Member
Joined 2009
Paid Member
kean - your sim file didn't run, gives unconnected node error; I'm running LTSpice 4.01K.
as for 80dB OLG with >100kHz bandwidth - I see a divergence in philosophy here - I'm heading for less gain-bandwidth and avoidance of gnf - I've been polluted by vacuum tubes, those things that John don't like 😀
as for 80dB OLG with >100kHz bandwidth - I see a divergence in philosophy here - I'm heading for less gain-bandwidth and avoidance of gnf - I've been polluted by vacuum tubes, those things that John don't like 😀
That's just it. 100KHZ is the open-loop bandwidth. Your BW is less than 100Hz. Add to this that my graph has a slope of 12db/octave while yours has one of 6db/octave.
- keantoken
- keantoken
Node unconnected usually means it can't find the model for the model...
Try adding a normal transistor to the schematic, with "Q2SC3423_e" where the transistor part number goes. This should cause SPICE to include the standard.bjt file I included in the simulation.
EDIT: there is already a redundant transistor you can use, shorted out on the negative rail. Replace "2N5089" with "Q2SC3423_e". This will make SPICE look at the file I included, hopefully...
- keantoken
Try adding a normal transistor to the schematic, with "Q2SC3423_e" where the transistor part number goes. This should cause SPICE to include the standard.bjt file I included in the simulation.
EDIT: there is already a redundant transistor you can use, shorted out on the negative rail. Replace "2N5089" with "Q2SC3423_e". This will make SPICE look at the file I included, hopefully...
- keantoken
Ok, good luck with stabilizing something like that. You do know that there is absolutely no point to it in an audio amp, right?
The highest I've done is 1Khz, my Abomination amp. Realistic goals are worth the effort.
Believe it or not it does work. I've been listening to the prototype for most of this year.
- keantoken
- keantoken
I could have sworn you said in some other thread that you can't get to run stable...
What do you believe the advantages of such an amp would be, Kean?
Please, don't say 'sonics'
🙂
Yes, there have been some troubles... It was stable and playing well for several months without oscillating into a 5.6ohm fullranger. I then built another which ran just as well. All my latest have been worse. The difference, as far as I can tell, is lead length. In my earlier prototypes it was below 1cm for the important parts. Since then it has grown to 3cm.
Advantages? Hmm, go to a fancy restaurant, flash it in front of the smartest looking guy.
smart looking guy: No, no way.
me: look again.
smart looking guy: no!
me: I built it. It works.
smart looking guy: ...really?
A week later I have lots of money and lots of women!
- keantoken
Advantages? Hmm, go to a fancy restaurant, flash it in front of the smartest looking guy.
smart looking guy: No, no way.
me: look again.
smart looking guy: no!
me: I built it. It works.
smart looking guy: ...really?
A week later I have lots of money and lots of women!
- keantoken
😀
Bragging rights are fine.
Considering that we only hear out to about 20Khz (some of us, myself included, substantially lower than that) it is an unworthy target for an audio amp.
I can consciously hear up to 18KHz, 16KHz on a bad day, which is about the time I gobble a handful of vitamins. As discovered through thoroughly unrigorous testing.
- keantoken
- keantoken
Bigun, my philosphy changes from time to time. My first poweramp will have an open loop gain of below 60db, I'm sure. I recognize the correlation between OLG and SQ is probably negative at best. I just like the intrigue of designing "high-performance" circuits.
- keantoken
- keantoken
Hello keantoken
In the past, I was sim my amps for high OLG and so it was needed a high GNFB, using buffered VAS, the thd was very low, the sound was good but it was missing something, it was not engaging and alive like it should be.
Few month ago, after a chat with Hugh, I've decide to re-sim my amps for a OLG of 65 db and a GNFB of 30 db... the sound are now much better.
Nelson Pass demonstrated, in a paper, that to much OLG and gain stages in a amp do create more IM thd with complex signal like musics.
Merry Christmas KT and all guys.
Bye
Gaetan
In the past, I was sim my amps for high OLG and so it was needed a high GNFB, using buffered VAS, the thd was very low, the sound was good but it was missing something, it was not engaging and alive like it should be.
Few month ago, after a chat with Hugh, I've decide to re-sim my amps for a OLG of 65 db and a GNFB of 30 db... the sound are now much better.
Nelson Pass demonstrated, in a paper, that to much OLG and gain stages in a amp do create more IM thd with complex signal like musics.
Merry Christmas KT and all guys.
Bye
Gaetan
Last edited:
What intrigues me is that it seems that the lower the number of transistors, the higher the OLG one can afford to have. This inreases stability, phase shift, and most other specs.
BTW, I think many times we use transistors in places where just resistors would do, if used creatively. This is a direction I may be heading.
- keantoken
BTW, I think many times we use transistors in places where just resistors would do, if used creatively. This is a direction I may be heading.
- keantoken
Member
Joined 2009
Paid Member
There is a correlation between these two, so it seems from the movies, but I think this forum demonstrates is the wrong place for this goal for a man of your age 😎
simple amps simply sounds better...,,,,
what if you can hear inter-modulation of very high frequencies...what does eg....40 and 42 KHz sine waves create in superposition...or 100 and 110 KHz....wavepatterns with much much lower frequency content.....Does it ever stop....???
merry X-mas
what if you can hear inter-modulation of very high frequencies...what does eg....40 and 42 KHz sine waves create in superposition...or 100 and 110 KHz....wavepatterns with much much lower frequency content.....Does it ever stop....???
merry X-mas
If I remember right there was one paper that described an experiment where people's brains were scanned while listening and it showed more brain activity when sound <20KHz were present. Unfortunately I lost the bookmark.
Not that any source can reproduce that, but there's nothing wrong with having support for the feature...
Bigun, did you ever get that simulation working?
- keantoken
Not that any source can reproduce that, but there's nothing wrong with having support for the feature...
Bigun, did you ever get that simulation working?
- keantoken
Did you mean >20Khz?
Not many well designed amps that can't reproduce up to and above ~100khz. I have squarewave tests on my Patchwork amp up that high (and higher). Real tests, not simulated. See HERE.
Stability and reliability is more important to me than bandwidth. Immunity to capacitive and inductive loads is more important also.
Member
Joined 2009
Paid Member
I'm glad to see there are amplifiers for us all to have fun with, it's what makes this hobby great.
Merry Christmas you lot !
Merry Christmas you lot !
- Status
- Not open for further replies.