Bad Transmit Signal

  • 2
  • Problem
  • Updated 3 years ago
  • In Progress
take a look at the pics.  I'm having a bad transmit.  Should I try to reset the unit?


The issue is across all bands...

Photo of Mike KD2CJJ

Mike KD2CJJ

  • 122 Posts
  • 33 Reply Likes

Posted 3 years ago

  • 2
Photo of Mark Gottlieb

Mark Gottlieb

  • 99 Posts
  • 3 Reply Likes
Is this Version 1.5?  If so, I would try deleting and reloading version 1.4.  There is a community thread in progress regarding a distorted, corrupted transmit signal with version 1.5.  FLEX is working on the solution.

Mark
WA2DIY
Photo of Mike KD2CJJ

Mike KD2CJJ

  • 122 Posts
  • 33 Reply Likes
Its 1.5.1 - the latest version which is suppose to fix that problem!
Photo of Mike KD2CJJ

Mike KD2CJJ

  • 122 Posts
  • 33 Reply Likes
PS - my unit came with 1.5 if I remember correctly.

I upgraded the moment 1.5.1 came out.
(Edited)
Photo of Mike KD2CJJ

Mike KD2CJJ

  • 122 Posts
  • 33 Reply Likes
I tried resetting doing everything.. its worse on 6M through 17 Meter.  20M isnt as  bad but still present.  Im at a loss.   I have never seen this before.  yesterday the only thing I did different was do a calibration - which I have done in the past but not in a while.  And that doesnt work either. Unit is how 4 weeks old!
Photo of Mark Gottlieb

Mark Gottlieb

  • 99 Posts
  • 3 Reply Likes
It should not be a hardware issue.  Version 1.5.1 should have the correction in the software that will solve the problem and that is a free download on the FLEX website.  If that doesn't solve it, download version 1.4 or 1.4.16 and see if that fixes the problem.

Mark
WA2DIY
Photo of Bill -VA3WTB

Bill -VA3WTB

  • 3002 Posts
  • 660 Reply Likes
When all this started, Flex began looking for the problem and found it, they fixed the problem in 1.5.1
Photo of Mike KD2CJJ

Mike KD2CJJ

  • 122 Posts
  • 33 Reply Likes
Yea.. well Im running 1.5.1!

Guess they didnt!
Photo of Bill -VA3WTB

Bill -VA3WTB

  • 3002 Posts
  • 660 Reply Likes
something eles going on then, I would open a help ticket.
Photo of Roy - W5TKZ

Roy - W5TKZ

  • 124 Posts
  • 13 Reply Likes
Mike,
Open a help desk ticket. Take the time to give as much detail about the problem as you can.

Roy - W5TKZ
Photo of Ken - NM9P

Ken - NM9P, Elmer

  • 4025 Posts
  • 1248 Reply Likes
A cold reboot is always a good first step. So is totally closing Dax and reopening.
Photo of Rob Fissel

Rob Fissel

  • 270 Posts
  • 48 Reply Likes
DAX gets wonky for me too every once in a while. Shutting down all associated software (DAX/CAT/SSDR, Log4OM, WSJT-X, FLDigi, CW Skimmer... basically anything tied to SSDR software) usually fixes it. 

Side note, Ken. Have you noticed in 1.5.1 anything strange about DAX VOX? When I first start up after a reboot, or restarting software/hardware, VOX won't trip when using a DAX source for somthing like JT-65. I find that I have enable TX so that audio is flowing, and click MOX on and off a few times. After that, VOX starts to trip as expected. 
Photo of Ken - NM9P

Ken - NM9P, Elmer

  • 4025 Posts
  • 1248 Reply Likes
Rob, I haven't noticed that particular VOX issue, since I don't use VOX for Digital modes.  I use CAT commands or hardware COM to Key PTT. 

