Could/would class A be a popular option ?

  • 10
  • Idea
  • Updated 2 years ago
Based on the exchanges in the recent thread about output power, I wonder how feasible it would be for FRS to introduce the option to run the 6x00 PA in class A via a firmware modification? If so, I suggest this might satisfy both camps in the power:cleanliness debate.

Select class AB if you want 100 solid watts but be aware that it will not be as clean. (eg when barefoot or on a quiet band)

Select class A if you are happy with (say) 50W maximum but with significantly reduced IMD. (eg on a busy band or when driving a linear)

Yaesu have this on the FTdx5000 and it does what is claimed:

http://www.manualslib.com/manual/3818...

An option like this on the 6x00 would put the onus on the operator to select the appropriate mode for the prevailing conditions and not reflect badly on Flex as some fear a fixed power increase might.
Photo of Paul

Paul

  • 434 Posts
  • 127 Reply Likes
  • hopeful

Posted 2 years ago

  • 10
Photo of G8ZPX

G8ZPX

  • 196 Posts
  • 110 Reply Likes
Yes please, love the idea.

50W may be little on the low side for driving some amps though. If we could get say 65-70W and still be fully class A then that would be a dream job.

....and something for FRS to shout about too !

The class A/B mode could go to say 105W (for 100% output on the scale) to ensure 100W is achievable on all bands.

When you give users real options it makes everyone happy instead of trying to reconcile different operating styles with a single fixed setting.
Photo of Stefano - W2WTZ

Stefano - W2WTZ

  • 82 Posts
  • 20 Reply Likes
For Solid State amps it's not a problem. I drive my Acom600s with 12-13W to remain into 500W legal limit. Nice idea Paul I agree.
Photo of Paul

Paul

  • 434 Posts
  • 127 Reply Likes
Agreed Steve & Stefano. The 50W I mentioned is a ballpark example that would need to be established by Flex. Yaesu cap their nominal 200W/50V PA at 75W in class A. The power is variable by the user up to that limit. Also, the actual bias point is adjustable via the menu to enable the user to tweak how far into class A they go.

As a footnote; this is posted as an "idea" so hopefully it will attract enough votes for Flex to give it consideration. Cheers.
Photo of Chris DL5NAM

Chris DL5NAM

  • 589 Posts
  • 128 Reply Likes
5W for 2K5 :-)
Photo of N4HY

N4HY, Elmer

  • 47 Posts
  • 19 Reply Likes
I think a better approach likely leading to better efficiency is not full class A with its inherent waste of power for being in the on condition 100% of the time to pushing the amplifier towards linearity by pre-distortion. This is already planned.
(Edited)
Photo of Ross - K9COX

Ross - K9COX

  • 335 Posts
  • 101 Reply Likes
The Flex 5000 listed class A as feature to be introduced early on however for whatever reason it never materialized.
Photo of Paul

Paul

  • 434 Posts
  • 127 Reply Likes
Correct me if I'm wrong but doesn't APD compensate for non-linearities in the PA rather than remove them by dynamically adjusting the input characteristics to give a net improvement? How about using both APD & class A for uber cleanliness?

True, class A is less efficient but I don't notice any detrimental effects when using my FTdx5000 in that mode - only benefits. The reduced power limit protects the output pair.
Photo of Bill -VA3WTB

Bill -VA3WTB

  • 2768 Posts
  • 610 Reply Likes
Remeber, as Gerald said, that to do that on the FTdx5000 is very easy to do because it is a 50v system. and the Flex PA does not work that way being 13.8v

Not the same thing at all.
Photo of Paul

Paul

  • 434 Posts
  • 127 Reply Likes
No, it actually IS the same thing. PA biasing principles are independent of supply voltage. A 13.8V PA can definitely work in ckass A. It's just the maximum output power for a given level of IMD is less for 13.8V vs 50V. It would be for Flex to prescribe the maximum permissible output if they did implement class A. It would definitely be less than 100W and maybe a little over 50W given the 13.8V design.
Photo of Mal

Mal

  • 47 Posts
  • 0 Reply Likes
43 watts into my 9500 gives me full legal output, so 50w class A would be fine for me.
Photo of G8ZPX

G8ZPX

  • 196 Posts
  • 110 Reply Likes
I converted an old Heathkit SB-220 to be a dedicated 6m amp. Works superbly but it needs 60-65W on 6m to get full output (2 x 3-500ZG)
Photo of Bill -VA3WTB

Bill -VA3WTB

  • 2769 Posts
  • 610 Reply Likes
Paul, ok that was Ken, whom I respect very much, I see what he is saying. Flex at this time has a very clean signal and pride themselves over it. to give people the option to dirty the TX would not be so good. For me? I would leave as is. It is just what we buy in Flex. 

