Measure the voltage drop across the 1k resistor while transmitting to determine current into the modulator, divide 9v by that current to determine the new resistor value.

开云体育

For those that want to use CW on the air without messing around in Raduino code, this should work: ?Regulated 9v goes to straight key, from straight key to a piezoelectric buzzer for sidetone (other side of buzzer to ground) and also to the red wire side of that pot (other end of pot to ground). ?The middle terminal of the pot is wired through the 1k resistor into the diode/cap junction on the Bitx40 board as previously posted. ?The toggle switch is double pole, one side in parallel with the mike PTT switch, the other side switches in a 100pf cap at C103 when transmitting. ?The cap brings the BFO down in freq a few hundred hertz so a responding signal is not at zero beat as Raj pointed out. ?The cap also brings the BFO down into the passband of the crystal filter, away from the skirt, for more consistent results. ?Should be consistent enough that the pot and 1k resistor could both be replaced with a single fixed resistor of some value that you must determine to give 5 Watts out, that same value should work well in other Bitx40's as well. ?I found that on my rig, a cap at C103 of 50pf or more was sufficient to put the BFO into the filter passband, and 300pf was still in the passband. ?Find a value such that a responding station zero beating your CQ has a received tone you like. ?Me, when I can get back to it in a few weeks, will be taking this scheme into the Raduino with a full iambic keyer, sidetone, RIT, and 2n3906 trapezoidal keying envelope.

As my Wiki Fu is lacking, I would recommend that a Wiki Master add the
following sections to the Wiki:

Main Page:
Raduino - An Si5351A DDS VFO with Display

Raduino Sub Page:
Functional Description(Hardware & Software & Pin Outs(spare pins))
Installing the Arduino IDE
Installing Libraries into the Arduino IDE
Retrieving the Raduino Code(needs to be a "Released" version to work
from)
Retrieving and Installing the Si5351A Library
Compiling and Downloading the Raduino Code
Software Hacks
Tuning Deadband
CW Operation
etc...
Hardware Hacks(probably need to split these between Raduino vs. Bitx
hardware hacks)
CW Operation
Optical Encoder Tuning
Speech Processor
High Power Heat Sinking
AD9850 VFO
etc...
Arduino References
C/C++ Language References
External Hack References(probably need to split these between
Raduino vs. Bitx hardware hacks)
Windows/Linux Quirks
etc...


James

开云体育

Yes. We should have RIT and sidetone for most any CW scheme. ?Easily provided by the Raduino. ?If you want to keep the analog bfo and don't want the carrier on the skirt, then the varactor diode scheme of my original post 20933 would do it. ?A piezo-electric buzzer could provide the sidetone if you want to keep it simple. ?I'll be going with an iambic keyer in the Raduino and BFO from the Si5351, I'm just showing how simple it can be and still send good CW.

A cap, resistor and diode with no raduino code (can be used with the analog vfo) is pretty darn simple and properly shapes the CW envelope. ?Don Cantrell claims no issue with key clicks with something similar to your method, but I have my doubts. ?Keying envelope will become important if we use this for PSK31.

I finished assembling my Bitx40 earlier today and was a little surprised that the Raduino code has USB functionality but that there didn't seem to be any way to enable it.? In the interests of getting something together I could play around with without soldering in another switch I added a quick hack to check if the tuning dial is set to max when powering up and if so go into USB mode.? It's not a great method, but it allows switching between LSB and USB with a twist of the knob and a power cycle.

I made a couple quick changes to operate with the newer version of the Si5351 library and fix a couple compilation warnings.? This should allow building with the newest Arduino IDE and the version of the SI5351 library installed by the built in Library Manager.?

I'm not sure if this will be useful to anyone else or not, but I figured I would throw it out there just in case.? The code is available via GitHub:

On a semi-related note has anyone who has modified the Raduino code to run in USB mode noticed that the frequency displayed seems a couple kHz lower than it should be?? I tried listening in for JT65 tones but had to tune down to 7.074 or so to find them.? Should there be an offset applied when switching sideband modes or is it a coincidence?

Greetings all,

? ?Could I get a sanity check please on my understanding of the board and my VFO?

? ? What I think I understand is that the BFO should be running about 1500 Hz below 12 MHz in order to create a sideband (upper, I think) that'll fit through the nominal 12 MHz crystal filter. ?Then that sideband is mixed with the VFO and filtered to make the 7.1-7.3 MHz LSB.

? ? SO...what should be measurable is a BFO freq of about 11,998,500, a VFO output that ranges between 4.7 and 5.0 MHz, an LCD display that ranges between 7.0 and 7.3, and an output sideband with a center freq between 6998500 and 7298500. ?Ok so far?

? ? Here's what I can measure using the counter built into a Rigol 1042E (hacked to 1102E) and 10x probe:

? ? BFO (at crystal top, L5, and C102): 11.9979 MHz

? ? VFO Display 7.0 MHz VFO output 4.99995 MHz ?RF output 6.99698 MHz

? ? VFO Display 7.3 MHz VFO output 4.69995 MHz ?RF output 7.29695 MHz

? ? (I've also measured the BFO with a piece of wire and my SDR-Play - non-contact - at 11,998,488 MHz. ?The BFO output, when received by the SDR, is also solidly between 4.7-5 MHz.)

