Date   

Re: Simple firmware instructions for USBASP #qcx #firmware

Ted 2E0THH
 

Thanks Alan

My method uses Extreme Burner and certainly has the ability to change fuse settings for those US supplied chips as you describe
I will have a read about it.

73s Ted
2E0THH.


Re: Simple firmware instructions for USBASP #qcx #firmware

Alan G4ZFQ
 

My method ignores anything to do with fuses and I have never had to bother with them.
I have used this upgrade method since t1.00e.
Ted,

Which is why I warned Kees.
Your PDF does not mention fuses but it is important especially as Hans discovered the American distributor was supplying chips with the wrong settings.

For others fuses probably do not need to be checked but this group is international.

73 Alan G4ZFQ


Re: Simple firmware instructions for USBASP #qcx #firmware

Ted 2E0THH
 

Hi Alan

My QCX is serial number 4902 and was purchased 2 years ago.
My method ignores anything to do with fuses and I have never had to bother with them.  
I have used this upgrade method since t1.00e.

Hope this helps

73s Ted
2E0THH


Re: T1 Winding/Soldering question

Paul Harrison
 

Looks fine from here. 30 Sec/3 Prim. is correct for 20m.

Paul DJ0CU.



Re: WSPR data analysis #qcx40 #wspr

Ted 2E0THH
 

Hi Hans

That would be terrific, thank you! I was just pondering where to start this weekend.

I have kept the same rig power/antenna setup for a few months now and I just want to compare my WSPR data against other propagation data available in the vain hope that I will begin to understand this particular branch of science a little more.

Thanks to all that have offered advice on this issue.

73s Ted
2E0THH


Re: Simple firmware instructions for USBASP #qcx #firmware

Alan G4ZFQ
 

I have attached a pdf in the hope that others might find them useful.
Ted, Kees,

Especially for anyone who may have had their QCX from the USA stockist before 27 May 2020 the fuse settings must be checked and changed if necessary.

https://sites.google.com/site/g4zfqradio/qrplabs_program_chip_with_USBasp

explains and also shows Avrdudess which along with Avrdude seems to be used by many.
(There are other programming softwares, Khazama is one dedicated to the USBasp.)

73 Alan G4ZFQ


Simple firmware instructions for USBASP #qcx #firmware

Ted 2E0THH
 

Some while ago for my own benefit, I wrote up some simple firmware upgrade instructions. 
The method uses the USBASP AVR which cost me £2 from ebay.

I upgraded to 1.05 last night and it took me less than 5 minutes.

I have attached a pdf in the hope that others might find them useful.
The are links to everything you need, including the QCX firmware.

73s Ted
2E0THH


Re: T1 Winding/Soldering question

 

Unless, of course, I misunderstood you, and the attached picture isn't yours, but from the manual... ;-)

Use my pic as a reference, though, for how it should look - in terms of turns, anyway.

Julian N4JO.


On 6/11/2020 11:44 PM, Julian Opificius wrote:

Jon,

First, welcome aboard!

The mantra for toroids is "every time the wire goes through the hole, that counts as one turn".

The holes in the circuit board are strtegically placed around either the inside or the outside edge of the toroid, so that for any winding the holes for the beginning and end are on the alternate concentric circle. If the hole isn't right where the wire is - or at least on the inside or outside ring as the wire goes down to the board,  you have probably not wound correctly.

I see that you're using the "modified way" of winding the coils, suggested for the 60m and 80m versions. When you cut the loop and fit the outer wire from each pair (i.e. the one coming UP through the middle and going DOWN the outside) you must push the end that comes out from under the outer edge back through to the inside and into the hole on the inner diameter of the toroid. Personally I don't find that method easier, and I suspect the manufacturer - Hans - didn't either, which is why it is only proposed for the 60 and 80m versions where there is so much wire that the coils overlap each other.

