Later, after the move and lots of house rebuilding, my man cave if almost finished, i installed wolverine in the main system and had few friends for listen.
All agreed this is the most transparent amp i own. I do have about 50 amplifiers (somewhere).
My main system is more complex and contains active crossover and pre. I like lush sound, and tube pre is a must.
As said before wolverine amp sounds sublime. Each instrument can be easily followed with no effort, even in very complex music. Top is the cleanest i heard. Chimes are spooky real. Like in the room.
Amp is never boring, never dry, never fatiguing. Have been listening for a month now, and it still is amazing.
I hope soon i will have time to build second one by myself.
All agreed this is the most transparent amp i own. I do have about 50 amplifiers (somewhere).
My main system is more complex and contains active crossover and pre. I like lush sound, and tube pre is a must.
As said before wolverine amp sounds sublime. Each instrument can be easily followed with no effort, even in very complex music. Top is the cleanest i heard. Chimes are spooky real. Like in the room.
Amp is never boring, never dry, never fatiguing. Have been listening for a month now, and it still is amazing.
I hope soon i will have time to build second one by myself.
Attachments
This is not for this post, but for something in your, "amazing," video series. You speak of checking for shorts with the beep function on the MM. I strongly suggest builders use the ohms function for this test and not rely on a full on dead short/beep for this test. When you have a potential short, it often first shows as a low R that does not trip the beep function on a MM. Like Infinite, same, same and 1k which will not trigger the short/beep. I know because I was on the test team and it happened to me.
Last edited:
Hello to all,
I wanted to thank all the team members who created this absolutely awesome DIY project!
I'm just starting my first DIY project that is so complex and has such high value when done properly.
This is my first post, and I was going trough this forum thread but I did not find an answer to my question about BOM list.
I´m a bit confused about resistors wattage values. IN BOM there is noted "RESISTORS 1/4W UNLESS NOTED" when I was going through this mouser part list then I saw that many resistors were .6W not 1/4W as referred. So is this ok to go with .6W? resistors if it is on the list?
I wanted to thank all the team members who created this absolutely awesome DIY project!
I'm just starting my first DIY project that is so complex and has such high value when done properly.
This is my first post, and I was going trough this forum thread but I did not find an answer to my question about BOM list.
I´m a bit confused about resistors wattage values. IN BOM there is noted "RESISTORS 1/4W UNLESS NOTED" when I was going through this mouser part list then I saw that many resistors were .6W not 1/4W as referred. So is this ok to go with .6W? resistors if it is on the list?
Your system looks amazing job well done! Glad to hear that you are really enjoying the Wolverine, guess it's really hard not too, to be honest.
Semi matching...need a bit a guidance as I did soem semi measurments using a DCA-55. I know about Q1, Q2 and Q3, Q4 as the need to be matched as much as possible. How about the pre-drivers Q105 and its complementary Q106? I reckon they need to have the same hFE as much as possible. Any other advice?
thx in advance
thx in advance
Last edited:
Bellow are two wave files ripped from CD. Test first the refrain of Turtles, Happy together. Are voices screaming or clear? The Frank's voice has a metallic grain, is it brass like or garbage?
HAYK , 0 distortion won't give you a "sound". I would say the wolverine's character would be "effortless" , thanks to it's Harmon-Kardon
EF3 and overbuilt DIYA power supplies. High frequency definition is determined by having sufficient slew and low damping factor.
BUT , low damping factor is only part of the equation. The speaker/Xover is the other part.
Bass definition is even more complicated. Woofers , ported vs. sealed .... crazy impedance swings. Here ripple rejection and damping factor
are the major players. Wolverine is "best of both worlds" with crazy damping and quite good PSRR.
An amp can only "guess" and correct what it sees at the main NFB takeoff. Its has no idea about the complex set of impedance changes from
speaker cable , Xover ... onward. Definitely not Zero at any speaker (driver) input terminal. Damping factor is limited by speaker cable gauge , as well.
So , actually .... one would not hear much difference from 100ppm down. You would hear differences in PSRR and the output stage , especially with the
bass .
I actually know a cute trick to get better PSRR <100hz , this would benefit linear supply users (50/60hz). Not to scare SMPS users , but the wolverine's
HF PSRR is a bit weak from 30-100Khz. The negative rail is the Wolverines "Achilles heel" . Larger negative ripple cap and increase the resistor values 2-3X.
Nice -100db at both 60hz and 50Khz.
EF3 and overbuilt DIYA power supplies. High frequency definition is determined by having sufficient slew and low damping factor.
BUT , low damping factor is only part of the equation. The speaker/Xover is the other part.
