Ubitx.net provides help for constructors: Preventing catastrophes and providing guides, fixes and modifications for your µBITx. We are an information site only and not associated with HF Signals.
WARNING: As shipped, your µBITx v3 or v4 board will probably not be compliant with emission requirements. It has been found to generate spurs and harmonics. There are now simple fixes available to address these issues. Make sure you apply them prior to operating the transceiver on an antenna.
CAUTION : If you power up the µBITX without the pull-up resistor the µBITX is likely to go into CW transmit. Make sure you install the 4.7k pull up resistor on the CW key pin!
v3 Board? – There are known issues with WX brand TDA2822 chips (U1): Read this. Earlier UCI and later socketed TDA2822 are not affected. You will probably want to do the audio pop mod at least.
v4 Board? – If your audio is distorted see this article. Note that audio output is reduced over the v3 board. If you use a low impedance speaker you may need an additional amplifier module (LM386, TDA2822, etc.) to get adequate volume. The audio pop mod is not required on v4 boards.
v5 Board? – It is too early to say as yet whether this is compliant, but early indications suggest that some boards on some bands may not be fully compliant, but they are a lot closer to compliance than v3 and v4 boards. At 30dB down or more your spurs and harmonics will be in milliwatts!
SHIPPING: µBitx orders were shipping within one or two working days of being ordered, but there seem to be delays with the v5 board currently of around 2 weeks. You should receive a shipping notification through PayPal once your product has been shipped. Estimated delivery times:
EU: IndiaPost: 2-5 days DHL: 2-5 days
US: IndiaPost: 10 days DHL: 2-5 days
Asia: IndiaPost: 2-4 days DHL: 2-4 days
AU/NZ: IndiaPost:10 days DHL: 5-7 days
BITX QSO DAY: Every Sunday – 3PM & 7PM Local Time – 7277 kHz in North America, 7177 kHz elsewhere.
Mark AJ6CU has added an updated version of the KD8CEC Nextion code for the 5″ screens (Basic and Enhanced) to the files sections of GROUP IO BITX20 list. (Only the 5″ Enhanced has been tested, as he doesn’t have a 5″ basic screen to test with. This work is an extension to the “heavy lifting” already done by Joe Puma (KD2NFC).
Mark couldn’t help himself and did some “improvements” to the user interface that you may or may not like. He has included the HMI file in the directory as well, so you can adapt it how you would like it to look if you know how to edit HMI files.
If you have a 7″ or 9″ screens, you should be able t load up the HMI, set the device to match your screen and regenerate the TFT file. If you have one of these larger screens and can’t figure it out, it is suggested you contact Mark.
Files: AJ6CU Nextion 5-inch files/5inch June 10
A summary of major changes that were made to Joe’s original effort”
Correction of scale for histogram (have more space, so let’s use it!)
Addition of direct access to memory STO/LOC screens (had to eliminate the Radio tower logo, sorry)
In case you had not noticed, the button to get “Home” was in various locations, although not done, I am in the process of moving it to a consistent place (upper right)
Scaling changes to Bandscan and Spectrum
Various pages required you to hit “read” or “refresh” to go get the data. I made this happen on page entry. Kept the refresh button as this seems to be a little flakey.
Tried to introduce a consistent color scheme. Blue background, white letters where you can change, yellow for titles.
All screens (except home screen) will eventually have the name of screen in upper right in red.
Button pushes go green when pushed and return to grey when released
Added a few more scrollbar slider sizes so that the slider fit the slide bar better.
Frequency entry (touch the middle 3 numbers on home page) is now a little more obvious. When you are in direct frequency entry, you only see numbers, When you are in Band select, you only see the bands.
Memory to VFO redesigned. Seemed easier to use for me.
VFO to Memory – Mark still doesn’t like this one. He would like to make it more like Memory to VFO where you can see all 10 memory slots and then select the one you want to overwrite. For now, he has just replaced the keyboard with a standard one that pops up when you click in the name and some fancy arrows to spin the mem#.
Although it is against his better judgement, he did make available KD8CEC’s debug screen. Tried to warn you off, but i am sure someone will do something awful to his/her poor ubitx using this screen. 😉
CW setup screen redesign (and it automatically fetches data on startup)
There are some issues remaining … The biggest issue being that when you go into a submenu (e.g. CW setup/MemtoVFO, etc.) the Spectrum Histogram in lower left corner doesn’t always automatically restart.
If you change any control VFO/Freq, mode, etc.) it restarts.
This is going to be a pain to debug because of the extensive use of timers and state machine in the Nextion code…. Would welcome anybody that can make a suggestion here.
I have also not tested it on a Basic 5″. There may be performance issues, especially since the scale of the graphs has increased.
Mark says there is still lots of work, as he also wants to get rid of some of thebackground pictures in the control screens that really don’t match the others. The bandscan is something he would really like to “adjust” because it is really not obvious how it works. And perhaps even a “?” in upper right of some screens for a little help? He plans to put together another update shortly.
