|
I was getting bugged by some squawking around 7.145 and got to looking around.
The audio matched with a station on 7.385 from Chennai. It was annoying loud
and present on my commercial rig.
Here is the math:
VFO's 4th harmonic minus IF
4.80 x 4 = 19.20 - 12 = 7.20
4.90 x 4 = 19.60 - 12 = 7.60
5.00 x 4 = 20.00 - 12 = 8.00
So the 4th harmonic receives tunes 7.2 to 8 Mhz while my dial goes from 7 to 7.2 !
I guess the only way to fix this is stick to a sine wave from the VFO or add a filter
between the VFO and mixer.
Any ideas guys & girls? Farhan?
Cheers
--
Raj, VU2ZAP
Bangalore, India.
|
Some typos crept in! I was getting bugged by some squawking around 7.145 and got to looking around.
The audio matched with a station on 7.385 from Chennai. It was annoyingly loud
and NOT present on my commercial rig. At 14-03-2017, you wrote: I was getting bugged by some squawking around 7.145 and got to looking around.
The audio matched with a station on 7.385 from Chennai. It was annoying loud
and present on my commercial rig.
Here is the math:
VFO's 4th harmonic minus IF
4.80 x 4 = 19.20 - 12 = 7.20
4.90 x 4 = 19.60 - 12 = 7.60
5.00 x 4 = 20.00 - 12 = 8.00
So the 4th harmonic receives tunes 7.2 to 8 Mhz while my dial goes from 7 to 7.2 !
I guess the only way to fix this is stick to a sine wave from the VFO or add a filter
between the VFO and mixer.
Any ideas guys & girls? Farhan?
Cheers
--
Raj, VU2ZAP
Bangalore, India.
|
Hi! Raj and others,
I feel that the mixer is being bit overdriven in transmit, this might result in some spurii...
EMRFD suggests a level of -10dbm at the transmixer however calculating the stage gain we get something like this -20dbm out at Bal mod add 3db loss in pad that makes it -23dbm. add to it +15db in amp Q6 gives a signal of -8dbm. 3db loss in filter makes it -11dbm yet another 15db in amp Q4 brings it to +4dbm. This is +14 db more than recommended -10dbm. I have been thinking of making Q4 a buffer a check the resulting output spectrum but then again I will need more gain in predriver stages..
73
Rahul VU3WJM
?
On Tuesday, 14 March 2017 6:04 PM, Raj vu2zap <vu2zap@...> wrote:
I was getting bugged by some squawking around 7.145 and got to looking around.
The audio matched with a station on 7.385 from Chennai. It was annoying loud
and present on my commercial rig.
Here is the math:
VFO's 4th harmonic minus IF
4.80 x 4 = 19.20 - 12 = 7.20
4.90 x 4 = 19.60 - 12 = 7.60
5.00 x 4 = 20.00 - 12 = 8.00
So the 4th harmonic receives tunes 7.2 to 8 Mhz while my dial goes from 7 to 7.2 !
I guess the only way to fix this is stick to a sine wave from the VFO or add a filter
between the VFO and mixer.
Any ideas guys & girls? Farhan?
Cheers
--
Raj, VU2ZAP
Bangalore, India.
|
Raj, As in post ??/g/BITX20/message/21997 Best bet is to move the VFO up to 19mhz, one of Allison's primary criticisms of the Bitx. ?Would be tough to get an analog VFO stable enough at 19mhz, though the Raduino could do it easily.? Transporting the VFO to the mixer as a sine wave might help a bit. ??A clean square wave doesn't have any even harmonics, only odd harmonics. ?I'd bet the fourth harmonic is likely due mostly to second order effects in the diode ring mixer, not due to what's coming out of ?the vfo, but I could be wrong. ?Worth trying. ?But no matter how perfect your sine wave is when going into that diode ring transformer, since the current through the diodes is an exponential function of the voltage across the diodes that current will be nothing like a sine wave. ?It is exactly there in the diode ring that the image is getting down converted to 12mhz.by the fourth harmonic of the vfo. ?A square wave is preferred around the diode mixers: ?/g/BITX20/message/22538 On the other hand, going to a sine wave on your 4.8mhz vfo would likely help on the birdie at 7.2mhz. ?There the 5'th harmonic of the vfo at 4.8*5=24, and that beats with the second harmonic of the bfo. ?With proper shielding around the BFO end of the filter and care taken to keep the 5'th harmonic of the vfo from radiating in from wires between your vfo and the Bitx40 mixer, the birdie should have nowhere to mix. > ?stick to a sine wave from the VFO or add a filter?between the VFO and mixer.
