Guess I'll just toss in my 2 cents worth. Converting a Drake to class C in CW mode only by simply changing the bias of the PA may not be a good idea depending upon how it is driven and to a lesser extent how much Q and harmonic filtering is in the tank circuitry. I have not studied the schematic so I'll just leave it at that. But I really don't think class C operation in general should get a bad rap. It's a perfectly legit class of operation and like any mode there is a right way and a wrong way to implement it. It has a major advantage in terms of improving efficiency and thus reduction of heat in the finals. I have measured my Viking II PA with good instrumentation at 79 to 80% over the core bands on multiple occasions. We're talking 144W out for 180W in. The final is parallel 6146's, fixed plus drive-induced bias, and a heavy duty tank with continously optimized Q due to slaving a rollerducter to the plate tuning cap. As for driving power, you generally need a little more with class C, but not a lot more (assuming beam power tubes and assuming we're talking CW, not plate mod AM). To control clicks one should shape the drive signal shoulders softer than what one would normally do when driving a linear amp because the turn-on and turn-off characteristic of class C is more abrupt.
There is nothing fundamentally wrong with class C for CW. In fact some cell phones use it. I think modern rigs avoid it simply because the PA is already linear for other reasons and already has a fan to get rid of heat. The maker would have to design the driver stage a little differently and the antenna coupling circuitry as well. I guess that is not worth adverstising additional 10's of watts of CW output power on the brochure. Dennis AE6C On 1/20/12, Garey Barrell <[email protected]> wrote: > Gary - > > Obviously the bias would remain in the Class AB1 area for SSB and AM. > > Applying a keyed signal to a non-linear amplifier is a bad idea. Every time > the drive is started or > stopped there is the opportunity for additional distortion products to be > generated. Some SSB > distortion is caused by biasing the amplifier too high or too low, resulting > in 'compression' or > limiting the waveform to less than it's peak. More distortion. > > Perhaps we have an amplifier expert on here who can definitively respond to > your question. There > are no doubt MANY reasons why all transmitters switched to linear operation > for CW, in spite of the > 'power/efficiency savings. If nothing else than it takes large amounts of > drive POWER, more > trouble, including greatly increased Grid dissipation. > > To me, turning an input signal on and off to a Class C amplifier is asking > for trouble. A Class C > amplifier is essentially a switch, ON or OFF, and relies on the Q of the > tank circuits to 'smooth > things out'. A lot of potential headaches for a few percent efficiency > increase. > > There are LOTS of 'ideas' on the internet that 'sound' good.... :-) > > Let us know what you find out!! > > 73, Garey - K4OAH > Glen Allen, VA > > Drake 2-B, 2-C/2-NT, 4-A, 4-B, C-Line > and TR-4/C Service Supplement CDs > <www.k4oah.com> > > > Gary Winblad wrote: >> Well yeahh Garey.... >> >> I am not suggesting we change the bias for SSB (or AM), just for CW (and >> real FSK). >> >> So, you would need a switch or relay added and USE it to change modes. >> Yes, >> bias it past cut-off, will result in no emission between code elements, >> but also over >> part of the cycle. >> >> Key clicks could be delt with... The reason they had key clicks was >> primarily because >> they were cathode keying high voltages with no shaping as I recall. no? >> Bias is only >> on the finals, we don't key those or do we? >> >> Junk? Isn't our driver pretty clean? If not, wouldn't we have the same >> spurs and junk >> with a linear final? >> >> We DO have a tuned output tank circuit for the flywheel effect... no? >> >> BUT, if we don't have enough (voltage) drive, that might be a problem, but >> we don't >> really need full power... >> >> Don't ya think? Changing bias for CW wasn't my idea, but it did sound >> good to me. >> TIA, I am trying to learn here... There probably IS some reason nobody >> did this >> originally! >> >> 73, >> Gary >> WB6OGD >> >> >> ----- Original Message ----- >> From: Garey Barrell <[email protected]> >> To: Gary Winblad <[email protected]> >> Cc: drakelist <[email protected]> >> Sent: Fri, 20 Jan 2012 15:31:44 -0000 (UTC) >> Subject: Re: [Drakelist] AC-4 (bias mod) >> >> Gary - >> >> In a word.... NO. >> >> The PA stage is operated as a LINEAR amplifier, Class AB1, meaning the >> output signal is a reasonable >> facsimile of the input signal. The BIAS voltage is carefully chosen to >> 'center' the operation of >> the amplifier in the linear portion of the tube's operating region. >> >> Increasing the BIAS voltage WILL reduce the IDLE current, (between code >> elements,) but will shift >> the PA to a non-linear region. This will result in distortion of the >> signal, with the potential of >> spurious signals, clicks, and other 'junk'. This is one reason clicks were >> so prevalent in the older >> transmitters. A true Class C amplifier requires considerably more driving >> power, and relies on the >> 'flywheel effect' of it's output network to 'smooth' out the 'pulses' of >> drive that exceed the bias >> potential. >> >> 73, Garey - K4OAH >> Glen Allen, VA >> >> Drake 2-B, 2-C/2-NT, 4-A, 4-B, C-Line >> and TR-4/C Service Supplement CDs >> >> >> >> Gary Winblad wrote: >> > All this talk of rebuilding AC-4 supplies had me thinking... >> > >> > I have seen a mod(?) that said to turn up(down?) the bias when running >> > CW >> > (and FSK?) to set the finals more to Class C operation to save power >> > dissipation. >> > >> > Is this a good idea? Wouldn't it be easy to add a switch or relay and >> > another potentiometer while you are in there rebuilding? >> > >> > 73, >> > Gary >> > WB6OGD >> > http://mailman.zerobeat.net/mailman/listinfo/drakelist >> > > _______________________________________________ > Drakelist mailing list > [email protected] > http://mailman.zerobeat.net/mailman/listinfo/drakelist > _______________________________________________ Drakelist mailing list [email protected] http://mailman.zerobeat.net/mailman/listinfo/drakelist
