Date   

Re: QCX-SSB: SSB with your QCX transceiver

jjpurdum
 

The Arduino IDE can compile code for the: STM32 ("Blue Pill"), the ESP32, and the Teensy 4.0. Indeed Al and I are using all of those in the new Projects book. All that is required is the installation of a software patch and off you go. The patch includes all of the core libraries and the requisite code generator. While there are a few quirks compiling with some of the microcontrollers, we have yet to find code that we couldn't compile within the Arduino IDE.

Jack, W8TEE

On Thursday, April 30, 2020, 11:49:22 AM EDT, Jim Koehler <jark@...> wrote:


I am gob-smacked at how much functionality Guido has been able to pack into the AtMEGA!  I would think the cheap little
'blue pill' using the STM32F103 would be a much more powerful substitute.  It has a very fast A/D, 128K of FLASH and 20K of RAM,
a 72 MHz clock and costs a few bucks.  There's an excellent free C compiler (GNU ARM) and a good, free library (libopencm3).
I can imagine a really small, inexpensive SSB QRP rig with just a few components using his ideas and this controller.
Jim, VE5FP


Re: QCX-SSB: SSB with your QCX transceiver

Jim Koehler
 

I am gob-smacked at how much functionality Guido has been able to pack into the AtMEGA!  I would think the cheap little
'blue pill' using the STM32F103 would be a much more powerful substitute.  It has a very fast A/D, 128K of FLASH and 20K of RAM,
a 72 MHz clock and costs a few bucks.  There's an excellent free C compiler (GNU ARM) and a good, free library (libopencm3).
I can imagine a really small, inexpensive SSB QRP rig with just a few components using his ideas and this controller.
Jim, VE5FP


Re: QCX for QRPp?

Mike Besemer - WM4B
 

FWIW, this tracks very close to my QCX-40.

 

Mike

WM4B

 

From: QRPLabs@groups.io [mailto:QRPLabs@groups.io] On Behalf Of Brent DeWitt
Sent: Wednesday, April 29, 2020 11:01 PM
To: QRPLabs@groups.io
Subject: Re: [QRPLabs] QCX for QRPp?

 

Hi Saku,

With my QCX-40 is a data point, the answer is yes.  I did a quick measurement of my rigs output power (scope method) at half volt steps from 14.5 to 7 volts supply. 0.5 Watts out was at 7.2 VDC supply.  Spreadsheet attached in .xls and .ods (LibreOffice Calc) formats.

Again, this is only the measurement of one example of the QCX and mine seems to run a bit low on output power.  Maybe a 1 turn mistake on inductor winding, but that's fine with me.

For WSPR I built a cheap and simple 30dB attenuator and left my normal supply voltage at 12 VDC (2.5 Watts).

--
Brent DeWitt, AB1LF
Milford, MA


Re: #qcx #problem #20m #qcx #problem #20m

HB9FIH
 

Of course Hans

Without C it's no Filter...

Next will have a equipment for checking.

73

btw in QCX Challenge I had seen you but no copy at all here.

with 40m session there was nothing  ...


--
---
73 de Erich

HB9FIH

HS0ZLS


Re: #qcx #problem #20m #qcx #problem #20m

Hans Summers
 

Hi Erich

Power measurements require a 50-ohm load impedance and a sinewave. If you remove C's and do not replace them - then any power increase is liable to be merely an illusion...

73 Hans G0UPL

On Thu, Apr 30, 2020 at 3:28 PM HB9FIH <erich.rieder@...> wrote:

Yesterday night I removed step for step the parts from LPF.

Each C which I removed a significant increasing of Power.

Unfortunally I have really no goods here for a quality check for the filter characteristics also not for the uH of L's.

I decided to wait until I have them.
---------------------------------------------------------------------------------------------------------------------

I have 2 QCX which runs.

TNX to all ho gave tips, I learned much.

--
---
73 de Erich

HB9FIH

HS0ZLS


Re: #qcx #problem #20m #qcx #problem #20m

HB9FIH
 

Yesterday night I removed step for step the parts from LPF.

Each C which I removed a significant increasing of Power.

Unfortunally I have really no goods here for a quality check for the filter characteristics also not for the uH of L's.

I decided to wait until I have them.
---------------------------------------------------------------------------------------------------------------------

I have 2 QCX which runs.

TNX to all ho gave tips, I learned much.

