Microphony in your uBITx

Peter G8FWY had a problem with his µBITx.  It picked up sound from touching the case and pushed this through his headphones.

If you have a similar problem it is most likely to be ceramic capacitors in the audio stages of your rig.

In this case it turned out to be C50 in the audio pre amp.  Peter replaced it with a leaded electrolytic capacitor and the problem went away.


Jackal Board for the uBITx released

Jack, W8TEE and Al, AC8GY have announced that QRP Guys is ready to take orders for the JackAl support board for the µBITX. A link to a list of its feature set, a video of it in action, and some photos can be seen at


Note that the AGC implements both audio and IF AGC. The appendix to the manual shows a plot of its characteristics. DSP filters are available, including user-defined filters (CUST in figure below). A CW keyer and decoder are also included. The main screen looks like this:

The decoded CW can be seen along the bottom. (In all honesty, that is a guy sending almost perfect code at about 22wpm.  Jack is still playing with the decoder and expects some really smart user to make it much better. It’s got real possibilities as it is based on the FFT bin counts.)

QRP Guys is offering the JackAl board for $40 for the first 50 boards.

After those are gone, the price will be $50.

Shipping in the CONUS is $5 and DX is $15. This is what the JackAl board you’ll get looks like:

The board has all of the SMD’s mounted (including the Si5351 and the display buffer IC). The user does have to supply additional parts: A Teensy 3.6 and associated audio board (both from PJRC.com) and either a 5″ or 7″ TFT touch screen display (BuyDIsplay.com). You should also consider case size when you select your display size. (Note: BuyDisplay.com has actual sizes for each display and they are measured like a TV, so the case can be smaller than you think.)  If addition, there are some IC’s, connectors, audio isolation xfmrs, and header pins that most of you will want to add. The cost of these additional parts should be less than $15 depending on your junk box. The board with these in place looks like this:

The connectors are not required but, since you are also give about a dozen free I/O pins for experimenting, you may want to add them. (The yellow pins above are test points.)

The assembly manual can also be found on the website.

This is an intermediate to advanced semi-kit project. Our intention is to not only give the µBITX additional capability, but to serve as a platform for experimentation, which we hope you will share with the rest of us. You will have approximately 700K of flash and 200K of SRAM free for your inventive side to exploit.

Al and Jack already have some (software) extensions we want to implement. Note JackAl does NOT fix the harmonic or spur issues. However, our test µBITX’s right out of the box were well within spec for 80 and 40 meters and spot on for 20M (15M and 10M, not so much). If you have doubts about your µBITX, check it out. It could well be that your favorite band is within spec. If you don’t have a spectrum analyzer, check with members of your club or the physics department of your local high school,. junior college, or university. Most will be glad to help.

You need to provide your own Teensy.  The reason for ordering a Teensy without pins is that you will need to use header sockets with long pins on the Teensy so that the Audio shield can plug into the Teensy and the Teensy plug into the JackAl board. The Teensy pins do not have the header sockets and are difficult to change without damaging the Teensy.

Note that the power jumper on the bottom of the Teensy 3.6 also has to be cut.


A chronology of mods

Dan W2DLC ordered his µBITx on December 29, 2017 and received it on March 2, 2018.

It was a version 3 with the WX branded TDA2822 audio chip.

(1) By the time he was ready to assemble it, Ron, W7HD had posted a fix using a voltage regulator.  This was his first mod and he has had good working audio since then.


(2)The second mod he implemented was installing the KD8CEC firmware which was more intuitive for me and solved some issues such as keying responsiveness.


(3) The third mod was to add a short USB cable from the Arduino Nano to allow data access outside the case.

(4) The fourth mod was adding the click-pop suppression kit from Don ND6T, Wayne VA7AT, and kitted by Kees K5BCQ.