Bass definition is even more complicated. Woofers , ported vs. sealed .... crazy impedance swings. Here ripple rejection and damping factor
are the major players. Wolverine is "best of both worlds" with crazy damping and quite good PSRR.
An amp can only "guess" and correct what it sees at the main NFB takeoff. Its has no idea about the complex set of impedance changes from
speaker cable , Xover ... onward. Definitely not Zero at any speaker (driver) input terminal. Damping factor is limited by speaker cable gauge , as well.
So , actually .... one would not hear much difference from 100ppm down. You would hear differences in PSRR and the output stage , especially with the
bass .
I actually know a cute trick to get better PSRR <100hz , this would benefit linear supply users (50/60hz). Not to scare SMPS users , but the wolverine's
HF PSRR is a bit weak from 30-100Khz. The negative rail is the Wolverines "Achilles heel" . Larger negative ripple cap and increase the resistor values 2-3X.
Nice -100db at both 60hz and 50Khz.
If I want to know how a zero distortion sound like with the files I posted, because of feedback theory. Grand masters St Bob, St Bruno etc, preconize either no feedback or very high NFB feedback. Why? Because of the open loop phase rotation in the audio band. Only the NFB is the parameter that can bring the 90° phase shift to near zero in time. St Bob in his evangel decides to design a MOSFET amp with 40khz open loop bandwidth with about 30db NFB only.
What goes wrong? When a single voice note is played, nothing is wrong, but with multiple voices with the same note but different frequencies from flat to sharp, the low NFB feedback provokes data loss on the beat after which the ear can no more reconstruct back each voice individually and you hear distorted sound. see bellow 13 voices of 1khz note equally ranged from flat to sharp. What I wondered if the nonlinearity is also taking part. I saw the loop gain on the first early version which is about 24db NFB at 20khz, this is same what chip amps do. You need 40db @20khz NFB to get the 16 voice of Happy together right, and 70db to get the Requiems of Verdi , 300 voices +70 instruments, and of Mozart sound celestial.
As was confirmed, The Requiem of Mozart can not be reproduced by your amp, but how about the Frank's voice, is this amp can be qualified as Audio or just in lab use for chemistry or Physics experiments.
What goes wrong? When a single voice note is played, nothing is wrong, but with multiple voices with the same note but different frequencies from flat to sharp, the low NFB feedback provokes data loss on the beat after which the ear can no more reconstruct back each voice individually and you hear distorted sound. see bellow 13 voices of 1khz note equally ranged from flat to sharp. What I wondered if the nonlinearity is also taking part. I saw the loop gain on the first early version which is about 24db NFB at 20khz, this is same what chip amps do. You need 40db @20khz NFB to get the 16 voice of Happy together right, and 70db to get the Requiems of Verdi , 300 voices +70 instruments, and of Mozart sound celestial.
As was confirmed, The Requiem of Mozart can not be reproduced by your amp, but how about the Frank's voice, is this amp can be qualified as Audio or just in lab use for chemistry or Physics experiments.
Attachments
Ha ha.... "phase rotation" ? in yer crossover network ? Wolverine actually is 32db OL gain over 26db actual - 6db "extra" NFB @ 20Khz. <1 degree phase shift
from 20-20Khz. I'm sure Mozart would be pleased. Xover and driver alignment would be much more of a factor.
"Franks voice" ? just think of the valve amps that created the master .... and then all the re-mastering since. phase shifts galore.
Frank would have to sing into my 200$ mic and be recorded at 32 bits (DSP) to get it right. Too bad we can't time travel. he he....
from 20-20Khz. I'm sure Mozart would be pleased. Xover and driver alignment would be much more of a factor.
"Franks voice" ? just think of the valve amps that created the master .... and then all the re-mastering since. phase shifts galore.
Frank would have to sing into my 200$ mic and be recorded at 32 bits (DSP) to get it right. Too bad we can't time travel. he he....
@wkloppen Theoretically, the more closely the transistors in each complementary pair match, the better. As the transistor size/operating current increases, it is more difficult to find a match over the entire operating conditions, so that any match will be a compromise. You might try matching at the bias current - ~5.5 mA pre-driver and ~15 mA for the drivers.
I'm not sure of the collector current controls on the tester you have. Generally, the base current is regulated to determine Hfe and Vbe at the base current operating point. If you can match the Vbes, then the emitter resistors compensate for modest differences in beta.
I'm not sure of the collector current controls on the tester you have. Generally, the base current is regulated to determine Hfe and Vbe at the base current operating point. If you can match the Vbes, then the emitter resistors compensate for modest differences in beta.