Auto Calibration Error Rev 1.5.1

  • 3
  • Problem
  • Updated 4 years ago
Well the 1.5.1 upgrade did not work for me on my 6300. Same problem. I am submitting a Help Desk Ticket because this sounds like I may have a hardware problem. 
Photo of Roy - W5TKZ

Roy - W5TKZ

  • 134 Posts
  • 13 Reply Likes

Posted 4 years ago

  • 3
Photo of Doug Hall

Doug Hall

  • 214 Posts
  • 60 Reply Likes
Didn't work for me either Roy. I tried a GPS Disciplined 10 MHz oscillator as well as WWV. Incorrect calibration with either source.
73,
Doug K4DSP
Photo of Doug Hall

Doug Hall

  • 214 Posts
  • 60 Reply Likes
Some additional info: the 10 MHz signal from the GPSDO was padded down to -40 dBm, while the 10 MHz WWV signal was fairly steady at around -80 dBm. I got very similar numbers for either source using auto-calibrate. The result returned by the auto calibration routine was about 1180 ppb off. This was after a 1 hour warmup.
Photo of EA4GLI - 8P9EH - Salvador

EA4GLI - 8P9EH - Salvador

  • 1780 Posts
  • 547 Reply Likes
Same for me. I end up doing the calibration manually by ear switching from LSB to USB on WWV 15 Mhz
Photo of Bruce - NØJX

Bruce - NØJX

  • 8 Posts
  • 1 Reply Like
I've tried the Auto Calibration several times using SmartSDR V1.5.1 and it is still giving an incorrect offset.  Using WWV at 5 MHz the Autocal routine gives results of -1100 (the same as the previous version V1.5.0), the correct offset should be +500 for my 6300.
(Edited)
Photo of Mike KD2CJJ

Mike KD2CJJ

  • 122 Posts
  • 34 Reply Likes
Me too..it gives me -1350 and for me +50 is near perfect. I don't think its working
Photo of Cal Spreitzer

Cal Spreitzer

  • 437 Posts
  • 111 Reply Likes
My 6300 is also off -1057 thru -1259 with ver 1.5.1

Cal/N3CAL
Photo of Bob Brown - N8OB

Bob Brown - N8OB

  • 90 Posts
  • 6 Reply Likes
I checked my 6300 on 5Mhz and Auto-cal showed -1446 and my manual correction is +640.  Sure sounds like they didn't get it right yet.

Bob/N8OB
Photo of James Kennedy-WU5E

James Kennedy-WU5E

  • 333 Posts
  • 34 Reply Likes
did my CAL at 10 MHz and it worked fine on my 6500

Jim/WU5E
Photo of Bill Roberts

Bill Roberts

  • 226 Posts
  • 46 Reply Likes
My results were the same.  I could not receive WWV at 15 MHz (solar flare?).  Set it to do auto-calibrate at 10 Mhz where WWV was strong and it worked fine.
Photo of Cal Spreitzer

Cal Spreitzer

  • 437 Posts
  • 111 Reply Likes
I also noticed while trying to do Cal on 5.0 Mhz with 6300 you can't use Xvtr or Ant 2 ports.  It switches to Ant 1 every time?? 

Cal/N3CAL
(Edited)
Photo of Tim - W4TME

Tim - W4TME, Customer Experience Manager

  • 9186 Posts
  • 3541 Reply Likes
Correct.  We only use ANT-1 for the frequency calibration.
Photo of Al / NN4ZZ

Al / NN4ZZ

  • 1848 Posts
  • 670 Reply Likes
Hi Tim,
I found that I can use Ant 2 for the frequency calibration.  After pressing the START button, it starts with ANT 1 but then I  select ANT 2 in the slice.  It seems to work for me.    

Any problem with doing it that way?

Regards, Al / NN4ZZ  
al (at) nn4zz (dot) com
6700 - HW......... V 1.5.1.70
SSDR / DAX...... V 1.5.1.152
CAT................... V 1.5.1.0
Win10


 
(Edited)
Photo of Tim - W4TME

Tim - W4TME, Customer Experience Manager

  • 9186 Posts
  • 3541 Reply Likes
I would not do it.  The calibration routine is designed to use ANT-1.  If you do that, YMMV on the accuracy due to the signal loss at the relay switch.
Photo of Al / NN4ZZ

Al / NN4ZZ

  • 1848 Posts
  • 670 Reply Likes
Hi Tim,
Ok on the YMMV.  Here is some background on what I found and why I use ANT 2. 

I did some testing and the offset I get is about the same within a few ppb when I switch between ANT 1 or ANT 2.  It seems to vary a few ppb when I repeat the test on a given antenna anyway so assume that is normal.  But the main reason I've been using ANT 2 is because of an issue I found.  

