Reply
 
Thread Tools Display Modes
  #1   Report Post  
Posted to rec.audio.tubes
Alex Pogossov Alex Pogossov is offline
external usenet poster
 
Posts: 95
Default AM detector, part 2

Modulation handling of an AM detector is determined by:
1. Carrier level and low level sensitivity;
2 AC impedance / DC load resistance ratio;
3. Slew rate.

If say we are aiming at 95% undistorted demodulation, then the carrier level
shall be about 20 times greater than the sensitivity threshold. For example,
for an unbiased detector with a silicon diode, which has a 0.6V knee, we
need at least 12V carrier. (Unless otherwise stated, amplitude is implied.)
For a properly designed vacuum diode (Vsens = 100mV) you need at least 2V
carrier.

For an optimum biased semiconductor diode (Vsens = 25mV) you need only 0.5V
of carrier for distortion free 95% demodulation. Thus a radio using a
PROPERLY designed p-n diode detector can run at low IF levels, have less IF
gain, less IF stages, would require less shielding and have better
sensitivity.

It is well known that for large signals maximum modulation index
approximately equals to AC / DC load impedance. The ways to make it close to
1 at lower frequencies a
- use a DC coupled hi-Z follower (cathode, source, op-amp buffer, high-beta
BJT with low base current, etc.);
- never AC couple a volume control to the detector. Instead, make the volume
control pot *the* load (or a part thereof);
- have a high (5...10M) input impedance of the audio amp, so that it does
not load the detector even at full volume. (This stuff has been mentioned in
a different thread);
- never connect AGC voltage RC filter to the detector load. Use a separate
AGC detector.

In an unbiased detector slew rate issues are also related to the AC/DC
impedance ratio. For example, a booooring detector with 500K || 100pF load
can handle 70% at about 3.5kHz only. To get things worse, a 47K+100pF ripple
filter is added straight after the detector. This cap virtually adds to the
effect of reducing the modulation handling at the highs. With 100pF load +
100pF filter the 70% cutoff comes down to the appauling 1.8kHz. No wonder
the sound of a booooring radio is so crappy.

There is another less known effect. Not only the heavy C distorts HF
modulation due to slew rate limitations, it also reduces original modulation
index at HF, working like sort of high-cut tone control. To understand the
phenomenon without a deep maths, consider that on a steeply rising RF
envelope the detector has to charge the load capacitance. This sucks extra
energy from the hi-Z IFT on top what is to be dissipated in the resistive
component of the load. On the steeply falling RF envelope slopes, R is being
fed from a discharging C instead of the diode. Thus the IFT gets unoaded on
the falling slopes. It is easy to see that the peaks are thus "cut and
rounded" and the troughs are "filled". Modulation virtually reduces. This
reduces the slew rate distortion, replacing it with a HF cut. In the end the
sound is still crappy. (Those who are familiar with the operation of a ratio
FM detector, as opposed to a Sheeley discriminator, will see many
parallels.)

The above HF unmodulation takes place only if the detector is directly fed
from a hi-Z IFT. If a buffer (cathode) follower is used, there is no HF
unmodulation phenomenon.

In unbiased detectors there is no high limit to carrier level (until the
diode breaks down) -- discharge current is proportional to the carrier
level. Slew rate perfomance does not depend on the RF signal magnitude.

This is not the case with biased detectors, where the discharge current is
(almost) constant. Here it is time to analyse the famous Partick's biased
detector.

Schematic values may vary, but here let us assume it is biased to 50V, uses
a cathode follower on 12AU7, a semiconductor diode, has 220pF of C,
pull-down resistor of 500K and a ripple filter of 100K+100pF. Thus the diode
bias current is about 100uA at no signal.

A DC biased diode, as I hope people know, has differential resistance of
(25 ohms / current, mA). In this case, at no signal and at very low signal
the diode acts as a 250R resistor. Capacitor of 220pF has about 1.6K
reactance at 455kHz. Thus the whole D+C circuit presents itself as about
1.7K impedance to the cathode of the cathode follower.

Now, what happens when a RF signal is applied to the D+C detector? As
someone wisely remarked, the detector will begin to detect if the diode in
NOT conducting continuously. This will happen when the AC component of the
current exceeds DC component (100uA). With 1.7K D+C impedance this will
happen when the RF signal reaches 170mV.

A 12AU7 tube has low transconductance and the followr probably has about
500R of output impedance. Thus it will probably require 200...220mV of the
signal on the grid for this detector to start working. Not really impressive
sensitivity. As has been shown above, to handle a 95% modulation, this
translates into at least 4V of carrier.