Pretty much the same thing. ?A 4.8mhz bandpass filter on the VFO output would permit only a sinewave through to the diode ring mixer. ?The mixer is better off with a square wave. You must be trying to get a rise out of me. ?All favorite topics to set me off into a rant. ? ?;-)
> Rahul: ?I feel that the mixer is being bit overdriven in transmit, this might result in some spurii.. ? Could be, I haven't looked. ?But that's a separate subject from Raj's images on receive.
Jerry, KE7ER
On Tue, Mar 14, 2017 at 05:34 am, Raj vu2zap wrote: So the 4th harmonic receives tunes 7.2 to 8 Mhz while my dial goes from 7 to 7.2 !
I guess the only way to fix this is stick to a sine wave from the VFO or add a filter
between the VFO and mixer. ?
|
I’m looking at the VFO (Raduino) output (not connected to mixer)
and the 4th Harmonic is about 33 db down from the main bang.
A low pass filter might be worth a try, but moving the VFO to 19
seems a better solution. Just my 2 cents…
?
?
toggle quoted message
Show quoted text
From: [email protected]
[mailto:[email protected]] On Behalf Of Jerry Gaffke via Groups.Io
Sent: Tuesday, March 14, 2017 9:30 AM
To: [email protected]
Subject: Re: [BITX20] The harmonics are out to play
?
Raj,
As in post ??/g/BITX20/message/21997
Best bet is to move the VFO up to 19mhz, one of Allison's primary criticisms
of the Bitx. ?Would be tough to get an analog VFO stable enough at 19mhz,
though the Raduino could do it easily.?
Transporting the VFO to the mixer as a sine wave might help a bit. ??A clean square wave doesn't have any even harmonics,
only odd harmonics. ?I'd bet the fourth harmonic is likely due mostly to
second order effects in the diode ring mixer, not due to what's coming out of
?the vfo, but I could be wrong. ?Worth trying. ?But no matter
how perfect your sine wave is when going into that diode ring transformer, since
the current through the diodes is an exponential function of the voltage across
the diodes that current will be nothing like a sine wave. ?It is exactly
there in the diode ring that the image is getting down converted to 12mhz.by
the fourth harmonic of the vfo. ?A square wave is preferred around the
diode mixers: ?/g/BITX20/message/22538
On the other hand, going to a sine wave on your 4.8mhz vfo would likely help
on the birdie at 7.2mhz. ?There the 5'th harmonic of the vfo at 4.8*5=24,
and that beats with the second harmonic of the bfo. ?With proper shielding
around the BFO end of the filter and care taken to keep the 5'th harmonic of
the vfo from radiating in from wires between your vfo and the Bitx40 mixer, the
birdie should have nowhere to mix.
> ?stick to a sine wave from the VFO or
add a filter?between the VFO and mixer.
Pretty much the same thing. ?A 4.8mhz bandpass filter on the VFO output
would permit only a sinewave through to the diode ring mixer. ?The mixer
is better off with a square wave.
You must be trying to get a rise out of me. ?All favorite topics to set
me off into a rant. ? ?;-)
?
> Rahul: ?I feel that the mixer is
being bit overdriven in transmit, this might result in some spurii.. ?
Could be, I haven't looked. ?But that's a separate subject from
Raj's images on receive.
?
Jerry, KE7ER
?
On Tue, Mar 14, 2017 at 05:34 am, Raj vu2zap wrote:
So the 4th harmonic receives tunes 7.2 to 8 Mhz while my
dial goes from 7 to 7.2 !
I guess the only way to fix this is stick to a sine wave from the VFO or add a
filter
between the VFO and mixer.
?
|
|
This email has been checked for viruses by Avast antivirus software.
|
|
Jerry and pals,
I forgot to mention that I switched off the VFO by shorting out the 2pin
connector in the middle of the board.
Then I fed my sig-gen at +17dbm into the mixer toroid. And all the ghost
images where gone. It was quiet like
a commercial rig. So a sine wave it is.
So before I get out the Si's and Radionos etc, I thought I will try to
tame the VFO to produce a sine wave.
First thoughts is to isolate the VFO Q9 and couple it to a buffer as
lightly as possible. Q9 may need some
work to JUST break into oscillation to produce a clean sine wave. Then
couple lightly and bring up the
power level required to the mixer.
Maybe I am too obsessed with this BITX40 analog VFO, thanks
Farhan.
73
Raj, vu2zap
At 14-03-2017, you wrote:
toggle quoted message
Show quoted text
Raj,
As in post??
