How to bias a MOSFET amp?

B

Bill N2CQR MOHBR

Guest
I'm trying to improve my limited understanding of how RF amps operate.
I've been studying "Solid State Design for the Radio Amateur" and
Experimental Methods in RF Design"

SSDRA has a very helpful section that asks the reader to calculate max
output power for a Class A RF amp (common emitter). They ask the
reader to consider maximum voltage and current swings in the collector
circuit that will keep the output linear. Basically, my understanding
is that (with an RF choke in the Vcc line) max peak signal voltage on
collector is Vc-Ve. Max peak current is the standing or quiescent
current. In this way when the collector voltage is hitting its peak
collector current is dropping almost to zero.

While the SSDRA example uses BJTs, I'm guessing that essentially the
same restrictions would apply to the drain circuit of a MOSFET amp.

Here's my problem: When I look at MOSFET amp circuits in the
literature, they often have linear amps with 12 volts on the drain,
but with standing currents of only around 40 milliamps. And they
claim 7 watts out. How can that be? Using the analysis outlined
above, I'd think that you'd get max output of 12*.040 = .48 watts.

Here's an example: I've been looking at Farhan's very FB SSB
Transceiver
http://farhan.net.co.nr/xcvr1.html

I'm trying to understand the biasing on his IRF510 final, and the RF
output he's getting.

He says he measures 20-24 volts peak RF across a 50 ohm load at the
output. That's about 8 watts peak output.

He's using 12 volt supply, and recommends setting the idle current
through the MOSFET at 80 ma. Can that be right? According to my
reading of Solid State Design for the Radio Amateur (SSDRA)(page 23)
with a 12 volt supply we can expect peak signal voltage at the Drain
of around 12 volts (with an RF choke in Vcc line). Peak current could
be max 80 ma.(maintaining Class A). Under these biasing conditions,
assuming Class A operation, max output power of .96 watts would be
provided by a load of 150 ohms.

Even if he were to be running this amp Class B (or close to it), I
can't see how he'd get 8 watts out with only 80 milliamps of standing
current

I'm very new to this kind of analysis, and strongly suspect that I'm
misreading either SSDRA or Farhan's excellent article.

Can someone please let me know where this apparent discrepency is
coming from.

Thanks and 73

Bill N2CQR M0HBR CU2JL
http://planeta.clix.pt/n2cqr
 
On 26 Aug 2004 22:40:36 -0700, meara.london@virgin.net (Bill N2CQR
MOHBR) wrote:

I'm trying to improve my limited understanding of how RF amps operate.
I've been studying "Solid State Design for the Radio Amateur" and
Experimental Methods in RF Design"
Click to expand...
Keep studying.

The static bias is the zero-signal drain current and your calculation
is for static DC loss. Signals applied to the gate can add to the
drain current, as well as subtract.

RF final amps are unlikely to be biased classA. Class B and C will
have the final acting more like a switch. Current is imited by the
load impedance - which in this case is a complex impedance that looks
like 50 ohms at the output only.

As to the answer for your posting - the fet in this case is biased by
varying gate static voltage using the variable resistor in the
schematic, as instructed.

RL
 
It's a bias voltage, rather than a bias current. The MOSFET enters its
linear conducting area after a certain threshold voltage is attained.

"Bill N2CQR MOHBR" <meara.london@virgin.net> wrote in message
news:22f6e3ee.0408262140.383c9581@posting.google.com...
I'm trying to improve my limited understanding of how RF amps operate.
I've been studying "Solid State Design for the Radio Amateur" and
Experimental Methods in RF Design"

SSDRA has a very helpful section that asks the reader to calculate max
output power for a Class A RF amp (common emitter). They ask the
reader to consider maximum voltage and current swings in the collector
circuit that will keep the output linear. Basically, my understanding
is that (with an RF choke in the Vcc line) max peak signal voltage on
collector is Vc-Ve. Max peak current is the standing or quiescent
current. In this way when the collector voltage is hitting its peak
collector current is dropping almost to zero.

While the SSDRA example uses BJTs, I'm guessing that essentially the
same restrictions would apply to the drain circuit of a MOSFET amp.

