Jump to content

CL-D Bi and Tri-amping (lots of photos and plots)


John Warren

Recommended Posts

The CL-D discussion got me thinking about buffers so I decided to put a thought to hardware.

 

The idea here is to split the small signal, say from the preamp or CD, into three "sub-channels".  Each split can then go to a LF, MID or HF active filter and a CL-D amplifier "sized" for that driver.  Alternatively, each split can go straight to a sized CL-D amp and then to a passive network connected its respective LF, MID or HF driver.

 

Doug Self designed a buffer to do the same thing but this one has less parts, on board +/-15VDC rails, XLR combo inputs and really nice Mill-Max IC sockets.  It might not be as quiet as Mr. Small's design claims to be but this design is pretty quiet, distortion is very low and it's got a ton of output current capacity (each "split" will provide +/-22mA across a 600Ohm load with +/-14VDC rails!).  

 

Ideally, having each split feeding an active filter for a woofer, mid and tweeter and then outputting the contoured small signal to a CL-D amp sized for the particular driver is where I want to take this idea (the filters will be a separate thread)

 

It's a stacked op-amp configuration, unity gain, voltage follower based on the LME49720 package.  I designed the circuit with RFI immunity built in.  Input impedance is between 9-10kOhms for unbalced and 18-20kOhm for balanced.  The power supply I'm testing it with is something I had from a previous job, a bel +/-24VDC, 600mA, linear supply (HAA24-0.6-AG).  They're made in China but good quality, low ripple and moderately priced at $60.  The +/-24VDC powers the +/15VDC supplies on the board.  Any DC supply capable of up to about 80V can be used.

 

First three photos are of the board, board with hardware and assembled.

 

buffer_5.jpg

 

buffer_8.jpg

Link to comment
Share on other sites

  • 2 weeks later...

Final (through hole and socketed) design shown below.  After working with the prototype I revised chassis and circuit ground paths, added RFI/EMI common and differential mode front end filtering, reduced feedback factor to about 2X, added additional filtering to supply rails and re-routed a few of the traces.  For this build bandwidth across RL=10kOhm is -3dB @ 80kHz.  

 

With headphones, a listen with either an LME49720 or NE5532 evidences nice, quiet background.  Buffer/splitter takes balance or unbalanced signal and splits it three-ways for tri-amplifier arrangement.  The board shown is two channel board so six splits in total, three/channel.  Each split can feed a separate amplifier or active filter.  Input impedance is about 18kOhm.  Output impedance is about 600Ohm with the LME package and lower with the 5532.  Resistors are 0.01% (lower will improve circuit CMRR, theoretically).  The plots are with through hole components and socket mounted packages, power supply is unit described above.  Might be a nice kit for builders.  The SMD version is much tougher to solder up.   

R3.0_build_1.jpg

bandwidth_RL=10kOhm.jpg

FFT_RL=10kOhm.jpg

%THD_RL=10kOhm.jpg

Link to comment
Share on other sites

  • 2 weeks later...
9 hours ago, CECAA850 said:

Me too.  So this is a splitter and each channel is identical to the original signal nor do they split into different frequencies?

 

Yup.  Each channel split up into three identical signals (essentially identical, there's always a bit of a difference).   There's some filtering at the front end for RFI but, other than that, no signal conditioning across the audio range.    

 

The webpage isn't complete.  I have some content related to common mode rejection and resistor selection that I'd like to publish.   I'd such a nice job of selecting resistors the CMR isn't easily measured however.  

 

Link to comment
Share on other sites

15 hours ago, John Warren said:

 

Yup.  Each channel split up into three identical signals (essentially identical, there's always a bit of a difference).   There's some filtering at the front end for RFI but, other than that, no signal conditioning across the audio range.    

 

The webpage isn't complete.  I have some content related to common mode rejection and resistor selection that I'd like to publish.   I'd such a nice job of selecting resistors the CMR isn't easily measured however.  

 

The advantages of using this over, say, a couple of RCA splitters?

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

×
×
  • Create New...