The All-band HF Amateur Radio Transceiver: Just waiting to be modified
Ashhar Farhan VU2ESE has announced on the IOGROUPS BITX20 list that
“HF Signals staff are taking a break over the New Year. The elverines have taken a break and the dwarfs have passed out. Any future orders will be shipped from January 20th onwards. We have also run out of the PCBs, and the new ones will arrive only by the 4th or 5th of January.”
Ashhar Farhan VU2ESE has indicated that there is a bug in the v6 code that prevents easy band switching to the 17m (18 MHz) band.
This is in ubitx_ui.cpp in line 746. It should read “17”, instead it reads “18” in the code as shipped with the first v6 µBITx off the line. This makes the button for 17 meters (18 Mhz) unusable, you have to use fast tune or direct frequency entry to get to 18 MHz.
Ashhar has patched the source code on https://github.com/afarhan/ubitxv6
He reminds those wanting to amend their code for uploading that the v6 source code compiles only when saved inside the sketches folder of the Arduino. This is an Arduino idiosyncrasy.
Anthony VU3JVX has shared details of his ubitx built from scratch. He has dedicated his description to Ashhar Farhan as he says, “Farhan has always motivated me to homebrew stuff.”
Anthony gained his license on March 2017. Passing the exam, and getting on HF bands are still two different things. He began by working on the local VHF repeater with a cheap Chinese Baofeng radio – probably the most economical way to start out as an amateur radio operator. However, he knew that he was missing out on something since I was unable to operate on the HF bands.
Around the time he got his licence, Anthony noticed that the Bitx40 was available as kit from HF Signals. That’s how he made his first HF rig. In joining the BITX20 forum he learned a lot from all the great people caring and sharing information.
He was already thinking of building BITx40 for another band (because these designs are easily modified) and then the µBITx design was announced. This changed everything.
He studied the µBITx circuit and begun collecting all the required components. He even ordered the indicated toroids from the W8DIZ website.
Anthony found it difficult finding sufficient time to put things together on the bench. By profession, he is a computer network & security engineer. However, he lost his job in July. Not sure what to do next, his XYL suggested that he should focus on something he had always wanted to do, but had never had sufficient time to do. That’s how he started on the journey of building the µBITx from scratch.
He started building the receiver segment first. He had his challenges during this time and at one point thought he had made a bad choice in deciding to build the rig the hard way. He wondered if he should have tried the bitx40 circuit first.
He was almost on the verge of packing up and shelving the project because he couldn’t hear anything from the receiver. However, h took a break for few days and then started troubleshooting each section of the receiver in sequence. Finally he found after reading through the BITX20 forum that the Q70 was likely to be defective. He decided to replace it with an audio type transistor 2SC945.
In this process Anthony read through most of the content of the BITX20 forum!
Anthony had all the version schematics but started out to build the V5 as he wanted to include the best new features. He kept the build approach modular and laid the modules out almost like the schematic for easy troubleshooting.
The next challenge was trying to be a good student and follow everything as suggested by Farhan. He made up the 11.059 Mhz crystal filter and after changing the X7 to 11.059 Mhz he was able to see the radio signal making it through the crystal filter. Then came the hurdle of fine tuning the USB and LSB and fiddling with the software for the right value.
After the receiver started working he took a break and started enjoying the receiver and checking out all the amateur bands.
He quickly moved on to building the ubitx transmitter. He was quite confident that he could get the transmitter built by now and thought it should be straight forward. He was so wrong. Following all the recommendations on the forum about harmonic issues and how to avoid them, he started building Band Pass Filters. Then he started work on the PA section He succesfully adjusted the bias current on the IRF510s. He calls this his “Bell the cat moment” since he didn’t get to blow either IRF510. However, the output power on 40 M was barely 2-3 Watts and on higher bands was less than 1 Watt.
He suspected something was wrong with those MOSFETs, but out of circuit tests on the MOSFETs suggested they were working normally. He had some other units from a different source, but these performed exactly the same.
