|
Hi all,
Calibration has been challenging for me.
In case the below will be helpful, here is the procedure that has ultimately worked well for me.
Thanks to Ian Lee (KD8CEC) for his nice firmware and uBITX_manager utility, and to others who pointed me in the right direction on a different thread.
Step 1:
Downoad:
Within the Zip file you'll find a file name UBITX_CEC_V1.073_16P.hex (This is the file you'll need with the standard display that ships with the board) - extract that out; you'll use it in Step 2.
Step 2:
You need to update your factory radio to Ian's firmware which is accompanied by a nice UI to load the parameters easily from uBITX_Manager utility.
Follow the instructions here (using xloader):
but ultimately pointing to the firmware that is referenced in Step 1 above.
Step 3:
Download and install the file here:
Note - when you run uBITX_Manager; the uart speed is 38400 and not the 57600 used by xloader.
Step 4:
Connect uBITX_Manager to the radio
"Read from uBITX"
Press "Decode =>"
"Load from file" using the .btx file that I attached as screen images and text file. These are my final numbers and they will get you close. My factory-fresh calibration values seemed to be out and it was very difficult for me to bring them to something that is "sane" simply by ear.
Press "<=Encode"
Press "Write to uBITX" - click through the next couple of dialogs
Press "Rest" (Reset)) - your radio is now running with the new firmware and my calibration figures. It should start to sound right.
Step 5:
See this post: /g/BITX20/message/47482 for the 7074 LSB/USB "trick".
With the attached file for my radio, the BFO calibration should be "close enough" so the calibration function accessible via the radio's menu will work ok now. Then, I adjusted the BFO frequency for the best unwanted sideband suppression.
Step 6
Now that the BFO is set properly, I then calibrated the main frequency by setting CW frequency (in uBITX_Manager) to 700 Hz which is the same as my commercial rig CW sidetone. I also ensured that the commercial rig was receiving CW on the same sideband as the uBITX ( lower sideband "CWL" for 40m ).
I then tweaked the main calibration setting until the CW note of the received signal was the same on the commercial rig and on the uBITX. I used an antenna switch to flip back and forth.
Step 7
Fine-tune the BFO setting by using the 7074 "trick" and the radio's BFO calibration setup menu item.
Good luck!
Dave
AE6RQ
-- Dave Beal AE6RQ
|
Dave, ? Thanks for this posting..I had the same problems.? One question:? in Step 5 you say, ¡°Then, I adjusted the BFO frequency for the best unwanted sideband suppression.¡±? Its not clear to me how you did this.? What was your test setup, what did you adjust, and what were you measuring? ? Thanks again. ? 73 John WI6P ?
toggle quoted message
Show quoted text
From: [email protected] <[email protected]> On Behalf Of David Beal
Sent: Monday, April 30, 2018 19:28
To: [email protected]
Subject: [BITX20] Calibration - my approach #ubitx ?
Hi all,
Calibration has been challenging for me.
In case the below will be helpful, here is the procedure that has ultimately worked well for me.
Thanks to Ian Lee (KD8CEC) for his nice firmware and uBITX_manager utility, and to others who pointed me in the right direction on a different thread.
