Timing of Ref Power measurement

Greg N8GD

By the way, I use a 20ms TX delay as specified by Ameritron, and it works just fine.

Greg - N8GD

Dave - WB5NHL

I use 20ms also.

Dave - WB5NHL

Definitive answer today, sort of. In the spirit of Rio, I now pass the baton to Flex radio. Test conditions; 2 different Flex 6500s, 2 different ALS-600s, 2 different radio shacks, multiple bands. Both  6500's with SmartSDR ver 1.8.3.167. Hardware ver 1.8.3.38. All combinations exhibited the Ref Pwr immediate jump and then fall back. When the Ref Pwr jump exceeded exceed 25 W the Fwd Pwr folded back. Ref Pwr reported by Flexmeter.  However there is a BIG BIG caveat. The phenomena occurs in CW mode and NOT LSB or USB!!!  So for everyone who has not seen the foldback occur -- What mode was your Flex in??   Try CW and the Tune button at full drive power??  Why Tune operation is mode sensitive is a question for the software guys to answer. Also, was this Ref Pwr phenomena in CW present in earlier SmartSDR versions??? It wouldn't have had any impact to operation in earlier versions.

I am now confident there is no RF problem in my shack. The ALS-600 is operating ok. Eric/Tim, I hope this info helps point you towards a fix. I am available for any addition testing.

Greg N8GD

Dave,

I am in agreement with you, now.  I just tried my Flex 6300 and ALS-600 combo with v1.8.3 in CW mode and voila - the power does indeed fold back.  You had not previously mentioned this occurred while using the TUNE function in CW mode.  LSB and USB, as well as RTTY seem to work without foldback,  AM, FM and most other modes I tried all exhibited the foldback issue.  This seems strange because I would think that TUNE is TUNE, no matter what mode you're in - you're basically throwing a carrier out (and I would think it would be the same as keying the radio in CW - apparently not).

So, I can confirm your foldback issue with the ALS-600.  That is now 4 systems exhibiting this problem.  FRS, the ball is back in your court.

Greg - N8GD

Dave - WB5NHL

I agree. I did not think mode would make a difference. We just discovered today.

Tom, WA9QQI

Dave,

I just wanted to chime in and tell you what a fine job you did.  I'm sure it wasn't easy shuffling your station from one location to another, or all the cable swapping etc you did as well.  You systematically peeled back the layers of the onion (so to speak) until you got as far as possible.  Nicely done!  Very nicely done!

I too have a fold back problem and, until I read this thread, chased it as an RF ingress problem.  My situation is different as I took my amp out of the system (don't have an ALS-600), disconnected all patch chords and even eliminated my Ethernet cable to my router (used an Ethernet to  wifi adapter.  Worked pretty well too...need to get back to that one day).  Have line isolators and enough iron on my inside coax to create a battle ship.  Still folded back.  Unlike you and Greg, I never even thought to try SSB as I operate mostly CW and like Greg said, "Tune is Tune", or so we thought.  So that was a revelation.

I may have some stray RF floating around my shack but, after reading your work, I think I'll wait and see if the folks at Flex can sort this out a little.

Thanks for your efforts!
Tom

Greg N8GD

I concur, Tom.  Dave did a great job of troubleshooting this issue and getting to the bottom of it along with the specifics.  I was not surprised that he did such a great job, though - according to his QRZ bio he's a retired electrical engineer.  I would expect that kind of Ham to dig, dig, dig and find a solution.  Thanks Dave!  Great troubleshooting - a man after my own heart.

Greg - N8GD

Dave - WB5NHL

Thanks for the kind words. What amp were you using when you got the foldback?

Tom, WA9QQI

It's a four 811A Homebrew amp.  I designed it with a pi network input for each band.  It's input SWR is 1.2:1 or less across each band.  At first I thought the amp could be the problem but it didn't seem to matter.  Ultimately, I left it out of the antenna circuit just to eliminate one variable.

Greg N8GD

I just realized that after Dave stated that this only occurred in CW mode, I had never checked my Flex 6300 without the amp engaged.  I just did a quick check on 40 & 80M with the 6300 barefoot and I have no foldback at low SWR (< 1.3:1) with the amp OFF.  It definitely appears when the amp is in line.  Strange!

Greg - N8GD

Tom, WA9QQI

Well, for what it's worth, I had foldback with the amp in line but off, in line and engaged, and out of the line.  Hey, at least I'm consistent!

All of this brings a couple of questions to mind.  I think most of us agree that the Flex radios are cutting edge gear.  Is there something about the PA in the FLEX that's unique and/or cutting edge too?  Are the folks at FLEX dealing with a different beast then we're used to seeing?  If so, maybe we're going to have to approach our transmission line and antenna circuits with a slightly different mindset.

