Objective Device Testing - Anything Electronic or Electro-machanical - Page 4 - AVS Forum
Forum Jump: 
Reply
 
Thread Tools
post #91 of 109 Old 06-06-2014, 03:21 AM
Newbie
 
electrovert's Avatar
 
Join Date: Jun 2014
Posts: 12
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10

Hello...First I must confess I haven't read the full thread yet!  Please for give me...I  will shortly.  But I've been looking for the schematic for this amp for nine years...Long story short.  I bought this amp and dove in blind modding it.  Yup no clue how it work.  Just knew I didn't like the sound.  Long story short ops said that.  Just finished the second go around and they sound fantastic.  If you want my blind theories as to why they sound great (yeah I have a pile of them) you can read my thread on Audio Asylum.  Here

 

http://www.audioasylum.com/cgi/vt.mpl?f=amp&m=194242

 

 

Happy tweaking

 

Thanks for posting the schematic.

electrovert

electrovert is offline  
Sponsored Links
Advertisement
 
post #92 of 109 Old 06-06-2014, 01:09 PM
Newbie
 
electrovert's Avatar
 
Join Date: Jun 2014
Posts: 12
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Maybe everyone got bored of this thread.  Correct me if I'm wrong but in a Class AB amp there is only a small amount of Class A operation, beyond which only half of the totem pole is conducting.  Which means you could consider it out of circuit or open.  This amp very obviously has no center tap from the transformer.  So all currents are coupled through the large caps.  So the large caps have to store charge and couple the signal. Double duty IMHO.  The question I would like answered is: Berhinger calls this a "SERVO-CONTROLLED" amplifier...can anyone shed some light on what that means? Look like global feed back and nothing much else from the schematic.
 
See my MOD/thouughts for this amp on the link above.
 
Thx

Last edited by electrovert; 06-13-2014 at 01:15 PM.
electrovert is offline  
post #93 of 109 Old 06-06-2014, 01:26 PM
AVS Special Member
 
Zen Traveler's Avatar
 
Join Date: Feb 2005
Location: Texas
Posts: 4,150
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 131 Post(s)
Liked: 318
I think everyone is waiting on this:
Quote:
Originally Posted by Guy Friendly View Post

Hi  Amir:

Yes, I have ordered the amp. I was told 2-3 weeks, the first week has almost passed, so just a while longer then I'll have it in my hands.

When it arrives, I plain to burn the amp in, so I won't be engaging it subjectively or objectively, for a day or so after burn in.

When it arrives, I'd be happy to take pictures of the PCB layout for you...? Yes - No?

Regards,

GF

PS After I'm done uploading my review, I'll be happy to perform any tests that you or others request; I'll even record some video footage, if you will.

Looking at the date it should be soon...
Zen Traveler is offline  
post #94 of 109 Old 06-06-2014, 04:47 PM
Newbie
 
electrovert's Avatar
 
Join Date: Jun 2014
Posts: 12
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10

Here is a more general question, but originated with the A500

 

Does negative feed back compensate for supply ripple and if so would reducing it improve signal detail?

 

Put another way.

 

 If there is a finite amount of correction resultant from a given amount of feedback will decreasing supply ripple increase signal resolution?

 

I think I worded that right...

electrovert is offline  
post #95 of 109 Old 06-06-2014, 06:59 PM
AVS Special Member
 
DonH50's Avatar
 
Join Date: Feb 2010
Location: Monument CO
Posts: 6,087
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 159 Post(s)
Liked: 262
Signal-path feedback does not generally change power-supply rejection, though it depends upon where the power-supply noise is coming from and where it is introduced in the circuit.

"After silence, that which best expresses the inexpressible, is music" - Aldous Huxley
DonH50 is online now  
post #96 of 109 Old 06-06-2014, 09:30 PM
Newbie
 
electrovert's Avatar
 
Join Date: Jun 2014
Posts: 12
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10


Here is what I am what I am getting at.

  A convention supply is a transformer, full wave bridge, and capacitors. 

With an AC line frequency of 60 Hz the caps get topped up at the peaks of the AC phase cycle which is 120 Hz. 