Worst of all though - and you're probably not going to like this - is that you appear to have wound 3 turns for the primary and S1, while the 40m version wants 5 turns :-(

Picture of my 40m T1 attached (I hope).


Julian N4JO.

On 6/11/2020 10:41 PM, kc3pds@... wrote:
Hi Everyone,

First I just wanted to say hello, I just recently got into amateur radio and have been very fascinated by all things HF and QRP.  I'm still learning CW, but I've successfully put together smaller/easier kits recently, made a decent (working) half wave end fed antenna, and and finally decided to try my hand at the 40m QCX kit. From the instructions for T1, I didn't have too much trouble winding it (other than snapping the wire 1/3 of the way through and barely having enough after unwinding and restarting), but after I mounted it to the board I'm second guessing the method I used.

The wires for holes 4, 6, and 8 come out from the bottom of the toroid, and I'm certain everything lines up correctly, but from the attached picture it looks like they are inserting the wire into hole 8 by passing the wire up, over, and through the center of the toroid, which I did (and did for all other "inside" holes), but now after having soldered and trimmed all of the connections I'm wondering if this was a mistake.  Did I just inadvertently add a 6th winding to 6, 4, and 8, or was this how it was intended? Was I supposed to squeeze the wire underneath the toroid to avoid this winding? If I did do this incorrectly, I'd rather know now before I get too much more on the board.

Thanks!!
Jon
KC3PDS


Re: T1 Winding/Soldering question

 

Jon,

First, welcome aboard!

The mantra for toroids is "every time the wire goes through the hole, that counts as one turn".

The holes in the circuit board are strtegically placed around either the inside or the outside edge of the toroid, so that for any winding the holes for the beginning and end are on the alternate concentric circle. If the hole isn't right where the wire is - or at least on the inside or outside ring as the wire goes down to the board,  you have probably not wound correctly.

I see that you're using the "modified way" of winding the coils, suggested for the 60m and 80m versions. When you cut the loop and fit the outer wire from each pair (i.e. the one coming UP through the middle and going DOWN the outside) you must push the end that comes out from under the outer edge back through to the inside and into the hole on the inner diameter of the toroid. Personally I don't find that method easier, and I suspect the manufacturer - Hans - didn't either, which is why it is only proposed for the 60 and 80m versions where there is so much wire that the coils overlap each other.

Worst of all though - and you're probably not going to like this - is that you appear to have wound 3 turns for the primary and S1, while the 40m version wants 5 turns :-(

Picture of my 40m T1 attached (I hope).


Julian N4JO.

On 6/11/2020 10:41 PM, kc3pds@... wrote:
Hi Everyone,

First I just wanted to say hello, I just recently got into amateur radio and have been very fascinated by all things HF and QRP.  I'm still learning CW, but I've successfully put together smaller/easier kits recently, made a decent (working) half wave end fed antenna, and and finally decided to try my hand at the 40m QCX kit. From the instructions for T1, I didn't have too much trouble winding it (other than snapping the wire 1/3 of the way through and barely having enough after unwinding and restarting), but after I mounted it to the board I'm second guessing the method I used.

The wires for holes 4, 6, and 8 come out from the bottom of the toroid, and I'm certain everything lines up correctly, but from the attached picture it looks like they are inserting the wire into hole 8 by passing the wire up, over, and through the center of the toroid, which I did (and did for all other "inside" holes), but now after having soldered and trimmed all of the connections I'm wondering if this was a mistake.  Did I just inadvertently add a 6th winding to 6, 4, and 8, or was this how it was intended? Was I supposed to squeeze the wire underneath the toroid to avoid this winding? If I did do this incorrectly, I'd rather know now before I get too much more on the board.

Thanks!!
Jon
KC3PDS


T1 Winding/Soldering question

Jon Karve
 

Hi Everyone,

First I just wanted to say hello, I just recently got into amateur radio and have been very fascinated by all things HF and QRP.  I'm still learning CW, but I've successfully put together smaller/easier kits recently, made a decent (working) half wave end fed antenna, and and finally decided to try my hand at the 40m QCX kit. From the instructions for T1, I didn't have too much trouble winding it (other than snapping the wire 1/3 of the way through and barely having enough after unwinding and restarting), but after I mounted it to the board I'm second guessing the method I used.