Here's my problem: When I look at MOSFET amp circuits in the
literature, they often have linear amps with 12 volts on the drain,
but with standing currents of only around 40 milliamps. And they
claim 7 watts out. How can that be? Using the analysis outlined
above, I'd think that you'd get max output of 12*.040 = .48 watts.

Here's an example: I've been looking at Farhan's very FB SSB
Transceiver
http://farhan.net.co.nr/xcvr1.html

I'm trying to understand the biasing on his IRF510 final, and the RF
output he's getting.

He says he measures 20-24 volts peak RF across a 50 ohm load at the
output. That's about 8 watts peak output.

He's using 12 volt supply, and recommends setting the idle current
through the MOSFET at 80 ma. Can that be right? According to my
reading of Solid State Design for the Radio Amateur (SSDRA)(page 23)
with a 12 volt supply we can expect peak signal voltage at the Drain
of around 12 volts (with an RF choke in Vcc line). Peak current could
be max 80 ma.(maintaining Class A). Under these biasing conditions,
assuming Class A operation, max output power of .96 watts would be
provided by a load of 150 ohms.

Even if he were to be running this amp Class B (or close to it), I
can't see how he'd get 8 watts out with only 80 milliamps of standing
current

I'm very new to this kind of analysis, and strongly suspect that I'm
misreading either SSDRA or Farhan's excellent article.

Can someone please let me know where this apparent discrepency is
coming from.

Thanks and 73

Bill N2CQR M0HBR CU2JL
http://planeta.clix.pt/n2cqr
Click to expand...
 
On 26 Aug 2004 22:40:36 -0700, meara.london@virgin.net (Bill N2CQR
MOHBR) wrote:


Even if he were to be running this amp Class B (or close to it), I
can't see how he'd get 8 watts out with only 80 milliamps of standing
current

I'm very new to this kind of analysis, and strongly suspect that I'm
misreading either SSDRA or Farhan's excellent article.

Can someone please let me know where this apparent discrepency is
coming from.

Thanks and 73

Bill N2CQR M0HBR CU2JL
http://planeta.clix.pt/n2cqr
Click to expand...

I think the RF guys (I'm not one!) call an amplifier "linear" if the
RF output amplitude follows the input drive amplitude. You can do this
with a transistor that has very low quiescent bias. So "linear" does
not mean "class A" to them. The key here is that an RF amp has a tuned
output, whereas an audio amp doesn't. So the lopsided bias would
normally produce intolerable distortion in something like audio, but
the tuned output circuit changes the pulsey-looking collector/drain
current back into a nice sine wave. So you don't need a lot of idle
current, and the transistor really amplifies half of the incoming sine
cycle.

Most mosfets are pretty nicely linear (ie, straight-line Ic/Vd curve)
beyond the initial knee. You could get gobs of watts at zero standing
current, but then you'd have some zero-clipping (no output) for the
smallest drive levels, so a little idle current helps.

John
 
On Fri, 27 Aug 2004 08:07:02 -0700, John Larkin
<jjlarkin@highlandSNIPtechTHISnologyPLEASE.com> wrote:

I think the RF guys (I'm not one!) call an amplifier "linear" if the
RF output amplitude follows the input drive amplitude. You can do this
with a transistor that has very low quiescent bias. So "linear" does
not mean "class A" to them. The key here is that an RF amp has a tuned
output, whereas an audio amp doesn't. So the lopsided bias would
normally produce intolerable distortion in something like audio, but
the tuned output circuit changes the pulsey-looking collector/drain
current back into a nice sine wave. So you don't need a lot of idle
current, and the transistor really amplifies half of the incoming sine
cycle.

Most mosfets are pretty nicely linear (ie, straight-line Ic/Vd curve)
beyond the initial knee. You could get gobs of watts at zero standing
current, but then you'd have some zero-clipping (no output) for the
smallest drive levels, so a little idle current helps.
Click to expand...
The only sensible way to do it AFAICS is to operate the MOSFET in
class C as a high speed switch and reconstruct the pulsed output into
a sine wave carrier by means of a suitable tuned circuit. I wouldn't
consider driving a MOSFET for RF use in any other way. The efficiency
should be pretty darned good, too.
--

"What is now proved was once only imagin'd." - William Blake, 1793.
 
