Go Back   Home > Forums > >
Home Forums Rules Articles diyAudio Store Blogs Gallery Wiki Register Donations FAQ Calendar Search Today's Posts Mark Forums Read

Solid State Talk all about solid state amplification.

The AMNESIS amp: a good amplifier, like a gentleman, has no memory.
The AMNESIS amp: a good amplifier, like a gentleman, has no memory.
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 Yesterday, 02:02 AM   #511
maxlorenz is offline maxlorenz  Chile
diyAudio Member
 
maxlorenz's Avatar
 
Join Date: Oct 2003
Location: osorno , Chile
Quote:
Originally Posted by Hans Polak View Post
Optimum value for R50 is 46 Ohm, which can be easily simulated with Vce*Ic for using a large signal, 47 Ohm only makes a slight increase.
At 20khz power fluctuates 30 times less as without R50.


Hans
Which proves that intuition is superior to logic!

No, seriously guys, I count on you for the rigorous analysis: if power (and thermal stress) only derives from Vce and Ic, then the choice of different BJT (or MOSFET for that matter) is irrelevant...

There is another potential problem which I did not address before and which I doubt I would be able to properly describe. Anyway, there it goes: apparently, for the VAS cascode (or any cascode) the base sees the emitter resistance (26R? inverted?) which is capable of influencing, for the worst, any resonant tank, which makes the subcircuit prone to oscillation. I wonder if the addition of R50 improves or deteriorates the immunity to oscillatory propensity...
I hope my wanderings were able at least in part to transmit the potential problem.

Dear Gaetan,

Soundstage, which is not a goal of mine per se, will not find critics, I hope, with some versions of the amp having it extended to the sides of the speakers, and with decent depth.
The salient feature I would say is the "concentration of energy" (correct sound envelope) ,in the time domain, of the notes, by which the attacks, the devellopment of the curves and the decays are properly conveyed, so one can relax on the natural description of the musical lines and suspend hypercritical judgement.
This is one of the reasons why the little amp sounds "powerful", meaning being satisfactory at normal listening levels. Added to this is the mid-bass rendition; at least for the ALBJT-BD139VAS amp as I did not have the time to test the JFET plus MPSA18VAS amp on other systems than the test system...
It is hard to describe the sound presentation, but when youŽll listen to it you will understand what I mean.

Sorry guys, I am robbing internet from my daughters' phone, so I will make short answers as time will allow.

Cheers,
M.

Last edited by maxlorenz; Yesterday at 02:10 AM.
  Reply With Quote
Old Yesterday, 10:13 AM   #512
Hans Polak is offline Hans Polak  Netherlands
diyAudio Member
 
Join Date: Mar 2005
Location: Blaricum
Quote:
Originally Posted by maxlorenz View Post
There is another potential problem which I did not address before and which I doubt I would be able to properly describe. Anyway, there it goes: apparently, for the VAS cascode (or any cascode) the base sees the emitter resistance (26R? inverted?) which is capable of influencing, for the worst, any resonant tank, which makes the subcircuit prone to oscillation. I wonder if the addition of R50 improves or deteriorates the immunity to oscillatory propensity...
I hope my wanderings were able at least in part to transmit the potential problem.
DearMax,

Hoping to understand you correctly, the base of the VAS Cascode sees Hfe[Q17]*(26)/Ic(ma)+R50+Ro[Q15] or approx 1.3+47+7k5.
As you can see, R50 will hardly play any role for the cascode transistor.
But for Q15 it will make a difference, because with R50 its collector will see a much larger voltage swing, combined with the effect that Vce will drop almost 1 volt (20mA*47ohm), both affecting the BW of the transistor and thus potentially stability.
As you have done all the time so far, listen to the beast and if no acoustical benefit is perceived from R50, leave it out.


Hans
  Reply With Quote
Old Yesterday, 10:58 AM   #513
Hans Polak is offline Hans Polak  Netherlands
diyAudio Member
 
Join Date: Mar 2005
Location: Blaricum
The brackets were lost in the hurry, base of Q17 sees Hfe*(1.3+47+7k5)


Hans
  Reply With Quote
Old Yesterday, 11:44 AM   #514
Symon is offline Symon  United Kingdom
diyAudio Member
 
Join Date: Mar 2002
Location: Herts, UK
Hi Max,
It's interestign to note that in Peufeu's testing schematics and prototype photo's the
magic resistor(R50) is not present. So as Hans says listening tests are required to
determine if it's doing anything useful.

And it's worth noting that this resistor partly defeats the Cascode because it
re-introduces in a small way the varying VCE across Q15 although the effect is
relatively small.

Regards,
Symon
  Reply With Quote
Old Yesterday, 01:06 PM   #515
indra1 is online now indra1  Indonesia
diyAudio Member
 
Join Date: Nov 2010
Location: Bogor
Quote:
Originally Posted by Hans Polak View Post
... As you have done all the time so far, listen to the beast and if no acoustical benefit is perceived from R50, leave it out. ...
Here is what Hugh Dean (Aspen Amplifier) thought about that particular resistor on Swordfishy/ASPEN FETZILLA power amp thread.
  Reply With Quote
Old Yesterday, 02:49 PM   #516
JOSI1 is offline JOSI1  Germany
diyAudio Member
 
Join Date: Mar 2006
Location: Bad Hersfeld
Default Update schematics and layout

I managed to combine the JFET version and the two AllBJT versions
from post #503 on the main board.
For the JFETs an additional pad for the gate connection was added.
(the gate pin of the JFET must be bended a little).
To replace JFET 2N5462 a footprint for the SMD version MMBF5462 (SOT23) was added.

