tube-like SS power amp

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Read O.H.Schade (RCA 1939) "Beam Power Tubes"

Same technique when applied to a MOSFET results
in what Nelson Pass calls "Zen". A triode emulator.

-----------------------------------

There are two factors that local negative feedback
might linearize. Voltage or Current. Take your pick.
Can shoot for something inbetween, but can't have
it "perfect" both ways in one stage.

NFB from resistance below the emitter/source/cathode
tends to increase current linearity collector/drain/plate
with resulting increase in dynamic resistance/impedance.
You gain control over current, but lose control of voltage.
Your curves tend to look like a Transistor or Pentode.

NFB from a network tied to the anode voltage has
quite the opposite effect. It linearizes output voltage,
but lose some control of output current in exchange.
This drastically lowers dynamic resistance/impedance.
And your adjusted curves tend to look like a Triode.

-----------------------------------

Here, Schade feeds back 1/10th of 6L6 plate swing
back into the G1 grid. The curves look very similar to
a Triode strapping. But the screen actually tied high,
just as a MOSFET today might behave...
 

Attachments

  • fig35.gif
    fig35.gif
    21.3 KB · Views: 897
Take a look at guitar amplifiers. Various examples of getting "tube-like" sounds (whatever that means in each case) have been made since early 1970's. Just too much stuff to discuss in a single post but basically you'll find:
- high output impedance (usually achieved with current feedback)
- various soft clipping circuits
 
In addition I found the following quote from Mr.Pass in the Feastrex thread:


The whole point of the F3 was to emulate a triode with a power JFET
operating in the "linear" triode-like region. A fine SET would be the
go-to choice, and by "fine" I mean something with .1% thd or less
at 1 watt.
 
ahhh, this is my 'next' project, a simple amp with complex sound.

I am on the same learning path.

I have become a believer in the opinions of others that the tube-sound can only be approximated with SS and that a key to this includes the nature of harmonics generated. In this case 2nd and 3rd are good and higher are bad.

There are several ways to achieve this, no one circuit topology has the monopoly. But of all the devices, the JFET has the closest operating characteristics to a tube.

Why not build a hybrid, with tube input stage and SS power output stage for best of both worlds?

AKSA55 is commercial, you can't have the schematic and component values nor can you buy the kit anymore !!! but the circuit topology is (I believe) very similar to many other amplifiers presented on this forum and the harmonic profile is controlled through the imbalance of current flow through the two legs of the input Long-Tail-Pair.

I'm trying to make an AKSA55 here: http://www.diyaudio.com/forums/showthread.php?s=&threadid=140461&pagenumber=1
 
i know that ......

you are going to kill me with what i am going to say ....but i am going to say it any way .......


been constructing a gozilion amps .... tubes fets bjts name it ive done it ( be aware now i said constructed not designed ) .....

a well constructed BJT sziklai bootstraped and over biased will be my suggestion ....

many will not agree but i going to place my point for anyone interested ....

i would go for an amp that has amazing sonics than an amp that is too square and too perfect ....

i would go for something more everlasting than an amp that has limited life

i would go for efficiency than an amp that is a toaster

just last night i finished a ZCA ( zero component amp ) by mark houston ..... took me about 6 hours to finish it all from scratch

then i listen to it for another six .......

it was the most nice 1.5 watt i ver listen to my life .... ( in few words totally pointless)

thats my opinion .....

an amp designed arround original 2SA1302-3281 with rugged psu abit of overdesign and a bit of overbiased is simply unbeatable !!!!

thanks
 
Bigun said:
Why not build a hybrid, with tube input stage and SS power output stage for best of both worlds?


Carefully designed, maybe. You'll have to design the tube stage to be the limiting mechanism to truly get the 'tube sound'. If you let the SS amp clip, it will still sound like a SS amp clipping. If your 12ax7 or 6sn7 or whatever front end soft limit below clipping you'll probably get the sound you're after.
 
Good point.

I still have a mountain of work to build the 5 channel TGM amplifier before I start playing with a hybrid. But my initial thought was to build two single-channel amps. Both with same PSU and same SS output stage. The output stage should not be a limiting factor as you say. The input stage of one will have a JFET biassed so that it works in 'triode mode' (I won't say more on that other than to refer to Nelson) and the other will have a tube.

The tube has some specific power supply requirements. But I have a good friend with a bag of special Russian tubes that are about the size of a transistor. They are in metal cans. They run off low voltage. I think Musical Fidelity had these once too. Anyhow, they may not have the reputation of a specific NOS glass valve, but they for sure have the response curve of a triode. At least this is the thought de jour.
 
rtill said:
Just curious if there is any SS power amp out there that may have a tube-like warm sound or qualities.


A good approxcimation of a valve sound can be got with a soft limiter. Use tthat in conjunction with a standard SS amp.

I no longer have any cct diagrams of one but a Google search will bring up loads of soft limiters. It can certainly be done with an op-amp, a couple of transistors and a few more components and is much cheaper than a valve pre-amp or power amp.
 
Soft limiters usually only interact with the source.
Where a tube final interacts mostly with the load.
And there are enough exceptions that neither is
a statement of unshakable rule.

Makes no diff one way or the other into an 8 ohm
dummy load. But significant difference when a real
loudspeaker is the load.

You can certainly do this entirely with SS, just don't
dumb down to a soft limiter up front and linear final.
The difference between an "effect" and an "amp".
 
kenpeter said:
Soft limiters usually only interact with the source.
Where a tube final interacts mostly with the load.
And there are enough exceptions that neither is
a statement of unshakable rule.

Makes no diff one way or the other into an 8 ohm
dummy load. But significant difference when a real
loudspeaker is the load.

You can certainly do this entirely with SS, just don't
dumb down to a soft limiter up front and linear final.
The difference between an "effect" and an "amp".


I used a soft limiter with a pot to vary the feedback and got great results. Certainly a lot cheaper than buying a valve amp.
 
rtill said:
looking for that natural tube-warmth sound. Guess I'm looking for maybe an emulation of a tube power amp using SS parts. Maybe a type of FET power amp....

I've built Flavio Dellepiane's 60W mosfet power amp with common IRF's and sounds excellent! I use it as guitar amp, and when I overload the input, crunches really nice! Try it, it's placed in www.redcircuits.com
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.