On Fri, 27 Aug 2004 19:28:39 +0100, Paul Burridge
<pb@notthisbit.osiris1.co.uk> wrote:

On Fri, 27 Aug 2004 08:07:02 -0700, John Larkin
jjlarkin@highlandSNIPtechTHISnologyPLEASE.com> wrote:

I think the RF guys (I'm not one!) call an amplifier "linear" if the
RF output amplitude follows the input drive amplitude. You can do this
with a transistor that has very low quiescent bias. So "linear" does
not mean "class A" to them. The key here is that an RF amp has a tuned
output, whereas an audio amp doesn't. So the lopsided bias would
normally produce intolerable distortion in something like audio, but
the tuned output circuit changes the pulsey-looking collector/drain
current back into a nice sine wave. So you don't need a lot of idle
current, and the transistor really amplifies half of the incoming sine
cycle.

Most mosfets are pretty nicely linear (ie, straight-line Ic/Vd curve)
beyond the initial knee. You could get gobs of watts at zero standing
current, but then you'd have some zero-clipping (no output) for the
smallest drive levels, so a little idle current helps.

The only sensible way to do it AFAICS is to operate the MOSFET in
class C as a high speed switch and reconstruct the pulsed output into
a sine wave carrier by means of a suitable tuned circuit. I wouldn't
consider driving a MOSFET for RF use in any other way. The efficiency
should be pretty darned good, too.
Click to expand...
---
That doesn't make any sense to me.

Unless things have changed pretty drastically from how they were when
I was doing RF, class "C" was pretty much relegated to FM, so that
when you hit PTT, you banged the hell out of the final and filtered
the hell out of the carrier, which went to maximum amplitude and
stayed there, and the information was put on the constant amplitude
carrier by varying its frequency (or phase).

AM and SSB finals were _always_ linear amps and, like John said, the
_amplitude_ of the carrier/sideband(s) followed the amplitude of the
modulating audio precisely.

Whether you use a MOSFET as a switch or as a resistive element
yielding a linearly varying output depends on how you tailor the
characteristics of the MOSFET to fit the application. After all,
there are lots of linear audio amps out there with MOSFET class A and
class B finals, aren't there? So why shouldn't there be linear MOSFET
RF amps as well?

--
John Fields
 
On Fri, 27 Aug 2004 19:28:39 +0100, Paul Burridge
<pb@notthisbit.osiris1.co.uk> wrote:

On Fri, 27 Aug 2004 08:07:02 -0700, John Larkin
jjlarkin@highlandSNIPtechTHISnologyPLEASE.com> wrote:

I think the RF guys (I'm not one!) call an amplifier "linear" if the
RF output amplitude follows the input drive amplitude. You can do this
with a transistor that has very low quiescent bias. So "linear" does
not mean "class A" to them. The key here is that an RF amp has a tuned
output, whereas an audio amp doesn't. So the lopsided bias would
normally produce intolerable distortion in something like audio, but
the tuned output circuit changes the pulsey-looking collector/drain
current back into a nice sine wave. So you don't need a lot of idle
current, and the transistor really amplifies half of the incoming sine
cycle.

Most mosfets are pretty nicely linear (ie, straight-line Ic/Vd curve)
beyond the initial knee. You could get gobs of watts at zero standing
current, but then you'd have some zero-clipping (no output) for the
smallest drive levels, so a little idle current helps.

The only sensible way to do it AFAICS is to operate the MOSFET in
class C as a high speed switch and reconstruct the pulsed output into
a sine wave carrier by means of a suitable tuned circuit. I wouldn't
consider driving a MOSFET for RF use in any other way. The efficiency
should be pretty darned good, too.
Click to expand...
That means, to get a linear amp, the input signal has to be converted
to PWM gate drive. That's hard to do at high frequencies. At 300 MHz,
a power mosfet doesn't much look like a high-speed switch any more.

John
 
John Larkin wrote:
On 26 Aug 2004 22:40:36 -0700, meara.london@virgin.net (Bill N2CQR
MOHBR) wrote:



Even if he were to be running this amp Class B (or close to it), I
can't see how he'd get 8 watts out with only 80 milliamps of standing
current

I'm very new to this kind of analysis, and strongly suspect that I'm
misreading either SSDRA or Farhan's excellent article.

Can someone please let me know where this apparent discrepency is
coming from.