The two AllBJT versions can be selected by solder bridges. For version 1 two or four transistors can be used.

To realize the simpler AllBJT version suggested by Hans
two current sources to the neg. supply were added using SMD components.

Nevertheless some pads for connection of a daughter board have been added.

For the VAS section BC5xx / 2N2222 / MPSA18 and BD139 can be used (footprint for BD139 added).

Remarks:
Acc. to the latest simulation schematics (post #503)
C10 ( || to the 15V zener diode) was reduced from 1000uF to 100uF.
This helped me to put all input versions to the main board.

Is C30=100n || the diodes D11/D12 still required?

Hans suggested
to connect C19=1000uF to R33-2 insteadt of R33-1
to change R52 from 2k2 to 10k

Please review the attached schematic. The layout proposal will follow soon.

Cheers
Attached Files
File Type: pdf Amnesis Schem 5.pdf (628.0 KB, 13 views)
  Reply With Quote
Old Yesterday, 04:07 PM   #517
Symon is offline Symon  United Kingdom
diyAudio Member
 
Join Date: Mar 2002
Location: Herts, UK
Hi Indra1,
Interesting to know that Hugh reports good results audio from adding this resistor.
Although it wasn't clear from the description that the considered the fact it increases
VCE of the driver to be greater than with just a simple cascode.
Although unless the current varies a lot it's difficult to believe that the VCE variation
is very much (47mV for 1mA change) so difficult to believe this causes instabliliy which was Max's concern.
Overall we are correcting small errors, which without correction hide detail in the sound. Or so we believe

- Symon
  Reply With Quote
Old Today, 02:48 AM   #518
maxlorenz is offline maxlorenz  Chile
diyAudio Member
 
maxlorenz's Avatar
 
Join Date: Oct 2003
Location: osorno , Chile
Quote:
Originally Posted by Hans Polak View Post
DearMax,

Hoping to understand you correctly, the base of the VAS Cascode sees Hfe[Q17]*(26)/Ic(ma)+R50+Ro[Q15] or approx 1.3+47+7k5.
As you can see, R50 will hardly play any role for the cascode transistor.
But for Q15 it will make a difference, because with R50 its collector will see a much larger voltage swing, combined with the effect that Vce will drop almost 1 volt (20mA*47ohm), both affecting the BW of the transistor and thus potentially stability.
As you have done all the time so far, listen to the beast and if no acoustical benefit is perceived from R50, leave it out.


Hans
Dear Hans, Symon and Indra,

Thank you for clarifying this interesting aspect of the so important VAS.
We are learning a lot.

Perhaps dear Hugh won't protest if we take the liberty of quoting him, for completeness' sake:

Quote:
If VAS and its cascode run at 6.5mA, then Zin of the cascode emitter is 4R (26/Ie). This corresponds to re, of course. If we interpose a resistor and drop 200mV across it, then the value becomes 200/Ie, or 7.7 times the value of Zin of the EF. This is 31R, npv of 33R. 7.7 is suspiciously close to the accepted engineering bridging ratio of 10.

It helps both with impedance transfer from collector to emitter and linearises the input impedance of the EF, drowning it in a constant value. There appears to be no penalty for this, since voltage drop is trivial and hardly affects rail efficiency.

Quote:
When the output (output impedance) Zout of a device (the source, ZS in illustration) is connected to the input (input impedance) Zin of another device (the load, ZL in the illustration), it is a bridging connection if the second device does not appreciably load the previous device.

Essentially no power is transferred. The second device is sensitive to the output voltage of the first device, and this is maximized when loading is minimized. All good line level audio connections are bridging, to ensure the high fidelity transfer of a voltage signal from one device in the chain to the next. If the bridging criterion is not satisfied, there may be distortion, altered frequency response or both.

A connection is commonly said to be bridged if the load impedance is at least ten times the source impedance.
Though this was not my fear, this aspect is now understood. I will try to find the reference, which I seem to remember comes from Bonsai...

http://hifisonix.com/wordpress/wp-co...mplifiers1.pdf

More Notes On Cascode Amplifier Oscillation

Dear JOSI1, brilliant work, as always...who would have thought at the beggining that the input section would end being so loaded with stuff!

Quote:
Is C30=100n || the diodes D11/D12 still required?
I believe it is. Why?

Dear Gaetan, with that one you probably need more jumpers than parts, hehe.

Best wishes to all,
M.

Last edited by maxlorenz; Today at 02:58 AM.
  Reply With Quote

Reply


The AMNESIS amp: a good amplifier, like a gentleman, has no memory.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

Forum Jump

Similar Threads
Thread Thread Starter Forum Replies Last Post
new memory for PC AndrewT Everything Else 36 24th October 2015 11:09 AM
down the memory hole? tomtt Everything Else 16 9th September 2014 02:27 PM
My memory is failing me I think?? Moondog55 Analogue Source 3 31st January 2011 04:57 AM
LCR and of course Memory Resistor impsick Parts 2 3rd May 2008 09:30 AM
Notebook Memory eag Swap Meet 4 24th April 2006 05:20 AM


New To Site? Need Help?

All times are GMT. The time now is 04:17 PM.


Search Engine Optimisation provided by DragonByte SEO (Pro) - vBulletin Mods & Addons Copyright © 2019 DragonByte Technologies Ltd.
Resources saved on this page: MySQL 15.00%
vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2019 DragonByte Technologies Ltd.
Copyright ©1999-2019 diyAudio
Wiki