Topics

Volume Limiting and Distortion Reduction for uBITX #ubitx


ND6P
 

I came up with this mod after testing my uBITX on the bench to determine how much signal it could handle before distorting at the speaker due to clipping of the audio signal.  What I found is that RF input signals greater than S9+5dB (-68 dBm) become distorted due to the audio output clipping on the negative side.  What the mod does is insert enough attenuation at the RF input to prevent the audio from clipping for signals that are stronger than S9+5dBm.  Signals S9+5dBm and weaker are not attenuated.  I've tested with RF input as strong as S9+40dBm (-33 dBm) and get clean output. Now I can listen to a weak signal without having to reach for the volume control when a strong signal comes on.  So effectively this is an automatic RF attenuation circuit to provide AGC functionality for the uBITX.

I did a similar mod for the Bitx40 in the past.  This one is tailored to work specifically for the uBITX.  The pin diodes (MA27B) are available at RF Parts Co.

73, ND6P



_Dave_ AD0B
 

Wow clever adjusting the bias on the pin diode. I assume that the points connected to the volume control are added in parallel to existing wiring? Where do a1 b1 a2 b2 c1 and c2 go? Is that the existing k1 and 7805? up in the corner by the antenna input on the main board?

Thank you


Jerry Gaffke
 

There have been dozens of threads in this forum about adding AGC to the Bitx40 and uBitx.
Most get hacked in (both detection and attenuation) after the rx audio preamp,
to prevent ear damage on a sudden strong signal.
Then perhaps add a manual RF gain pot at the antenna port to avoid audio preamp overload.

Some used PIN diodes to attenuate the signal at RF (or IF) frequencies,
and thus get around the dynamic range limitations of that 2n3904 audio pre-amp stage.
Here's the earliest thread to identify the dynamic range issue of that Bitx/uBitx audio pre-amp stage:
:    https://groups.io/g/BITX20/topic/5945215#32066

Kees (K5BCQ) sold nearly 1000 kits of an AGC design from Don (ND6T).
This was audio detection with RF attenuation at the antenna port using 2n7002 FETs.
Nik (VK4PLN) was planning to sell a kit with audio detection and attenuation
after the audio preamp, I don't know what came of that.
    http://ubitx.net/category/hardware/agc-mod/
 

Sensing signal levels is probably best done after conversion to audio.
Must sense after the 12mhz crystal filter, or strong signals elsewhere in the band
will bring on the AGC action.  But if we sense between that 12mhz crystal filter
and the demodulator, then we will likely be seeing mostly the unshielded 12mhz bfo.

Attenuation could be anywhere between the antenna port and the input to the audio preamp.
Putting attenuation at the antenna port avoids overloading the mixers and IF amps,
but they generally have enough dynamic range that it doesn't really matter where we pot it.
Putting attenuation after the audio preamp works, but does not fix the dynamic range
issue of that preamp.

Various ways to implement attenuation at RF, here's a good one that 
tries to maintain a 50 ohm impedance match, and is quite cheap:
    https://www.nxp.com/docs/en/data-sheet/BAP64Q.pdf

Allison recommends using a 1n4007 as a pin diode to set the gain in one (or both)
of the bidi IF amps, many of us have a 1n4007 in the junk box:
  https://groups.io/g/BITX20/message/45984

Jerry, KE7ER


Jack, W8TEE
 

There are pros and cons to each approach, which is why JackAl has both RF and audio AGC. Indeed, the AGC section takes up a good portion of the JackAl's nanoacres. BTW, we just received the assembled Jack boards which has the SMD parts in place and looks like this:

Inline image

This is what QRP Guys will be selling after we spend a little more time testing everything. After adding the off-board parts (less than $13), the board looks like this:

Inline image

All of the yellow pins are test points to encourage experimentation. You also have about a dozen free I/O pins to experiment with.

Jack, W8TEE

On Wednesday, December 26, 2018, 11:26:29 AM EST, Jerry Gaffke via Groups.Io <jgaffke@...> wrote:


There have been dozens of threads in this forum about adding AGC to the Bitx40 and uBitx.
Most get hacked in (both detection and attenuation) after the rx audio preamp,
to prevent ear damage on a sudden strong signal.
Then perhaps add a manual RF gain pot at the antenna port to avoid audio preamp overload.

Some used PIN diodes to attenuate the signal at RF (or IF) frequencies,
and thus get around the dynamic range limitations of that 2n3904 audio pre-amp stage.
Here's the earliest thread to identify the dynamic range issue of that Bitx/uBitx audio pre-amp stage:
:    https://groups.io/g/BITX20/topic/5945215#32066

Kees (K5BCQ) sold nearly 1000 kits of an AGC design from Don (ND6T).
This was audio detection with RF attenuation at the antenna port using 2n7002 FETs.
Nik (VK4PLN) was planning to sell a kit with audio detection and attenuation
after the audio preamp, I don't know what came of that.
    http://ubitx.net/category/hardware/agc-mod/
 

Sensing signal levels is probably best done after conversion to audio.
Must sense after the 12mhz crystal filter, or strong signals elsewhere in the band
will bring on the AGC action.  But if we sense between that 12mhz crystal filter
and the demodulator, then we will likely be seeing mostly the unshielded 12mhz bfo.

Attenuation could be anywhere between the antenna port and the input to the audio preamp.
Putting attenuation at the antenna port avoids overloading the mixers and IF amps,
but they generally have enough dynamic range that it doesn't really matter where we pot it.
Putting attenuation after the audio preamp works, but does not fix the dynamic range
issue of that preamp.

