Topics

Cec v1.2 software question

barry halterman
 

I have a question with the CEC v1.2 firmware that I am running in my V4 ubitx. Can someone explain to me how or what happens when the ATT (attenuator) is enabled in software? What would cause receiver sensitivity to be better with the ATT on and adjusting the slider half way to the right, then when the ATT is turned off?
Thanks
K3BO

Evan Hand
 

Barry,

It is my understanding that the ATT function of the CEC software is using the "skirt" of the first 45 MHz filter to "Attenuate" the signal.  This works by adjusting the second VFO or Clock 1 to move the received section of the band to be just off of the center of the crystal.  If the signal is getting stronger or clearer, then the BFO might be slightly out of adjustment 

The way the uBITX works is the dial or received frequency is added to the first VFO (Clock2 of the Raduino) to get the 45 MHz first IF.  This is passed through the 45 MHz single crystal filter.  Because it is a single crystal, the bandwidth (15 Mhz) is wider than the second 12 crystal SSB filter (2500 Hz).  Also, the slope of the sides of the 45 MHz filter are relatively shallow.  To attenuate the received signal, the second VFO is adjusted up or down to move the portions of the spectrum that are then passed through the second IF and the second 12 crystal filter that has a much steeper slope and bandwidth.  If the alignment of the second VFO and BFO are not correct, then instead of attenuating the signal on the skirt, the recieved part is moving towards the center of the 45 MHz filter.

The one thing to be careful of when adjusting the BFO is to not get to the point where the carrier is not suppressed.  I have had this happen to me when I was trying for the "best" reception and sound on receive.  I then started using an audio spectrum analyzer to adjust the BFO.  This seemed to work the best for me.  Adjusting the BFO when tuned to a clear frequency allows you to use the white noise to adjust the signal "plateau" to start around 500 Hz and drop off at 3000 Hz.

The software can be gotten on the internet for free.  I would try it to see if the received part either too high or too low.

The full description of the uBitx operation is on the HFsignals web page.  There are a couple of errors on the block diagram in the clock1 frequency for upper and lower sideband (they are reversed) and that the v5 board SSB crystal frequency is 11.0592 MHz and not the original v3 and v4 12 MHz.
http://www.hfsignals.com/index.php/ubitx-circuit-description/

Just my thoughts, please be sure to check your rig for proper operation and out of band signals after making adjustments.
73
Evan
AC9TU

barry halterman
 

Evan, thank you for the information and you got me looking in the correct area, the 45 MHz filter response. Both calibration and BFO were properly adjusted, so I ruled that out. Apparently,
When the ATT was on the LO was shifted to have the signals more in the passband. What I did was remove the two 680uh inductors and installed two 330uh. That adjusted the filter response to compensate for the LO. The 45mhz filter must be slightly off.
All is good, no loss in transmit power, etc.
Barry

On Wed, Nov 27, 2019, 7:38 AM Evan Hand <elhandjr@...> wrote:
Barry,

It is my understanding that the ATT function of the CEC software is using the "skirt" of the first 45 MHz filter to "Attenuate" the signal.  This works by adjusting the second VFO or Clock 1 to move the received section of the band to be just off of the center of the crystal.  If the signal is getting stronger or clearer, then the BFO might be slightly out of adjustment 

The way the uBITX works is the dial or received frequency is added to the first VFO (Clock2 of the Raduino) to get the 45 MHz first IF.  This is passed through the 45 MHz single crystal filter.  Because it is a single crystal, the bandwidth (15 Mhz) is wider than the second 12 crystal SSB filter (2500 Hz).  Also, the slope of the sides of the 45 MHz filter are relatively shallow.  To attenuate the received signal, the second VFO is adjusted up or down to move the portions of the spectrum that are then passed through the second IF and the second 12 crystal filter that has a much steeper slope and bandwidth.  If the alignment of the second VFO and BFO are not correct, then instead of attenuating the signal on the skirt, the recieved part is moving towards the center of the 45 MHz filter.

The one thing to be careful of when adjusting the BFO is to not get to the point where the carrier is not suppressed.  I have had this happen to me when I was trying for the "best" reception and sound on receive.  I then started using an audio spectrum analyzer to adjust the BFO.  This seemed to work the best for me.  Adjusting the BFO when tuned to a clear frequency allows you to use the white noise to adjust the signal "plateau" to start around 500 Hz and drop off at 3000 Hz.

The software can be gotten on the internet for free.  I would try it to see if the received part either too high or too low.

The full description of the uBitx operation is on the HFsignals web page.  There are a couple of errors on the block diagram in the clock1 frequency for upper and lower sideband (they are reversed) and that the v5 board SSB crystal frequency is 11.0592 MHz and not the original v3 and v4 12 MHz.
http://www.hfsignals.com/index.php/ubitx-circuit-description/

Just my thoughts, please be sure to check your rig for proper operation and out of band signals after making adjustments.
73
Evan
AC9TU

Erwin Serlé
 

I just found and applied this http://www.hamskey.com/2018/04/improve-ubitx-receive-performance-by.html
and for me checking in uBITx MM software the IF function and then setting ATT=70 made a big difference in receive performance.

The number of FT8 signals received just exploded, at least doubled or tripled.

At the moment I am receiving WSPR and the system with the uBITx is better then my YAESU FT450D.

Of course they are on different antenna's

All this was after noting that my base frequency was allright, about 8 Hz different from FT450D (2ppm) on 80 meter.
And correcting the 'location' of my USB sideband, as this was too far away from the dial frequency.
I calibrated the BFO by carefully changing the big number in MM for USB calibration. In my case factory setting 11054600 and now 11056500. This took me 7 iterations watching the change in the 2 FT8 results: dial frequency and active bandwidth.

This started at 7.072.000 between 2300 and 4500 and ended where it should be at 7.074.000 between 300-2600.
Both VFO and BFO can still be improved I know.

Oh, I am currently using the latest CEC firmware 1.20
-
PE3ES - F4VTQ - Erwin
73