So the supply would have to deliver ~10 cycles of 1.2KHz signal before it gets topped up. 

If the transistor conducts the same for every output cycle (tracks only the input) then the output should fall off. 

 

Point>

 

 Negative feed back is being abused to cover up poor regulation in the supply and hurting the signal quality... But is never heard as 120Hz or hum in the output because in the case of the A500 it has no ground other than the input ground.

 

The mod I added solves this problem.  IMHO

 

These are generalities taken for simplicity...Giant high voltage cap supplies can store huge amount of power which regulate well at low power, and don't have to (as much) at high power.

electrovert is offline  
post #97 of 109 Old 06-06-2014, 09:37 PM
Newbie
 
electrovert's Avatar
 
Join Date: Jun 2014
Posts: 12
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10


I could put it another way.  If you want to fly straight through curved space you have to fly in a curve.  Space being the energy reserve, and the path the signal.  The transistors have to be compensating for the supply fluctuations and that has to be in the negative feed back...which has a finite resolution...therefore the signal detail is eroded.  IMHO 

electrovert is offline  
post #98 of 109 Old 06-06-2014, 10:09 PM
AVS Special Member
 
DonH50's Avatar
 
Join Date: Feb 2010
Location: Monument CO
Posts: 6,087
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 159 Post(s)
Liked: 262
I am not sure I followed that, and what i could follow does not match my engineering knowledge base. In any event enjoy your mods. - Don

"After silence, that which best expresses the inexpressible, is music" - Aldous Huxley
DonH50 is online now  
post #99 of 109 Old 06-06-2014, 11:49 PM
Newbie
 
electrovert's Avatar
 
Join Date: Jun 2014
Posts: 12
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10

I looked at the schematics posted earlier in this thread and it's worse than I thought.  R6 & R7 (A500)   & TR23 (QSC) are the ripple compensation applied directly with the negative feed back to the front end of the amp.  But what I put forward above should hold for most conventional amps... 

electrovert is offline  
post #100 of 109 Old 06-07-2014, 06:35 AM
AVS Addicted Member
 
arnyk's Avatar
 
Join Date: Oct 2002
Location: Grosse Pointe Woods, MI
Posts: 14,381
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 747 Post(s)
Liked: 1161
Quote:
Originally Posted by electrovert View Post

Here is a more general question, but originated with the A500

Does negative feed back compensate for supply ripple and if so would reducing it improve signal detail?

Put another way.

 If there is a finite amount of correction resultant from a given amount of feedback will decreasing supply ripple increase signal resolution?

I think I worded that right...

Power supply ripple has two major undesirable outcomes:

(1) Maximum undistorted output varies with the ripple. When the amp is running at full output with a sine wave, there are usually at least a few volts of ripple. That means that the power supply voltage is varying by several volts, and thus the peak undistorted output is varying by several volts.

(2) The ripple feeds though the amplifier to the speaker terminals and reduces the SNR and increases THD+N.

As long as the amplifier output stage is operating in its linear range, the ripple feeding through to the speaker terminals is reduced by the power supply rejection of the output stage. If there is loop feedback, then power supply rejection at power supply ripple frequencies is high - 90 dB or more.

If the power amplifier clips then it is not operating in its linear range and far more of the power supply ripple can feed through to the speaker terminals.
arnyk is offline  
post #101 of 109 Old 06-07-2014, 07:09 AM
AVS Special Member
 
DonH50's Avatar
 
Join Date: Feb 2010
Location: Monument CO
Posts: 6,087
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 159 Post(s)
Liked: 262
Quote:
Originally Posted by electrovert View Post

I looked at the schematics posted earlier in this thread and it's worse than I thought.  R6 & R7 (A500)   & TR23 (QSC) are the ripple compensation applied directly with the negative feed back to the front end of the amp.  But what I put forward above should hold for most conventional amps... 

