The wide majority of modulation distortion problems occur in the audio driver stage. By using either a push-pull transformer coupled or two direct coupled cathode/source followers overall distortion is essentially eliminated. We rarely can make enough audio in AB1 class and when the modulator tube grids draw the slightest amount of grid (AB2) current the resulting sudden very low impedance reflected to a single stage driver's distortion is easily heard ! The little push-pull 12AU7 is not the least bothered by the wide changing transition between AB1 and AB2 as we speak along ! Open loop corner frequencies are properly established for extremely stable closed loop operation. I.E. No in-band low frequency motorboating or out-band ultrasonics. JFET phase splitter at bottom of this page. k1deu
A Speech Amplifier, Modulator For up to 300 Watt AM
Published in Electric Radio November 1995 , Number 79
The problem many of our old rigs have with audio can be found in the speech amp and driver section.
The preamp section, usually a 12AX7, should be set up according to the "Radiotron Designer's Handbook" and the RCA tube manual's "Resistance Coupled Amplifiers" charts or the 1963 ARRL Handbook, page 260, "Resistance - Coupled Voltage Amplifier Data".
The driver is usually a single-ended power amp and this can be a major defect. The driver transformer must carry the audio signal and the DC plate current of the driver. When a sufficient amount of audio is needed for 100% modulation the transformer saturates and distortion develops. The use of a push-pull driver circuit and a good driver transformer will eliminate saturation and provide plenty of audio for the modulator grids. The ideal transformer is a multi-tap interstage transformer part number PT20A14, for $21.95. This is an excellent transformer made by Thordarson. ( The Hammond model 124E price $26 or more is also excellent. Use the 135K secondary as the primary (12AU7 Plate to Plate) with the 15 K primary to the modulator grids for stepping down. added 13 Jan 2008).
The PT20A14 is used as a 3 to 1 turns ratio step down driver transformer and with 58 Henrys per plate it has very good frequency response. An 18K 1-watt resistor parallels the full secondary and the full secondary is also paralleled by a 1000 pF silver mica and a 1.8K resistor in series. These load the secondary and help to limit distortion.
This transformer will drive a modulator in class AB1, AB2 or in class B. The secondary current determines the only limit to what the transformer will drive.
The mike input at pin 2 of the 12AX7 can be 2.7M for a plain D-104 or 270K for an amplified mike. At pin 2 use a 470 micro-Henry choke and a 30 to 60 pF bypass cap to stop RF from getting into the audio input. The leads connected to the tube socket should be kept very short so they do not become RF antennas.
A 100K resistor is used to block RF at the audio gain wiper arm, again very short wire length is a must. If the old type shielded hookup wire is still in use in the gain pot circuit it should be replaced with RG-174U. Ground the shield at the tube socket end only, grounding both ends may cause ground loop induced hum.
The first cathode is bypassed with a large electrolytic to increase gain and to reduce hum. To decouple the first plate a 10 MF electrolytic will be large enough if the filter cap in the power supply is 80 to 100 MF or more.
In the audio shaping network, resistor Rh and cap Ch are used to cut the highs 6 db per octave from 4 kHz to 8 kHz and 8 kHz to 16 kHz. The network R1 and C1 is used to cut the lows 6 db per octave 120 Hz to 60 Hz and 60 Hz to 30 Hz, a 6 db cut. The audio shaping network from Radiotron Designer's Handbook is used to assure a stable speech amp, an amplifier that will not oscillate or produce ultrasonics. Oscillations and ultrasonics produce considerable side channel interference and can destroy the modulator section of your transmitter.
The plate and cathode resistors of the 6C4 phase inverter ( see below * for alternate option ) must be matched pairs to prevent an unbalanced signal to the driver tube.
The driver tube cathode resister will be determined by the value of the B+ voltage. The resistor will have to be a value that will keep the plate dissipation less than the maximum value. The driver is a class A amp so the grid bias (voltage across Rk) can be calculated.
Class A bias = .66 Ep/mu (12AU7 mu approx. 14 when the total plate dissipation is 4.5 watts.) The grid bias, determined by the self-bias cathode resistor Rk, would be 16.5 volts (.66 x 350/ 14=16.5V.) The cathode current would be 4.5=(350-16.5)( I, 1=13.5 mA. The resistance of Rk is (16.5V=.0135 x R) or 1222 ohms; use a 1.2K ohm 1 watt resistor.
