[Amps] LDMOS availability

[Jim Thomson]

Date: Sat, 10 Jun 2017 05:51:41 -0400
From: "Roger (K8RI)" <k8ri at rogerhalstead.com>
To: amps at contesting.com
Subject: Re: [Amps] LDMOS availability

<Like Bill and Jim(s) I disagree.  You really aren't running class C all 
<the time. You are varying the bias from class to change from class C to 
AB 1 or 2 either continuously or in steps. I question the effect this 
would have on the actual linearity.  With the bias / class changes, the 
drive required changes and the gain changes.

Even with normal amps, is the output a faithful reproduction of the 
input? With the current state of the art for ham gear, I doubt it even 
in class AB1. Many of the signals I hear on the bands went through some 
stage that wasn't linear. Dynamic predistortion we have finally 
surpassed the signal quality of the old tube rigs. HOWEVER, as I've 
asked before,  how much deviation can we accept in the output from being 
a faithful reproduction of the input before we no longer see it as being 
linear.

73, Roger (K8RI)

###  The concept of sliding bias  has been around in the audio world for quite
some time. Krell  uses it to slide their audio amps  from Class A....down to Class AB.
That way, their  Class A amps are not running like blast furnaces, with low drive levels,or 
during pauses etc.   Audio amps are not like RF amps, where on ham  RF amps, the idle is cut off
on RX.   Audio amps are running Class A  all the time, as soon as you turn it on. 

##  My Denon stereo amp uses sliding bias, but in the opposite fashion.  With real low signals, it
runs Class A....  up to aprx 5-7 watts.   Any more, and it slides into Class AB, where its more eff, and
less heat... like  from 8-120 watts. 

##  Yaesu shoulda done that on their   Class  A  xcvrs, when in Class A mode.   IE:  Slide it into class AB,
between words, pauses  etc, so its not sitting there...sucking 10A  @  30 vdc =  300 watts.   Key the ptt,
say nothing, and you end up with  300 watts being dissipated.   Talk..and its now 75 watts out..and  225 watts diss. 
They coulda dumped a lot of heat using a simple sliding bias scheme.   On the yaesu 5000 + 9000, that front panel
adjustable bias control only works while in Class A mode.    Crank it on way and its pure class A.   Crank it the other
way, and its adjustable from Class A....down to class AB.     And either way, its still only 75 watts out.   You have to switch
to Class  AB, to get the full 200 watts out.   That was yaesus method to reduce heat while in Class A mode.   Which 
sorta defeats the Class A concept. 

##  I have  not done this ..yet, but I will try either reducing the idle current, when in Class A mode.... or increasing the idle when in Class AB
mode.   Im certain that a compromise can be made.   Class A is over kill...while  AB  is  less than optimum.   I have 8 x bias adjustments
in total on my yaesu MK-V.   4 for the driver, and 4 for the final.      2 are for driver, 2 for final..for Class  A,  then repeat for AB. 

##  right now its sucks 2A  in AB...and  10A  in A.   I will experiment with  say 3-5 A  in class AB.   The idea being   a compromise between
oem A  +  AB.     The mating Driver bias  will be tweaked in the same ratios. 

##  when listening on a 2nd MK-V...3 ft away....and off freq, the difference in splatter between  A + AB is astounding...and thats  running 75 w pep
out in either mode.   The MK-V  in AB, when run at 75 w , is a lot cleaner vs  200w  in AB. 

##  Y/ I /K  needs to introduce xcvrs with the pre-distortion built in...without the 20 msec latency.  That 20 msec latency on the ANAN 8000  would
drive me nuts.... no way to monitor yourself when using headphones.... which kills it for me right then and there. 

later... Jim   VE7RF