I use DDUtil to control my most of my antennas.  DDUtil sends the frequency data to an Array Solutions BandMaster switch.  The ANT 1 input is then connected to the appropriate antenna.  I have a separate receive antenna directly connect to ANT 2 that does not go thru the BandMaster switch.  The data is sent to the BandMaster via a serial port and in the snapshot below you can see where I'm monitoring it.    When the calibration START button is pressed, the incorrect frequency data is being sent to the switch.   There is an extra zero in the frequency data.   The BandMaster switch decides I don't have an antenna for that frequency and disconnects them all and of course WWV is no longer heard.  

You may remember that a while back there was an identical issue where DDUtil had to make a change related to packet timing to fix this same problem during normal frequency change operations.  

https://community.flexradio.com/flexradio/topics/v1_3_ddutil_doesnt_seem_to_be_getting_band_data_on_...

I suspect the root cause I see during the calibration is similar.   I haven't reported it to Steve (DDUtil Author)  because I doubt many others have the same antenna switching setup.  And since using ANT 2 seems to work, I didn't want to make extra work for something that may not a problem for most folks.   If anyone else has this issue I can forward on my notes and results to Steve.

It sounds like as long as the results I get are acceptable (i.e. YMMV) my workaround is okay. 

Regards, Al / NN4ZZ  
al (at) nn4zz (dot) com
6700 - HW......... V 1.5.1.70
SSDR / DAX...... V 1.5.1.152
CAT................... V 1.5.1.0
Win10


The first 6 lines of frequency data below are from monitoring WWV in a slice.   The issue arises when I press the START button to calibrate.   After the calibrate is complete the data returns to normal although that is not captured below. 

Photo of Tim - W4TME

Tim - W4TME, Customer Experience Manager

  • 9186 Posts
  • 3541 Reply Likes
Official Response
Thank you for these reports that the frequency calibration is not working as intended in 1.5.1.  I have added this to our bug tracker and we are actively investigating the issue.  It looks like for a subset of 6300 users, the frequency calibration  is not working and we are trying to determine why.  If you have this issue, in the mean time, you can manually calibrate your radio by entering the correct offset.  There are several methods for doing this.  One is by using the Fldigi Frequency Analysis feature (http://www.wb1gof.org/files/UsingFldigiForRigCalibrationAndFrequencyMeasurement.pdf) or by 
(Edited)
Photo of Ross - K9COX

Ross - K9COX

  • 345 Posts
  • 107 Reply Likes
Had to wait until today to test since the propagation has been so poor. Worked fine on a 6300 whereas it did not with 1.5
Photo of Tim - W4TME

Tim - W4TME, Customer Experience Manager

  • 9186 Posts
  • 3541 Reply Likes
It seems to effect some radios and not others. A most frustrating type of defect, it is.
Photo of Rick  WN2C

Rick WN2C

  • 293 Posts
  • 41 Reply Likes
Hey Tim, can you explain how the calibration works? Why would I get so different a reading from 1 band to another if the signals so close in strength? And what is the radio comparing frequency to when it goes thru the calibration cycle...WWV frequency... or the different tones from WWV as the clock ticks away?

Rick  WN2C
Photo of Tim - W4TME

Tim - W4TME, Customer Experience Manager

  • 9186 Posts
  • 3541 Reply Likes
What is the standard deviation of the "so different readings"?

We calibrate on the carrier tone on the exact WWV frequency.  
Photo of Jim, KJ3P

Jim, KJ3P

  • 93 Posts
  • 14 Reply Likes

So, it doesn't matter what WWV is transmitting for audio...you reference only the carrier?

FWIW, my 6300, vintage August 2014, will not calibrate, either.  Readings of -400, -232, etc., when correct value should be around +400.  It's behaved like this for the past few updates.  Nice strong 20 MHz carrier.

Photo of Tim - W4TME

Tim - W4TME, Customer Experience Manager

  • 9186 Posts
  • 3541 Reply Likes
We do not use the audio portion of WWV; just the carrier.

Please see my comments above about some 6300s continuing to have an issue with frequency calibration.  We are investigating it.
Photo of Bob Brown - N8OB

Bob Brown - N8OB

  • 90 Posts
  • 6 Reply Likes
According to my calculations, the frequency spec is  0.5ppm which is 7.5 hz at 15Mhz.  Auto Cal on my 6300 shows 12.3 Hz low using Auto Cal.  After the supposed fix in v1.51, it still didn't work correctly. I know the problem is being looked into, but is there any fix coming in the near future?  I use the Flex 6300 for Frequency Measuring Test and it would be nice if this feature would at least bring the radio within Spec.
Bob/N8OB
Photo of Tim - W4TME

Tim - W4TME, Customer Experience Manager

  • 9186 Posts
  • 3541 Reply Likes
This issue has been fixed for the upcoming SmartSDR v1.6 version.