/g/BITX20/message/21997
Best bet is to move the VFO up to 19mhz, one of Allison's primary
criticisms of the Bitx.? Would be tough to get an analog VFO stable
enough at 19mhz, though the Raduino could do it easily.
Transporting the VFO to the mixer as a sine wave might help a
bit.?? A clean square wave doesn't have any even harmonics,
only odd harmonics.? I'd bet the fourth harmonic is likely due
mostly to second order effects in the diode ring mixer, not due to what's
coming out of? the vfo, but I could be wrong.? Worth
trying.? But no matter how perfect your sine wave is when going into
that diode ring transformer, since the current through the diodes is an
exponential function of the voltage across the diodes that current will
be nothing like a sine wave.? It is exactly there in the diode ring
that the image is getting down converted to 12mhz.by the fourth harmonic
of the vfo.? A square wave is preferred around the diode
mixers:?
/g/BITX20/message/22538
On the other hand, going to a sine wave on your 4.8mhz vfo would likely
help on the birdie at 7.2mhz.? There the 5'th harmonic of the vfo at
4.8*5=24, and that beats with the second harmonic of the bfo.? With
proper shielding around the BFO end of the filter and care taken to keep
the 5'th harmonic of the vfo from radiating in from wires between your
vfo and the Bitx40 mixer, the birdie should have nowhere to mix.
>? stick to a sine wave from the VFO or add a filter between the
VFO and mixer.
Pretty much the same thing.? A 4.8mhz bandpass filter on the VFO
output would permit only a sinewave through to the diode ring
mixer.? The mixer is better off with a square wave.
You must be trying to get a rise out of me.? All favorite topics to
set me off into a rant.??? ;-)
> Rahul:? I feel that the mixer is being bit overdriven in
transmit, this might result in some spurii..?
Could be, I haven't looked.? But that's a separate subject from
Raj's images on receive.
Jerry, KE7ER
On Tue, Mar 14, 2017 at 05:34 am, Raj vu2zap wrote:
- So the 4th harmonic receives tunes 7.2 to 8 Mhz while my dial goes
from 7 to 7.2 !
- I guess the only way to fix this is stick to a sine wave from the VFO
or add a filter
- between the VFO and mixer.
|
Doesn't the AD9850 put out a decent sine wave?
Jack, W8TEE
From: Raj vu2zap <vu2zap@...>
To: [email protected]
Sent: Tuesday, March 14, 2017 10:59 AM
Subject: Re: [BITX20] The harmonics are out to play
Jerry and pals,
I forgot to mention that I switched off the VFO by shorting out the 2pin
connector in the middle of the board.
Then I fed my sig-gen at +17dbm into the mixer toroid. And all the ghost
images where gone. It was quiet like
a commercial rig. So a sine wave it is.
So before I get out the Si's and Radionos etc, I thought I will try to
tame the VFO to produce a sine wave.
First thoughts is to isolate the VFO Q9 and couple it to a buffer as
lightly as possible. Q9 may need some
work to JUST break into oscillation to produce a clean sine wave. Then
couple lightly and bring up the
power level required to the mixer.
Maybe I am too obsessed with this BITX40 analog VFO, thanks
Farhan.
73
Raj, vu2zap
At 14-03-2017, you wrote:
Raj,
As in post??
/g/BITX20/message/21997
Best bet is to move the VFO up to 19mhz, one of Allison's primary
criticisms of the Bitx.? Would be tough to get an analog VFO stable
enough at 19mhz, though the Raduino could do it easily.
Transporting the VFO to the mixer as a sine wave might help a
bit.?? A clean square wave doesn't have any even harmonics,
only odd harmonics.? I'd bet the fourth harmonic is likely due
mostly to second order effects in the diode ring mixer, not due to what's
coming out of? the vfo, but I could be wrong.? Worth
trying.? But no matter how perfect your sine wave is when going into
that diode ring transformer, since the current through the diodes is an
exponential function of the voltage across the diodes that current will
be nothing like a sine wave.? It is exactly there in the diode ring
that the image is getting down converted to 12mhz.by the fourth harmonic
of the vfo.? A square wave is preferred around the diode
mixers:?
/g/BITX20/message/22538
On the other hand, going to a sine wave on your 4.8mhz vfo would likely
help on the birdie at 7.2mhz.? There the 5'th harmonic of the vfo at
4.8*5=24, and that beats with the second harmonic of the bfo.? With
proper shielding around the BFO end of the filter and care taken to keep
the 5'th harmonic of the vfo from radiating in from wires between your
vfo and the Bitx40 mixer, the birdie should have nowhere to mix.