Because of the discharge current (100uA) is constant, HF performance depends
on the carrier level. For example,
- at 4V carrier it can handle 70% modulation to 15kHz and 95% to 10kHz;
- at 10V carrier it can handle 70% modulation to 6kHz and 98% to 4kHz;
- at 20V carrier it can handle 70% modulation to 3kHz and 99% to 2kHz, etc.

(Here 220pF+100pF was considered as an audio frequency load).

The performance is remarkably better than of a booooring unbiased detector!
The only disappointing thing is that the stronger the signal (from a local
station), the worse is the modulation HF margin. You would expect the
opposite from a detector for hi-fi application. The only remedy is to use a
strong amplified delayed AGC to maintain 4...5V carrier for a station of any
strength.

(I use an active integrator based AGC amplifier for that purpose, so that
all the stations are levelled up to the optimum level, but Partick's AGC
into the mixer only is very primitive and inefficient. Besides it introduces
more distortion by nonlinearly loading the IFT. However it may be a separate
thread to discuss.)




  #2   Report Post  
Posted to rec.audio.tubes
Alex Pogossov Alex Pogossov is offline
external usenet poster
 
Posts: 95
Default AM detector, part 2

Unfinished. Posted by accident. Please ignore.


  #3   Report Post  
Posted to rec.audio.tubes
Engineer[_2_] Engineer[_2_] is offline
external usenet poster
 
Posts: 209
Default AM detector, part 2

On Jul 16, 12:12*am, "Alex Pogossov" wrote:
Modulation handling of an AM detector is determined by:
1. Carrier level and low level sensitivity;
2 AC impedance / DC load resistance ratio;
3. Slew rate.

If say we are aiming at 95% undistorted demodulation, then the carrier level
shall be about 20 times greater than the sensitivity threshold. For example,
for an unbiased detector with a silicon diode, which has a 0.6V knee, we
need at least 12V carrier. (Unless otherwise stated, amplitude is implied..)
For a properly designed vacuum diode (Vsens = 100mV) you need at least 2V
carrier.

For an optimum biased semiconductor diode (Vsens = 25mV) you need only 0.5V
of carrier for distortion free 95% demodulation. Thus a radio using a
PROPERLY designed p-n diode detector can run at low IF levels, have less IF
gain, less IF stages, would require less shielding and have better
sensitivity.

It is well known that for large signals maximum modulation index
approximately equals to AC / DC load impedance. The ways to make it close to
1 at lower frequencies a
- use a DC coupled hi-Z follower (cathode, source, op-amp buffer, high-beta
BJT with low base current, etc.);
- never AC couple a volume control to the detector. Instead, make the volume
control pot *the* load (or a part thereof);
- have a high (5...10M) input impedance of the audio amp, so that it does
not load the detector even at full volume. (This stuff has been mentioned in
a different thread);
- never connect AGC voltage RC filter to the detector load. Use a separate
AGC detector.

In an unbiased detector slew rate issues are also related to the AC/DC
impedance ratio. For example, a booooring detector with 500K || 100pF load
can handle 70% at about 3.5kHz only. To get things worse, a 47K+100pF ripple
filter is added straight after the detector. This cap virtually adds to the
effect of reducing the modulation handling at the highs. With 100pF load +
100pF filter the 70% cutoff comes down to the appauling 1.8kHz. No wonder
the sound of a booooring radio is so crappy.

There is another less known effect. Not only the heavy C distorts HF
modulation due to slew rate limitations, it also reduces original modulation
index at HF, working like sort of high-cut tone control. To understand the
phenomenon without a deep maths, consider that on a steeply rising RF
envelope the detector has to charge the load capacitance. This sucks extra
energy from the hi-Z IFT on top what is to be dissipated in the resistive
component of the load. On the steeply falling RF envelope slopes, R is being
fed from a discharging C instead of the diode. Thus the IFT gets unoaded on
the falling slopes. It is easy to see that the peaks are thus "cut and
rounded" and the troughs are "filled". Modulation virtually reduces. This
reduces the slew rate distortion, replacing it with a HF cut. In the end the
sound is still crappy. (Those who are familiar with the operation of a ratio
FM detector, as opposed to a Sheeley discriminator, will see many
parallels.)

The above HF unmodulation takes place only if the detector is directly fed
from a hi-Z IFT. If a buffer (cathode) follower is used, there is no HF
unmodulation phenomenon.

In unbiased detectors there is no high limit to carrier level (until the
diode breaks down) -- discharge current is proportional to the carrier
level. Slew rate perfomance does not depend on the RF signal magnitude.

