Yep used that before. ?
Dr.?William J. Schmidt - K9HZ J68HZ 8P6HK ZF2HZ PJ4/K9HZ VP5/K9HZ PJ2/K9HZ ? Owner - Operator Big Signal Ranch ¨C K9ZC Staunton, Illinois ? Owner ¨C Operator Villa Grand Piton - J68HZ Soufriere, St. Lucia W.I. Rent it: www.VillaGrandPiton.com
email:??bill@... ?
toggle quoted message
Show quoted text
On Sep 3, 2023, at 1:57 PM, Larry Acklin <acklin@...> wrote:
?
Little more work but proven.?
Larry
|
±á³¾³¾³¾¡
I get the opposite. With the audio hat plugged into the qse I have the hum frequencys. Unplug and I get clean (residual) carrier. Secondary grounds don¡¯t help.?
My hypothetical answer is there are unwanted products created in the audio hat and a brute force external filter may be needed. As you proved the library answer doesn¡¯t work.?
I have fiddled with earlier versions of the audio adapter but never really tried to get critical.
I¡¯m looking at the SGTL5000 data sheet and it has very specific recommendations about the analog ground is treated. I¡¯m going to puzzle that next.?
Larry
|
On 2023-09-03 10:57, Larry Acklin wrote:
Little more work but proven.
*** Looks like the actual high pass filter is a pair of 39nF capacitors, a 10K resistor and a 20K resistor. The op-amps are just unity-gain buffers. I tried hooking the cable from the QSE audio input to my scope...saw nothing. Interestingly, with the QSE input unconnected, I get a lot MORE NOISE when I hit transmit. The noise goes down to almost nothing if I short out the ends of the cable. If I unplug and plug the QSE audio cable, the noise goes up and down. Much more with the cable unplugged. I think the noise is being created inside the QSE. - Jerry
|
Little more work but proven.?
Larry
|
OK, I added a pair of biquad highpass filters, all four sections cutoff at 300Hz, Q=5 at the input
of the output device. No difference whatsoever.
Next thing to try - unplug the IQ cable from the QSE and pipe those signals direct into an oscilloscope.
Then maybe try a simple RC highpass filter in line. The phase should be OK, as long as any change is the same in both channels.
- Jerry, KF6VB
toggle quoted message
Show quoted text
On 2023-09-03 10:02, jerry-KF6VB wrote: On 2023-09-02 16:39, Larry Acklin wrote:
But since I¡¯m spitballing here- the standard SSB audio bandwidth is
300-3000 Hz. Sounds like an audio hipass filter kneed at 250-300 would
fix the hum. Placed in the I/Q line to the qse- Since graphic eq introduces phase shifts that may be an issue. That sounds like a good idea. At least as a diagnostic step.
The T41 mostly uses the PJRC audio library.
On the left-channel output we have
Q_Out_L_Ex port 0
patchcord15
modeSelectOutExL
patchcord19
i2s_quadOut port 0
On the R/H channel output we have
Q_Out_R_Ex port 0
patchCord16
modeSelectOutExR
patchCord20
i2s_quadOut port 1
I think I can insert a filter and a patchcord between PatchCord19
and i2s_quadOut port 0. And another one between PatchCord20 and
i2s_quadOut port 1. A single biquad highpass with all four knees set
to 300Hz should do nicely.
- Jerry, KF6VB
On Sat, Sep 2, 2023 at 6:36 PM K9HZ <bill@...> wrote:
Ohhhh maybe loose connection!
DR. WILLIAM J. SCHMIDT - K9HZ J68HZ 8P6HK ZF2HZ PJ4/K9HZ VP5/K9HZ
PJ2/K9HZ
Owner - Operator
Big Signal Ranch ¨C K9ZC
Staunton, Illinois
Owner ¨C Operator
Villa Grand Piton - J68HZ
Soufriere, St. Lucia W.I.
Rent it: www.VillaGrandPiton.com [1]
email: bill@...
On Sep 2, 2023, at 12:21 PM, Larry Acklin <acklin@...>
wrote: ?Curious. I was fiddling, same setup, keyed the rig and got a
perfect clean residual carrier. QSY up band, have the BCI again.