The wires for holes 4, 6, and 8 come out from the bottom of the toroid, and I'm certain everything lines up correctly, but from the attached picture it looks like they are inserting the wire into hole 8 by passing the wire up, over, and through the center of the toroid, which I did (and did for all other "inside" holes), but now after having soldered and trimmed all of the connections I'm wondering if this was a mistake.  Did I just inadvertently add a 6th winding to 6, 4, and 8, or was this how it was intended? Was I supposed to squeeze the wire underneath the toroid to avoid this winding? If I did do this incorrectly, I'd rather know now before I get too much more on the board.

Thanks!!
Jon
KC3PDS


Re: Upgrading firmware on the QCX to Ver 1.05 ....where is the .hex file

Kees T
 

Finally have it (ATMEGA328P firmware reloader) working well. The two problems I had were:
 
1) getting the correct USBtiny ISP driver installed for this board (found it on Adafruit) and
2) finding out, with this particular programmer, you have to program the ATMEGA328P standalone, not in the circuit.
    In circuit gives an error, out of circuit does not. Anyway, Ver 1.05 loaded and running.

You also have to use the "-F" command on the AVRDUDESS because there is a slight mismatch between ATMEGA328P IDs .....then it tells you that it assumed it's a ATMEGA328P and loads/verifys the firmware.
It will not detect the ATMEGA328P automatically due to the ID difference.

73 Kees K5BCQ


Re: Tx->Rx click once again (QCX)

Wolfgang DK4RW
 

Dan,

Good point. Indeed, component tolerances in the front end could cause differences between different QCX. The spikes originating from the RX front end are in both the I and Q channel. Depending on the I-Q balance setting (R27) the spikes may or may not cancel.

In my other QCX, where I just changed C20 and C21, clicks are almost gone. The keying just “sounds harsher” (is this the right wording?) compared to the QCX with the new modifications.
Btw, both QCX are on firmware 1.03

73 Wolfgang DK4RW


Re: Si5351: P1, P2, P3, what's it really all about?

Milt
 

Make that P3 and P2 are 20 bit and P1 is 18 bit.

Milt

On Jun 11, 2020, at 3:13 PM, Milt <w8nue@...> wrote:

Is there a problem with the fact that P1 and P2 must fit in 20 bit registers and P3 in an 18 bit register?

Milt W8NUE

On Jun 10, 2020, at 4:44 AM, Guido PE1NNZ <threeme3@...> wrote:


Sorry Phil,

I misread your question, so my answer does not make sense, and please reject my previous email.
It is a good question what happens for value in between 128 for P2, no idea...never tried.

On Wed, Jun 10, 2020 at 11:30 AM Guido PE1NNZ via groups.io <threeme3=gmail.com@groups.io> wrote:
Hi Phil,
This is mathematically valid, does it work? Is there anything in the datasheet or application notes which says we can't do this? Is there any effect on the jitter or other performance?
My understanding is that this allowed and works. as it meets the criteria that "a+b/c has a valid range of 15 + 0/1,048,575 and 90"  [AN619 p.3 par. 3.2].
A zero in b means that there is no fractional part involved, and improves the jitter performance, when you set FBA_INT/FBB_INT [AN619 p.4 par. 3.2.1].

BUT!! then do not start using fractional divisions on the Multi-synth dividers as the jitter is worse compared to fractional dividers in the PLL Multisynth.


Re: Si5351: P1, P2, P3, what's it really all about?

Milt
 

Is there a problem with the fact that P1 and P2 must fit in 20 bit registers and P3 in an 18 bit register?

