Evening up power output across bands

There has been discussion recently about how to even up the power output of the uBITx across bands.   Maximum output (in excess of 10w PEP) is achieved on 80m, while output on 10m can be as low as 1w PEP.

Method 1:  Preset replaced by Pot on Front/Back Panel

John  G0UCP says “Drive to the finals is controlled by the preset RV1. This could be replaced by a variable pot, perhaps located on the back panel.  However, I think it was Farhan who pointed out a long time ago that the only problem with this is the temptation to keep turning it up! “
[NB:   as RF is carried on RV1, so if you are locating a replacement potentiometer some distance away from it’s location on the main board, it would be advisable to use coax to/from the board to the potentiometer.]

Method 2:  Mike Preamp with Mic Gain Control (works on SSB only)

Jerry Gaffke says “Alternately, build a new mike preamp with audio gain control, though that solution will not affect CW power level.”

Method 3:  Relays to different driver pots by band group

Bill Schmidt K9HZ has relays driven by the TX-A, TX-B and TX-C control lines that connect with 4 different multi-turn pots to set drive level by band grouping (80m/60m, 40m/30m, 20m/15m, 12m/10m.   Details of this straight forward solution can be found here.

Method 4: Vary the Power Supply Voltage to the finals
Marco – KG5PRT suggests,  “It’s kind of crude, but you could power the entire radio with 24v. Use a 12v regulator for the receiver and a LM350 for a variable regulator to the PA. You could set it up to be variable for only a few volts change to vary the output. You alternately could use a 12 v source and the LM350 to vary the voltage to the PA. Again, you could design something that would vary by a few watts.  Running from 24v lets you get the higher output but allows you to turn things down, as needed “
A number of alternative means of lowering voltage were suggested by list members.
Method 5:   Attenuator method with PIN diodes

Karl Heinz – K5KHK introduced another suggestion, “There is an HP application note about PIN diode attenuators.   The device used in this particular configuration reached it’s end of life, and may be hard to get soon. There are alternatives:


The data sheet has a chart that shows insertion loss based on frequency and control voltage.

Jerry KE7ER refers constructors to old threads that are well worth reading that touch on pin diode attenuators, mostly in regard to AGC for the receiver.    https://groups.io/g/BITX20/topic/5945215

Here’s another old thread:

Method 6:   Use a digital attenuator

Carl, K0MWC  ran across this web page  where  a 6-bit, 50 ohm, 1-4000 MHz digital RF attenuator chip from Peregrine Semiconductor is used to vary RF attenuation from 0dB to 31.5dB in 0.5dB steps via an SPI serial interface.   The newer PE4312 chip (the replacement for the obsolete PE4306 used in the design above) goes for less than $5 in small quantities (less than 100 units).

Carl observes, “The PE4312 datasheet has a maximum allowed input RF power that is much lower for HF frequencies, going from roughly 11.5db @ 1MHz to about 23.5dB @ 50MHz (see Figure 4).  The PE4312 also allows control via a parallel interface for those that would rather control it that way, perhaps using an I2C digital I/O expander chip to control the attenuation rather than using SPI to save pins on the Raduino.”


An alternative to using the PE4312 could be to use the HMC470 module (pre-built) ex China for around US$13.    Mike ZL1AXG suggests that when inserted between the pre-driver stage and the bandpass filter the attenuator could be activated on both TX and RX under the control of the Raduino.  It would require 5 i/o lines. 

On TX: the digital attenuator would control drive level to the pre-driver stage, allowing RF output for both Phone and CW on all bands to be near uniform.  Phone and CW output could be controlled for on each band to within 1dB.

On RX it would be able to be used as:

an RF gain control (selectable in the menu and using the encoder to control 32 steps from 0dB to 31dB of attenuation)  AND

as part of an AGC circuit to reduce gain for strong signals.

An S-meter sensor  taken off the input to the volume control could be captured on the A7 analogue line and software could be used to display S level and control the attenuator to reduce gain on stronger signals. The setting of the RF gain control would adjust the “floor” of the S-meter.    Some work would need to be done in calibrating an S meter for typical uBITx sensor readings with the floor varying between s1 (no attenuation) and s6 (maximum attenuation) according to the (virtual) RF gain control setting.  If you want a larger AGC range, you could use two attenuators in series, to give 62dB of AGC (at a cost of 10 digital lines).

Using an i2c display would release 6 digital I/O ports, or an i2c port expander could be used to get the additional I/O lines required.

There are many options to choose from, but Mike likes the ultra-inexpensive CD74H4067 16 channel analogue/digital multiplexer module (US$0.50 or thereabouts on AliExpress). This takes 4 digital i/o ports and a single analogue port, but switching between these multiplexed ports can be done very quickly by turning on/off the four digital ports used to select 1 of 16 ports. Just read or write via the analogue port.

Even after grabbing an analogue port and 5 digital lines for the attenuator from the multiplexer, you are left with 9 digital or analogue I/O ports for future use in the µBTIX.   Analogue out is another bonus feature of this module.

Because the RV1 drive level control (in the uBITx circuit) is in the RF line, it is not recommended to use a DAC variable potentiometer, as this is likely to get RF back into the raduino.    PIN diode attenuators should work, and could be used with a DAC I2C chip or similar.   However, a digital attentuator is an easier arrangement.