Step 1: Downoad: ? ? Within the Zip file you'll find a file name UBITX_CEC_V1.073_16P.hex?? (This is the file you'll need with the standard display that ships with the board)? - extract that out; you'll use it in Step 2. ? ? Step 2: You need to update your factory radio to Ian's firmware which is accompanied by a nice UI to load the parameters easily from uBITX_Manager utility. Follow the instructions here (using xloader): ? but ultimately pointing to the firmware that is referenced in Step 1 above. ? ? Step 3: Download and install the file here: ? Note - when you run uBITX_Manager; the uart speed is 38400 and not the 57600 used by xloader. ? ? Step 4: Connect uBITX_Manager to the radio "Read from uBITX" Press "Decode =>" "Load from file" using the .btx file that I attached as screen images and text file.? These are my final numbers and they will get you close.? My factory-fresh calibration values seemed to be out and it was very difficult for me to bring them to something that is "sane" simply by ear. Press "<=Encode" Press "Write to uBITX" - click through the next couple of dialogs Press "Rest"? (Reset)) - your radio is now running with the new firmware and my calibration figures.? It should start to sound right. ? Step 5: See this post: ?/g/BITX20/message/47482 for the 7074 LSB/USB "trick". ? With the attached file for my radio, the BFO calibration should be "close enough" so the calibration function accessible via the radio's menu will work ok now.? Then, I adjusted the BFO frequency for the best unwanted sideband suppression. ? ? Step 6 Now that the BFO is set properly, I then calibrated the main frequency by setting CW frequency (in uBITX_Manager) to 700 Hz which is the same as my commercial rig CW sidetone.? I also ensured that the commercial rig was receiving CW on the same sideband as the uBITX ( lower sideband "CWL" for 40m ). I then tweaked the main calibration setting? until the CW note of the received signal was the same on the commercial rig and on the uBITX.? I used an antenna switch to flip back and forth.?
Step 7
Fine-tune the BFO setting by using the 7074 "trick" and the radio's BFO calibration setup menu item.
Good luck!
Dave
AE6RQ --
Dave Beal
AE6RQ
|
Hi John -?
Under the uBITX menu, I selected "Setup On"?
Exit the menu
Re-enter the menu
Scroll past the "Exit menu" option until you get to "Set the BFO"?
Press tuning knob, tweak BFO setting, press PTT (not the Function switch) to confirm and store.?
Do one final "read from uBITX" in uBITX_Manager and you should see the new value.? ?Press uBITX_Manager "decode =>" and then save your configuration file for future tweaking.?
--
Dave Beal
AE6RQ
|
Oh - and I was tuning by ear for lowest audio level on unwanted sideband (LSB) whilst also receiving the wanted USB.?
--
Dave Beal
AE6RQ
|
On Tue, May 1, 2018 at 05:11 pm, David Beal wrote:
Press tuning knob, tweak BFO setting,
John should also have asked here is - what do you mean by "tweak BFO setting" - what does that mean and how is it accomplished in the 'Set the BFO' menu?
|
In the "Set the BFO menu", you can directly modify the BFO frequency while listening to the audio output of the radio.? So, you have real-time feedback. You can not however tune the radio as it's in the BFO adjust mode.??
--
Dave Beal
AE6RQ
|
Dave, ? We¡¯re almost there.? What are you listening for?? A null? A peak? When I tried this before it was not obvious what I was supposed to be listening for¡ ? Thanks again ? John ?
toggle quoted message
Show quoted text
From: [email protected] <[email protected]> On Behalf Of David Beal
Sent: Tuesday, May 1, 2018 19:59
To: [email protected]
Subject: Re: [BITX20] Calibration - my approach #ubitx ? In the "Set the BFO menu", you can directly modify the BFO frequency while listening to the audio output of the radio.? So, you have real-time feedback. You can not however tune the radio as it's in the BFO adjust mode.??