Went back down band and still have it. So for a moment there it
was without the noise.
Larry
KB3CUF Links:
------
[1]
[2] /g/SoftwareControlledHamRadio/message/24086
[3] /mt/101082780/243852
[4] /g/SoftwareControlledHamRadio/post
[5] /g/SoftwareControlledHamRadio/editsub/243852
[6]
/g/SoftwareControlledHamRadio/leave/10484476/243852/1943518115/xyzzy
|
On 2023-09-02 16:39, Larry Acklin wrote: But since I¡¯m spitballing here- the standard SSB audio bandwidth is
300-3000 Hz. Sounds like an audio hipass filter kneed at 250-300 would
fix the hum. Placed in the I/Q line to the qse- Since graphic eq introduces phase shifts that may be an issue. That sounds like a good idea. At least as a diagnostic step. The T41 mostly uses the PJRC audio library. On the left-channel output we have Q_Out_L_Ex port 0 patchcord15 modeSelectOutExL patchcord19 i2s_quadOut port 0 On the R/H channel output we have Q_Out_R_Ex port 0 patchCord16 modeSelectOutExR patchCord20 i2s_quadOut port 1 I think I can insert a filter and a patchcord between PatchCord19 and i2s_quadOut port 0. And another one between PatchCord20 and i2s_quadOut port 1. A single biquad highpass with all four knees set to 300Hz should do nicely. - Jerry, KF6VB On Sat, Sep 2, 2023 at 6:36 PM K9HZ <bill@...> wrote:
Ohhhh maybe loose connection!
DR. WILLIAM J. SCHMIDT - K9HZ J68HZ 8P6HK ZF2HZ PJ4/K9HZ VP5/K9HZ
PJ2/K9HZ
Owner - Operator
Big Signal Ranch ¨C K9ZC
Staunton, Illinois
Owner ¨C Operator
Villa Grand Piton - J68HZ
Soufriere, St. Lucia W.I.
Rent it: www.VillaGrandPiton.com [1]
email: bill@...
On Sep 2, 2023, at 12:21 PM, Larry Acklin <acklin@...>
wrote: ?Curious. I was fiddling, same setup, keyed the rig and got a
perfect clean residual carrier. QSY up band, have the BCI again.
Went back down band and still have it. So for a moment there it
was without the noise.
Larry
KB3CUF Links:
------
[1]
[2] /g/SoftwareControlledHamRadio/message/24086
[3] /mt/101082780/243852
[4] /g/SoftwareControlledHamRadio/post
[5] /g/SoftwareControlledHamRadio/editsub/243852
[6]
/g/SoftwareControlledHamRadio/leave/10484476/243852/1943518115/xyzzy
|
Re: T41 Electronic Filter Prototype for QSE Output to Power Amplifier
Hi Greg,
I have used these in my receiver and transceivers over the years with good results. Tuning is done without variable tuning caps by adjusting the spacing of the turns on the toroids. This is the RF band-pass filter board I use in the analog and SDR transceiver: with response curves. Let me know if you cannot access the link in Sabin's article to the build data and I will post the file.
73, Rod AD5GH
|
Re: T41 Electronic Filter Prototype for QSE Output to Power Amplifier
So the one BPF made/sell (boards currently being made) is a 9 pole chebychev arrangement. ?Why? Well after using Elsie, i found out that i could get 100db+ attenuation with only 1.1 db insertion loss in that arrangement by using very common coilcraft or murata parts (less than 50 cents each) and tune the filters with cheap ten cent caps. ?And of course, like everything i sell, build what you need¡ up to 11 bands and up to 9 poles each. ?Instructions provided for 5 and 7 pole filters on the same board. All 9 pole filters cover the exact frequency range of the selected band only.?