I do know that if you engage DAX while in SSB mode that VOX does not work on the mike inputs.  I wish this was a selectable option.  When I am contesting and using the voice keyer program I cannot use VOX from my mike.  I wish I had the option to engage vox in SSB/AM/FM modes when DAX is active.

I understand why they don't have it engaged in DIGI modes, because you don't want to sneeze or holler at your cat while working PSK31 and have your voice transmitted all over the digital bands!

Ken - NM9P
Photo of Mike KD2CJJ

Mike KD2CJJ

  • 122 Posts
  • 33 Reply Likes
I did not try rebooting my PC - I did however trying power cycling the unit and resetting the persistence DB (hold down power button till its white).  That did not help.

I will see a recycle of my PC helps.

But why would that help?  I thought transmit all happens within the radio and the PC is just a dumb client?

Help Desk ticket submitted.

The only thing I did different very recently (yesterday) was do a calibration - which doesnt work clearly.  

PS - now I know why no one could hear me during the contest - even with my 1kw amp on.  I was constantly being asked what my call sign was after much repeating.  
(Edited)
Photo of Bill W2PKY

Bill W2PKY

  • 461 Posts
  • 79 Reply Likes
Hello Mike
Sounds like a good time to try the FDX feature to hear your transmitted signal.
See if the distortion is present at low power levels and if not increase power to see what happens. I have had issues with RFI getting into the earphone circuit that caused distorted audio. Good luck I'm sure you will find a fix.
Photo of Mike KD2CJJ

Mike KD2CJJ

  • 122 Posts
  • 33 Reply Likes
I spend probably 2 hours yesterday tweaking my transmit audio because of the very poor results on the contest (many many time I had to say my call - with my FTDXF i never had this issue)...   The audio sounds pretty good even with the default profiles.  However I suspect the issues were there as I was having significant issues during the contest.

The issue is not as apparent under 20 watts... on the XVR port we are pushing less than 1 watt.
(Edited)
Photo of Mike KD2CJJ

Mike KD2CJJ

  • 122 Posts
  • 33 Reply Likes
Can anyone post a comparative screen shot of their 6 meters and 18 meters using the Tune Function at 100 watts into their antenna?

I dont know what a clean signal should look like and thus that would help me.
Photo of Cal Spreitzer

Cal Spreitzer

  • 352 Posts
  • 64 Reply Likes
Mike,

Here is a quick snapshot of my 6300 tune on 17M.  Looks almost the same. 

 

Note:  I've been having TX issues with 1.5.1 also.  Trouble ticket has been generated but no resolution yet..

Cal/N3CAL
Photo of Mike KD2CJJ

Mike KD2CJJ

  • 122 Posts
  • 33 Reply Likes
Cal... Dudly is eluding to that this is a normal signal. He said he will get back to me with a definantive answer. I'm no expert so I could not say.

Thanks for the image. I feel better to know normal or not I'm not the only one.
Photo of Mike KD2CJJ

Mike KD2CJJ

  • 122 Posts
  • 33 Reply Likes
Any update on your ticket Cal? Not too much here. Now just was told its maybe an issue by dudley....
Photo of Cal Spreitzer - N3CAL

Cal Spreitzer - N3CAL

  • 36 Posts
  • 3 Reply Likes
Mike,

My issue appears to be only when using DAX (DIGU) with my 6300.  Flex has told me they are adding a fix to the next release of SmartSDR (v1.6) that manages the DAX ring buffers better, preventing this from happening and lowering the propagation latency through the DAX ring buffer.

I've sent my 6300 back to Flex and they are allowing me to upgrade to a 6500 which will better suit my needs.  Flex has top notch customer service! 

Cal/N3CAL
(Edited)
Photo of Jim Best

Jim Best

  • 46 Posts
  • 9 Reply Likes
Got the same problem. On SSB I still have an ugly looking display somewhat like the old one. I'm not sure if it's just the display or if the signal is actually that wide.
Photo of Dave Dave

Dave Dave

  • 135 Posts
  • 28 Reply Likes

6300 here on 1.5.1 and my signal same settings looks on the display very good, very different then ones posted here.