Feedback by itself does not reduce power supply ripple (unless in the power circuit). In your example, if you assume the input divider couples power noise into the input, feedback will not help reduce it. The feedback has no way of distinguishing noise at one side of the input from signal at the other. This is true in general. It is also worth noting that some designs may use dividers from the power rails to help the input reject power supply noise by tracking it out. You can't really look at pieces of a design in isolation.

"After silence, that which best expresses the inexpressible, is music" - Aldous Huxley
DonH50 is online now  
post #102 of 109 Old 06-08-2014, 02:32 AM
Newbie
 
electrovert's Avatar
 
Join Date: Jun 2014
Posts: 12
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
I see R6 / R7 doing two things.  They set the the output to split the supply rail...and track the supply rail ripple to "conceal it".
 
On another note when I got the A500 nine years ago I found the volume control distorted in any position, and replaced it(them)...It might be a more direct approach to finding out if it's a design flaw or just a cheap part. Simulation is great but in this case swapping out the pot is pretty easy...The detent mechanism jars the wiper so it doesn't make proper contact.   Well what do you want for the money.  Some people spend more on pots than this amp costs...I bought it because it had good specs and the picture with the lid off look like the money was spent in the right places...What I got was a crazy cap coupled amp...but I learned a lot...and you guys helped...
 
Another point made in the thread...was the power transistors squeezes the stored energy into the load.  If you simplify the circuit down to: the load, the transistor, and the cap...this is exactly what is happening.  The QSC description of the signal riding on DC leads to misunderstanding.  This is a capacitor coupled amp with no center tap from the transformer.  All output currents have to be squeezed out of the capacitor.  I say squeezed because it is a closed loop.  A conventional amp has stored charge above ground which is conducted by the output transistor through the load.  This amp connects the load directly to the one side of the stored charge and then closes the loop with the output transistor.  It might seem like semantics but with no center tap from the transformer this is the only source/path of energy flow.  Which brings up another difference.  With no other source of stored energy the "off" cycle cap gets charged relative to the phase of the AC line voltage...or doesn't...Thus my mod to add a secondary bridge cap.  With that there are now two sources of stored charge so the off cycle cap can get topped up every zero crossing of the signal, and by doing so support the "on" cycle cap.  But I'm repeating myself.  I said all this in my thread at Audio Asylum.  
 
any way I now take R6/ R7 to be the "Servo-Controlled" part of Berhinger A500.
 
 
All said and done I hated these amps stock for their bouncy low end and veiled midrange.  If you follow what I did noted at Audio Asylum you end up with a very good sounding amp with an exception noise floor.  I'm listening to them now biamped and they match good enough.
 
I want to thank everyone involved in this thread.  The schematic was all I needed to confirm what I had done and why it worked so well.
 
Happy listen.
 
Electrovert. 

Last edited by electrovert; 06-13-2014 at 01:28 PM.
electrovert is offline  
post #103 of 109 Old 06-08-2014, 02:44 AM
Newbie
 
electrovert's Avatar
 
Join Date: Jun 2014
Posts: 12
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10

http://usr.audioasylum.com/images/5/51996/IMAG0117.jpg

 

Here is what it looks like without the three  30uF/100v caps.

electrovert is offline  
post #104 of 109 Old 06-08-2014, 02:46 PM
Newbie
 
electrovert's Avatar
 
Join Date: Jun 2014
Posts: 12
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10

I have a couple more A500 to retrofit.  When I get those done I will stick them on the bench and take some objective measurements as best I can with my limited test gear.  It is my hope that I can buy down my loses, and pay my way, by sharing what I have learn about these amps.  Most people wouldn't invests the material cost, time, effort, have the tools and skills to double the cost of a device like this.  For me it's been a bit of a hobby.  My brother needed some amps and I kept them after they faded away.  So it was just a matter of going though the extensive DigiKey Cap selection to optimize for the best cap/s for the job.  The first array (coupling caps) got them functional with a good bottom end.  Solen had 30uf/100V poly caps on sale so they were bridged in to smooth out the mid range.  The top was fixed with 2.2uF/250V.  The second array was added to the bridge.  It was just an idea that came to me after simplifying the circuit and looking at how the stored energies were flowing.  Lastly another 30uF/100V poly was added to the bridge array...to smooth out the midrange even more.  The fine stranded pure silver wire made a critical improvement and I would recommend trying it as a replacement in any amps internal wiring.  I looked at the specs of the 4580 IC used in the front end and it 5V/mS is really slow in my book.  I would have liked to have seen 15V/mS+.  But the other specs, and requirement of this application probably make it difficult to cross to a higher slew part...all said and done.  I probably would have steered clear of this amp had I know it was cap coupled.  Finding out it was conceived by QSC is a good thing and probably why the all the A500 I have died without taking a speaker with them.  The ripple amps just killed the output caps.  So there you have it. my journey abridged and condensed pun intended of these amps.