--
Dave Beal
AE6RQ
|
ok - no problem. One step at a time, right?? Quoting from:?? /g/BITX20/message/47482" Another member of the group gave me this advice.Tune to 7074 using LSB. You should hear nothing. Then change to USB andyou should hear all kinds of psk and other digital modes.If you hear these signals in LSB then your BFO frequency is notcorrect. My BFO was way off after I first tried to calibrate it. I then startedover with mine at 996.4 and ended up at 996.7. If yours is a long wayoff from that then I'll bet you hear psk at 7074 when in LSB."For my interpretation of the above, I Step 0: Confirm that the BFO is set to 11997.00 (via the radio menu which is accessible after you enable "Setup On" ) Step 1: Set frequency to 7074 Step 2: Set mode to USB? Step 3: Confirm that there is audible digital activity Step 4: Set mode to LSB Step 5:? Return to the "Set the BFO" menu (at this time in this menu, you can not adjust mode or frequency)? Step 6: while listening to 7074 on LSB, adjust the BFO for the lowest audible level of the digital signals.? (in my case, the default 11997.00 was good. I don't expect that you'll need to adjust this much at all). Press the PTT button to write to uBITX memory Step 7: Set mode to USB - ensure that you can still hear the digital signals well.? Note that I did have a very strong, lower audio frequency digital signal that still got through on LSB,? but all of the others all but disappeared with BFO adjustment.? ? Good luck!? -- Dave Beal AE6RQ
|
Dave¡ªthat¡¯s perfect! ?Again, thanks and 73
toggle quoted message
Show quoted text
On May 1, 2018, at 21:41, David Beal < aardvarq@...> wrote: ok - no problem. One step at a time, right?? Quoting from:?? /g/BITX20/message/47482" Another member of the group gave me this advice.Tune to 7074 using LSB. You should hear nothing. Then change to USB andyou should hear all kinds of psk and other digital modes.If you hear these signals in LSB then your BFO frequency is notcorrect. My BFO was way off after I first tried to calibrate it. I then startedover with mine at 996.4 and ended up at 996.7. If yours is a long wayoff from that then I'll bet you hear psk at 7074 when in LSB."For my interpretation of the above, I Step 0: Confirm that the BFO is set to 11997.00 (via the radio menu which is accessible after you enable "Setup On" ) Step 1: Set frequency to 7074 Step 2: Set mode to USB? Step 3: Confirm that there is audible digital activity Step 4: Set mode to LSB Step 5:? Return to the "Set the BFO" menu (at this time in this menu, you can not adjust mode or frequency)? Step 6: while listening to 7074 on LSB, adjust the BFO for the lowest audible level of the digital signals.? (in my case, the default 11997.00 was good. I don't expect that you'll need to adjust this much at all). Press the PTT button to write to uBITX memory Step 7: Set mode to USB - ensure that you can still hear the digital signals well.? Note that I did have a very strong, lower audio frequency digital signal that still got through on LSB,? but all of the others all but disappeared with BFO adjustment.? ? Good luck!? -- Dave Beal AE6RQ
|
|
|
I had a horrible time trying to calibrate using the procedure in Rod's manual and KD8CEC version 1.072.? Part of the problem was that with current
ionospheric conditions WWV's 10 MHz signal has not been good in my locations at the times I have been free to work on this. And as I tried to zero beat
WWV my uBitX seemed to jump from USB to LSB or back.? That is of course what is supposed to happen in general use, although I didn't expect it to
happen during the calibration procedure.? I went back and reinstalled Ashhar's? original firmware, and was frustrated that I could not get the memory
manager to work. in looking for instructions on using the memory manager the instructions said that changes were needed to Ashhar's? source to
make the memory manager work. I decided that was not worth my time, I am a CW person and I think Ian's firmware works much better on CW
I am now thinking of going to get Ian's v1.073 and try again.? I am also considering modifying the source to change the frequency at which
USB/LSB change automatically to something slightly different, like 9.900 MHz. Does anyone else think that might be? worthwhile?
Also, I wonder if somewhere there is an up-to-date manual on using the memory manager with Ian's current firmware.
73,
Al N1AW
|
Hi Al,?
I too was completely unsuccessful to calibrate against WWV.? I had multiple zero-beat points (not sure what they were due to) and I couldn't determine which was the correct one to use for zero-beat, and I ended up with opposite sidebands and other annoying oddities that I don't recall.? Perhaps these would have been avoided if I had just written the default BFO values; but I wasn't that smart then.? The first thing I would do, with a new radio is to update the factory firmware with Ian's firmware () ; explicitly download the default configurations, and then go through the calibration steps that I have above.?
Here are some links to help you use uBITX_Manager. These were written a few releases ago so the details may be incorrect (baud rates) or new features not listed:?
Note that the recent addition of the Reset ("Rest") button is very helpful as you won't have to plug/unplug your USB cable as you modify the settings.??
