Go Back   Home > Forums > Source & Line > Analog Line Level

Analog Line Level Preamplifiers , Passive Pre-amps, Crossovers, etc.

Please consider donating to help us continue to serve you.

Ads on/off / Custom Title / More PMs / More album space / Advanced printing & mass image saving
Reply
 
Thread Tools Search this Thread
Old 23rd March 2011, 01:56 AM   #91
Account disabled at member's request
 
Join Date: Aug 2007
Nice job ! Looks much better than mine ;-)

You were right about the noise above 0dB. As soon as the chip goes into gain the noise ramps up dramatically; from normal snr to audibly hissy. Horrible.

The other thing I've found is that there seems to be some noise-decay (noise like the ticking of a clock) when the signal is suddenly interrupted at high gain. I get a slight thock-thock-thock that decays to nothing very rapidly. It's very quiet but audible at high gain. This is nothing to do with changing settings. It's simply the source being paused. I wonder if it might be the source and volume control interacting in a weird way, or if it is the PGA itself turning itself down to control offset or something, and that digital noise is getting into the analogue cos of the poor PCB.

Anyway, the chip does a good job up to 0dB. It's only above 0dB when it is a poor performer. So if anyone needs gain, use a separate op amp as this chip is dire for gain.

Also, I forgot to mention, the relays in the kit took a tap to get them to work initially, but have worked flawlessly since. So upgrading them might be worthwhile.

I had a Mini years ago that had the same fault on the starter solenoid. To my friends great amusement, I used to have to open up the hood/bonnet and hit it with a tap hammer to get the ignition to switch on. Not funny with electronics though.

Tom
  Reply With Quote
Old 23rd March 2011, 02:17 AM   #92
diyAudio Member
 
Join Date: Jul 2008
Location: santa clara, CA
I hear the ticking as you go above 0db into gain but only when you vary thing. once you leave the knob alone things are ok.

maybe you have dc offset coming thru? or not good (perfect) grounding?

if you are using the controller that came with the ebay toy, that could also be kicking back junk into the psu. my board had the luxury of having a private 7805 for the cpu and a set of 7805/7905 for the pga analog/digital path. if yours does not have a cpu regulator, maybe try that. and play with grounds.
__________________
My Photostream:http://www.flickr.com/photos/linux-works/
  Reply With Quote
Old 23rd March 2011, 02:43 AM   #93
diyAudio Member
 
Join Date: Jul 2008
Location: santa clara, CA
pic of my perf-board port expander:

Click the image to open in full size.

took me a few hours to build that. its essentially half of a delta1 controller, just 1 port expander chip, some debug led segments and address jumpers.

I added a 2nd i2c set of headers as a pass-thru port. in case you want to daisy chain a switch or some other i2c controlled device to this.
__________________
My Photostream:http://www.flickr.com/photos/linux-works/
  Reply With Quote
Old 23rd March 2011, 02:55 AM   #94
Account disabled at member's request
 
Join Date: Aug 2007
Quote:
Originally Posted by linuxworks View Post
I hear the ticking as you go above 0db into gain but only when you vary thing. once you leave the knob alone things are ok.

maybe you have dc offset coming thru? or not good (perfect) grounding?

if you are using the controller that came with the ebay toy, that could also be kicking back junk into the psu. my board had the luxury of having a private 7805 for the cpu and a set of 7805/7905 for the pga analog/digital path. if yours does not have a cpu regulator, maybe try that. and play with grounds.
Hi,

No offset and the grounding is starred so I'm going with mcu noise and/or needing independent grounding.

cheers
  Reply With Quote
Old 23rd March 2011, 03:16 AM   #95
diyAudio Member
 
Join Date: Apr 2008
Quote:
Originally Posted by KlipschKid View Post
Hi,

No offset and the grounding is starred so I'm going with mcu noise and/or needing independent grounding.

cheers
Sounds like a grounding or offset issue to me.




