F5 power amplifier

Re: Speculative fiction

wayness_tamm said:


Source degeneration is diffinitely feedback. The ratio of the undegenerated gain verses gain with source degengeration has everything to do with how your SSTART measures and sounds.
Your circuit also has no power supply rejection and you are listening dierectly to the supply in your topology. Try a folded cascode to eliminate the influence of the power supply. I read article again and found no reference to something as basic as what the gain was, but do remember something about Martians landing in ones backyard.... and name dropping of Nelson Pass of course.



This goes beyond a passing resemblance to Fred's posting style.
Source degeneration = feedback:
Kindly use the search function to locate one of the many threads arguing the semantics of what does and does not constitute feedback. Take your complaints (and attitude problem) there. This isn't the thread for it.
No PSRR:
Use a decent power supply, preferably regulated. Given your posturings of expertise, surely you'll be able to whip something up in no time. When in doubt, consult one of your textbooks. Be sure to use plenty of feedback.
Folded cascode:
That would take the circuit farther from being a solid state analog of the CJ ART. Surely, this rates as a DUH! to anyone who actually bothered to read the writeup. I allowed myself the one departure in topology (a cascode) due to the voltage limitations of the JFETs, but stopped there. I mean, gee, after you've cascoded, then added a folded cascode, you might as well put in a follower, then make the front end a differential, then...Surprise, surprise, surprise!...you've built yourself a discrete opamp.
Boring!
Specs of various sorts:
I abase myself, O Great One! Steven and I talked about projects. He offered to put one up. I said okay. Then a few days later, he kinda dropped the bomb; he put out a last call for articles for the September/October DIY, the deadline being August 10th. He hadn't warned me that I was on that short a tether. This was right at the end of July, the 28th or 30th, perhaps. Now, given that the twins have yet to demonstrate proficiency in changing their own diapers, the fact that the Dow Industrials were at that time rapidly approaching some critical price levels, the fact that I work a full time job, and a few other odds and ends, like the fact that the window in my car spontaneously detached itself and fell into the door, I was a wee bit busy. Nonetheless, I managed to pull the circuit and writeup together with Logan sitting on my lap trying to help Poppa type. (Gotta teach that boy to use the Shift key. Reading his typing reminds me of Archy jumping on the keys to make the typewriter activate.)
The punchline?
After all the hurry and scurry, Steven renamed the July/August DIY to Summer and delayed the September/October until just a few days ago, calling it the Fall DIY. So in hindsight, I had all the time in the world (excepting the twins, the fact that the market--as I anticipated--fell off a cliff [WHEEEE!], having to pack my oldest off to continue graduate studies in Munich, Germany, my five year-old entering kindergarten [meetings with teachers and such], etc.) and could have done all sorts of things...possibly an entirely different circuit.
Oh, yeah. I also like to get at least four or five hours of sleep per night when possible. This week I'm averaging a less than four. Ugh.
No, I can't cut out watching TV. Gave that up thirty years ago.
Name dropping:
If you don't like Nelson, don't post in this Forum. If you know of a better one-stop-shop for instructions on how to match parts, feel free to post same. I could have puffed out the writeup with matching advice, but I was not being paid by the word and I was on a short lead, time-wise.
If you don't like Martians, stay the hell away from me...I don't want to get zapped if they take offense at your attitude and blast a couple of city blocks into glowing, radioactive dust in the process of relieving you of the burden of Earthly existence.
If you want to build a SSTART, please do. If you don't want to, then don't. What could be simpler? Why all this moaning and groaning?
Again, this is a thread about Nelson's F5. Go whine elsewhere.

One final point: Given the number of typos in your posts, you might want to invest in a good dictionary. If you are in an impecunious way, you could always try www.m-w.com or any of the other online dictionaries. It's hard to take someone seriously when they're pounding their chest in Alpha male fashion, but writing at a third grade level.
Unless you want to be mistaken for Fred, that is. He never did master written English.


Nelson Pass said:
Nice piece, Grey.

:cool:


Thanks.

