20WPA

Jump to Latest
#41
It seem like this circuit was designed by working out the math for specific devices.
Ahhh ... I like designs that allow for junkbox parts to do PPM.
wahab said:
Bob Cordell s BC1 has a VAS that drain 10mA and yet it is loaded
by an EF3 that has at least a 500 000 gain, even its EF3 is weird since the pre drivers are at 10mA and the drivers at 27mA,
Click to expand...

Wow , too hot !! wolverine does 3mA pre - 10mA driver. Vas is 5mA. My input modules run 3.2-3.8 mA VAS.
NO heatsinks ! except for driver/output. SMD likes to run super cool. Still all is <5PPM.
All this makes for reduced cost , smaller circuit . CHEAP PPM. Might last longer , as well.
OS
 
#42
wg_ski said:
Biggest problem with these darlingtons is the fact that you’re stuck with a 3 MHz driver transistor.
Click to expand...
Harder to find darlingtons , they go obsolete quickly - replaced by 21'st century low Rds class D FET's.
PS - darlingtons are tubelike , only sought by a small market of audiophiles.
AB is still "semi-mainstream" , if it can be enhanced by exquisite circuit design or next gen class G/H (Topping B200).
OS
 
#43
The following Darlingtons were used in the first original:
BDV65B & BDV64B.


60W into 8Ohm,
that's just under 4A_peak, let's just take a decent beta_drop as a given, so the 32mAdc for the VAS could be explained quite easily.

ft of the BD139 is estimated to be around 50MHz, I can't find any information on the TIP14x at the moment ..!
 
#44
Not much data, but already a good indication:



1728548586560.png


1728548627948.png


1728548661474.png




Different manufacturers -> different data!
 

Attachments

#45
hbtaudio said:
The following Darlingtons were used in the first original:
BDV65B & BDV64B.


60W into 8Ohm,
that's just under 4A_peak, let's just take a decent beta_drop as a given, so the 32mAdc for the VAS could be explained quite easily.

ft of the BD139 is estimated to be around 50MHz, I can't find any information on the TIP14x at the moment ..!
Click to expand...

BD139 Ft vary with the manufacturer, 50MHz for Siemens and likely most manufacturers, faster seems to be Phillips with 190MHz.
TIP140-147 serie is the same as BDV64/65 set apart for the internal resistances wich are lower for the latter.
 

Attachments

#46
No simulation yet? Is this ancient design so uninteresting today? Did I already say that the end product (my little monos) is really convincing? Driven by a preamplifier with low output resistance (<<< 100Ohm) !


So I like this crazy thing - "the 20WPA", simply put together.


Look also in the attached file (the second original version) ... my reduction /scaling is slightly different.



regards,
HBt.
 

Attachments

#48
The amplifier is interesting in the same way that news coverage of disasters is interesting. 😉

The 0.47uF shunt capacitor (C3) is huge. The reactance is only 17 ohms at 20KHz. The dominant pole is at 4KHz. The pre-amp had better be able to drive low impedance. This is the first time I have seen a tantalum capacitor used for frequency compensation (!).

You did a better job by reducing the value. 🙂

Your amplifier also has much better thermal stability due to increasing the emitter resistors from 0.2 to 0.5 ohms.

I am critiquing the amplifier by modern standards. The amplifier does work and its faults are likely below audibility.
Ed
 
  • Like
Reactions: hbtaudio
#50
wahab said:
The value of R7/C4, 82R + 220uF is also a nice "feature", a 220uF impedance at 20Hz is 36R.
Click to expand...
I really had to read the article several times and look at the schematic a few times - “the design has a few special features”. To put it kindly.
With something so slightly crazy in my hand, I no longer worry about abstruse ideas like the AB-BA gm-doubling compensation ..! I also immediately stumbled across all the points mentioned in this thread - but what the hell?!

The amp is a joy to use (and the measured values are not "so" bad). Let's see what @ostripper
's simulation will produce, I hope he's still simulating the 20WPA.
 
#52
@EdGr, Actualy they are enough with 6-7mA, i once used these BDV to replace the quasi of a Roland Jazz Chorus 2 x 40W guitar amplifier that has way lower VAS current, this ETI has a 32maA VAS current, wich is overkill but not
as much as the LTP wich is at 15-16mA and dissipate 0.5W, wich is also the TDP of its CCS, that s 0.25W/transistor.