Same settings as far as I can tell with pix posted.

Photo of Mike KD2CJJ

Mike KD2CJJ

  • 122 Posts
  • 33 Reply Likes
Dave can you post an image?
Photo of Dave Dave

Dave Dave

  • 135 Posts
  • 28 Reply Likes
Photo of Dave Dave

Dave Dave

  • 135 Posts
  • 28 Reply Likes
Here you go, least I think its a normal signal however when I monitor it using FDX its almost too perfect on display.
Photo of Douglas Maxwell

Douglas Maxwell

  • 96 Posts
  • 16 Reply Likes
I am running SSDR v1.5.1 and have a cold start issue with the FPGA which results in distorted TX audio which can be immediately identified (when zoomed out to view whole band 200KHz or so of any band) by seeing multiple evenly spaced spurii of diminishing magnitude from the centre transmit frequency. If you see spurii at or below -100dBm then this isn't a problem. When you see evenly spaced spurii at a much higher level than this during transmit then you have this problem. I have monitored my audio with another trx when this happens and it sounds garbled/digital. Not all 6300s behave the same due to Process Voltage and Temperature. FPGAs are manufactured/bought by Flex in batches from Xilinx, each batch behaves slightly different from the last and designers are meant to ensure sufficiently tight timing constraints are placed on the firmware design to ensure the same behaviour across these batch variations. I recon my cold start problem is due to a firmware interface calibration that is only carried out at startup. It seems that this calibration on my radio fails without the user being alerted and allows severe crud to be transmitted over the entire band....very embarrasing! The solution is that you never transmit on first power on, instead re-start the 6300 then operate as normal. My 6300 always shows dirty tx on cold start, then perfectly acceptable tx after a restart. I have only owned my 6300 for 3 weeks or so and it came direct from the factory. I downloaded the latest SSDR version and upgraded the firmware to the latest so my problem is very current. Flex know about the problem but things have gone awfully quiet from support. I am very dissatisfied with the situation, but realise that many 6300s out there are operating trouble free. Anyway I thought I'd spell out my problem and findings to others that may have the same problem. P.S. My restart fix works with a 100% success rate so far. I imagine this is due to the calibration on re-start being successful as now a warm FPGA (as heat changes interface timings). Anyway YMMV.
(Edited)
Photo of Douglas Maxwell

Douglas Maxwell

  • 96 Posts
  • 16 Reply Likes

Cold start result, tuning into a dummy load with 10w...-40dbm level spurii.
(Edited)
Photo of Douglas Maxwell

Douglas Maxwell

  • 96 Posts
  • 16 Reply Likes

Warm start result, tuning into a dummy load with 10w...-95dbm spurii.
Photo of Douglas Maxwell

Douglas Maxwell

  • 96 Posts
  • 16 Reply Likes
Both above pictures show the same portion of the band and the same magnitude range..just that the second picture is the result of a restart of the 6300. Welcome to my world!
(Edited)
Photo of Mike KD2CJJ

Mike KD2CJJ

  • 122 Posts
  • 33 Reply Likes
Doug that is pretty bad -  both dont look like clean signals to me.
Photo of Douglas Maxwell

Douglas Maxwell

  • 96 Posts
  • 16 Reply Likes
Mike, I recon they have biased an FPGA interface timing window towards high temperature operation. My 6300 will probably perform to higher temperatures better than old working 6300s. The trouble is my cold start temperature is outside the working range for my 6300. Older 6300s may see my problem if they try a cold start at low temperatures i.e. field day in the UK hi! I can only hope that the firmware engineers at FRS will find a compromise solution that lowers the working temperature range to allow old and new 6300s to all cold start properly. If I'm right in the above assumptions, it doesn't say much for the reliability of a 6300s if it's FPGA has an interface on a timing knife edge at room temperature! Anyone like to try doing a cold start of their working 6300 at lower temperatures (using the latest firmware 1.5.1) to see if they can mimik the above test?
Photo of Tim - W4TME