BTW, Linuxworks, which port expander chip did you use, and how did you wire it up? I'm new to this sort of digital business.
  Reply With Quote
Old 23rd March 2011, 05:30 AM   #96
diyAudio Member
 
Join Date: Jul 2008
Location: santa clara, CA
pcf8574. I'm using those a lot since they are multi sourced, cheap, fast and don't have too bad a layout. no chip overhead; each byte over i2c is the byte you get. no setups needed, like the mcp chips need.

pinout is data sheet. or use this for a ref:

AMB Laboratories DIY Audio • View topic - ?1 links

and this

The LCDuino-1 I/O processor - Page 23 - Head-Fi.org Community

the schematic for the delta1 and 2 use the same controller.

my bargraph shows amazing similarity (lol) and is exactly half that controller, minus the relay driver ULN chip.

you can bit-bang over i2c well enough. that means you send an i2c *message* for each BIT you want to change. when you send the typical dialog, chip-select, data, clock, data, clock - each of those is a bit 'wiggle' and that means an i2c byte is sent over. the byte is the new bitmask and you just replace old values with new values. if a bit does not change, you send its old value and it never 'bounces' until and unless you send a new bit value in the mask, to the PE.

Code:
void 
pga2311_set_volume (byte left, byte right)
{


    // strobe chip-select
    pcf.write(6, PGA_I2C_CS_HI | PGA_I2C_MUTE_HIGH);        // start from high (logical-NOT on chip-select)
    pcf.write(6, PGA_I2C_CS_LOW | PGA_I2C_MUTE_HIGH);       // begin getting the chip's attention

    // write 2 bytes of data
    pga23xx_write(left);                      // left value (0..255)
    pga23xx_write(right);	              // right value (0..255)

    // unstrobe chip-select
    pcf.write(6, PGA_I2C_CS_HI | PGA_I2C_MUTE_HIGH);
  return;
}



void 
pga23xx_write (byte out_byte)
{
  int i;

#ifdef DEBUG_PGA
  Serial.print("PGA:");
  Serial.println((int)out_byte);
#endif


  // loop thru each of the 8-bits in the byte
  for (i=0; i < 8; i++) {

    // strobe clock
    pcf.write(6, (PGA_I2C_CS_LOW | PGA_I2C_SCK_LOW | PGA_I2C_SDATA_LOW | PGA_I2C_MUTE_HIGH) );

    // send the data_bit (we look at the high order bit and 'print' that to the remote device)
    if (0x80 & out_byte) {  // MSB is set
      pcf.write(6, (PGA_I2C_CS_LOW | PGA_I2C_SCK_LOW | PGA_I2C_SDATA_HI | PGA_I2C_MUTE_HIGH) );
    
    } else {
      pcf.write(6, (PGA_I2C_CS_LOW | PGA_I2C_SCK_LOW | PGA_I2C_SDATA_LOW | PGA_I2C_MUTE_HIGH)) ;
    }


    // unstrobe the clock
    if (0x80 & out_byte) {  // MSB is set
      pcf.write(6, (PGA_I2C_CS_LOW | PGA_I2C_SCK_HI | PGA_I2C_SDATA_HI | PGA_I2C_MUTE_HIGH) );
    } else {
      pcf.write(6, (PGA_I2C_CS_LOW | PGA_I2C_SCK_HI | PGA_I2C_SDATA_LOW | PGA_I2C_MUTE_HIGH) );
    }
    

    // get the next bit  
    out_byte <<= 1;   // left-shift the byte by 1 bit
  }

  return;
}
__________________
My Photostream:http://www.flickr.com/photos/linux-works/
  Reply With Quote
Old 23rd March 2011, 07:09 PM   #97
diyAudio Member
 
Join Date: Jul 2008
Location: santa clara, CA
better photo of the current test bed. 4 channels (out of 6) are wired up. and chipped up, too, I guess

Click the image to open in full size.

the box is just too small to do 4 more rca's on the back. right now, I have my dcx2496 splitting into 2-way with my subwoofer taking the LP and my tannoys taking what's above it. this 2-way volume control works fine for that. and I don't feel like redoing the back plastic work - it was already a re-use of a previous project

2 chips seem to be working ok. I have not tried the 3rd yet and I've recently seen an ebayer (I think the only one with this board) show this board with a wire on the back and it looked like it was going toward the 3rd chip. maybe that's the time bomb (lol) that I haven't come across yet. maybe I'll stop at 2 and consider myself lucky.