--
Dave Beal
AE6RQ
|
If you use something like FLdigi to watch the modulation, which can graphically show you the frequency of the modulation, you can probably calibrate pretty accurately.
toggle quoted message
Show quoted text
Dave? AE6RQ
Using WWV or WWVH for frequency calibration is a bit confusing unless you know
about their modulation scheme.? They transmit carrier with 500 Hz modulation during
part of the even minutes and carrier with 600 Hz modulation during part of the odd
minutes.? Minutes 43 through 52 are un-modulated carrier.? These are the times when
you can zero your Si5351a with WWV.
Scroll down in the page for this URL and you will find a chart that explains all this.
Arv? K7HKL
_._
|
Thank you Arv,?
I just happened to read your post at 43 minutes past the hour and it seems to have worked well.? I ended up adjusting my radio +17 from my initial setting.? ?
Immediately after calibrating against WWV with the radio menu, something odd with the BFO calibration setting did happen though; the BFO also changed (it was up around 11999 and I think it took effect when I pressed PTT to confirm the new calibration setting).
After realizing that the radio's BFO calibration menu would take me forever to wind back to 11997, I connected uBITX_Manager as I could just type in the number and then write to the radio.? However - as soon as uBITX_Manager connected, it seems to have reset the radio and the BFO value reverted to 11997.? ?
So I additionally suggest to immediately following your WWV frequency calibration with a radio power cycle before checking the BFO Calibration setting.? If anyone else can confirm this behavior, perhaps we can flag it for Ian.?
Finally, tonight there was some Olivia or other MFSK traffic that was very audible on LSB at 7074.? So for those doing the calibration using the "by ear digital signal" technique I summarized above, be sure that you're listening for the 15 second interval FT-8 transmissions and not the arbitrary length MFSK modes.??
--
Dave Beal
AE6RQ
|
Dave: Many thanks for your calibration post. I was able to load your data file to bring my uBitx back to life after screwing it up trying to perform the Cal using WWV. My radio is now about 30 Hz high. Have you or anyone figured out the relationship between the Master Cal numbers and frequency? as would like to adjust the 30 Hz frequency offset by manipulating the data directly.
73
N4AYE
|
The calibration factor adjusts the nomially 875000000 value for si5351bx_vcoa down in ubitx_si5351.ino
in Farhan's stock ubitx code.
Any change to vcoa? (which is the frequency of the VCO inside the si5351 in Hz, nominally 875mhz)
is exactly proportional to the change you can expect in the operating frequency of the rig.
So if you are receiving a 10mhz WWV signal and find that your radio dial reads 10000030 when at zero beat,
then reduce vcoa by? ?875000000*(30/10000000) = 2625
With that, your rig should be exactly on frequency for anything from 500khz to 30mhz.
One of the joys of having all oscillators being based on a single 25mhz reference crystal.
Jerry, KE7ER
toggle quoted message
Show quoted text
On Wed, May 16, 2018 at 04:52 am, <alans77@...> wrote:
Dave: Many thanks for your calibration post. I was able to load your data file to bring my uBitx back to life after screwing it up trying to perform the Cal using WWV. My radio is now about 30 Hz high. Have you or anyone figured out the relationship between the Master Cal numbers and frequency? as would like to adjust the 30 Hz frequency offset by manipulating the data directly.
|
Does anyone have a simple way to calibrate the uBITX when a good service monitor is at hand ???
I know the unit will not be super stable and don't expect it to be.? I have a service monitor that I know is accurate to less than 1 HZ at 10 MHz. Every time I try to get the ubitx set I wind up messing up the calibration, and have to start back from a loaded set of parameters.? I have it working fine now, but not too sure how accurate it is set.? I usually feed a 1 khz tone into the mic and read the frequency on the counter to see if it is that far above or below the ubitx LCD.?
On Wed, May 16, 2018 at 10:29 AM, Jerry Gaffke via Groups.Io <jgaffke@...> wrote: The calibration factor adjusts the nomially 875000000 value for si5351bx_vcoa down in ubitx_si5351.ino
in Farhan's stock ubitx code.