Grey
 
Thank you Juma for posting something about actually constructing this amp, rather than all the other B.S.
I had given up on reading this thread.
Take all your ego's and shove them where the sun doesn't shine(or in some cases, where some of you think it does)............sorry for being so crass.

I have even forgotten which schematic your using for this board Juma, but it looks really good.
 
Formerly "jh6you". R.I.P.
Joined 2006
I re-opened my text book to clarify myself.

Yeah. . . the source R is a negative feedback resistor. This feedback is for a game of design compromise for stability by trading-off voltage gain. As far as I understand, Papa is also playing the source R as a feedback to find the "sweet spot", playing not only for the stability. . . Am I wrong, Papa?

YEAH. . . the book also talks about negative feedback (in OP-amp). In a negative feedback amplifier, the output is sampled and part of it is returned to the input. As far as I understand, when the output voltage is sampled by a voltage devider, we call it "voltage feedback". Meanwhile, when the output current is sampled to get a feedback voltage, we call it "current feedback". Wow, it says that the negative feedback signal produces remarkable changes in circuit performance. Advantages are stable gain, less distortion, and higher frequency response.
I believe, when we are mentioning negative feedback, we in general remind us of this type of feedback :dodgy: . . .

F5 is having a voltage feedback? I think so. But, the feedback is returned to the source of the input Jfets, not to the gate, with an effect of adjusting the value of the source R :confused:

Ah... Ya... I am at low level so that I have to always open the text books. . .

Cheers,

:)
 
The one and only
Joined 2001
Paid Member
Re: posturings of expertise?

wayness_tamm said:
Mr. Pass, why two negative feedback loops instead of one returned to the common connection of the jfet source resistors?

Assuming the use of pair of feedback resistors seperate from the jfet source resistors; what happenens as the feedback resistors impedance is made higher ( for the same closed loop gain) and approaches the approximate 30 ohms source impedance of the jfet pair, in terms of sonics and measurements. I am interested in the differences between voltage and "current" ( i.e. feedback resistance greater than the source impedance of the jfet pair)

It's a tweak. Try it both ways and see what you get.

When the feedback resistance is higher, the open loop gain of the
amplifier goes down.

:cool:
 
Yeah! Yeah!..........:censored:

Schematic is done, we will have a new project next week

.......30w Class A
........no global feedback
........comp.symmetry
........cascodes
........Stasis topology(with Mr.Pass's permission )
........no animals were harmed in the development of the design
........non-tree hugging design

I left out the current mirrors this time

What more do you want?

So where is my lobster dinner?

Jam
 
modification

Hi, what I have done is the following. I wanted at least 40 watt power (80 peek). so, increased the PSU voltage to +/- 29 volts with two toroids of 400 VA. I also increased the bias current to 1.6 A for both mosfets (two mosfets per channel). The gain I left it the same (5,7, i do not mess with the feedback) but I increased my preamp gain (I use class a bjt preamp) to about 4.4, in order to achive full power with 1 volt input. I sure have more power, now, more heat (I can deal with that) and everything goes well until now. My question is do I stress the output devices under this conditions? the dissipation is about 93 watt per devise. Is this acceptable if I use a sink of 0,5 c/w?
Thanks
 
modification

Thanks for your answers.
If we see the IRFP' specs we can see that max ratings for ~29 Volt is about 6 Ampere (for 25 vlt the ratings is about 7 Ampere). We approach this value only with 4ohm speakers. with 8 ohm speakers we consume max ~ 3.2 ampere in my situation. If we asume that we will drive only 8 ohm load then I think we are in the safe side. I am totaly wrong here?
Else I could dicrese the bias to 1,4 ~ 1,5 A for 30 ~ 35 class A watts.
 
jam said:
...........unless you cascode the second stage. I seem to recall Mati Otala did this in his groundbreaking amp design (bipolars here).

F5 is not usual design with followers at the output, therefore driving voltage of its output stage is very tiny compared to amps Vout.
You don't win much holding Vgd of my additional SF stage truly constant, just add unnecessary complexity to this otherwise very elegant design.
I don't know that Otala's design though.