Optimizing TDA7294 Output - Page 47 - diyAudio
Go Back   Home > Forums > Amplifiers > Chip Amps

Chip Amps Amplifiers based on integrated circuits

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 15th May 2013, 07:52 PM   #461
diyAudio Member
 
danielwritesbac's Avatar
 
Join Date: Nov 2007
Thank you so much for the testing.
In the audio band, the testing shows:
1). Almost flat with 1n
2). Droops inside the audio band with no comp (and a droop at 20khz will damage the harmonics that support 10khz, thus supporting Bob's comment about inferior treble described as "no sparkle").

The "Here is 2.7nF and 1nF" chart shows that 1N+18R or 470p+15R is likely to have a flat response within the audio band. . .
On your board, not everyone's board.
Need of compensation changes with board layout. It is a "moving target" problem. The schematic that you found is for the Jim's Audio TDA7294 single layer mess. However, it is noticeable that the Parallel TDA7293 board doesn't use lead lag comp at all.

Here's a TDA7294 schematic schematic without added comp.
Attached Images
File Type: gif TDA7294-Adjustable2.gif (33.5 KB, 458 views)
__________________
Tools, Models & Software for DIYClipNipper boostLM1875 TurboPowerful TDA7293 kitTDA7294 pt2pt ♦ My post has opinion.

Last edited by danielwritesbac; 15th May 2013 at 07:56 PM.
  Reply With Quote
Old 15th May 2013, 09:13 PM   #462
tyger23 is offline tyger23  United States
diyAudio Member
 
Join Date: Nov 2010
Quote:
Originally Posted by danielwritesbac View Post
Thank you so much for the testing.
In the audio band, the testing shows:
1). Almost flat with 1n
2). Droops inside the audio band with no comp (and a droop at 20khz will damage the harmonics that support 10khz, thus supporting Bob's comment about inferior treble described as "no sparkle").
You need to spend more time looking at the scale of my drawings.

With no compensation: -0.004dB down at 20Hz, -0.1dB down at 20KHz.

With 15r+1nF = the boost OUTSIDE the audio band will greatly affect the harmonics inside the audio band.

Note: the droop at 20KHz is due to the preamp section inside the AUDIOSOURCE, not due to the TDA7294.

Additionally the overdamped filter created by the "snubber" circuit is yanking quite a bit on the phase, throwing the treble section out of phase by several degrees in BOTH directions (even when using the 15r+1nf).

Perhaps more experiments can be run with a lower resistance factor, but my opinion remains - why in the world would you want to ADD a filter that will only negatively affect the purity of the audio signal? Any filter that gets added cumulatively adds to phase deviations, even if it makes the actual response "flatter".

IMHO, the audio should be as pure as possible. Any "colorations" that make the treble seem brighter or the midrange more muddy is likely due to phase deviations, distortion, or other non-PURE audio signals. Now, that's not to say that one person may or may not like the way those phase deviations sound. SRS makes all of their money by yanking on the phase in the treble region to make things seem brighter. Some people like this. That's why SRS stays in business. I personally find phase deviations to be fatiguing over time, and I turn off any SRS, Dolby, DTS, or Beats processing that I find on any of my systems.
  Reply With Quote
Old 15th May 2013, 10:02 PM   #463
JMFahey is offline JMFahey  Argentina
diyAudio Member
 
JMFahey's Avatar
 
Join Date: Mar 2009
Location: Buenos Aires - Argentina
Clearly lead-lag compensation is a bad idea in this case.
Thanks for measuring and posting
  Reply With Quote
Old 15th May 2013, 10:17 PM   #464
diyAudio Member
 
keantoken's Avatar
 
Join Date: Aug 2006
Location: Texas
Blog Entries: 2
What is your source impedance and how does this affect the measurements? Is response okay if the volume pot is removed?

I was wondering why a chipamp like this would need any kind of external compensation when it is supposed to already be stable.

Since Dan thought the chip was unstable, I gave him instructions on how to apply this kind of compensation. I was worried he would have trouble without a scope but he was willing to try it out. I wonder how much of its effect comes as an EQ control and how much comes from an effect on stability or RFI tolerance.
  Reply With Quote
Old 15th May 2013, 11:19 PM   #465
tyger23 is offline tyger23  United States
diyAudio Member
 
Join Date: Nov 2010
Keane,

To answer your question, I quickly drew up what the preamp section of this HEAVILY modified AudioSource amp looks like.

The topology of the factory amp is the same, but I've heavily modified the values to achieve a much flatter response than normal.

The droop you see in my non-compensated response is actually due to two different things:
1. The R2/C2 combination
2. The C5 value through the U2 op-amp

These values are what is currently populated and being measured. I have no idea why audiosocure wanted to boost the output about +14dB through the op-amp, then immediately yank it back down to 0dB with R10, but they did.

Anyways, I hope this helps. The TDA7294 appears to be waaaay more than stable (at least in this design).
Attached Images
File Type: jpg preamp.jpg (94.5 KB, 211 views)
  Reply With Quote
Old 15th May 2013, 11:55 PM   #466
diyAudio Member
 
keantoken's Avatar
 
Join Date: Aug 2006
Location: Texas
Blog Entries: 2
So output impedance is at most 1.5k. With such a large RC across the inputs, input impedance can be changed enough to cause problems with reasonable source impedances. I'm not sure how much that applies in this case.