The only reason the subject came up is that people were suggesting Flex is misleading in there specs and not honest about it. And that the testing Flex does at the factory does not count because it does not reflect real working conditions. but many along with myself get 100W on some bands, so I don't really understand the commotion.
Photo of G8ZPX

G8ZPX

  • 196 Posts
  • 110 Reply Likes
This is not the same thread as the power output. Paul asked if we can get class A by accepting a further reduced output. Lets not kill the idea before FRS has even had  chance to review the possibilities. 
Photo of Paul

Paul

  • 434 Posts
  • 127 Reply Likes
There may have been a commotion in the original post. My idea here is just that, no commotion necessary.
Photo of John AD6NR

John AD6NR

  • 34 Posts
  • 8 Reply Likes
Having a class A option is an excellent idea.   The cleaner the signals we emit the better we all are. 
Photo of Bill -VA3WTB

Bill -VA3WTB

  • 2769 Posts
  • 610 Reply Likes
lets see what Gerald thinks. if they don't have to re design the radio...
(Edited)
Photo of Tim - W4TME

Tim - W4TME, Customer Experience Manager

  • 9148 Posts
  • 3467 Reply Likes
Our stated approach for improving transmitter IM is to use APD.  We are not considering in the near term modifying the radio (hardware or software) for class A operation.
Photo of Ken - NM9P

Ken - NM9P, Elmer

  • 3971 Posts
  • 1226 Reply Likes
Well, that answers my #4 below!  [grin]
Keep up the good work!

Ken - NM9P
Photo of Paul

Paul

  • 434 Posts
  • 127 Reply Likes
Fair enough Tim, thanks for making this clear. Thanks to everyone who "voted" for the idea and followed the post. I'm sure we'll all be interested to see how APD develops.
(Edited)
Photo of Ken - NM9P

Ken - NM9P, Elmer

  • 3971 Posts
  • 1226 Reply Likes
Caveat:  I am NOT an RF engineer, and have not seen the schematics of the 6000 Power deck, so take this as personal conjecture.

1) For those who never need more than 50 Watts and desire the absolute cleanest RF output possible with a 13.8 V Bipolar RF deck, Class A operation might be an intriguing idea.

2) Though almost ANYTHING is technically possible, I do not know if the biasing adjustments could be EASILY accomplished in the 6000 RF Deck. 

The course I took in solid state devices many, many years ago in college told me that biasing was easier in FET and PowerFET devices because it only involved adjusting the voltage on the gate, whereas bias adjustments on Bipolar devices also requires careful consideration of both base voltage and base current in order to get it right and not screw up the input impedance.  I recognize that this may have changed since then as devices and techniques have changed.  If one of the RF Engineers online needs to gently update me, that wouldn't bend my ego. 

3) Another question that would need to be considered is, what the added thermal load would be on the RF Deck?  Increased bias means more standby current, and that means more heat. This might be offset by reduced power drive, etc.  These are questions that I don't know and only the RF design engineers would be able to answer.

4) When APD arrives, will any of this make any significant difference? 

Again, it is an intriguing idea, but there are many unanswered questions before I would jump very hard on the bandwagon.  If it could be done easily, without sidetracking other important innovations, then it could be a win-win option.  But I wouldn't do a major overhaul of the RF deck for it.

Ken - NM9P
Photo of EA4GLI - 8P9EH - Salvador

EA4GLI - 8P9EH - Salvador

  • 1752 Posts
  • 534 Reply Likes
Is ADP possible in the current batches of 6000s without hardware changes? Don't you need to feed the signal after attenuating it, back into the radio? 
Photo of Bill -VA3WTB

Bill -VA3WTB

  • 2768 Posts
  • 610 Reply Likes
The radio is ready to go for ADP, they planed it into the radio from the start. it is just software implementation. Most the connections will be internal. Gerald explained this in detail before,,it we can find the post.
(Edited)
Photo of Tim - W4TME

Tim - W4TME, Customer Experience Manager

  • 9148 Posts
  • 3467 Reply Likes
Salvador - yes.  However if you have an amp, you will need to have a tap on the output of the amp that is fed back into one of the ports on the radio (XVTR, RX-A or RX-B)
Photo of Bill -VA3WTB

Bill -VA3WTB

  • 2768 Posts
  • 610 Reply Likes
Photo of EA4GLI - 8P9EH - Salvador

EA4GLI - 8P9EH - Salvador

  • 1752 Posts
  • 534 Reply Likes
Nice!! Thank you guys!! I am looking forward to that and making some use of the USB ports. :)