--
---
73 de Erich

HB9FIH

HS0ZLS


Re: Suggestions for soldering PL-259s in the field

David Griffiths
 

I can recommend M0MAT plugs.  No crimp tool needed and soldering only of the centre conductor.  Very good product.  m0mat.co.uk  and on ebay https://www.ebay.co.uk/str/m0mat/

David
G4DMG


Re: QCX for QRPp?

Saku JP3OUG
 

Dear Tim, Manuel, and Brent

Thank you for your detailed advice.
Well, I don't think that I should remove some of the Finals.
I will try to give 7.0~7.2 Volts to my QCX.

And I will continue looking for a good battery!

73 DE JP3OUG Saku


Re: QCX LPF 60m adjustments

ian liston-smith
 

Thanks for the info guys.

I did measure all the LPF toroids on a peak LCR meter to be on or slightly under the inductance value in the table for 60m. I tweaked the windings a bit with the display module removed, but it didn't make much difference.

QCX inbuilt RF power meter says 4.54 W (but also 2.58 W on receive!) but both 50MHz scope and power meters give about 1 watt RF output on 60m.

Some of the LPF caps get very warm...

73/72.

Ian, G4JQT


From: QRPLabs@groups.io <QRPLabs@groups.io> on behalf of Hans Summers <hans.summers@...>
Sent: 29 April 2020 21:51
To: QRPLabs@groups.io <QRPLabs@groups.io>
Subject: Re: [QRPLabs] QCX LPF adjustments
 
Hi Ian

Yes, you can remove the LCD and power-up, once you are confident everything is working fine. I do this all the time, for making adjustments to the LPF toroid spacing. QCX works fine without the LCD; it only talks at the LCD, it does not check for any answer back. So if the LCD is not present, the firmware doesn't notice or care. 

73 Hans G0UPL


On Wed, Apr 29, 2020 at 11:34 PM ian liston-smith <ian.ls@...> wrote:
Quick question:

Can the display module be removed so the LPF coils can be poked and tweaked for adjustment? Does removing the display module and powering up cause any problems? Will the rig still work so adjustments can be made?

If so it will make spacing/squashing the toroid turns with a cocktail stick for maximum power much easier.

Thanks.

Ian


Re: QCX LPF adjustments

Hans Summers
 

One point to note...

If you plug an LCD into an already powered up QCX, then your LCD will show the dreaded row of black blocks. This is not a fault. This is just because the LCD requires a certain initialization sequence, which is executed by QCX at power-up. So if you plug in an LCD module afterwards, it hasn't had the initialization sequence and it just does nothing. So remember, when re-connecting the LCD module, you need to power-cycle the QCX. 

73 Hans G0UPL
http://qrp-labs.com

On Thu, Apr 30, 2020 at 12:23 AM Skip Davis via groups.io <skipnc9o=yahoo.com@groups.io> wrote:
Yes Ian you can remove the display and make adjustments to the LPF windings. This will not hurt the QCX functionality and when you re-plug in the display all should be good again.

Skip Davis, NC9O




Re: QCX for QRPp?

Brent DeWitt
 

Hi Saku,

With my QCX-40 is a data point, the answer is yes.  I did a quick measurement of my rigs output power (scope method) at half volt steps from 14.5 to 7 volts supply. 0.5 Watts out was at 7.2 VDC supply.  Spreadsheet attached in .xls and .ods (LibreOffice Calc) formats.

Again, this is only the measurement of one example of the QCX and mine seems to run a bit low on output power.  Maybe a 1 turn mistake on inductor winding, but that's fine with me.

For WSPR I built a cheap and simple 30dB attenuator and left my normal supply voltage at 12 VDC (2.5 Watts).

--
Brent DeWitt, AB1LF
Milford, MA


Re: QCX-SSB Help what do I do now? #qcx

Guido PE1NNZ
 

Hi Christiaan,

This is a very good question.

Yes. with class-E we are trying to digitally switch with RF (as non-linear it can be). And.. because SSB (ideally) needs a variable amplitude, the QCX key-shaping circuit was used to accomplish this, which effectively changes the supply-voltage to the class-E PA.

Originally I used the unmodified QCX key-shaping circuit. In this case Q4 is switching with fast PWM signal on-off, but because of C31 some analog voltage occurs and then opens Q6 in some resistive mode: this not efficient at all, Q6 heats-up a lot, and due to these thermal issues it is also difficult to control the envelope reliably due to the varying bias-point. The supply-voltage is also not very linear with the PWM duty-cycle (so this gave the idea to start experimenting with pre-distortion, but the thermal-unreliability was too much of an issue and caused the linear response to drift-away during a QSO).