Curt WB8YYY has been pleased with the somewhat unusual VK3YE AGC circuit, that uses a LDR and LED pair, as it nicely removes the top of large signals.
VK3YE has suggested measuring current in parallel with the LED that drives the AGC action, but he found this gave little indication of relative signal strength. In fact, it works much better measuring current in series with the LED.
Curt is using a small meter movement supplied by Sunil, about 250 uV peak current. A shunt resistor across the meter is necessary since the LED current at peak is at least 20 mA.
The approximate value of the shunt resistor can be found using the formula Rsh = ( Im x Rm ) / Ish. Rm was unknown but Curt was able to measure it with his DVM at around 500 ohms. Inserting the two known values gives a shunt resistance value of 6.25 ohms.
Curt found a resistor of around 5 ohms and it working nicely.
He says “Its not a real S-meter response – let’s call it a signal strength meter.” It can discern signals from approximately S5 to S9. for signals that do not result in meter movement, the LED itself could be observed to sense signal strength – but the meter represents a nice touch.
A number of constructors have eyed up adding an afterburner to their µBITx. The usual cautions apply: make sure your µBITx has clean output with earlier v3 and v4 boards having been upgraded to remove harmonics and spurs, before even considering adding a power amplifier.
There are a range of cheap Chinese and Russian kits available to give you between 40w and 70w output. However, these may be better avoided, since you can build a WA2EBY power amplifier without a lot of effort. This is a very solid design, and well proven.
Some list members suggested it may be hard to source boards for this amplifier design. However, take a look at:
http://www.golddredgervideo.com/kc0wox/wa2ebyamp/ (pointer to availability of boards) and
Its a good design and allows for getting good performance at higher frequencies. Mine with a little effort does 37W on 10M (1.8W drive) and with the same drive at 40M about 55W. 80 and 20 are about 50. Never though to try 15 or 17m but I’d expect about 44w. The 1.8W is because I use an attenuator at the amp input as most of my HB radios do 4W which is excessive power or the amp. FYI I run it at 28V.
It is a good amp. Mine is now 13 years old and still running the same set of IRF510s. I did use a large heatsink (4×8″ with 1″ fins and the base thickness was .300″). Some call it overkill, but with no fan running and with a brick on the key for 10 minutes, the amp doesn’t fail.
At least one of the “70W DIY AMP” I’ve seen did produce that much power for about 1 minute into a dummy load before it blew up. Failure was likely due to self oscillation or overheating of the supplied heatsink. The heatsink was maybe pentium II vintage with mounting points for
a fan and not at all large or having many fins.
Doug Wilner finally got around to putting KD8CEC’s outstanding firmware on his V3 µBITX. When he is not using his rig for other things, he leaves it on rx on WSPR. He finds using CAT control for band hopping is fantastic.
He did not want power over the USB cable going to the raduino when the rig is otherwise switched off, so he stripped back the insulation on the USB cable, parted the shielding, and like a B movie bomb tech, he cut the red wire.
With a little heat shrink on the cut ends, some copper foil to repair the shielding, and electrical tape to replace the outer jacket it looks good as new…from 100 yards 🙂
This may not be a revolutionary break through but Doug figured it was worth sharing with µBITX constructors.
Please note that there are still some changes that he plans to make to the code, but there are no plans to change the hardware of the MCT.
Al and Jack consider the MCT to be Open Source, subject to the MIT Open Source agreement. (This more-or-less means that you need to leave my headers in the files.)
Note that there are multiple files in the zip file and those files must all appear in the same project directory. There are also some non-standard headers used that are not part of the standard Arduino IDE. The URL for these non-standard headers are given after their #include directives in the MorseTutor.h header file. This is to be expected because the code uses the STM32F103 (“Blue Bill”) microcontroller.
This requires you to install the STM32F patch, and there are plenty of places to find help doing that. I used “Using the Arduino IDE with the STM32F” as my search in DuckDuckGo (I’m done with Google) and found plenty of entries from which to choose, including video and written guides.
Jeff N0KAI shows off what he calls his “fun build”.
Mods to his µBITx include the Nextion 4.3″ screen , a 2.5 MHz high-pass filter for BCI, a Kit Projects AGC board, and fidelity mods to the LM386 amp.
He says that the Uniden CB mic works, but its gain is a bit low, which needs to be fixed. CW, however, works fantastically apparently.
Jeff brought the RF gain control forward to the front panel on a concentric potentiometer with the AF gain control. He really likes that, but direct soldering shielded cable to the SMT potentiometer pads on the board was a recipe for disaster. Within hours, the ground pad lifted even though he was gentle as hell with it.
Jeff suggests that it would be nice to see a future revision of the Kit Projects AGC board include better solder points or even headers for all the wires, as he tacked header pins to the pads for the switch connections. He notes, however, that the AGC circuit itself works very well.
Jeff concludes by saying “This is gonna be a fun QRP rig for Field Day…”