Milt W8NUE

On Jun 10, 2020, at 4:44 AM, Guido PE1NNZ <threeme3@...> wrote:


Sorry Phil,

I misread your question, so my answer does not make sense, and please reject my previous email.
It is a good question what happens for value in between 128 for P2, no idea...never tried.

On Wed, Jun 10, 2020 at 11:30 AM Guido PE1NNZ via groups.io <threeme3=gmail.com@groups.io> wrote:
Hi Phil,
This is mathematically valid, does it work? Is there anything in the datasheet or application notes which says we can't do this? Is there any effect on the jitter or other performance?
My understanding is that this allowed and works. as it meets the criteria that "a+b/c has a valid range of 15 + 0/1,048,575 and 90"  [AN619 p.3 par. 3.2].
A zero in b means that there is no fractional part involved, and improves the jitter performance, when you set FBA_INT/FBB_INT [AN619 p.4 par. 3.2.1].

BUT!! then do not start using fractional divisions on the Multi-synth dividers as the jitter is worse compared to fractional dividers in the PLL Multisynth.


Re: WSPR data analysis #qcx40 #wspr

Hans Summers
 

Hi Ted, all

I have taken a different approach whenever I needed to analyze a download of a month's worth of WSPRnet data. I used Excel with some VBA code, that opens the WSPRnet downloaded csv file and reads it sequentially one line at a time. Then I implement a filter for whatever rows I want to look at, and any row that matches, I write that out into another csv file. 

I've always done it this way. 

To an extent it depends what software you have available and what you're familiar with. I had Excel available and am familiar with VBA having used it professionally for many years. So that was the solution I chose. 

Ted let me know if you're interested and I can find the VBA code I used and email it to you.

73 Hans G0UPL 

On Thu, Jun 11, 2020, 16:51 Graham, VE3GTC <colonelkrypton@...> wrote:
I do just what Michael has described.

You don't need a computer with Linux however but it is easier to get going.

On a Windows computer you can install many of the common *NIX command line tools available in 



and there are other WIN32 GNU command line tools sources as well.

BUT, these are not for everyone.

an alternative already exists with Windows.

There is a command line tool called FINDSTR.

Open a cmd.exe window and enter:  help findstr and you fill a simple listing of how to use this tool

findstr will search a specified file for a specified string just like using grep in Michaels example.

for example, at the c: prompt:   findstr callsignToSearchFor wsprFile.csv > results.csv

This will search wsprFile.csv for the string CallsignToSearchFor and will put the results in file results.csv

This will reduce that very large .csv file down to something more manageable that Excel can import.

Also, SQLite is a very good alternative to using Microsoft Access if you wished to go that way.

cheers, Graham ve3gtc


On Thu, Jun 11, 2020 at 1:24 PM Michael Babineau <mbabineau.ve3wmb@...> wrote:
Ted :

The simple approach that I have used to deal with these massive WSPR .csv files is to pre-process the file on a Linux machine
to extract the relevant data into a new file, discarding all of the records that I don't care about. Then I transfer that  new file to my
Windows machine and then open it with Excel.

The grep utility on Linux/Unix is line-based so it reads in one line at a time and processes that very efficiently. It can search using any 
regular expression so it is a simple matter to grep the .csv searching for your callsign and then use ">" to redirect the output
to a new text file.   Something like :  grep "VE3WMB"  wspr.csv  > newcsvfile.csv    would search wspr.csv for VE3WMB and output
only those lines. The '>' takes those output lines and redirects them to a new file named newcsvfile.csv . 

You can do this on any Linux/Unix box, even a Raspberry Pi.  If you don't have a Linux machine you could create a USB stick
with a live Linux distribution and boot from that without installing Linux on your computer. Or if this seems to difficult find someone
you know who is a Linux user and get them to pre-process the file for you.  

The advantage of this approach is that you separate the "wheat from the chaff" before you try to do anything with the data.  
Using a database you are still loading in a pile of information that you really just want to discard.