I forgot to mention the 5.5R + 0.940uF Zobel network, otherwise this amp has both H2/H3 at -66dB for 23V pk/8R.
 
Last edited:
#55
.SUBCKT TIP142_FC 1 2 3
Q1 1 2 4 Q1model
Q2 1 4 3 Q2model 5.888
D1 3 1 Dmodel
R1 2 4 8.000E3
R2 4 3 120
.MODEL Dmodel D
  • IS=4.721E-12 RS=1.083E-6 N=1.068
  • CJO=4.822E-10 VJ=0.5 M=0.34
  • FC=0.5
.MODEL Q1model NPN
  • IS=8.075E-14 BF=7.827E3 NF=1
  • VAF=100 IKF=1.094 ISE=4.463E-13
  • NE=1.297 BR=0.1001 NR=1
  • VAR=100 IKR=0.4891 ISC=2.302E-13
  • NC=3.197 RB=2.222 NK=0.5471
  • RE=0.001 RC=0.990 EG=1.110
  • CJE=4.389E-10 VJE=0.77 MJE=0.3101
  • CJC=4.852E-10 VJC=0.4855 MJC=0.3379
  • XCJC=0.469 FC=0.5
.MODEL Q2model NPN
  • IS=8.075E-14 BF=7.827E3 NF=1
  • VAF=100 IKF=1.094 ISE=4.463E-13
  • NE=1.297 BR=0.1001 NR=1
  • VAR=100 IKR=0.4891 ISC=2.302E-13
  • NC=3.197 RB=2.222 NK=0.5471
  • RE=0.001 RC=0.990 EG=1.110
  • CJE=4.389E-10 VJE=0.77 MJE=0.3101
  • CJC=4.852E-10 VJC=0.4855 MJC=0.3379
  • XCJC=0.469 FC=0.5
.ENDS TIP142_FC


.SUBCKT TIP147_FC 1 2 3
Q1 1 2 4 Q1model
Q2 1 4 3 Q2model 5.845
D1 1 3 Dmodel
R1 2 4 8.000E3
R2 4 3 120
.MODEL Dmodel D
  • IS=2.896E-18 RS=1.000E-6 N=0.1001
  • CJO=5.833E-10 VJ=0.7067 M=0.321
  • FC=0.5
.MODEL Q1model PNP
  • IS=1.832E-13 BF=6.370E3 NF=1
  • VAF=75.95 IKF=1.199 ISE=1.907E-12
  • NE=1.334 BR=.1769 NR=1
  • VAR=100 IKR=11.87 ISC=9.990E-6
  • NC=1.051 RB=12.43 NK=0.5279
  • RE=0.001 RC=0.17 EG=1.110
  • CJE=3.647E-10 VJE=0.8826 MJE=0.3255
  • CJC=3.964E-10 VJC=0.6635 MJC=0.2578
  • XCJC=0.469 FC=0.5
.MODEL Q2model PNP
  • IS=1.832E-13 BF=6.370E3 NF=1
  • VAF=75.95 IKF=1.199 ISE=1.907E-12
  • NE=1.334 BR=.1769 NR=1
  • VAR=100 IKR=11.87 ISC=9.990E-6
  • NC=1.051 RB=12.43 NK=0.5279
  • RE=0.001 RC=0.17 EG=1.110
  • CJE=3.647E-10 VJE=0.8826 MJE=0.3255
  • CJC=3.964E-10 VJC=0.6635 MJC=0.2578
  • XCJC=0.469 FC=0.5
.ENDS TIP147_FC
 
#56
hbtaudio said:
I hope he's still simulating the 20WPA.
Click to expand...
I suppose I could do so for some "mental masterbation".
When I have time , I need to build and access a China regulator kit first.
hbtaudio said:
In particular, the power losses of the BJTs were a decisive criterion for reducing the rails of the PSU by 15Vdc each (from 40 to a maximum of 25).
Click to expand...
That's a real cripple , most of my amps will do 25 -70V rails - no adjustments.


OS
 
You must log in or register to reply here.
Top Bottom