Dr.?William J. Schmidt - K9HZ J68HZ 8P6HK ZF2HZ PJ4/K9HZ VP5/K9HZ PJ2/K9HZ ? Owner - Operator Big Signal Ranch ¨C K9ZC Staunton, Illinois ? Owner ¨C Operator Villa Grand Piton - J68HZ Soufriere, St. Lucia W.I. Rent it: www.VillaGrandPiton.com
email:??bill@... ?
toggle quoted message
Show quoted text
On Sep 3, 2023, at 9:23 AM, Albert Peter <albertfpeter@...> wrote:
?Greg,
Very nice. The only thing I would suggest is to go with band-pass filters between the QSE and PA to reduce some lower frequency products that could add to ultimate IM distortion in the PA. I tried it both with LP filter like yours - I prefer modified Butterworth to get flatter pass band performance, as well as band pass filters.. The band-pass filters gave a cleaner output from the PA compared to the LP filters. I would say, however, that the LP filters will provide an improvement in the IM distortion, as well.
Al
AC8GY
|
Re: T41 Electronic Filter Prototype for QSE Output to Power Amplifier
Hi Al-
Yes, I am going to move the filters closer to Butterworth to flatten out the response.? I've got some modeling and adjusting to get done.
The problem with the bandpass is they will probably require variable tuning caps.? I'm going to see what bare minimum circuitry is required to achieve decent performance.
--
73 Greg KF5N
|
Re: T41 Electronic Filter Prototype for QSE Output to Power Amplifier
Greg,
Very nice. The only thing I would suggest is to go with band-pass filters between the QSE and PA to reduce some lower frequency products that could add to ultimate IM distortion in the PA. I tried it both with LP filter like yours - I prefer modified Butterworth to get flatter pass band performance, as well as band pass filters.. The band-pass filters gave a cleaner output from the PA compared to the LP filters. I would say, however, that the LP filters will provide an improvement in the IM distortion, as well.
Al
AC8GY
|
Yes that should be very easy to add¡. In fact it might already be there. ?
Dr.?William J. Schmidt - K9HZ J68HZ 8P6HK ZF2HZ PJ4/K9HZ VP5/K9HZ PJ2/K9HZ ? Owner - Operator Big Signal Ranch ¨C K9ZC Staunton, Illinois ? Owner ¨C Operator Villa Grand Piton - J68HZ Soufriere, St. Lucia W.I. Rent it: www.VillaGrandPiton.com
email:??bill@... ?
toggle quoted message
Show quoted text
On Sep 2, 2023, at 7:39 PM, Larry Acklin <acklin@...> wrote:
? The BCI is not present now, my stack of attenuation may have had a loose connection.?
But since I¡¯m spitballing here- the standard SSB audio bandwidth is 300-3000 Hz. Sounds like an audio hipass filter kneed at 250-300 would fix the hum. Placed in the I/Q line to the qse- And depending on if the Teensy audio library was used in development (don¡¯t know) it could be relatively simple to put in the chain. I might try a piece of audio gear to see if my theory works.?
Since graphic eq introduces phase shifts that may be an issue.? On Sat, Sep 2, 2023 at 6:36 PM K9HZ < bill@...> wrote: Ohhhh maybe loose connection!
Dr.?William J. Schmidt - K9HZ J68HZ 8P6HK ZF2HZ PJ4/K9HZ VP5/K9HZ PJ2/K9HZ ? Owner - Operator Big Signal Ranch ¨C K9ZC Staunton, Illinois ? Owner ¨C Operator Villa Grand Piton - J68HZ Soufriere, St. Lucia W.I. Rent it:
email:??bill@... ? On Sep 2, 2023, at 12:21 PM, Larry Acklin <acklin@...> wrote:
?Curious. I was fiddling, same setup, keyed the rig and got a perfect clean residual carrier. QSY up band, have the BCI again. Went back down band and still have it. So for a moment there it was without the noise.?
Larry
KB3CUF
|
The BCI is not present now, my stack of attenuation may have had a loose connection.?
But since I¡¯m spitballing here- the standard SSB audio bandwidth is 300-3000 Hz. Sounds like an audio hipass filter kneed at 250-300 would fix the hum. Placed in the I/Q line to the qse- And depending on if the Teensy audio library was used in development (don¡¯t know) it could be relatively simple to put in the chain. I might try a piece of audio gear to see if my theory works.?
Since graphic eq introduces phase shifts that may be an issue.?