>? stick to a sine wave from the VFO or add a filter between the
VFO and mixer.
Pretty much the same thing.? A 4.8mhz bandpass filter on the VFO
output would permit only a sinewave through to the diode ring
mixer.? The mixer is better off with a square wave.
You must be trying to get a rise out of me.? All favorite topics to
set me off into a rant.??? ;-)
> Rahul:? I feel that the mixer is being bit overdriven in
transmit, this might result in some spurii..?
Could be, I haven't looked.? But that's a separate subject from
Raj's images on receive.
Jerry, KE7ER
On Tue, Mar 14, 2017 at 05:34 am, Raj vu2zap wrote:
- So the 4th harmonic receives tunes 7.2 to 8 Mhz while my dial goes
from 7 to 7.2 !
- I guess the only way to fix this is stick to a sine wave from the VFO
or add a filter
- between the VFO and mixer.
|
A DDS is decent enough, though will have some spurs that the Si5351's PLL's square wave would not have. ?Both are based on a crystal reference oscillator, so plenty stable to run at 19mhz where Raj's images would not be an issue. ?A $1USD Si5351 on the main board would be a lot cheaper than a separate board with a DDS chip. ?The Si5351 has an additional channel to drive the BFO. ? And has a third channel for a signal generator during testing, or to drive the extra mixer on the uBitx. ?Capable of higher frequency operation than an AD9850 which will be important when supporting upper bands. The Si5351's PLL's have more phase noise than a DDS but less phase noise than the analog VFO, so likely more than acceptable. ?Using multiple Si5351 channels does raise the possibility of crosstalk between channels. ?Overall, in this application, I'd say the Si5351 is a clear win.
On Tue, Mar 14, 2017 at 08:12 am, Jack Purdum wrote:
Doesn't the AD9850 put out a decent sine wave?
?
|
Raj: How about an EXTERNAL filter? If that's the only station troubling you then you could trap it out. A bandpass filter for just the 40m amateur band would clean up your signal even further and eliminate all the rest. Even a sharp cut-off low pass on the antenna side might do the trick. -Don, ND6T
|
Unfortunately the images he's getting are in the 40m amateur band. ?They cross over at 7.2 mhz, so if you tune into a LSB station at 7.2 mhz you will also be receiving it as a USB station through the magic of ? (12-7.2)*4-12 = 7.2mhz. ?A sharp notch filter might knock out a particularly troublsome SW broadcaster but is not a general solution here.
On Tue, Mar 14, 2017 at 08:51 am, Don Cantrell wrote: A bandpass filter for just the 40m amateur band would clean up your signal even further and eliminate all the rest. ?
|
I moved my VFO to 19MHz a couple weeks ago and have had none of that nasty SW broadcast interference since. Of course you have to move your BFO to the other side of the filter. That was no problem since I am generating that signal in the Si5351. I suppose it is a little more difficult with the analog BFO.
|
I'm glad I am not alone!? The move to 19MHz vfo sounds good but my 12MHz xtal filter has a very gentle slope on the LF side so moving the carrier oscilator down 3kHz to output USB from the balanced modulator would hit the carrier and oppsite sideband rejection badly. (As another comment, the carrier suppression is not very good anyway but that is what you get with a single board and no screening.) As for sine wave or square wave drive to the final balanced mixer, I note that Wes Hayward used an analogue vfo, buffer, low pass filter in the Ugly Weekender. Jum Kortge used a low pass filter in the ss-40 vfo .... I am going back to an analogue vfo for ssb, built as a separate module. Feeding the Raduino into the original vfo with the coil removed was never going to be clean solution. After three transistors, I bet we do not have a 1:1 mark/space square wave. All sorts of harmonics are going into the balanced mixer. Brian.
|
Don,
A bandpass for 40M will not be sharp enough to rapidly chop off after 7.2 MHz.
There are many stations images between 7.1 and 7.2 that make a mess of the band on BITX40
I am going to try a filter for the VFO or a hack to see if I can get rid of 19-20 Mhz harmonics.
73 Raj
toggle quoted message
Show quoted text
At 14-03-2017, you wrote: Raj: How about an EXTERNAL filter? If that's the only station troubling you then you could trap it out. A bandpass filter for just the 40m amateur band would clean up your signal even further and eliminate all the rest. Even a sharp cut-off low pass on the antenna side might do the trick. -Don, ND6T
|
Raj vu2zap
Don, ND6T