To fix this, C31 was removed. Now Q6 also switching with a fast PWM signal on-off: very efficient, because it is either nearly 0R or open no current dissipates in Q6, heating and inefficiency problem solved! Another good thing is that the key-shaping voltage is now much more linear (compared to the original circuit), the PWM duty-cycle now very much aligns with the voltage over C32, which helped also in getting a better IMD3 response. This "digital" key-shaping actually work pretty well for generating SSB envelope, also the potential mixing products were well enough attenuated by C32. This was the design used in R1.01d and onwards, until I got a new idea around R1.01x that made the key-shaping circuit redundant:

If the class-E BS170 PA anyway is hard-switched with 0V-5V TTL logic to make class-E, then... why not moving this 5V square-wave a bit up and down with a bias voltage so that the amplitude of the PA can be controlled? At least that was the thought... BS170 FETs start switching on at around 2.7V, so a bias of 0V means that the BS170 see a -2.5V/2.5V square-wave meaning PA is OFF; a bias of 2.5V means that BS170 see a 0V/5V square-wave, so PA is back ON in class-E (as it was in original QCX). Applying a bias voltage in-between (e.g. 1.5V) produce a square-wave of -1V/4V which is basically still switching the BS170 on and off (still in class-E), but not that hard-switched so less RF is produced. 

So, basically the PWM signal that was originally controlling the key-shaping circuit, can (after passing it through a low-pass filter), be directly applied as a bias-voltage to the BS170 gates to control the envelope. The RF signal produced by the TTL logic is also passed to the gates with a decoupling capacitor still switching the BS170s fully on-off to the level the bias-voltage allows the FETs to be closed. This works very well in practice, and you can get even more power out then possible with the original QCX circuit by driving them even harder: (apply a 5V bias; basically switching the BS170s with 2.5V/7.5V). The two theoretical disadvantages of this approach is that it is a less linear (esp. in the lower range) as the digital key-shaping approach, and during the SSB low-envelope transitions the FETs are resistive hence a small portion of the  power is dissipated during these moments, but in practice for normal voice-operation the FETs stay cool and do not see the distortion-effects due to the non-linearity of the bias-voltage/amplitude-envelope. I did not use the pre-distortion corrections yet with this technique, but it might be interesting to see if there are further improvements possible. 

So yes, I think you are right, and I think that the response is relatively linear for the higher-power outputs range. I think the non-linearity in envelope control actually helps implementing what normally a speech-compressor is implementing.

Thanks for the interest in this project, and enjoy the experiments.

73, Guido


On Wed, Apr 29, 2020 at 10:26 PM Christiaan PA3FUN <christiaan@...> wrote:

Hi Guido.

I'm reading my way through the QCX-SSB documentation and came across something I don't understand. One of the design-aspects of the QCX(-SSB) is that the PA is Class-E. Basically meaning a switched "amplifier". As non-linear as it can be. As we know a Class-E linear can be perfectly envelope-modulated by super-imposing the audio/envelope onto the supply voltage of the PA-stage. As such I see you using the original QCX- key-shaping circuit, where Q6 basically acts as a variable resistor that's now modulated by the 32KHz PWM signal. From an efficiency perspective not the most optimal but since we're talking QRP the power-loss is not really a problem.

Now to what I don't understand. In the GITHUB-documentation I read that "the PA MOSFETs can be directly biased by the PWM envelope signal, basically making the key-shaping circuit redundant". Maybe I'm wrong, but as far as I know this can only (properly) work if the output-stage is operating in a linear mode, where there is a linear relationship between input voltage (G) and output voltage. And that's most certainly not the case for a switched/class-E PA-stage.

What am I missing? Help..

And thanks for helping me getting on-board of this inspiring project!

73's Christiaan PA3FUN


Re: 15M QCX? #cat #18m #lpf

DL2ARL
 

Hello group, Hello Hans,

Hans says:
"Anyone who wishes to try it, accepting the experimental nature of the enterprise, please just order a 17m QCX and email me a note, or add a note to the order form, about which band 10, 12 or 15m you really want. "

Yes, I really want. Always dreamed of a 12m qrp rig with it's short antenna, never came to one. But if I start building one, I think of heaving read somewhere in this group something about "duo-banding" the QCX. Would it be feasible to wish a 12 & 15 m rig? Would it be better to wish a 10 & 12 m one? Would the main band be the higher one and the lower the compromised one or vice-versa?
There was a discussion on this topic, but the search function of this group is an intellectual challenge I can not cope with. This is why I ask instead of searching myself.
Yours friendly, Razvan DL2ARL


Re: Fixing "Yet Another QCX 20 with Low Power Output"

Steve in Okinawa
 

FYI Digikey and Mouser both have free international shipping for orders over $40 US.  I know it's frustrating when you just need some small parts, but can enjoy the challenge of creating a nice stock of goodies.