On Sat, Sep 2, 2023 at 6:36 PM K9HZ < bill@...> wrote: Ohhhh maybe loose connection!
Dr.?William J. Schmidt - K9HZ J68HZ 8P6HK ZF2HZ PJ4/K9HZ VP5/K9HZ PJ2/K9HZ ? Owner - Operator Big Signal Ranch ¨C K9ZC Staunton, Illinois ? Owner ¨C Operator Villa Grand Piton - J68HZ Soufriere, St. Lucia W.I. Rent it:
email:??bill@... ? On Sep 2, 2023, at 12:21 PM, Larry Acklin <acklin@...> wrote:
?Curious. I was fiddling, same setup, keyed the rig and got a perfect clean residual carrier. QSY up band, have the BCI again. Went back down band and still have it. So for a moment there it was without the noise.?
Larry
KB3CUF
|
Ohhhh maybe loose connection!
Dr.?William J. Schmidt - K9HZ J68HZ 8P6HK ZF2HZ PJ4/K9HZ VP5/K9HZ PJ2/K9HZ ? Owner - Operator Big Signal Ranch ¨C K9ZC Staunton, Illinois ? Owner ¨C Operator Villa Grand Piton - J68HZ Soufriere, St. Lucia W.I. Rent it: www.VillaGrandPiton.com
email:??bill@... ?
toggle quoted message
Show quoted text
On Sep 2, 2023, at 12:21 PM, Larry Acklin <acklin@...> wrote:
?Curious. I was fiddling, same setup, keyed the rig and got a perfect clean residual carrier. QSY up band, have the BCI again. Went back down band and still have it. So for a moment there it was without the noise.?
Larry
KB3CUF
|
Yep thats probably it¡
Dr.?William J. Schmidt - K9HZ J68HZ 8P6HK ZF2HZ PJ4/K9HZ VP5/K9HZ PJ2/K9HZ ? Owner - Operator Big Signal Ranch ¨C K9ZC Staunton, Illinois ? Owner ¨C Operator Villa Grand Piton - J68HZ Soufriere, St. Lucia W.I. Rent it: www.VillaGrandPiton.com
email:??bill@... ?
toggle quoted message
Show quoted text
On Sep 2, 2023, at 11:51 AM, Larry Acklin <acklin@...> wrote:
?My SA plot looks identical on battery. On Greg¡¯s suggestion, I used the regular SDRPlay app and when the rig is keyed, I hear broadcast on the sideband. Clear as a bell. I live near an AM station tower, but since the BCI is following the suppressed residual carrier, my inclination is that the T41 is picking up the AM station and adding it to the transmission.?
Larry
KB3CUF
|
The noise burst at key down and key up is due to QSE only. Unplugged the I/Q baseband cable, the hum spikes disappeared. The bursts remained. Hooked up an amplified speaker to the QSE I/Q from the main board but can¡¯t hear anything. Probably because the speaker has no low end since it is small in size or low level- the speaker is expecting consumer line level.
So two issues in fact¡
Larry
KB3CUF
|
The transmit equalizer seems to have no effect on the hum.? Turn it on, turn it off, set the bottom two bands all the way down, set them all the way up... no difference.
Which makes sense, seeing as there's no audio coming in - no mic connected, and the mic gain set to -40.
|
T41 Electronic Filter Prototype for QSE Output to Power Amplifier
Here is a little filter board I have had for weeks, and I finally evaluated it:
--
73 Greg KF5N
|
per above post¡  
|
Thanks. I moved things all to new locations on the work desk and the BCI is now gone. However a spectrum display of the residual SSB carrier shows the same issues with the noise spikes at 60, 150, etc below the frequency. So perhaps the SSB artifacts (Al¡¯s term) are still present. They seem to be in the same locations and frequencies as reported by Jerry and my rather coarse SA plot. There is also a burst of audio hash that is present at PTT key down and then at key up. No mic connected. Both power supply and battery behave the same.?
Larry
KB3CUF
|
Hi Larry, the clock to the QSD is disabled during transmit.? That was taken care of start with V049.
There is a special function which handles the Si5351 configuration for calibration only.
--
73 Greg KF5N
|
K9HZ