I really hate the layout of this board. digital signal real close to analog ones; no groundplaning at all near the sensitive chips, no bypassing other than a SINGLE electrolytic, just not much good to say other than it does electrically work and lets a software guy play around I would not deploy this kind of board in production, though. its a POS. but for software, even POS is fine (lol).
__________________
My Photostream:http://www.flickr.com/photos/linux-works/
  Reply With Quote
Old 27th March 2011, 05:47 AM   #98
diyAudio Member
 
Join Date: Jul 2008
Location: santa clara, CA
today, AMB (my co-design partner) and I did an rmaa on this box. we were both surprised. "its clean"

I'll post a link to the rmaa results but IIRC, the noise was down something like 100db or more. definitely 'clean and quiet' for most ears. even with this 'badly laid out' PGA board, we were getting very decent specs at 0db and even 3db (yes, with gain). not much distortion, IM or anything bad. this is without even doing the 'required' bypassing that BB and cirrus recommend.

so I guess this 'test bed' is actually quite a bit more. its almost production sound quality

leaving out the caps and the level shifting resistors was not a bad idea. leaving out the relays also didn't hurt (I've heard these are really cheap relays that ship with this kit).
__________________
My Photostream:http://www.flickr.com/photos/linux-works/
  Reply With Quote
Old 27th March 2011, 06:41 AM   #99
diyAudio Member
 
Join Date: Apr 2008
Very nice, i've been thinking of designing a board for this chip that has room for decent caps and utilises proper grounding, nice to know even a sub-standard pcb can give good results.
  Reply With Quote
Old 27th March 2011, 07:46 AM   #100
diyAudio Member
 
Join Date: Jul 2008
Location: santa clara, CA
surprised the hell out of me, it did

by all accounts, it should suck. and yet it did not. I will repeat the rmaa test but once we got windows to finally see the firewire audio (test) box, its test run showed really nice results.

I think its easy enough to upgrade the 7805 series chips to 317, at the very least. it can't hurt. and proper bypass (electro AND a .1 of some kind) should also be done. again, I can't imagine it would hurt and could only help.

I did use all my own parts, though. I used the board from ebay but skipped using his caps, his chips and his sockets. I will try his chips later on and see if they test any differently, but I wanted to use known good chips, first, to at least give this critter a fighting chance.

as I posted on a 'digital line level' thread, I also have my DIY spdif switch integrated with this entire system:

Click the image to open in full size.

the 'pod' on the right is a collection of 3 toslink inputs and 1 toslink output. you go into the LCD menu, select 'spdif' as the i/o engine and whammo - your input ports now do spdif routing on that remote pod. locate the pod near your dac or other digital things and connect the pod to the front-facing lcd/ir box via an i2c cable. details to be worked out later as to connector types, but I'm thinking of overloading the firewire (6pin) connector and that gets me pre-made easy and cheap cables and less soldering for me to worry about there should be enough 'specs' in the firewire cable to carry my i2c connection of clock+data+power over to the pod box.

this is what I'm using now as my main audio/video system. this little box and its pod have replaced my previous AVR. (note, I don't do multichannel at home; I downmix all my movies to 2.0 so that they are spdif compliant and not dd or dts stuff). when I select 1 input, it picks the right spdif port and also adjusts volume to that port's last used setting. switch to another port and you get the spdif-feed from my HTPC or media streamer and it corrects the volume to last-set value for that port, etc.

I have a hack for linking in an hdmi switch, too; but more about that later.
__________________
My Photostream:http://www.flickr.com/photos/linux-works/
  Reply With Quote

Reply


Hide this!Advertise here!
Thread Tools Search this Thread
Search this Thread:

Advanced Search

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off


Similar Threads
Thread Thread Starter Forum Replies Last Post
Volume control on DVD player remote. Lose resolution at low volume? Circlotron Digital Source 2 10th June 2011 07:20 PM
PGA 2311 Test Circuit HarryK Chip Amps 13 15th March 2011 09:42 AM
Integrated AMP with PGA 2311 preamp kpero Chip Amps 5 26th April 2009 07:05 PM
Remote volume control for existing remote? theAnonymous1 Parts 7 6th August 2007 08:22 AM


New To Site? Need Help?

All times are GMT. The time now is 04:47 AM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2014 DragonByte Technologies Ltd.
Copyright ©1999-2014 diyAudio

Content Relevant URLs by vBSEO 3.3.2