Cheers

Michael VE3WMB 


Re: TCXO for U3S CPU clock #u3s #clock

 

Halden,

I believe I saw something of this nature on the QCX when I first stared playing with WSPR. As I understand it there is some common code between the two products.

Julian N4JO.

On 6/11/2020 10:57 AM, HF via groups.io wrote:
Hi Hans,
Thanks for that definitive information!
Last night and this morning, I measured the 20 MHz oscillator.  It's running at 20,005,717 kHz at the beginning of the transmission, dropping to 20,005.414 at the end of the transmission.  Then it shifts up to 20,005.419.  It then drifts down to 20,005.418, turns around and drifts up to 20,005.423.  Then it shifts back down to 20,005.717 when TX starts.  Summarily, the heat causes downward drift on the frequency and being in TX mode shifts it down by 5 Hz.  I'm guessing that the supply voltage drops in TX mode, causing the 5 Hz shift.  I've also seen about 1 Hz drop in frequency when the display backlight is on - also presumably a supply voltage drop.
Looking at the characteristics of the drift during TX, I'm guessing that adding the second transmission to the cycle would drop the average frequency by about 4 Hz.  That's 0.2 ppm and would change the timing drift by 1 second in about 58 days if I calculated that right.  I'm seeing much more than that.
So...anyone have other ideas on what could cause this?
FYI, the average measured frequency is 20,005.718.  The setpoint I determined experimentally and entered into the U3S many months ago is 20,005.470.  The 248 Hz difference is 12.4 ppm, corresponding to a drift on 1 second per day.  I'll enter that in to the U3S now and see what happens.
But before I go, I have a related question.  I have found that every time I go into the menus and change something other than the time, I seem to lose about a half a second on the internal clock. Is anyone else seeing this?  Is there a programmatic reason for this?
Cheers
Halden


Re: TCXO for U3S CPU clock #u3s #clock

HF
 

Hi Hans,
Thanks for that definitive information!
Last night and this morning, I measured the 20 MHz oscillator.  It's running at 20,005,717 kHz at the beginning of the transmission, dropping to 20,005.414 at the end of the transmission.  Then it shifts up to 20,005.419.  It then drifts down to 20,005.418, turns around and drifts up to 20,005.423.  Then it shifts back down to 20,005.717 when TX starts.  Summarily, the heat causes downward drift on the frequency and being in TX mode shifts it down by 5 Hz.  I'm guessing that the supply voltage drops in TX mode, causing the 5 Hz shift.  I've also seen about 1 Hz drop in frequency when the display backlight is on - also presumably a supply voltage drop.
Looking at the characteristics of the drift during TX, I'm guessing that adding the second transmission to the cycle would drop the average frequency by about 4 Hz.  That's 0.2 ppm and would change the timing drift by 1 second in about 58 days if I calculated that right.  I'm seeing much more than that.
So...anyone have other ideas on what could cause this?
FYI, the average measured frequency is 20,005.718.  The setpoint I determined experimentally and entered into the U3S many months ago is 20,005.470.  The 248 Hz difference is 12.4 ppm, corresponding to a drift on 1 second per day.  I'll enter that in to the U3S now and see what happens.
But before I go, I have a related question.  I have found that every time I go into the menus and change something other than the time, I seem to lose about a half a second on the internal clock. Is anyone else seeing this?  Is there a programmatic reason for this?
Cheers
Halden


Re: Clock Alarm with Speaker within Case #clock #case #mods

KEN G4APB
 

Yes Chip,
you will need to programme a second alarm to come on at 8am and go off at 8:01am. You also need to set the sense of the output you use for ‘active low’ to only enable the speaker ‘gnd’ connection during this period.
73 Ken G4apb


Re: WSPR data analysis #qcx40 #wspr

Graham, VE3GTC
 

I do just what Michael has described.

You don't need a computer with Linux however but it is easier to get going.

