开云体育

In searching for documentation for the W8TEE VFO, I came up with this for the
ATMega2560 board:

There are two versions of the USB board (which is the one Jack uses). One is
called the "Meduino" and the other is branded as the English Creoqode board.
The Meduino is most often unbranded. Apparently these are the same board;
both were probably made in China; the English board was branded by them.

The documentation on the boards is scarce. I did not find a schematic, but presumably
they are smaller versions of the ATMega2560 A3 board, so the schematics should be
about the same. There are 2 "datasheets" on the web for these boards. But be
careful -- the sheet on the Epalsite Wiki is about the USB version but the
one on the AliBaba site (and others) is about the "Inhaos" version, which has
no USB interface and relies on the USB/Serial converter with the six pin connector.
It is called the "Mega 2560 Core Datasheet"

One of the things to watch out for are the ICSP connections. There are 2 on the USB
board; only one is noted as ICSP. The one near the USB plug is also ICSP but is for the
ATMega16U2. Unless you know what you are doing, don't even think about that connection
because the 16U2 takes a special firmware which interfaces the USB and the RX0/TX0
of the 2560. Messing the firmware up destroys the board.? If you are brave and understand ICSP
you can use the labeled ICSP connector to interface with the2560.

Another thing I noticed is that the crystal oscillators for the 2560 and the 16U2 are the same.
The one for the 16U2 is for the USB interface and presumably is 12 MHz. So the one for the 2560 is
also presumably 12 MHz. Another frequency is possible, but would not make sense for a USB interface.

The Epalsite wiki is about the best documentation on this particular board. It has a nice picture
of the general pinout of the board (which has been duplicated a lot elsewhere). Also the Arduino
site has some nice pinout and layout pictures for the ATMega2560 board.

Most of the available boards are from Wideem991, either from China or the US. The Creoqode board
is available in the UK, of course, but also from some US sources. Prices range from about $US13.00
to $US20.00. This for the same board. Prices seem to be coming down somewhat since they were
generally higher a few weeks ago.

john
AD5YE

Hi Ashhar.

?

I assume you use the G3UUR test circuit?shown in Experimental Methods in RF design or something similar.

If so, when calculating the values of Cm and Lm, did you actually measure and include the stray capacitance of the crystal holder and the wiring to the crystal switch circuitry?

Were the results significant?or are they so minimal in effect as to be ignored?

?

Many thanks for your time

?

Slim Haines G4IPZ

?

----- Original Message -----

From: Ashhar Farhan

To: [email protected]

Sent: Thursday, June 22, 2017 5:50 PM

Subject: Re: [BITX20] uBITX - A reboot of the old BITX

We have to first characterize the crystal. that's measure their motional inductance and capacitance. i do this using the G3URR's method. it is a very simple two transistor oscillator. after that, i use the same oscillator to measure the frequency of each crystal and sort them into bins.

thanks to the bitx40 production, at least for 12 mhz crystals, i don't have to match them. they are kept prematched in separate pouches. i just dip into one!

- f

?

Virus-free.

Hi all,

Based on the great description from Ashar I have started to build the uBitx from scratch. Even as far as not using a Raduino, but using standard (Ebay) Arduino Nano board, a standard LCD display from the shelf and a locally bought Adafruit SI5351 board. I have changed some aspects of the design. It's a very nice project to change and work onto. One of the planned changes is to include 6 meters. I expect to need a preamp but that is not my first concern. To suit the tuning range to 52 MHz I have changed the IF to 75 MHz and decided to use tuneable coils for that IF section. The second oscillator is now on 87 MHz to bring the second IF back to 12 MHz.? The first oscillator has a tuning range from 90 to 139 MHz to cover all the HF bands including 6 meters.?

Off course the low pass filter at the input has been redesigned to drop off above 52 MHz. I plan to make four band pass filter sections for the transmitter stage, switched by the software. This will include a dedicated 50 MHz band pass filter. I expect to replace the IRF's by VHF types (RD's). The TX part is not ready yet apart from the bidirectional sections. First I want to have the receiver in working order.