You may think about trying to lower the source impedance. With many poweramps a large source impedance will bring out more distortion.
  Reply With Quote
Old 16th May 2013, 04:23 AM   #467
tyger23 is offline tyger23  United States
diyAudio Member
 
Join Date: Nov 2010
The LM4562 has a closed-loop impedance of 0.01 ohms. I don't think I can lower it much past that. .
  Reply With Quote
Old 16th May 2013, 04:41 AM   #468
diyAudio Member
 
danielwritesbac's Avatar
 
Join Date: Nov 2007
Quote:
Originally Posted by JMFahey View Post
Clearly lead-lag compensation is a bad idea in this case. Thanks for measuring and posting
Absolutely true and totally expected--Compensation needs can change with board layout.
Much earlier, I had mentioned that the lead lag comp was put on because of one specific board with bad layout; specifically it was the Jim's Audio single layer TDA7294 board which behaved so badly I thought the chip was fake. Yes, that board has a rather annoying influence--Removing the board also removes the problems.
Testing results seem to indicate:
I need to retry the Jim's Audio Single Layer TDA7294 Board, remove the lead lag effect, and then try to work out the rest with power circuit mods for correcting the layout issues. Maybe a full scale implementation of Tom's and KSTR's iteration array daughtercards could help remove/dampen the sonic weapon coloration of that one particular board layout--the Jim's Audio single layer TDA7294 that needs either a timely disposal or much more help.

Compare:
There's no lead lag comp used on the TDA7293 Parallel board. Likewise, the AudioSource AMP-100 board doesn't need lead lag comp. Apparently, those two amplifier boards have a useful layout.
Quote:
Originally Posted by tyger23
The droop you see in my non-compensated response is actually due to two different things:
1. The R2/C2 combination
2. The C5 value through the U2 op-amp
There is one more case, where I'd expect to see exactly the treble droop shown, and that case is: If the "1u or 0.47u electrolytic" bypass cap were omitted (at the nfb-shunt cap area), then I'd expect a small treble droop. I'd suggest to test drive some bypass caps to see if you get a flatter response and clearer sound. If your large size NFB-shunt cap is of most excellent quality then adding the "1u or 0.47u electrolytic" might be overkill, in which case a little green polyster dip cap with values such as 22n or 10n or smaller may do.
Quote:
Originally Posted by tyger23
Anyways, I hope this helps. The TDA7294 appears to be waaaay more than stable (at least in this design).
Yes, I like cool and efficient. Thank you so much for checking the stability and amplifier behavior!
SO, my question is, once the comp (that your board didn't need) was removed, then did you find the rest of the design suitable? And, what changes, if any, would you suggest?
__________________
Tools, Models & Software for DIYClipNipper boostLM1875 TurboPowerful TDA7293 kitTDA7294 pt2pt ♦ My post has opinion.

Last edited by danielwritesbac; 16th May 2013 at 04:57 AM.
  Reply With Quote
Old 16th May 2013, 06:24 AM   #469
diyAudio Member
 
keantoken's Avatar
 
Join Date: Aug 2006
Location: Texas
Blog Entries: 2
Quote:
Originally Posted by tyger23 View Post
The LM4562 has a closed-loop impedance of 0.01 ohms. I don't think I can lower it much past that. .
Look at your own schematic. It has a 10k resistor in series with the output, only mitigated by the 1.5k shunt resistor!
  Reply With Quote
Old 16th May 2013, 06:56 AM   #470
tyger23 is offline tyger23  United States
diyAudio Member
 
Join Date: Nov 2010
Quote:
Originally Posted by danielwritesbac View Post
SO, my question is, once the comp (that your board didn't need) was removed, then did you find the rest of the design suitable? And, what changes, if any, would you suggest?
The one suggestion I would recommend is to change that input cap. You currently show a 1uF with a 27K resistor. If you check out the bode plot for this filter, you'll see that it will cause a 0.35dB droop at 20Hz (which isn't too bad), but the bad part is that you're 17 degrees out of phase.

I would recommend 22uF or greater. I personally use a 33uF when there's a 27K input impedance. Attached are the simulations for a 1uF, 10uF, and 33uF input cap. Alternatively, I guess you could change the input impedance. Enjoy!
Attached Images
File Type: jpg 1uF.jpg (146.6 KB, 189 views)
File Type: jpg 10uF.jpg (169.9 KB, 166 views)
File Type: jpg 33uF.jpg (173.4 KB, 158 views)
  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
OPtimizing the VBE MUltiplier fglabach Solid State 45 17th September 2014 08:22 PM
Optimizing active crossover atledreier Multi-Way 27 19th August 2011 03:53 PM
optimizing the VBE multiplier hienrich Solid State 80 29th November 2010 11:25 PM
Interesting linestage optimizing ideas PixelPlay Tubes / Valves 5 23rd June 2008 11:18 PM
optimizing for efficiency Dwiel Solid State 13 10th May 2005 10:31 PM


New To Site? Need Help?

All times are GMT. The time now is 01:51 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