Thanks and 73

Bill N2CQR M0HBR CU2JL
http://planeta.clix.pt/n2cqr



I think the RF guys (I'm not one!) call an amplifier "linear" if the
RF output amplitude follows the input drive amplitude. You can do this
with a transistor that has very low quiescent bias. So "linear" does
not mean "class A" to them. The key here is that an RF amp has a tuned
output, whereas an audio amp doesn't. So the lopsided bias would
normally produce intolerable distortion in something like audio, but
the tuned output circuit changes the pulsey-looking collector/drain
current back into a nice sine wave. So you don't need a lot of idle
current, and the transistor really amplifies half of the incoming sine
cycle.

Most mosfets are pretty nicely linear (ie, straight-line Ic/Vd curve)
beyond the initial knee. You could get gobs of watts at zero standing
current, but then you'd have some zero-clipping (no output) for the
smallest drive levels, so a little idle current helps.

John

RF guys call the amplifier "linear" if the output, after filtering,
Click to expand...
looks like a bigger version of the input -- basically the same criterion
as any other amplifier. The reason that you can get away with half as
many active elements as with an audio amplifier is because if the
modulation is narrow compared to the carrier each half of the waveform
looks the same, so amplifying half of it then filtering reconstructs the
half you didn't play with.

Class A amplifiers (and push-pull class AB or B amplifiers) are used in
RF work, but mostly because they cut down on the harmonics that must be
filtered out.

--

Tim Wescott
Wescott Design Services
http://www.wescottdesign.com
 
Paul Burridge wrote:

On Fri, 27 Aug 2004 08:07:02 -0700, John Larkin
jjlarkin@highlandSNIPtechTHISnologyPLEASE.com> wrote:


I think the RF guys (I'm not one!) call an amplifier "linear" if the
RF output amplitude follows the input drive amplitude. You can do this
with a transistor that has very low quiescent bias. So "linear" does
not mean "class A" to them. The key here is that an RF amp has a tuned
output, whereas an audio amp doesn't. So the lopsided bias would
normally produce intolerable distortion in something like audio, but
the tuned output circuit changes the pulsey-looking collector/drain
current back into a nice sine wave. So you don't need a lot of idle
current, and the transistor really amplifies half of the incoming sine
cycle.

Most mosfets are pretty nicely linear (ie, straight-line Ic/Vd curve)
beyond the initial knee. You could get gobs of watts at zero standing
current, but then you'd have some zero-clipping (no output) for the
smallest drive levels, so a little idle current helps.


The only sensible way to do it AFAICS is to operate the MOSFET in
class C as a high speed switch and reconstruct the pulsed output into
a sine wave carrier by means of a suitable tuned circuit. I wouldn't
consider driving a MOSFET for RF use in any other way. The efficiency
should be pretty darned good, too.
Click to expand...
Don't operate SSB much, do you?

--

Tim Wescott
Wescott Design Services
http://www.wescottdesign.com
 
John Fields wrote:

On Fri, 27 Aug 2004 19:28:39 +0100, Paul Burridge
pb@notthisbit.osiris1.co.uk> wrote:


-- snip --
The only sensible way to do it AFAICS is to operate the MOSFET in
class C as a high speed switch and reconstruct the pulsed output into
a sine wave carrier by means of a suitable tuned circuit. I wouldn't
consider driving a MOSFET for RF use in any other way. The efficiency
should be pretty darned good, too.


---
That doesn't make any sense to me.

Unless things have changed pretty drastically from how they were when
I was doing RF, class "C" was pretty much relegated to FM, so that
when you hit PTT, you banged the hell out of the final and filtered
the hell out of the carrier, which went to maximum amplitude and
stayed there, and the information was put on the constant amplitude
carrier by varying its frequency (or phase).

AM and SSB finals were _always_ linear amps and, like John said, the
_amplitude_ of the carrier/sideband(s) followed the amplitude of the
modulating audio precisely.

Whether you use a MOSFET as a switch or as a resistive element
yielding a linearly varying output depends on how you tailor the
characteristics of the MOSFET to fit the application. After all,
there are lots of linear audio amps out there with MOSFET class A and
class B finals, aren't there? So why shouldn't there be linear MOSFET
RF amps as well?