On a Windows computer you can install many of the common *NIX command line tools available in 



and there are other WIN32 GNU command line tools sources as well.

BUT, these are not for everyone.

an alternative already exists with Windows.

There is a command line tool called FINDSTR.

Open a cmd.exe window and enter:  help findstr and you fill a simple listing of how to use this tool

findstr will search a specified file for a specified string just like using grep in Michaels example.

for example, at the c: prompt:   findstr callsignToSearchFor wsprFile.csv > results.csv

This will search wsprFile.csv for the string CallsignToSearchFor and will put the results in file results.csv

This will reduce that very large .csv file down to something more manageable that Excel can import.

Also, SQLite is a very good alternative to using Microsoft Access if you wished to go that way.

cheers, Graham ve3gtc


On Thu, Jun 11, 2020 at 1:24 PM Michael Babineau <mbabineau.ve3wmb@...> wrote:
Ted :

The simple approach that I have used to deal with these massive WSPR .csv files is to pre-process the file on a Linux machine
to extract the relevant data into a new file, discarding all of the records that I don't care about. Then I transfer that  new file to my
Windows machine and then open it with Excel.

The grep utility on Linux/Unix is line-based so it reads in one line at a time and processes that very efficiently. It can search using any 
regular expression so it is a simple matter to grep the .csv searching for your callsign and then use ">" to redirect the output
to a new text file.   Something like :  grep "VE3WMB"  wspr.csv  > newcsvfile.csv    would search wspr.csv for VE3WMB and output
only those lines. The '>' takes those output lines and redirects them to a new file named newcsvfile.csv . 

You can do this on any Linux/Unix box, even a Raspberry Pi.  If you don't have a Linux machine you could create a USB stick
with a live Linux distribution and boot from that without installing Linux on your computer. Or if this seems to difficult find someone
you know who is a Linux user and get them to pre-process the file for you.  

The advantage of this approach is that you separate the "wheat from the chaff" before you try to do anything with the data.  
Using a database you are still loading in a pile of information that you really just want to discard.

Cheers

Michael VE3WMB 


Re: Tx->Rx click once again (QCX)

Daniel Conklin
 

I'm wondering why this is not happening on all QCXs. Could it be component tolerances?  I use QSK with my stock QCX-20 and have not had this issue. 
-- 
73, Dan - W2DLC


On Thu, Jun 11, 2020 at 08:47 AM, Wolfgang DK4RW wrote:
Hi Steve,

Thank you for your modification of C24. It helped a lot.
Nevertheless I still had the same ringing Kuba documented in his post #44926. It turns out that T/R switching transients generated in the QSD mixer enter the audio chain. These fast spikes cause ringing of the CW filter.
A simple RC low pass filter, fitted after R27 reduces these transients. Ringing is reduced close to the RX noise level. Even with max. audio gain clicks or thumps are no longer audible anymore. The attached oscillograms show the results of tests on a dummy load.

This is a summary of all my modifications for click/thump reduction:

• C21, C22 changed to 0,47uF – 1uF; standard on PCB Rev 4, QCX assembly Rev. 1.11 and beyond
• C24 added 100uF parallel to existing 10uF capacitor; will be standard standard in QCX+
• R59 changed to 33k; reduces CW filter overload at high sidetone settings
• added C54 10nF; eliminates small spike
• added RC low pass (3.3k + 47nF) between wiper of R27 and input of CW filter

One additional modification, which may or may not apply to a standard QCX:
In my QCX I replaced T1 by a broadband transformer (5 turns trifilar on FT37-61) for multi-band operation. With this non-selective input the sidetone became distorted with RF output of more than 3.5 W. Residual RF at the input of the QSD caused overload in the mixer and/or in the pre-amp stages. Obviously the isolation of the T/R switch Q5 was insufficient. A 47 Ohm resistor parallel to the primary side of T1 solved the problem, without significantly reducing the RX sensitivity.

Enjoy QSKing.

73 Wolfgang DK4RW