(5) The fifth mod was the RF output peaking mod suggested by Howard Fidel WB2VXW



(6) The sixth mod was the ND6T AGC kit that he got from K5BCQ at the same time as the click-pop kit.  When he installed the AGC kit, he used Kees’s installation suggestions for mounting it on the main board.  He also added an RF attenuator/gain control as suggested by ND6T, and hooked a wire from J1 to the purple analog wire on the Raduino to provide an S meter.


(7) While he had the board on the table, R250 was changed to 100K to reduce the sidetone level and C1 cap was replaced with a 1μF to shape the CW note.


(8) Then  the relays were changed out for  Axicoms  as suggested by Mike W0MNE, to reduce harmonics when using SSB on the lower HF bands.


(8) Finally, he replaced L7 and L5 with SMT inductors as suggested by Raj VU2ZAP to reduce spurs on the higher HF bands.


Dan is very happy with how well my uBITX functions at this point.  He has placed it in a small Apache case behind an aluminium faceplate that he fabricated.  The whole thing is pretty rugged and works well with a three cell 18650 battery pack.

Dan says he finally has the spy radio he has always wanted.

Overall Reference 

Antuino under design

Ashhar Farhan VU2ESE and designer of the µBITx has announced that he is working on a new kit – the Antuino.

The PCB version of the Antuino (The Arduino based Antenna Analyzer) is still under development, but we will all be looking forward to its release.


ND6P Volume limited and distortion reducer

ND6P came up with this mod after testing his uBITX on the bench to determine how much signal it could handle before distorting at the speaker due to clipping of the audio signal.  What he found is that RF input signals greater than S9+5dB (-68 dBm) become distorted due to the audio output clipping on the negative side.

What the mod does is insert enough attenuation at the RF input to prevent the audio from clipping for signals stronger than S9+5dBm. Signals S9+5dBm and weaker are not attenuated. He’s tested with RF input as strong as S9+40dBm (-33 dBm) and gets a clean output.

He says he can now listen to a weak signal without having to reach for the volume control when a strong signal comes on. So effectively this is an automatic RF attenuation circuit to provide AGC functionality for the uBITX.

Points connected to the volume control are in parallel with existing wiring.

The pin diodes (MA27B) are available at RF Parts Co.


Having fun with the uBITx

Woody KZ4AK tell us that he is very happy with his µBITx.
He has completed many of the board mods, added RX amp/attn/agc, changed and added to the RX audio chain, added forward and reverse power meter & S meter, VOX, sound card interface, and fan.
Woody says the TX power is still about 6db down from what he expected and he plans to address that later.  He is achieving  about 3 watts out on 80m SSB which is a bit of a challenge, but it works!  On the other hand he is getting near 10 watts on 30M FT8.

Most of the details are in two ZIP files  that can be downloaded.  The uBITX is at the top of the list.


RF preamp

Woody KZ4AK says “My µBITX has always been a little deaf (couldn’t figure out why) so I built a RX front-end addition”.

Woody also wanted AGC so it became an amplifier followed by an attenuator. As with prior AGC / RF Gain mods, it inserted in the RX antenna trace (Ripley, et.al.). As built, it provides a maximum gain of ~ 8 db (+12db from mmic – 4 db from atten) with a 3.5 db NF, which can then be attenuated to more than (about) 50 db down.  Woody has not measured this exactly… but it really wakes up the RX!

Attenuator Control: >5 volts for minimum attenuation, approaching zero volts for maximum attenuation (at 22 to 5 ma, depending on attenuation setting). For manual control would use a potentiometer and voltage follower for current buffer.

Woody made no attempt to minimize the current draw. It is a bit of a pig, consuming 70 ma + control current (22ma).  Also, no attempt was made to minimize the board physical size.

The circuit uses a MAR3 SM MMIC followed by a HSMP-3816 pin diode array attenuator.

If anyone is interested,  he has a to-scale PDF for the board layout, a parts placement image, full resolution schematic, etc.  Woody can ZIP up a package for you …

He is also currently working on an audio derived AGC amp/controller for this thing. Have it sort of working, but he is not happy with the performance yet.