Anthony played around with different PA transformer settings and rechecked and traced the RF signal. Everything looked normal until the IRF510 linear amplifier stage. The chances of receiving fake IRF510 is generally low, as this is not an expensive RF part like the RD16HHF1.
Anthony happened to have some new RD16HHF1 parts. He was not very hopeful thinking that they may well prove to be fake since they were purchased online. However he gave these a try. The pin layout was easy, given the design approach (see the photos).
He was surprised to find he was getting a whopping 10+ Watt output on 40M, 5W on 20, 10W on 17M, 10W on 15M, and 5W on 12M and 10M. It appears there are a lot of fake IRF510 on the internet!
Now it was time to test the homebrew rig on air. He checked in that evening to the All India Net and got a 59 report. The net controller assumed he was using a commercial rig. Finally, despite difficult times in his personal life he was able to smile and sleep well that day.
The icing on the cake will come with installing the Nextion 3.5 display.
Does that mean everything was 100% with his home brew ubitx ? No! He is still trying to figure out how to fix a feedback issue from the speaker during transmit. He found this seemed to be audio leaking from the emitter of Q6. He has still not solved the audio leak on TX problem.
Anthony has tried out all the audio circuits used in the µBITx. He finally settled for the TDA2822 circuit [presumably with the authentic part!].
Anthony is happy to share any information related to his build or software settings or tuning. He is very familiar with programming in C++.
Neil, W2NDG has received his v6 uBITx and has it up and running. He has yet to have a QSO, but has some observations and questions:
It is not difficult to add a 2.8″ TFT display to your existing µBITx. Make sure you order a 2.8″ ILI9341 touch panel display. These are readily available on eBay and AliExpress for under US$10, which is half the cost or less than a Nextion display. This display, however, has its downsides … more on that later.
Wiring changes required
Ashhar Farhan VU2ESE has clarified that on the new V6 mainboard, the connector to the Raduino has two extra pins. These don’t disturb the original configuration of the original connector, they bring the Keyer pin and the PTT from the Raduino to the main board to be routed to the keyer and the mic connectors. You could still plug in an old 16×2 LCD raduino into the new v6 main board and it will work all the same.
Pictures will be posted shortly.
It is recommended that you add a heatsink to the regulator on your Raduino board, as feeding the ILI5341 will result in additional current draw over the 1602 standard display.
Installing the new firmware
The factory firmware can be downloaded here.
Note that if you have a v3 or v4 board, the firmware will require a modification to change the IF oscillator frequency. This is because the v5 and v6 boards use a different IF filtering frequency (12MHz). Only one line of code will need changing. Which line? Anybody?
Upsides and downsides of this display
The upsides of the ILI 9341 display is that it is very affordable and supported by factory uBITx firmware.
The downside of this display, however, as can be seen on the demonstration video from HF Signals, is that it is very slow in operation. This is the result of the slow Arduino Nano processor having to communicate via serial port with the processor-less ILI9341 display.
The Nextion display mod from Ian Lee KD8CEC results in a much more responsive and attractive display, updating very quickly and being able to display S-meter readings and has other features not found on the v6. However, it does come at a greater cost.
Ashhar Farhan VU2ESE has announced today the release of the v6 uBITx transceiver. The circuit is exactly the same on the main board. However, there will be no soldering required as all of the connectors and related componentry either plugs in or has been built in to the board. It is available in two formats:
Here is what the second option looks like after assembly:
And of course, you can buy it on hfsignals.com. Shipping starts onTuesday with a limited supply of the first 200 boards prior to Christmas.
The most important thing about this revision is that the radio circuitry is almost unchanged. HF Signals incorporated connectors on the PCBs so this kit needs no soldering. You snap in the TFT Raduino onto the main board, plug the power and antenna into the back, plug in headphones, plug in the mic (supplied with the kit), and off you go!