Feedback voltage is furnished by at least 3 resistors in series. Use at least three to withstand the high voltage from the modulator. Total value of 300K to 900K and a .047 MF @ 1600V cap. It is desirable to have as much feedback as possible. The signal must be in the proper phase or the amp will oscillate. If oscillation does occur then reverse the plate connections of the modulator tubes or increase the value of the feedback resistors. The feedback resistor and the 330-ohm resistor in the cathode form a voltage divider. With three 220K in the feedback line and 330 ohm in the cathode circuit, then about 1/2000 the modulator voltage is introduced in the cathode circuit as negative feed back.
I would suggest that if you are not familiar with feedback circuits you should consult someone that is. Wrong connections could destroy the speech amp and the modulator.
There is no secret to good audio, it is a matter of good design, proper coupling between stages and decoupling between B+ feed points and operating the tubes within their parameters. This circuit is being used in many Rangers, Valiants, Vikings, and DX-100's and can be used in any AM plate modulated rig.
Written by Bill,W1CKI and edited by K1DEU and W1ECO.
John Shaw, K1DEU and Paul Frelick, W1ECO, designed and engineered this circuit, and K1DEU proofed this article w/JFET in his DX-100, While W1ECO proofed this w/JFET in his Ranger . Bill, W1CKI proofed this circuit without JFET in his valiant and wrote and Published it in Electric Radio Nov 1995 , Number 79. Schematic Here K1DEU also cross laminated the little push-pull driver transformer. Requires about 20 minutes !
Hum reduction in tube, Speech Amplifiers Its nice to
have a JFET impedance converter in the base of a D-104 to force the mike input
to a low impedance for less hum pick-up. But
In all transmitters I use new braid RG-58A/U from mike in jack to first audio grid. Also make sure you do not have ripple on any Push to talk parallel line. Next, in All tube transmitters, I use balanced filament wiring to the first speech amp tube usually a AX7 or AU6.
This makes a night and day difference in High impedance mike hum pickup. Do the standard stuff. Twist two equal small insulated conductors together to run from New floating (at tube) filament pins (neither grounded to chassis) to the filament transformer winding where one side is grounded to chassis (fine). No current is required to light one tube so use tiny wire. Make sure RF, filament bypass caps at first speech amp filaments are .01 from each side to ground. Now, we cannot have a ground loop bringing 60 Cycle hum directly up through our grid leak resistor to our control grid ! This great technique came from Peter W1VZR many years ago.
Next to develop proper positive going
150 % modulation peaks to maintain the linearity and Asymmetry of the male
human voice, here is a method I, K1DEU prototyped using my Valiant with 811A
modulators around 1984 and passed to WA1HLR, Tim & WA3WDR, Bacon & many hams in New England on 75 meter AM.
This Higher B+ (bootstrap) for 811a modulators has been implemented in DX-100's
Apache's and many other 100 watt class transmitters. The early version only
added the second, full wave rectifier capacitive input to allow much higher B+
for the modulator tubes, with the lower choke input B+ for the RF Finals. Later
he added the option of using a modulator coil instead of modulation transformer.
* The 6C4 Phase splitter In the space cramped space rigs ( Ranger, DX-100, Apache, etc. ) may be directly replaced ( gate for grid, source for cathode, and drain for plate ) with no component changes with a tiny ( TO-92 ) 1 watt 500 volt N-Channel Depletion Mode MOSFET Triode, a Supertex Inc. Type # LND150N3 <<< specs >>> mounted under the chassis. These MOSFET's are Transconductance devices very identical to their older tube counterfellows. <<< Mouser >>> carries this MOSFET for 51 Cents each. To locate sellers for these MOSFET's go to Google and type in part # and " price* ". K1DEU Note we left this out of the published article to not over excite, old narrow minded farts with a solid state, filament less, JFET Triode.
Parts supply sources are:
Antique Electronic Supply - (602) 8205411 Digi-Key Corp. - (800) 344-4539 Ocean State Electronics - (401) 596-3080 Mouser Electronics - (800) 346-6873
Negative Feedback design in Audio Amplifiers and AM Modulators.
Using an Audio Reactor De Coupling Choke
I, K1DEU strongly recommend adding The ( so called ) 3 diode negative peak limiter by Steve, WA1QIX for Plate modulated Transmitters.
And Especially using the
WA1QIX Modulation Monitor
for prototype/testing and or normal modulation monitoring !!!
07/30/2009 Back to K1DEU Ham Index Page