Well, AM tube finals were often operated class C with the modulation
Click to expand...
applied to the plate supply. This is harder to do with silicon because
the varying collector voltage modulates the collector-base capacitance
and causes weird phase shifts.

And there are linear MOSFET RF amps; they're necessary for single-sideband.

--

Tim Wescott
Wescott Design Services
http://www.wescottdesign.com
 
On Fri, 27 Aug 2004 08:07:02 -0700, John Larkin
<jjlarkin@highlandSNIPtechTHISnologyPLEASE.com> wrote:


Most mosfets are pretty nicely linear (ie, straight-line Ic/Vd curve)
^^^^^
Click to expand...

Oops, I meant Id/Vg. But you all knew that.

John
 
On Fri, 27 Aug 2004 13:00:56 -0700, John Larkin
<jjlarkin@highSNIPlandTHIStechPLEASEnology.com> wrote:

That means, to get a linear amp, the input signal has to be converted
to PWM gate drive. That's hard to do at high frequencies. At 300 MHz,
a power mosfet doesn't much look like a high-speed switch any more.
Click to expand...
Certainly not at that kind of frequency! But for the lower HF bands,
it's *perfectly* feasible.
--

"What is now proved was once only imagin'd." - William Blake, 1793.
 
On Fri, 27 Aug 2004 15:40:15 -0700, Tim Wescott
<tim@wescottnospamdesign.com> wrote:

Don't operate SSB much, do you?
Click to expand...
Nope. I'm a CWer. But the use of MOSFETs at RF for Anything other than
SSB (FM & AM in particular are ideally-suited) is as Kosher as Jim
Thompson's Saturday afternoon lunch of salt beef sandwiches with extra
dill.

--

"What is now proved was once only imagin'd." - William Blake, 1793.
 
On Fri, 27 Aug 2004 15:42:53 -0700, Tim Wescott
<tim@wescottnospamdesign.com> wrote:


Well, AM tube finals were often operated class C with the modulation
applied to the plate supply.
Click to expand...
---
Funny, I never considered plate modulation to be class C; that is if
we're talking about the same thing. What I'm thinking about is when
you key the transmitter on and it starts putting out a carrier at some
level, then you modulate the plate supply with audio so that at the
low peaks of the audio waveform the output of the TX is zero, but at
the output of the high peaks it's twice (?) what it was with no
modulation. Is that class C?

--
John Fields
 
On Sat, 28 Aug 2004 00:28:21 +0100, Paul Burridge
<pb@notthisbit.osiris1.co.uk> wrote:

On Fri, 27 Aug 2004 13:00:56 -0700, John Larkin
jjlarkin@highSNIPlandTHIStechPLEASEnology.com> wrote:

That means, to get a linear amp, the input signal has to be converted
to PWM gate drive. That's hard to do at high frequencies. At 300 MHz,
a power mosfet doesn't much look like a high-speed switch any more.

Certainly not at that kind of frequency! But for the lower HF bands,
it's *perfectly* feasible.
Click to expand...
Have you actually built a class C linear RF power amp? Tell us how it
works.

John
 
John posted:
<< On Fri, 27 Aug 2004 15:42:53 -0700, Tim Wescott
<tim@wescottnospamdesign.com> wrote:


Well, AM tube finals were often operated class C with the modulation
applied to the plate supply.
Click to expand...
---
Funny, I never considered plate modulation to be class C; that is if
we're talking about the same thing. What I'm thinking about is when
you key the transmitter on and it starts putting out a carrier at some
level, then you modulate the plate supply with audio so that at the
low peaks of the audio waveform the output of the TX is zero, but at
the output of the high peaks it's twice (?) what it was with no
modulation. Is that class C?
Yes, that's class C. AM, CW, and FM (and NBFM) almost invariably used Class C
Final due to the high efficiency (about 80%).

For 100% modulation the audio reqired from the modulator (for plate modulation,
aka high level modulation) is 50% of the RF input power. For 100 Watts RF
input, 50 Watts of audio will modulate 100% and result in 25 Watt USB and a 25
Watt LSB.

Don
 
On Fri, 27 Aug 2004 19:01:37 -0500, John Fields wrote:

On Fri, 27 Aug 2004 15:42:53 -0700, Tim Wescott
tim@wescottnospamdesign.com> wrote:

Well, AM tube finals were often operated class C with the modulation
applied to the plate supply.