Any change to vcoa? (which is the frequency of the VCO inside the si5351 in Hz, nominally 875mhz)
is exactly proportional to the change you can expect in the operating frequency of the rig.
So if you are receiving a 10mhz WWV signal and find that your radio dial reads 10000030 when at zero beat,
then reduce vcoa by? ?875000000*(30/10000000) = 2625
With that, your rig should be exactly on frequency for anything from 500khz to 30mhz.
One of the joys of having all oscillators being based on a single 25mhz reference crystal.
|
The technique using WWV is more than good enough if you keep one thing in mind.
The mater osc referenced to the crystal osc is the source for the HFOm IFLO and BFO
so they all move if the master OSC? moves.
IF you do the cal you may have to move the other two to have it all line up.? however
In my heart 30hz is not a significant error.? Must have been awful for those guys using?
old school VFOs with dials calibrated to 1khz.? ;)? ? Then again I manage to talk to people
with my Tempo-one (old school analog tube).
I use a crystal source (20mhz) that's dialed into WWV 30 years ago and checked as
a very accurate source.? That is all and then go through the process.
I do it on receive. just as accurate ans the RX and tX should be the same.
For 20MHz source, set RX for 1khz tone (20.001 or 19.999 depending on selected sideband).
Use counter to measure tone out of the RX (1000.0 hz).? For the 5351 based systems I set the?
correction for the crystal osc by measuring the the its highest frequency output first with a
counter and make that correction as others have noted.? Then I only need to touch up the BFO
so its at the right point on the filter slope. at that point if there is error if the control has the
option I correct the IF offset.? Over all I do it 2 or three times for high accuracy or call it good
enough.? Usually if its it is within about 10hz or less and that better than most radios out there
I stop there.
For frequency stability its all tied to the 25mhz osc and that is usually 20 or 50ppm part?
over the temperature range so over time It drifts a few Hz to aging and the rest is
temperature.? For room temp plus or minus less than 10C change its a small error
and a generally reproducable one.
Allison
|
Jerry: Thank you! That did the trick.
73
N4AYE
|
Ha! ?The HW ¨C 100 that we used were ?calibrated only to 5 kHz, and I have seen errors of Thirty KiloHertz ?on some of the ones I have renovated.
My sb-102 was calibrated at 1 kHz. When I saw ?this rig for only $109 with digital ?synthesis I thought I had died and gone to heaven.
On May 16, 2018, at 13:56, ajparent1/KB1GMX < kb1gmx@...> wrote:
toggle quoted message
Show quoted text
The technique using WWV is more than good enough if you keep one thing in mind.
The mater osc referenced to the crystal osc is the source for the HFOm IFLO and BFO
so they all move if the master OSC? moves.
IF you do the cal you may have to move the other two to have it all line up.? however
In my heart 30hz is not a significant error.? Must have been awful for those guys using?
old school VFOs with dials calibrated to 1khz.? ;)? ? Then again I manage to talk to people
with my Tempo-one (old school analog tube).
I use a crystal source (20mhz) that's dialed into WWV 30 years ago and checked as
a very accurate source.? That is all and then go through the process.
I do it on receive. just as accurate ans the RX and tX should be the same.
For 20MHz source, set RX for 1khz tone (20.001 or 19.999 depending on selected sideband).
Use counter to measure tone out of the RX (1000.0 hz).? For the 5351 based systems I set the?
correction for the crystal osc by measuring the the its highest frequency output first with a
counter and make that correction as others have noted.? Then I only need to touch up the BFO
so its at the right point on the filter slope. at that point if there is error if the control has the
option I correct the IF offset.? Over all I do it 2 or three times for high accuracy or call it good
enough.? Usually if its it is within about 10hz or less and that better than most radios out there
I stop there.
For frequency stability its all tied to the 25mhz osc and that is usually 20 or 50ppm part?
over the temperature range so over time It drifts a few Hz to aging and the rest is
temperature.? For room temp plus or minus less than 10C change its a small error
and a generally reproducable one.
Allison
|
Al Woodhull