let's keep going! Input switching? Gain balancing? Balanced vs single ended?
-dave
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
This is not a sig. This is a duck. Quack.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Chris Savage
Parallax Tech Support
Now I have to put some time aside to breadboard this thing. I've been reading up on the 2311 and it looks like a really nice piece. I think I'll S/W mute it and ramp it back up to the last volume when I un-mute. I'll keep you posted on how it works out.
Thanks for all of the feedback and discussion.
Andrew
Glad you got your parts. You’ll probably have it wired up and tested before me. I still need to get a dual supply up and running for this. Take care.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Chris Savage
Parallax Tech Support