This is not the case with biased detectors, where the discharge current is
(almost) constant. Here it is time to analyse the famous Partick's biased
detector.

Schematic values may vary, but here let us assume it is biased to 50V, uses
a cathode follower on 12AU7, a semiconductor diode, has 220pF of C,
pull-down resistor of 500K and a ripple filter of 100K+100pF. Thus the diode
bias current is about 100uA at no signal.

A DC biased diode, as I hope people know, has differential resistance of
(25 ohms / current, mA). In this case, at no signal and at very low signal
the diode acts as a 250R resistor. Capacitor of 220pF has about 1.6K
reactance at 455kHz. Thus the whole D+C circuit presents itself as about
1.7K impedance to the cathode of the cathode follower.

Now, what happens when a RF signal is applied to the D+C detector? As
someone wisely remarked, the detector will begin to detect if the diode in
NOT conducting continuously. This will happen when the AC component of the
current exceeds DC component (100uA). With 1.7K D+C impedance this will
happen when the RF signal reaches 170mV.

A 12AU7 tube has low transconductance and the followr probably has about
500R of output impedance. Thus it will probably require 200...220mV of the
signal on the grid for this detector to start working. Not really impressive
sensitivity. As has been shown above, to handle a 95% modulation, this
translates into at least 4V of carrier.

Because of the discharge current (100uA) is constant, HF performance depends
on the carrier level. For example,
- at 4V carrier it can handle 70% modulation to 15kHz and 95% to 10kHz;
- at 10V carrier it can handle 70% modulation to 6kHz and 98% to 4kHz;
- at 20V carrier it can handle 70% modulation to 3kHz and 99% to 2kHz, etc.

*(Here 220pF+100pF was considered as an audio frequency load).

The performance is remarkably better than of a booooring unbiased detector!
The only disappointing thing is that the stronger the signal (from a local
station), the worse is the modulation HF margin. You would expect the
opposite from a detector for hi-fi application. The only remedy is to use a
strong amplified delayed AGC to maintain 4...5V carrier for a station of any
strength.

(I use an active integrator based AGC amplifier for that purpose, so that
all the stations are levelled up to the optimum level, but Partick's AGC
into the mixer only is very primitive and inefficient. Besides it introduces
more distortion by nonlinearly loading the IFT. However it may be a separate
thread to discuss.)


Alex, there's quite a lot in the Internet about "Hi-Fi AM detectors" -
I started to keep these but need to take another look... BTW, could
you please post a schematic somewhere (NOT in the binaries - many of
us can't get them) of a recommended and most easily implemented AM
detector improvement (even if not Hi-Fi!) for the common AA5/6 (and
other) radios?
Thanks and cheers,
Roger
  #4   Report Post  
Posted to rec.audio.tubes
Patrick Turner Patrick Turner is offline
external usenet poster
 
Posts: 3,964
Default AM detector, part 2


Alex, there's quite a lot in the Internet about "Hi-Fi AM detectors" -
I started to keep these but need to take another look... *BTW, could
you please post a schematic somewhere (NOT in the binaries - many of
us can't get them) of a recommended and most easily implemented AM
detector improvement (even if not Hi-Fi!) for the common AA5/6 (and
other) radios?
Thanks and cheers,
Roger- Hide quoted text -


Ah, you would also like to see schematics, and tested schematics with
a fully written up appraisal, but here at r.a.t., most posters just
like talking about stuff and they are allergic to doing the WORK of
preparing schematics after testing them etc.

There is a lotta stuff on AM detectors and not much using tubes that
is any different to glorified 1950 techniques.

So the world which won't WORK can glide right past me and i think I'd
not really miss anything they have to offer.

Patrick Turner.

Reply
Thread Tools
Display Modes

Posting Rules

Smilies are On
[IMG] code is On
HTML code is Off


Similar Threads
Thread Thread Starter Forum Replies Last Post
AM detector, part 1 Alex Pogossov Vacuum Tubes 22 July 20th 11 05:49 AM
How to built a dolby amplifier part by part ? AMBROSE Tech 3 January 25th 07 09:05 PM
Improved AM Detector John Stewart Vacuum Tubes 94 July 22nd 04 01:53 AM
Yet Another DC Coupled AM Detector John Stewart Vacuum Tubes 10 July 16th 04 03:47 AM
Infrasound detector David Skinner Tech 15 June 24th 04 09:40 AM


All times are GMT +1. The time now is 10:14 AM.

Powered by: vBulletin
Copyright ©2000 - 2024, Jelsoft Enterprises Ltd.
Copyright ©2004-2024 AudioBanter.com.
The comments are property of their posters.
 

About Us

"It's about Audio and hi-fi"