 

Happy tweaking, and happier listening.

 

electrovert.

electrovert is offline  
post #105 of 109 Old 06-08-2014, 03:05 PM
AVS Special Member
 
DonH50's Avatar
 
Join Date: Feb 2010
Location: Monument CO
Posts: 6,087
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 159 Post(s)
Liked: 262
Since this thread is about objective testing I would love to see your measurements before and after, especially the critical improvement switching to silver wire made.

"After silence, that which best expresses the inexpressible, is music" - Aldous Huxley
DonH50 is online now  
post #106 of 109 Old 06-08-2014, 03:58 PM
Newbie
 
electrovert's Avatar
 
Join Date: Jun 2014
Posts: 12
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10

I guess I did take my turn hijacking this thread...Ops sorry...

 

Here what I have for test gear:

 

Potomac Instruments AG-51

 

Potomac Instruments AA-51

 

Creative Labs E-MU 204 DAC

 

Soundforge 6

 

I could take basic THD, IMD, and bandwidth. with the PI test set.

With a padded "probe" I could use the E-MU to feed Soundforge FFT display for harmonic spectrum.

 

I should be able to get a real spectrum analyzer on it eventually.

 

I don't have a dummy load that could take much more than 10 watts?

 

Would that be useful?  If it wasn't such a huge shipping expense I would send to someone to have it measured.  Just for the sake of an objective document.

electrovert is offline  
post #107 of 109 Old 06-08-2014, 06:39 PM
AVS Special Member
 
DonH50's Avatar
 
Join Date: Feb 2010
Location: Monument CO
Posts: 6,087
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 159 Post(s)
Liked: 262
Build up a higher-wattage test load using parallel-series combinations of resistors. Make sure there is airflow around them. Check out Stereophile's speaker test load, or build a few of your own to make it a more realistic test. You need an ADC to capture waveforms, but most sound cards include one.

"After silence, that which best expresses the inexpressible, is music" - Aldous Huxley
DonH50 is online now  
post #108 of 109 Old 06-08-2014, 07:42 PM
Newbie
 
electrovert's Avatar
 
Join Date: Jun 2014
Posts: 12
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10

The E-MU 204 is an out board converter 24-bit/192kHz A/D and D/A converters  (A/D: 113dB SNR, D/A: 113dB SNR)  so I don't think I need a sound card.

 

I have no interest in building a 200 watt dummy load.  But I will try to accurately measure what I can and post it here.  I have to order another sack of cap form DigiKey first.  I don't have a working stock amp to measure, and not sure what repeating what numerous other have done...other than to confirm my measuring techniques.  I think there is an older Audio Precision around I can get my hands on.  So I'll share the raw data, and measuring methods when I get them.

electrovert is offline  
post #109 of 109 Old 06-08-2014, 08:52 PM
AVS Special Member
 
A9X-308's Avatar
 
Join Date: Mar 2008
Location: Australia; now run by adults.
Posts: 5,273
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 88 Post(s)
Liked: 72
Quote:
Originally Posted by DonH50 View Post

Since this thread is about objective testing I would love to see your measurements before and after, especially the critical improvement switching to silver wire made.
So would I.
A9X-308 is offline  
Reply Audio theory, Setup and Chat

User Tag List

Thread Tools
Show Printable Version Show Printable Version
Email this Page Email this Page


Forum Jump: 

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