Mark,
First let's define classes of operation:
Class A, plate current flows over 100% of the cycle of the input signal.
Class B, plate current flows over exactly only 50% of the cycle of the
input.
Class C, plate current flows over less than 50% of the cycle of the input.
Class AB1, plate current flows over greater than 50% of the cycle of the
input with no grid current ever drawn.
Class AB2, plate current flows over greater than 50% of the cycle of the
input WITH grid current drawn.
True or (almost) class B amplifiers are only found in push pull audio
amplifiers where one tube amplifies the positive half of the drive signal
and the other tube amplifies the negative half. Even then often the tubes
are biased for a small amount of idle plate current.
Most grounded grid amplifiers are called class B amplifiers but are really
class AB2 amplifiers as they draw grid current and the plate current
conducts greater than 50% of the cycle.
In AB2 the tubes bias is set to allow a certain amount of plate current to
flow with no signal input. Less bias voltage the greater the plate current.
In the case of the 3-500, a pair will idle at between 120 to 230 ma of plate
current depending on the plate voltage and the amount of bias used. At lower
plate voltages in the range of 2000 to 2500 volts these tubes are often run
with zero bias as the idle current is in the desired range that way.
The reason for operating the tubes with some idle plate current is to keep
the tube within its desired plate current load line, where it is the most
linear. At low plate current values the desired straight load line starts to
bend or curve and that curve causes the amplification to go into a non
linear range (distortion).
A true class B amplifier has its bias set so that the plate current is just
zero with no signal applied. When drive is applied plate current flows on
50% of the cycle.
As discussed above, this allows operation in the non linear part of the
plate current curve for part of the drive cycle which will cause some
distortion.
By raising the idle plate current up to some point above the non linear part
of the plate current range the tube operates much cleaner.
As a result of having some idle plate current the tubes are actually
operating in class A at low levels of drive. This keeps distortion very low
in the part of the load line that is otherwise non linear.
Some tubes draw a substantial amount of power at idle. The 3-500 with 2500
volts on the plates if I remember right calls for around 230 ma of idle
current which is around 575 watts for a pair.
Some people operate them at much lower idle current which runs them cooler
but distortion products are greater.
Back to your question of changing bias and class of operation.
The heathkit SB220 changed the plate voltage (CW/SSB) to keep the amp legal
during tune up.
I don't have the voltages/currents in front of me but essentially this is
how it works: Plate voltage of 1800 volts at 550 ma = 1000 watts input.
That is a plate load impedance of around 1925 ohms. E/(IK) where K is 1.7
At 2600 volts and 800 Ma that is 2080 watts input. This plate voltage and
plate current supply the same plate load impedance as the lower plate
voltage and plate current as the above so the amp is still tuned properly at
the higher power level. 2600/(800 x K)
Nowadays you can tune up legally at full power so this type of CW/SSB
function was eliminated on modern amps. But some amps still had/have
"tune/operate" voltages to allow tune up at lower power input as above but
to save wear on the tubes and power supplies
Some of the later amps that have a CW/SSB switch are there for an entirely
different reason.
As discussed above SSB operation requires a very linear amplifier and a
rather high idle current to get that linearity.
CW/FSK does not require the amp be linear.
The switch on those amps change the bias to allow for much lower idle
current, some go to zero idle current.
Lowering the idle current on an amp allows for a lower plate dissipation as
there is little to no plate current flowing between CW characters where as
in SSB with the higher idle current there is a constant amount of plate
dissipation even when you don't talk, until you un-key the PTT of course.
The amp could be biased high enough to get it into class C operation for CW
and efficiency could be increased some. However there are problems with
transient voltages when driving an amp in a mode such as this that can
generate undesirable artifacts (such as key clicks) that can cause
interference.
Even lowering the idle plate current to close to zero as do some CW/SSB
modes, can cause undesirable artifacts.
A note on amplifier efficiency:
If an AB2 amplifier has an efficiency of say 60%, it is only 60% at FULL
OUTPUT POWER when tuned properly for SSB. As input (drive) level is reduced
efficiency falls off. As an example for AM operation amp output level has to
be reduced to 25% of full output power (so that the modulation peaks have
room) that the amp was tuned for. At 25% output level the amp will have an
efficiency of about30% or around one half of full power efficiency.
This is normal amplifier operation.
This also means that with low drive levels in SSB operation where you are
operating under the plate idle current range the amp is near the class A
range and yes efficiency is lower also.
73
Gary K4FMX
> -----Original Message-----
> From: Mark Bitterlich [mailto:markbitterlich@embarqmail.com]
> Sent: Thursday, October 31, 2019 8:30 AM
> To: 'Gary Schafer'; 'Joe Subich, W4TV'; 'Joe'; amps@contesting.com
> Subject: RE: [Amps] SB-220 HV Transformer Wanted ; thread stolen for
> bias question
>
> Hi Gary, thanks for your reply.
>
> Yes, I was around when regulations were based on input power (1968) and
> have
> heard your explanation before but I tend to question it simply because
> the
> Henry 3K and 8K series of amps also came stock with the CW/SSB selection
> and
> either of those two could easily exceed 1000 watts DC input in the CW
> position. FSK and AM modes were also advised to use CW and not SSB
> modes
> possibly with plate dissipation limits in mind as well as legal issues
> of
> input power.
>
> But based on the typical pair of 3-500's say in the 3K original designs,
> in
> CW mode idling current was 90-120 ma, and in the higher plate voltage
> SSB
> position 170-220 ma.
>
> My logic was simply:
>
> Bias controls the class of amplifier.
> A change in bias means the class of amplification must have also
> changed.
> Switching from CW to SSB causes an indicated bias change from 120 ma to
> 220
> ma.
> The impact of a bias change of 120 to 220 ma would lead me to believe
> that
> the class of amplifier shifted slightly in a more linear direction with
> also
> the result of slightly lower efficiency.
>
> Right or wrong or maybe just trivial ... if wrong could you please
> explain
> your logic?
>
> Mark
> Wa3jpy
>
> -----Original Message-----
> From: Gary Schafer [mailto:garyschafer@largeriver.net]
> Sent: Wednesday, October 30, 2019 10:13 PM
> To: 'mark bitterlich'; 'Joe Subich, W4TV'; 'Joe'; amps@contesting.com
> Subject: RE: [Amps] SB-220 HV Transformer Wanted
>
> No. The amp runs AB2 meaning it draws grid current.
>
> The reason for the CW/SSB switch was to allow tune up at 1 KW DC input
> on
> CW.
> This was the regulation at the time the amp was designed. You were not
> allowed to ever exceed 1 KW DC input, even during tune up.
> Switching to SSB after CW tune up allowed higher PEP input by
> maintaining
> the same plate load impedance (where the amp was tuned)for the higher
> power.
> In other words the voltage to current ratio remained the same when the
> higher voltage was selected.
>
> 73
> Gary K4FMX
>
> > -----Original Message-----
> > From: Amps [mailto:amps-bounces@contesting.com] On Behalf Of mark
> > bitterlich
> > Sent: Wednesday, October 30, 2019 4:23 PM
> > To: Joe Subich, W4TV; Joe; amps@contesting.com
> > Subject: Re: [Amps] SB-220 HV Transformer Wanted
> >
> > Ok, this brings up a technical question that pertains to the topic
> > amazingly enough. When you shift from single sideband to CW on this
> amp
> > are you or are you not moving the class of operation from AB1 towards
> > AB2 and thus improving efficiency?MarkWA3JPY
>
>
>
_______________________________________________
Amps mailing list
Amps@contesting.com
http://lists.contesting.com/mailman/listinfo/amps
|