Tim - W4TME, Customer Experience Manager

  • 9152 Posts
  • 3478 Reply Likes
We have address this in the upcoming SmartSDR v1.6 release.
Photo of Douglas Maxwell

Douglas Maxwell

  • 96 Posts
  • 16 Reply Likes
Does this mean you have found the root cause but are waiting for other firmware fixes to go into the v1.6 release? If so, that is great news. Any details on what you think is causing my problems would be very much appreciated.
Photo of Dave Dave

Dave Dave

  • 135 Posts
  • 28 Reply Likes

Tim can you post a pix of how the display should look?

Also Squelch...... 1.6???

Photo of Tim - W4TME

Tim - W4TME, Customer Experience Manager

  • 9152 Posts
  • 3478 Reply Likes
Doug - we found an internal communication timing issue that ends up close to the edge on some radios with the latest firmware.  Thermal factors can move that point to be over the edge, so we changed tolerances to improve reliability.

Dave Dave - the display should look like Doug's 2nd picture.
Photo of Douglas Maxwell

Douglas Maxwell

  • 96 Posts
  • 16 Reply Likes
If working temperature range isn't the root cause, then I would suspect the effect of temperature on similar frequency but asynchronous clocks used over an interface i.e. rx and tx clocks over an RGMII interface such as that used for transferring 1Gb Ethernet. 
Photo of Mike KD2CJJ

Mike KD2CJJ

  • 122 Posts
  • 33 Reply Likes
Glad to hear it was found!   I was getting from my contacts that I sounded "Digital" some times and other times not.

I would suggest during the support process that this information be shared to those who have support cases open!  So fare I have not received an update on my ticket with this good news or even acknowledged it was an issue!  But clearly some have it and some dont.

I would ALSO suggest trying out patches with those customers who have an issue rather than wait till a major release.  This beta testing maybe happening but clearly those of us who have open support tickets are in the dark when it comes to issue resolution.

I am NOT happy with customer support at this point in handling this specific issue.
Photo of Mike KD2CJJ

Mike KD2CJJ

  • 122 Posts
  • 33 Reply Likes
In fairness I just received a reply from support acknowledging the issue with the same explanation.  When 1.6 comes out I will post the results.
Photo of Rob Fissel

Rob Fissel

  • 270 Posts
  • 48 Reply Likes
Any reason why you haven't rolled back to 1.4 at all? 

I have to defend Flex against your customer service handling complaint. I've dealt with Yaesu and ICOM in the past regarding hardware/software/firmware issues on different radios, and let's just say that their responsiveness and day to day customer interaction is severely lacking when stacked up against Flex. 

Also, this is an issue that's going to get fixed in an updated release. What's amazing is that this fix is doable. I say this, because as an IC7100 owner, I'm stuck with the permanent based hardware audio issues (TI chipset problem) that many radio owners are facing right now. ICOM has remained completely silent. Also, the low SSB average RF power output issues that the 7100 is known for... not a peep from ICOM publically, and many emails from reps to a fellow ham deny the issue even exists. No FW updates or anything...

At least with the Flex, they respond to us in a public, collaborative forum, usually within the same day. 
Photo of Mike KD2CJJ

Mike KD2CJJ

  • 122 Posts
  • 33 Reply Likes
Rob, because there are OTHER issues that are fixed in 1.5.1 I have hesitated from rolling back.  

And frankly who are you to defend the company. Customer satisfaction is not a science but an art.  One persons interaction when dealing with a company on an issue can be a perfect experience whereas another persons same dealings would be considered a nightmare.  

There are fundamentals that all customer service organizations should follow as a best practice.  Frequent communication, followup, etc. is one of those.   In this instance, for this type of severe problem that in fact degrades the performance of our transmitters I believe extra attention should be provided to those customers with such a problem.  All issues should need not be treated the same - especially when one may think there is a hardware issue apparent.

