"blown out" blameless ...
OK ,
got it .. 🙂
Sub PPM .. all the way (below 2).
Many models .. still the same (topology driven result).
Schema (below 1 ) - For the PCB , I might have to ditch the huge input cap ..
Need room.
14 devices for a IPS ... 😱
OS
OK ,
got it .. 🙂
Sub PPM .. all the way (below 2).
Many models .. still the same (topology driven result).
Schema (below 1 ) - For the PCB , I might have to ditch the huge input cap ..
Need room.
14 devices for a IPS ... 😱
OS
Attachments
I haven't quite gotten there yet. I would have to go the way of GURU
E. Stuart with many devices/Hec .. etc.
Naf's is close at (below 1) .Just a "bump" of H2 at -130db. Used
zetex , BCxxx , and fairchild - within 1 -2 ppm.
OPS bias/ IPS interaction determines those last few PPM's.
(the OPS's switching distortion is "whats left")
Run either IPS into a class A OPS - both are sub PPM. 😕
OS
Attachments
There will be ... for ... now these are split up (OPS/IPS).
One could just read off the OPS or IPS and make a quick BOM,
The schema's are complete , it up to the builder to choose the quality level
of the devices (mje or KSA ot toshiba).
SO MANY devices will work in these designs.
For instance , for the IP/CCS - KSA/P , sa970/sc2240 ,d667/b647.
ALL to-126 can be mje340/350 , sc3503/sa1381 (or similar). VAS is
most crucial for a Hi-perf pair.
PS - for now , make your own ... ask me about device selection - you
will be happy.
OS
Hi Ostripper
greetings P-V1.2 IN#500 this schematic i have all trs and parts PLEASE can you share its output pcb input pcb i have is wolverine v1.0 is it ok
warm regards
andrew
greetings P-V1.2 IN#500 this schematic i have all trs and parts PLEASE can you share its output pcb input pcb i have is wolverine v1.0 is it ok
warm regards
andrew
You have the standard wolverine 1.0 - no cascode - it is good/simple/rock
solid...
Tomorrow will be wolverine - C v1.2 ... C is for Cascode.
I have the "simple blameless" IPS (it was posted) now ... but I don't want 2 versions.
Wolverine "C" will be like the badger where you can omit the cascode.
"all in one" ...
You might like the C .. it is a nastier wolverine ... 😀
ps - not as much room for the audiophile CAP on "C"
OS
Lets make "plug and play" IPS 😀.
We can standardize VAS current.
The IPS's PCB must small as posible, we can use SMD parts and multilayer PCB. The IPS's PCB can plug into OPS's PCB without soldering 😎
We can manufacturing here and make some money 😀
We can standardize VAS current.
The IPS's PCB must small as posible, we can use SMD parts and multilayer PCB. The IPS's PCB can plug into OPS's PCB without soldering 😎
We can manufacturing here and make some money 😀
The Vssa would be a good candidate.
Hey , lookie what I have (below) .... a leach -
😀
Another "victim" 😀
OS
Attachments
Last edited:
My ''baby'' version output stage based on version 1.1 is ready...😀 Now i am going out for the parts.
Ostripper what are the values of c-b driver capacitors on your later version?
Can you suggest other alternatives for transistors ksc3503,ksa1381,ksa992,ksc1845,mje340-350 because i can find them but from different manufacturer the pnp and the npn part!! 😡 can 1n5406 be used instead of mur460?the second one is a fast recovery diode while the first is a general purpose....
Ostripper what are the values of c-b driver capacitors on your later version?
Can you suggest other alternatives for transistors ksc3503,ksa1381,ksa992,ksc1845,mje340-350 because i can find them but from different manufacturer the pnp and the npn part!! 😡 can 1n5406 be used instead of mur460?the second one is a fast recovery diode while the first is a general purpose....
Attachments
Last edited:
Smaller amp , the 3a 1n5406 should be fine.
Global sourcing of replacements is not a real headache.
Just X- reference your part to the specified one.
Vceo can be rail + rail and 20V or more derated.
Gain (hfe) is not critical on the EF3 , driver and OP are typically low gain
(30-100hfe) predriver a little higher.
If you are building a CFA (balanced) , the small signal N/P matching is more
important. Most vfa just need matched same gender pairs.
OS
Very hard "blameless" ...
This VFA has a lot of IPS parts (63 to be exact - 31R/12C/5D/14Q).
2 freakin' trimmers for tweakers 😀 .
Flirting with sub PPM is a touchy thing ... just the right IP current , VAS I ,
and exactly ! 72ma OP bias will get <1ppm@ 20k.
With both the ON models (NJW/MJL) , jumps above 5ppm below 60 or over 80
mA per device.
To preserve this goodness on a real amp would require a VERY stable
bias .... I was wise to choose the dual Vbe H/K.
PSRR plot is different with Dado's cascode "trick" (notch is at 15K - but
-120db ... whole audio band)- The BEST i've seen. 🙂
Below is final schema for the "Wolverine-C" (asc too)
OS
This VFA has a lot of IPS parts (63 to be exact - 31R/12C/5D/14Q).
2 freakin' trimmers for tweakers 😀 .
Flirting with sub PPM is a touchy thing ... just the right IP current , VAS I ,
and exactly ! 72ma OP bias will get <1ppm@ 20k.
With both the ON models (NJW/MJL) , jumps above 5ppm below 60 or over 80
mA per device.
To preserve this goodness on a real amp would require a VERY stable
bias .... I was wise to choose the dual Vbe H/K.
PSRR plot is different with Dado's cascode "trick" (notch is at 15K - but
-120db ... whole audio band)- The BEST i've seen. 🙂
Below is final schema for the "Wolverine-C" (asc too)
OS
Attachments
Last edited:
49810 incompatible with EF3. EF2/lateral-BJT or darlington only...
Junk ... For a VFA , a joke .... ASTX's amp (9ppm) badger (20-30ppm)
wolverine "C" (1-5ppm) .... 49810 = 100-200+ !
4 device VSSA even has it beat (20-30ppm).
LM PSSR rises with frequency ,barely -65db @ 10k , wolverine is -120db.
The LM's noise and offset are good - that's about it... 😱
Chip amps , belong at walmart. 😀
PS - looked at the LM3886 datasheet ... it is a (blameless)wolverine , it's PSRR is -105-120db
....reflecting it's internal topology. A good chip !!
OS
Last edited:
