Arny Krueger wrote:

"Patrick Turner" wrote in message


They are only as linear as a triode with regard to
voltage gain when external loops of NFB have been applied,
as in the case of the emitter follower connection or
having a few BJTs with a loop of NFB around the
lot of them to correct all their mistakes.

BJTs are inherently non linear for voltage gain and MUST
rely on large amounts of externally connected loops of
NFB.

Simply not true. Local feedback works well with BJTs.

Local loops are externally applied loops of NFB.

Bjts are so damned non linear with regard to voltage gain that large amount of
NFB, local or otherwises must be applied
to linearize the outcome, as well as raise input impedance.



This applies regardless of whether you build a preamp
with an opamp, a discrete transistor preamp,
or any power amp.

Triodes don't, they have a small amount of NFB built into
them.

Compared to BJTs, triodes have massive amounts of local NFB built into them,
and they are often used with addtiional local and/or loop NFB.

You don't know how much NFB is within any given triode now do you?

Please provide some calculations and details to prove to the group that you have
the slightest idea about what you are talking about.

Keep your answer to less than 5,000 words.



I would ask Trevor to set up a simple 1 transistor preamp
with a BC109, and without any
NFB applied, not even an emitter current feedback
resistor, and as a common emitter voltage amplifier with
a supply of 20V.

An apples-to-apples comparison would involve comparing circuits with gain
and impedances, both input and output, that are as typical and similar as
possible. However, this comparison is totally ludicrous, because nobody in
their right mind uses a stand-alone BC109 as a high grade audio amplifier
any more. They'll use some kind of an op amp, perhaps a NE 5532 or its
equivalent.

It isn't ludicous to drag people screaming back to the basics.
The basic fact is that bjts WITHOUT ANY NFB, LOCAL OR OTHERWISE applied anywhere
have hopelessly poor
voltage gain linearity.
Statements have been made on the group that what i am saying is untrue.

All opamps also must be used with NFB, and they contain many interconnected bjts
like many
discrete component amplifier, so the non linearity of all these adds together
and must be linearized with NFB.
opamps usually have way too much gain without NFB; and usually poor open loop
bandwidth,
something also flattened out by NFB.


To understand his own utter stupidity, I ask Trevor to
also set up a 6SN7 with a 250V supply to act as a preamp
and without any loop NFB, and he can then tell us all
about the results which will prove what a
jackarse he is.

Illogical comparison. Everybody knows that 6SN7s have a ton of internal
feedback, and that BJTs compare most closely to pentodes, not triodes.

A "ton" of NFB is an extremely uninformative unit applied to NFB.

Please try be specific and factual lest we consider you to be utterly stupid
like Trevor.

Exactly how much NFB is within a 6SN7?

I will also say that a 6AU6 pentode would be more linear at a volt of output
than a BC109 at a volt of output,
with both devices tested without NFB of any kind.

Unless you have done the experiments and comparisons, you remain uneducated
about the basics.

When people talk about the linearity of devices, what is being stated is the
linearity of the devices without
any external loops of NFB applied. That is what engineers understand when
somebody talks of device linearity.

bjts offer fair current linearity between base current and collector current but
offer hopeless
voltage linearity.
Amplifiers are built to provide voltage linearity regardless of current
variations due to load variations.

External loops in circuit applications of NFB are never included in discussions
defining device linearity.

You cannot say a bjt is linear if you are using an emitter follower as an
example of the device use.
The term "emitter follower" is a description of a circuit where a large amount
of series voltage NFB is
employed so that the voltage gain of the bjt is reduced from a disgustingly non
linear large number, say
100 times to say 0.99, and the open loop THD is reduced 100 times.

I will say to that where you have a 12AX7 set up as a cathode follower then
there is a gain reduction
of say 75 to just less than 1.0, so the linearity of the device cannot be known
without more provisos.

if you have an EF bjt circuit and THD = 0.01% at 1Vrms and if open loop gain was
100,
then the THD without the follower connection with the load used will be 100 x
0.01 = 1%.

The 1/2 12AX7 in a CF will typically produce only 0.0005% THD at 1V out.
If the same tube was working without the CF connection but in common cathode
the THD = 75 x 0.0005 = 0.0375%, and a small fraction of what is produced by the
BC109 without any NFB.




Patrick Turner.