Various ways to implement attenuation at RF, here's a good one that 
tries to maintain a 50 ohm impedance match, and is quite cheap:
    https://www.nxp.com/docs/en/data-sheet/BAP64Q.pdf

Allison recommends using a 1n4007 as a pin diode to set the gain in one (or both)
of the bidi IF amps, many of us have a 1n4007 in the junk box:
  https://groups.io/g/BITX20/message/45984

Jerry, KE7ER


ND6P
 

Yes, points connected to volume control are in parallel with existing wiring.  The relay and regulator are new, not the existing uBITX ones.   Here is a photo of the mod that I made for mine.


Bob Lunsford
 

RE:
Allison recommends using a 1n4007 as a pin diode to set the gain in one (or both)
of the bidi IF amps, many of us have a 1n4007 in the junk box:
https://groups.io/g/BITX20/message/45984

Jerry, KE7ER
~~~~~~~~~~~~~~~~~
The link refers to replacing R13 with a 1N400X diode to assist
in leveling the audio and avoiding the booming effect. The "fix"
is simplicity itself even if not totally effective.

This relates to lack of AGC for the UBITX.

A few times listening to a net where some powerhouse station blasts
the audio line is enough to make this a consideration. Seeing the link
above shows that a simple "fix" may be effective but ass to just how
effective it may be is another factor.

My question is whether merely bridging R13 with the 1N400X would
bring about positive results or if adding the diode would be useful
only if R13 is removed? I know the mechanics of removing an SMD
device but bridging the resistor with rhe diode would be much simpler.

Also, it seems obvious that the diode's cathode would be connected
to ground. Is it so obvious that polarity was not mentioned?

Bob — KK5R

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

On Wed, 12/26/18, Jerry Gaffke via Groups.Io <jgaffke=yahoo.com@groups.io> wrote:

Subject: Re: [BITX20] Volume Limiting and Distortion Reduction for uBITX #ubitx
To: BITX20@groups.io
Date: Wednesday, December 26, 2018, 11:26 AM

There have been dozens of
threads in this forum about adding AGC to the Bitx40 and
uBitx.
Most get hacked in (both detection
and attenuation) after the rx audio preamp,
to prevent ear damage on a sudden strong
signal.
Then perhaps add a manual RF gain
pot at the antenna port to avoid audio preamp overload.

Some used PIN diodes to
attenuate the signal at RF (or IF) frequencies,
and thus get around the dynamic range
limitations of that 2n3904 audio pre-amp stage.
Here's the earliest thread to identify the
dynamic range issue of that Bitx/uBitx audio pre-amp
stage:
:    https://groups.io/g/BITX20/topic/5945215#32066

Kees (K5BCQ) sold nearly 1000
kits of an AGC design from Don (ND6T).
This
was audio detection with RF attenuation at the antenna port
using 2n7002 FETs.
Nik (VK4PLN) was planning
to sell a kit with audio detection and attenuation
after the audio preamp, I don't know what
came of that.
    http://ubitx.net/category/hardware/agc-mod/
 

Sensing
signal levels is probably best done after conversion to
audio.
Must sense after the 12mhz crystal
filter, or strong signals elsewhere in the band
will bring on the AGC action.  But if we sense
between that 12mhz crystal filter
and the
demodulator, then we will likely be seeing mostly the
unshielded 12mhz bfo.

Attenuation could be anywhere between the
antenna port and the input to the audio preamp.
Putting attenuation at the antenna port avoids
overloading the mixers and IF amps,
but they
generally have enough dynamic range that it doesn't
really matter where we pot it.
Putting
attenuation after the audio preamp works, but does not fix
the dynamic range
issue of that preamp.

Various ways to implement
attenuation at RF, here's a good one that 
tries to maintain a 50 ohm impedance match, and
is quite cheap:
    https://www.nxp.com/docs/en/data-sheet/BAP64Q.pdf

Allison recommends
using a 1n4007 as a pin diode to set the gain in one (or
both)
of the bidi IF amps, many
of us have a 1n4007 in the junk box:
  https://groups.io/g/BITX20/message/45984

Jerry,
KE7ER


Jerry Gaffke
 

There's a lot more going on here than just replacing R13 with a 1n4007.
We need to send a current through the 1n4007 from a control circuit,
the more current through the 1n4007 the lower its resistance and 
the greater the gain in that IF stage.

If you bridge R13 with a 1n4007, then gain in that IF stage
can never be less than that of a stock radio (not good).
Removing R13 from the board will be the least of your worries.

To get any AGC design to work well, you either have to be
extremely lucky or really know what you are doing. 
Or copy a design that somebody else got working.

If you want AGC, I'd recommend dead-bugging the ND6T scheme here:
   http://ubitx.net/category/hardware/agc-mod/
Use the circuit under the heading "New AGC System", April 16 2018.
That is what Kees shipped out as a kit (he has no more available).
Works on both the Bitx40 and the uBitx.

Kees' version uses jumpers to adjust R6 (in the ND6T schematic)
for values between 330k and 2meg, not just frozen at 1meg.
This allows the AGC hold time to be set to your preference.
Making R6 be 0 ohms would shut down the AGC, so it works like a stock rig.
Kees also added some 0.01uF and 1.0uF bypass caps to the 5v rail.

Allison's 1n4007 would more or less replaces Kees' two 2n7002 FET's.
Details of how to implement Allison's scheme are left as an exercise for the reader.
Good luck! If you want easy, I suggest you use ND6T's scheme.

Jerry, KE7ER


On Wed, Dec 26, 2018 at 08:07 PM, Bob Lunsford wrote:
My question is whether merely bridging R13 with the 1N400X would
bring about positive results or if adding the diode would be useful
only if R13 is removed?