Zen-V8.pdf

(532 KB) Pobierz
The Power JFET Amplifier – Zen Variations #8
By Nelson Pass
Introduction
Thanks to a nice person on the Pass Labs Forum (
www.diyaudio.com
), I
became aware that high current power JFET transistors are again available.
You can check them out at
www.lovoltech.com
which offers a small variety of
high current N-channel vertical JFETs in TO-251 and TO-252 packages. They
don’t hold a lot of voltage 20-24 volts) or dissipate lot of power (69 watts,
derated at 1.8 watts/degree), but they do deliver up to 100 amps peak. Clearly
designed as switches, they nevertheless offer a linear operating region and are
attractive for audio amplifier design.
Small signal JFETs are routinely used in the input stages of the finest solid-state
amplifiers and preamplifiers, where they give very high input impedance, high
linearity, and very low noise. JFETs are routinely extolled in promotional
literature for their “tube-like” qualities. Their characteristic curves, like
MOSFETs, largely resemble pentodes.
In the 1970’s, Sony and Yamaha offered a few JFET power amplifiers, but
discontinued them for unknown reasons. Figure 1 shows a simplified schematic
of the Yamaha B1, remains a highly regarded product.
Why power JFETs you ask? Don’t we have enough transistors already? We do
have lots of choices in devices, but a vertical JFET brings some particular
characteristics to the table that we don’t get elsewhere. When we look at
characteristic curves for devices we see what I would call the “triode character”
and the “pentode character”, as illustrated in Figure 2.
Here we see that the triode character has an partially exponential curve of
current versus voltage across the device, while the pentode has a more
logarithmic shape. You could think of the triode as “concave” and the pentode
as “convex”, and by comparison, a resistor would have a straight line.
Since no gain device does such a good job of approximating a resistor, triode
designers linearize the tube circuit by operating the voltage/current load line so
as to get some cancellation between competing voltage and current distortion.
This can be done with either single-ended or push-pull circuits.
The guys with pentode type devices, which includes solid state as well as non-
triode tubes, take a different approach, which is to try to work them in the more
voltage-invariant area and use other methods to correct for remaining non-
linearities, usually through negative feedback.
I’m not here to argue the merits of one approach over another, but as I look at
the characteristic lines for these new power JFETs I notice that there is a
narrow region at the bottom where there is less convex curvature. In fact it
looks concave,
kind of like a triode.
In the case of JFET part LU1014, we note that with a gate voltage of –1 volt, the
curve is concave below about 5 amps and 4 volts. In this range it has that
triode character, and this is the area of interest to us here.
There are some other potential advantages to these JFETs. For a given bias
current, they are more linear than comparable MOSFETs, and while they have
comparable capacitances, the values are more linear over the operating
voltages. Depletion mode is convenient, as it offers self-biasing with no
particular circuitry to set the bias point, and the devices have a small negative
temperature coefficient, so they hold a stable bias point easily. Last but not
least, JFETs are not sensitive to static discharge,
The Single Approach
If we have a device that behaves a bit like a triode, then it is natural to try it out
in a popular triode amplifier topology. For many aficionados that topology would
be single-ended Class A operation. 300B’s and 211’s operated single-ended
and coupled to an output transformer have been held by many as a low power
musical standard. Single-Ended Class A tube amplifiers are not very powerful,
and their measurements are nothing to write home about, but there is no
denying that they have strong musical appeal to a sizable segment of the
audiophile population.
Well, it seems easy enough, we’ll just pop one of these self-biased little buggers
in series with a transformer, as shown in Figure 4. The gain and input
impedance of the transistor are such that we won’t even need more than this
single stage – a lot easier than the three gain stages of your better tube amps
and no fussing around with filament supplies, either.
The Problems
There are some reasons why the circuit of Figure 4 might not be appropriate.
First, as shown the JFET will draw a tremendous amount of current, and quickly
take up smoking. We can fix that with some resistance on the Source to
ground. Looking at the curves, maybe an ohm or so.
While we’re fixing things, perhaps we can do without the transformer at all. In
tube circuits the transformer converts the high voltage / low current of the tube’s
terrain to the relatively low voltage / high current of the speaker. Again looking
at the JFET’s curves, we see that perhaps it isn’t necessary, and we can
replace the transformer with a resistor and output capacitor, as in Figure 5. If
you think that a capacitor is not an improvement over a transformer, then you’re
welcome to your opinion, and in fact the transformer will raise the efficiency of
the amplifier, but we are going to leave that for another time.
As we slowly crank up the supply voltage, we see that this certainly works, but
the distortion is kind of high, and when we raise the voltage enough to get a
good listen, we notice that the high end is a little deficient. After a few exciting
seconds, the amplifier makes a popping noise, and then:
Nothing.
OK, we’re making progress anyway. Our problems (hereafter referred to as
opportunities) are fairly simple. First, the transistor in question has limited
dissipation. Even though it’s rated at 69 watts, realistically I’d be afraid to
operate the package at more than about 10 watts or so. A TO-251 is a
small
device, and you have to work to get the heat away from it.
Second, the transistor is rated at only 24 volts. That just about enough to make
a 2 watt amplifier with this circuit, and under these circumstances the transistor
will be dissipating something approaching 16 watts. We might get away with
this, but we still have pretty high distortion, and the top end is still down.
We can fix the top end – just get a preamp with a low output impedance. The
400 pF capacitance between the Gate and the Drain of the JFET is being
amplified by the gain of the amplifier, and looks like more than 3200 pF. What?
You say your preamp has a 10K output impedance? We can fix that.
Even with a low impedance preamp, the distortion is still high. Some
opportunities just seem insurmountable.

Plik z chomika:

Inne pliki z tego folderu:

Inne foldery tego chomika:

Zgłoś jeśli naruszono regulamin