---
Funny, I never considered plate modulation to be class C; that is if
we're talking about the same thing. What I'm thinking about is when
you key the transmitter on and it starts putting out a carrier at some
level, then you modulate the plate supply with audio so that at the
low peaks of the audio waveform the output of the TX is zero, but at
the output of the high peaks it's twice (?) what it was with no
modulation. Is that class C?
Click to expand...
It can be. Depends on the biasing. It *is* AM, however.

Class C just means that the conduction angle is less than 180 deg. I
recall seeing it specified as a certain amount less than 180 once,
but I don't recall how much. The idea is to give the output tank
enough kick to get it to swing.

So you amplify the carrier with class C and modulate the plate. I've
seen projects which do this with bipolars. One that comes to mind is
an old Radio Electronics article for a uWave ATV commo system using
gunplexers, but that can't be right since gunnplexers are modulated
at the gunnplexer. I think it was another ATV project. If I ever
find that old article I'll letcha know.
--
Best Regards,
Mike
 
John Fields wrote:
On Fri, 27 Aug 2004 15:42:53 -0700, Tim Wescott
tim@wescottnospamdesign.com> wrote:



Well, AM tube finals were often operated class C with the modulation
applied to the plate supply.


---
Funny, I never considered plate modulation to be class C; that is if
we're talking about the same thing. What I'm thinking about is when
you key the transmitter on and it starts putting out a carrier at some
level, then you modulate the plate supply with audio so that at the
low peaks of the audio waveform the output of the TX is zero, but at
the output of the high peaks it's twice (?) what it was with no
modulation. Is that class C?

In a tube setup the RF amplifier should be operating in class C and the
Click to expand...
power audio amplifier should provide nice linear modulation to the RF
amp's plate supply. In fact* one needs to provide sufficient excitation
to the RF final, too, lest the thing go into a current limited mode on
the modulation peaks.

Presumably you could make a REALLY EFFICIENT setup with transistors by
operating the RF final in class E, but you get that pesky capacitance
problem back...

* So I understand, I'm just a tube wannabe.

--

Tim Wescott
Wescott Design Services
http://www.wescottdesign.com
 
Paul Burridge wrote:

On Fri, 27 Aug 2004 15:40:15 -0700, Tim Wescott
tim@wescottnospamdesign.com> wrote:


Don't operate SSB much, do you?


Nope. I'm a CWer. But the use of MOSFETs at RF for Anything other than
SSB (FM & AM in particular are ideally-suited) is as Kosher as Jim
Thompson's Saturday afternoon lunch of salt beef sandwiches with extra
dill.

Motorola used to list RF MOSFETs for linear amplifier use. I don't know
Click to expand...
if they went to On semi or FreeFall.

--

Tim Wescott
Wescott Design Services
http://www.wescottdesign.com
 
On Fri, 27 Aug 2004 18:09:43 -0700, Tim Wescott wrote:

John Fields wrote:
On Fri, 27 Aug 2004 15:42:53 -0700, Tim Wescott
tim@wescottnospamdesign.com> wrote:

Well, AM tube finals were often operated class C with the modulation
applied to the plate supply.

---
Funny, I never considered plate modulation to be class C; that is if
we're talking about the same thing. What I'm thinking about is when
you key the transmitter on and it starts putting out a carrier at some
level, then you modulate the plate supply with audio so that at the
low peaks of the audio waveform the output of the TX is zero, but at
the output of the high peaks it's twice (?) what it was with no
modulation. Is that class C?

In a tube setup the RF amplifier should be operating in class C and the
power audio amplifier should provide nice linear modulation to the RF
amp's plate supply. In fact* one needs to provide sufficient excitation
to the RF final, too, lest the thing go into a current limited mode on
the modulation peaks.

Presumably you could make a REALLY EFFICIENT setup with transistors by
operating the RF final in class E, but you get that pesky capacitance
problem back...

* So I understand, I'm just a tube wannabe.
Click to expand...
I don't know how well it works, but I saw a class E schem using a
section of line to take out the odd harmonics.
--
Best Regards,
Mike
 
You must log in or register to reply here.

Welcome to EDABoard.com

Sponsor

Back
Top