For those using the 16×2 display at present with a v5 kit (or perhaps earlier kits with some software modifications) and who would like to upgrade, you will have to do three things:
Ashhar has been hacking away on the task of adding a TFT display for the Arduino for some time. He uses the inexpensive 2.8 inch TFT display based on the ILI9341 controller. The display updates slowly, but the display is very useable. The TFT display uses the same pins that earlier connected to the 16×2 LCD display. This display is available everywhere for a few dollars.
All the cables and connectors of v5 will work with the new kit. The only thing you need to do is to swap the 16×2 LCD for a TFT display.
The cases are only available as a part of the kit. Much of the extra cost relates to increased freight costs for the case.
Buy the v6 uBITx from the HF Signals home page: http://www.hfsignals.com/
There are the two pricing options shown each taking you to their own order page.
UBITX V6 – BUY THE BASIC KIT
http://www.hfsignals.com/index.php/ubitx-v6-buy-the-basic-kit/
UBITX V6 – BUY THE FULL KIT
http://www.hfsignals.com/index.php/ubitx-v6-buy-the-full-kit/
Tom N8TPN notes that the New PCB Layout offers some obvious and some not-so-obvious advantages.
The integration of the connectors onto the Main PCB greatly speeds and simplifies assembly and hookup for builders – hopefully fewer wiring errors and requests for assembly/troubleshooting help.
The layout has some extra space at the rear of the chassis for larger heatsinks – excellent for the digital mode crowd.
The inclusion of the USB extension cable in the full package means easier firmware upgrades.
The integrated keyer and paddle jack are a nice feature – no need to add that messy pull-up resistor…hopefully there is some way to de-activate it to allow external contest-type keyer inputs.
The included fully-wired microphone is very convenient – The choice of standard 1/8″ (6.35mm) stereo connector will allow the user to easily switch to digital-mode inputs. Just swap the cable.
The custom case in the full version, and front panel templates in the base version, provides for mechanically stronger and more secure assemblies with less chance of failure over time due to the movement of connections in the PCB headers and Molex (Relimate) connectors. The simple control layout and available unused front panel space should allow adequate customization of controls such as AGC and separate digital inputs.
It wasn’t clear in the video if the menu functions can be accessed from the front panel tuning encoder knob as in previous versions.
Talmadge Huckabee suggests that a good source for a Nextion noise reducer choke (2.5mh) is to get a toroid (FT114_43) from kitsandparts.com and wind 70 turns of #22 wire on it. This has about 1 ohm of resistance so it will not have much voltage drop (and will not dim the brightness of the Nextion display).
Joel KD6AGW has been working on his UBiTx V5 for a few months now and added the KD8CEC update with Nextion screen and the I2C update. He had all the features working (CW decode etc) except the SWR meter.
Joel also built a Kits and Parts SWR bridge and hooked it up to the second Arduino. However, the meter never operated correctly. The home brew Stockton bridge that Ian (CEC) used apparently did not have this problem.
After hooking up an oscilloscope to the forward and reverse outputs, Joel noticed that the reverse voltage would bounce up to 4 volts or so before settling down after a few seconds. This was while the uBITx RX input was hooked to a 50 ohm dummy load. When the probe was hooked up, the meter on the Nextion display would go high, and then settle down to a good swr reading. However, as soon as he removed the probe to the oscilloscope, the meter would just stay pegged at very high SWR.
He ran across a blog item here (http://www.kk5jy.net/swr-meter-v1/) about building a digital SWR meter and noted that:
“In order to prevent stray RF from getting into the A/D circuits, a 0.05uF ceramic disc capacitor was placed between each line and the GND pin. Since the A/D inputs are very high-impedance, and the SWR sensor outputs were meant to drive current devices (e.g., analog meters), a 75k resistor was placed across each of the 0.05uF caps, to drain the accumulated charge when no RF was applied”.
After doing these modifications, Joel has found the SWR meter to work great.
For information, here is a sketch of the circuit modification:
Following a question about power wastage in the Stockton Bridge SWR circuit Joel checked the output with the bridge in and out. There was no perceptible difference in power out on the oscilloscope.