Re: Fixing "Yet Another QCX 20 with Low Power Output"

Shirley Dulcey KE1L
 

Vishay parts are available from many distributors. Hans has chosen to get his from Digi-Key, but getting them from any major distributor (Mouser, Newark, RS, Future, and many more) should also get reliable components. Buying capacitors from secondary sources, especially sources in Asia, is what needs to be done cautiously.


On Wed, Apr 29, 2020 at 11:06 AM Arv Evans <arvid.evans@...> wrote:

Vishay has manufacturing plants in Israel, Asia, Europe and the 
Americas.  Seems that with this geographical diversity there should
be sales outlets in Europe, Asia, and the Americas.  Surely DigiKey 
is not the only retailer of Vishay components?


Arv
_._


On Wed, Apr 29, 2020 at 2:11 AM <LX2KD@...> wrote:
Does anyone have a source in Europe for these apart from DigiKey?  Shipping to Luxembourg is €18, which seems a bit much for 54 cents-worth of capacitors.

In fact, the whole LPF from QRP-Labs including shipping would be only $8.11.  Can QRP-Labs guarantee that a LPF ordered now would contain good caps?

73
Kevin Dorrell
LX2KD / G4AZO


Re: #qcx #problem #20m #qcx #problem #20m

HB9FIH
 

Attachement with 2*40m / 2*20m Voltages / Power checked.

Next start with the LPF
--
---
73 de Erich

HB9FIH

HS0ZLS


Re: QCX LPF adjustments

Skip Davis
 

Yes Ian you can remove the display and make adjustments to the LPF windings. This will not hurt the QCX functionality and when you re-plug in the display all should be good again.

Skip Davis, NC9O


Re: #qcx #problem #20m #qcx #problem #20m

HB9FIH
 

Thanks to all,

So I checked now 3 of the QCX and compared te Voltages.

Especially the 40m which I got assembled direct from QRP-LAbs (when I was in QTH TA3 - and I had no time to assemble)

Following the tip from Allison I checked several times these values - but all my QXC have not the Value in the manual.
In the Table with the Values compared Manual against the QRP-Labs assembled and all mine you see the difference. The QRP-Labs ass. has also much lower V on these pins.
The 2nd 20m will soon also check and fill the  the Values into the table.

But just in the afternoon I had success with the 40m - It runs and is not bad - why. adjusting adjusting adjusting. The audio is low in headphones, but with external audio amplifier its vy ok.
Also no cklick

Now looking for the output power ... next
Here I have a Vertical which has at 7030 SWR 1:1.26 by Z 45.6 and a -j 10.3
With Antenna Power is only 0.09 Watts, on a 50 Ohms Dummy = 3.6 Watts
Direct by Transistors 5.2 Watts also by entry of C29 - BUT after C29 (node C27-L3) only 0.6 Watts !!! Obviously I did not see visually any mistake - so I will unsolder and check.

 

In the attachement the table with the values. (asap I fill these with the 4th QCX (2nd 20m))
--
---
73 de Erich

HB9FIH

HS0ZLS


Re: QCX LPF adjustments

Hans Summers
 

Hi Ian

Yes, you can remove the LCD and power-up, once you are confident everything is working fine. I do this all the time, for making adjustments to the LPF toroid spacing. QCX works fine without the LCD; it only talks at the LCD, it does not check for any answer back. So if the LCD is not present, the firmware doesn't notice or care. 

73 Hans G0UPL


On Wed, Apr 29, 2020 at 11:34 PM ian liston-smith <ian.ls@...> wrote:
Quick question:

Can the display module be removed so the LPF coils can be poked and tweaked for adjustment? Does removing the display module and powering up cause any problems? Will the rig still work so adjustments can be made?

If so it will make spacing/squashing the toroid turns with a cocktail stick for maximum power much easier.

Thanks.

Ian


QCX LPF adjustments

ian liston-smith
 

Quick question:

Can the display module be removed so the LPF coils can be poked and tweaked for adjustment? Does removing the display module and powering up cause any problems? Will the rig still work so adjustments can be made?

If so it will make spacing/squashing the toroid turns with a cocktail stick for maximum power much easier.

Thanks.

Ian

15781 - 15800 of 61122