I am by no means denigrating the customer service as a whole but for this specific incident I have voiced my concerns with very specific feedback.  In general I believe Flex has OUTSTANDING customer service - but the bar is always raised; no matter how high the bar is.  And as a customer, I will push them to raise that bar.

In my professional experience, I always want to hear from my customers and understand what I am doing right and what I am doing wrong.  We are never perfect.   The moment you become complacent and believe your infallible is the day you start playing catch-up to your competitors.  

Your defense in fact hurts a company in the long run.  Let me as the customer voice my opinion based on my own interactions - this time they are doing wrong in my point of view.    And you have every right to create a post of praise to tell them what they are doing right.

Edit:
And let Flex defend how they handled this incident.  Satisfying a customer after an unhappy experience truly differentiates between good customer service and world class customer service.
(Edited)
Photo of Rob Fissel

Rob Fissel

  • 270 Posts
  • 48 Reply Likes
Frankly, I'm a customer, just like you. I don't see why you feel I can't defend the company, yet it's ok for you to do the opposite. We'll just agree to disagree. 

Sorry we can't see eye to eye. Hope you enjoy your rig as much as I do!

73,

Rob
(Edited)
Photo of Mike KD2CJJ

Mike KD2CJJ

  • 122 Posts
  • 33 Reply Likes
I love my Flex and never going back!!!!
Photo of Douglas Maxwell

Douglas Maxwell

  • 96 Posts
  • 16 Reply Likes
The fix to this problem is in firmware. I see the previous fix for this issue caused SSDR to go from v1.5 to v1.5.1. Now the new fix will cause SSDR to go from v1.5.1 to v1.6.

1) Why (this time round) is there a major version number jump in SSDR?

2) Is the third number in the version number to signify a 'beta' SSDR version?

3) What is the firmware version number?
Photo of Tim - W4TME

Tim - W4TME, Customer Experience Manager

  • 9152 Posts
  • 3478 Reply Likes
1) Why (this time round) is there a major version number jump in SSDR?

It isn't a major release number change.  That would be going from 1.x.x to 2.x.x.  New minor versions (1.5->1.6) indicate a different focus or theme for the minor release features.  For 1.5 it was noise mitigation.  For 1.6 it will be contesting.

2) Is the third number in the version number to signify a 'beta' SSDR version?

We do not release beta versions to the public; everything is a general release or a maintenance release..  The third number in the version string starting with v1.6 will be an alpha release sequence.  For example when SmartSDR v1.6 is initially released it may have a release number of  something like 1.6.6 indicating there were 7 alpha releases of 1.6 (0-6) needed before the general release was ready.  This is a change we will be making for v1.6  and all releases going forward to more easily allow for an interim maintenance release after we have started working on a new minor release.

3) What is the firmware version number?

It is synchronized with the SmartSDR for Windows release.  The first three digits of the version number will match.  Example, if the SmartSDR for Windows release is 1.6.6 the firmware will be 1.6.6.x where x is a build number.
(Edited)
Photo of Douglas Maxwell

Douglas Maxwell

  • 96 Posts
  • 16 Reply Likes
Thanks for the detailed response. Will the new firmware version be tested to ensure clean transmit capability both at the top and bottom of the specified operational temperature range 0 to +50C, +32 to +122F?
Photo of Tim - W4TME

Tim - W4TME, Customer Experience Manager

  • 9152 Posts
  • 3478 Reply Likes
While we do our very best to ensure that any software fix is solid before releasing by testing on a sample of radios, there can be no absolute guarantees since the FLEX-6000 is a very complex embedded system.
Photo of Douglas Maxwell

Douglas Maxwell

  • 96 Posts
  • 16 Reply Likes
Caveats aside, surely in light of the recent firmware instability due to temperature, a transmission test should be carried out at least within likely operational temperatures plus a margin. This is the kind of test that will ensure all 6300s will work and not just a select few as we saw last time round.
(Edited)