Yesterday the receiver side has started to come to life. My first impressions:

The overall gain is not good yet. Signals of 10 uV are just audible (tested around 7 MHz). I think the circuit around Q70 (audio preamp) needs to be adjusted. I use a BC550 and already needed to increase R50 up to 470k, Biasing of this transistor needs to be further adjusted as the collector voltage is currently down to 2V.. Optimizing this will increase the gain of the audio chain which will also increase the receiver sensitivity.

There are still a LOT of digital spurs received all over all bands, so I need to work on further shielding and grounding of supply leads etc. This will probably become a major part of the time spent to optimize the rig. I might need to use shielded wiring for the I2C connections to the SI5351 and the leads to the LCD.

On my LCD display, the + and - connection of the background leds are swapped. Don't know why but it's good to know there are different connected types out there.

I'm already amazed about the crystal filter. After setting the master calibration and the LSB and USB carrier frequencies, tuning on the band sounds quite nice. Space reservation and plans for an additional 5MHz narrow CW filter are already made.

Beware, on the uBitx INO file, the are some mistakes in the comments. For example the description of the A6 and A7 lines of the Arduino are swapped. A0 is not the PTT line.Took me some time to figure this out as my rig was behaving very strange to the tuning pot which was connected to the keyer input..... I do not have a C++ background but have worked in assembler for PIC and ATMEL processors before. It's a nice learning curve to dig into the software and figure things out. This is definitely necessary in order to do my changes.

My housing is small, about the same size in width and depth as the one used in Ashar's youtube video but about half the height. My copper clad PCB is mounted about half the way in height so I have both sides to build on. This is necessary as I also plan to include a 12V battery back from 10 NiMH penlights in the box also.?
Don't know if it's going to work at the end, but hey, that's part of the hobby.

73, Alex
PA1FOX

8-)

Sent from my Verizon, Samsung Galaxy smartphone

-------- Original message --------

From: Dexter N Muir <dexy@...>

Date: 8/29/17 6:18 PM (GMT-07:00)

To: [email protected]

Subject: Re: [BITX20] OT

Dang! Spied and tried "chats" in the sidebar (not obvious, but will have to learn that!) with a reply composed but not sent, and now can't get it back :( Here goes...

I use RSS for alerts (Akregator) and by link to Firefox and website. Not missing functionality or content, can copy and paste at will ... Just that Groups.io doesn't do emoticons (they add expression to communication) or interwork with (the likes of) Messenger for OT Chat (could be a right-click on name?). Haven't tried email client for RSS - that'd mean leaving Thunderbird running where Akregator runs as a daemon.
Still immensely grateful for what we have / get..
73
Dex ZL2DEX

These questions and answers al all very interesting. Now I have mine working but only 2 watts RF output when I shout at it that is. Im making an proved mic amp to get a bit more audio drive as it looks a bit low. If I can get 4 watts with 12 volt FC input I will be happy.
but I think the 510 is a bit soft and that's the main reason for a low output.
kind regards
Paul.(HS0ZLQ G0MIH)

Reducing that front end RF gain could well help.
But Henning is suggesting that there's more than 15 or 20 dB at stake here:
? > ?a dynamic range of 75 dB is possible, due to the audio preamp this range will be only 43.5 dB now!! We are loosing 31.5 dB of dynamic range!

I'm by no means an RF kind of guy. ?
But I like his argument, and I like the idea of a PIN diode attenuator for AGC between the crystal filter and modulator.
That NXP BAP64Q quad PIN diode device is $0.25 in quantity, $0.50 for a single unit from Mouser.
With that AGC in place perhaps we can leave the first RF amp as is, so the rx still works with a poor antenna.?

Jerry


On Mon, Sep 4, 2017 at 01:12 pm, Ashhar Farhan wrote:

The bitx was sketched on a flight. I didnt know muchP about dynamic range. In retrospect, the rf amp severely compromises the dynamic range. Someone should measure the iip3 with and without the rf amp.

I thought of redoing the bitx design for Bitx40 but in the end, decided to stick to the original for it to be called a bitx.

It is a sum of many tradeoffs and lazy choices (like the LM386).

- f