Or maybe I'm just overthinking this

Tom

Dave - WB5NHL

Latest summary data from my experiments from multiple Flex 6500s, multiple ALS-600 and different shacks.
1. Using Tune in CW mode with the ALS-600 results in frequent foldbacks at modest drive powers. As reported by Flexmeter the Ref Pwr immediately jumps up , then settles down to typical small value after 2-3 seconds.
2. The Tune Ref Pwr  phenomena does NOT occur in SSB (LSB or USB). Other people have report some modes exhibit the behavior and some don't
3. Ref Pwr effect occurs on all bands and changing TX delay from 20ms to as much as 700ms made no change.
4. Ref Pwr effect does NOT occur with barefoot Flex or with amplifier in standby.
5. Ver 1.7.30 of SmartSDR exhibits the same Tune Ref Pwr behavior, of course without foldback.
6. Keying with a manual straight key why monitoring Flex Meter Ref Pwr showed NO EFFECTS and no foldback at full drive power. 

It appears to me that this issue is a Tune phenomena and does not appear in regular CW mode operation. Proper operation in CW mode at full drive needs to be confirmed by other ALS-600.

Timo - OH5KW

There may be DC voltage at amp input. That can be eliminated with suitable coil. 
73, Timo OH5KW

Dave - WB5NHL

My workaround until Flex corrects the issue with Tune in CW is to Tune in SSB regardless of final operating mode. In SSB use Tune for checking the SWR and tuning any antenna tuners. Then switch to CW etc whatever is the operating mode.

Greg N8GD

It's interesting that this anomaly only affects TUNE when the selected mode is CW (or AM/FM from my testing) - operation in CW works FB.  I was thinking of using your workaround for my own situation, Dave.  Hopefully Flex can find the reason for this odd operation.  Fortunately it doesn't prevent operation with v1.8.3.  It's certainly not a reason to downgrade to v1.7.30.  Again, thanks for all your efforts, Dave.

Greg - N8GD

Greg N8GD

I'm still of the opinion that the 35 ohm swamping/matching non-inductive resistor in the input of the ALS-600 might be playing a role in this behavior.  That, coupled with some programming that is making the TUNE function differently amongst the various modes makes this a real head scratcher.

I was thinking that similar problems might occur with the ALS-1300 series amps, but they have completely different inputs, utilizing a power divider to split the incoming RF between the two 600 W modules that are similar to the single module in the ALS-600.  I doubt the ALS-1300 owners are having problems, though.

Greg - N8GD

Terry K7NY

I just checked my ALS-1300 in cw tune set to 47 and flex meter shows reflected as high as 50.87 watts. Then the fold back works, it is funny that it does not do it that high all the time. I have not seen the fold back problem anywhere else.
Terry K7NY

Mark NS9N

I just confirmed that my 6500 and ALS600 tunes perfect in SSB but in AM and CW folds back. But interesting, 1 out of about 20 tries, the RF does not fold back. Switching the amp to bypass and everything is normal. All testing was done into a dummy load. There is clearly a transient response issue and I really can not believe that all the ALS600s have an intermittent 35 ohm input resistor?

km9r.mike

@ Dave wb5nhl...just curious. After getting the system (amp / possible turner) tuned in the ssb mode, will foldback still occur if you now immediately select cw mode and then tune ? Of course there is no reason to retune, but was just curious what resulted.

Greg N8GD

I agree that the 35 ohm resistor is possibly not involved, especially since Terry (above) reports that his ALS-1300 exhibits the foldback issue as well, and its input circuitry is completely different from the ALS-600.  I, too, have found the foldback on my ALS-600 to be intermittent, with no foldback occurring perhaps 10% of the time in CW/TUNE mode.  The input circuitry in both of these amps does incorporate some inductive devices in the form of various transformers that may be involved in the weird reflected power being seen.  My ALS-600 probably has less than 2 hours of total key down time, so degradation of the 35 ohm resistor or physical contacts seems unlikely.

Greg - N8GD

Eric-KE5DTO

Thanks for all the feedback.  We will definitely take this into consideration as we work towards a fix.

Terry K7NY

After more testing I only see the fold back in the CW/TUNE mode not in AM or FM/TUNE.  My old SB-221has Rich Measures QSK mod and I don't see any fold back in any mode.
Terry K7NY

Carl K5HK

Just got back to reading posts and lots of great info.  To Greg/N8GD's comment about the input resistor on the ALS-600, I have 20+ year old unit and have had problems with that resistor before.  I will be checking and cleaning it as a PM item again soon after many years since last checking/cleaning.  It has caused other problems with the amp before so know that clamp in method a problem.

Very interesting something about "Tune" in CW different from ssb somehow - it will be interesting to hear from flex as to just what that might be about.  Anyway oddly enough going back to v1.6.21 cured that problem for me here.

I'll stay tuned.


Mni TKS to all!

73,
Carl / K5HK

W7NGA

You're overthinking ...

W7NGA
Bordeaux, France

Tim - W4TME

This issue will be resolved in the next release of SmarSDR.