? ?I think I've got good audio/BFO getting through the crystal filter and have an intelligible signal getting out and allowing for a solid 7W of output. ?I also think that I have a 1500 Hz mismatch between the actual transmit signal and the VFO display. ?I think I'm transmitting about 1500 Hz too low.

? ?Am I reading this right?

? ? Thanks and 73,

? ? ?Andy KG5RKP

? ??

Never mind, I found another message which explained the diode arrangement and reviewing the schematic and datasheet for K1 realized that there are only 8 pins but the pin count is based on a 16 pin DIP.

OK now.

Thanks.

Might be cleaner to use the brown wire from the center pin of the pot back into the rig using the provided 3 wire harness. ?With L4 out, that node only has some capacitance on it. ?Add the new 1k resistor from that center pot pin node at R94 over to the unused pin of C93 (the nearest in to R94). ?Then a wire from that pin of C93 to your straight key jack on the front panel, and a wire from the straight key jack to the junction of the new diode and cap at C103. ?Note that your straight key is isolated from DC ground. ?A single transistor (such as that 2n3906 circuit mentioned in the previous post) would allow for a grounded key.

I tried the modulator injection scheme proposed in my post ?/g/BITX20/message/20933? ? Turns out that when just leaving the Bitx40 in stock LSB mode, the carrier is positioned far enough up the skirt of the crystal filter that we don't really need to move the BFO. ?So all we have to do is inject a DC current into the top of C107 to unbalance the modulator whenever we wish to send a dot or dash. ?I describe here a very simple mod for clean CW which can be left in place when operating SSB phone. ?(The diode doesn't start to conduct till the forward voltage exceeds 0.6 volts, when using SSB phone the modulation level there is around 0.1 volts.)

I'll assume we have a Bitx40 with Raduino so the analog VFO tuning pot connection into the board is not used, if you are using the analog VFO you can still do this, you just need to find some other way to get regulated power (anything between 5 and 12 volts) into the top of the pot. May not want the key on the same regulated supply as the analog vfo, as that could give you a bit of chirp.

Instructions: ? Need a 10k pot (logarithmic or linear), 1k resistor, 1n4148 diode, 0.1uF cap, toggle switch, and a straight key. ?Also the spare 3 wire analog tuning pot harness that came with the Bitx40.

Variable capacitor C103 is not stuffed, and conveniently close to C107. ?The C103 pin furthest from the board edge is ground, you can verify this by inspecting the bottom of the board. ?The C103 pin closest to C107 is not connected to anything. ?Solder the 0.1uF cap between those two C103 pins. ?Solder the anode end of the 1n4148 diode to the C103 pin closest to C107. ?Solder the cathode (banded) end of the diode to the side of C107 that is nearest the board edge.?

Wire up the pot to the 3 wire harness, black (ground) wire on one side and red (9 volts DC) on the other side, we don't use the brown wire. ?Add a 1k resistor to that center pin of the pot, and from the other end of that resistor run a wire to one terminal of the straight key. ?Attach a wire from the other terminal of the straight key into the junction of the diode and cap that we previously mounted at C103.

Add the toggle switch in parallel with the PTT switch on the microphone.

Put a dummy load and wattmeter on the rig. ?Power up the rig. ?Set the pot for minimum voltage from the middle pin of the pot. ?Flip the toggle switch to transmit, then adjust the pot for 5 Watts out to the dummy load. ?The 0.1uF cap shapes the transmitted waveform nicely to avoid key clicks, and also prevents RF from entering or exiting the Bitx40.?

We could do better. ? I have an iambic keyer working on the Raduino. ?I believe Don Cantrell does as well. ?The Raduino can drive a 2n3906 set up as an integrator, much like the schematic in the top right of figure 6.114 on p6.63 of EMRFD. ?I have this simulated in LTSpice, creating a nice balanced trapezoid shape, should also be suitable for PSK31 transmissions. ?(We can invert the VFO from an Si5351 register to do the PSK31 phase reversal.) ?Running the BFO from channel 0 of the Si5351 allows us to move the carrier out of the skirt and into the center of the crystal filter passband for more repeatable results, allowing the use of fixed resistor values instead of the pot. ?Some brave soul might want to play with moving the final into class C operation for higher power and better efficiency.

Seems to work, though haven't spent any time using it. ?I'll be on the road for a couple weeks. ?Need to leave this parked on the bench for now. ?

Jerry, KE7ER

?

Al,

Finishing my new unit and saw this thread. ? ?I did not see the two diodes (to protect Q13) in the mods or hacks sections. ?Could you explain further how the two diodes are connected and where? ?Are these from pin 1 to ground (my K1 only has 8 pins not 12)??

Thanks,

Gene

N2IJF

Before the fix, was zero volts. ?After the fix, seeing 200mv.