Date   

Re: QCX PTT Mod

Hans Summers
 

Hi Paul
 
In the paper Hans put out on the PTT mod to the QCX for use with the 50W PA, he describes a hardware mod
 by adding a 10K resistor (value entirely uncritical) between the RX signal and +5V. My question is this:
- my junk box has 100, 100k and 1k resistors.
 - so Hans says value uncritical- so would any of these I have work in this application?

Even the mod itself is not particularly critical. I did find it prevented a particular glitch startup condition with a particular power supply, which once in 6 months of operation, caused me to kill the IRF510s. 

What happens is that on power-up, the microcontroller resets itself. Then starts the program. Early on in the program, the processor pin 14 is set to high, which puts the QCX into RECEIVE mode. But until this happens, pin 14 is high impedance and IC3 pins 12 and 13 are left floating; the only things on this signal are IC3 pins 12 and 13, and the gate of Q5 (Tx/Rx switch) and C34, 0.1uF to ground. Regrettably C34 will cause the voltage to be - momentarily at least - pulled to ground; thereby putting QCX into TX state and activating the PTT. 

To avoid this, a pull-up resistor from pins 12 and 13 to +5V keeps everything in a determined state (Receive) during this short interval. Ok so it still has to charge up the 0.1uF capacitor. I chose 10K. 1K would be fine too. If you go too low, then during TX the microcontroller I/O pin will have to work too hard to pull current through it, which could exceed the rating of the I/O pin. 1K will draw 5mA which doesn't harm anything but is a bit of a waste. 10K, 0.5mA. If the resistor is too high, then it will be slow to charge up the 0.1uF capacitor and we will be back to a momentary activation of PTT. 

If your junkbox only has 1K and 100K resistors then 

1) Please stock up on some more junk ;-)

2) I'd tend to use the 1K resistor, accepting the 5mA higher current draw on TX, which won't be noticeable, in preference to the unknown quantity of any risk of lengthening the momentary PTT pulse.

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


Re: Compressing or expanding turns on a toroid.

Timothy Fidler
 

 
Re - sparse wound toroids and the drop off of measured L, there is no black magic here -
 
Re T50-6 sparse wound coils  at  220 degrees for an L of approx 2.9uH as per diagram quoted on VK Ham.
Toroid data is from  tables provided by the Toroid King aka Diz of Florida USA.
 
Vkham article is at https://www.vkham.com/Info/ferro/tut_5a.html
 
Od is 12.7, Id is 7.7  
 
Mean D which is the nominal D for flux run purposes is 10.2mm
 
The amount of run len which is avail for wiring at 220 deg worth of cover is  D * pi * 220/360 
which is 19.6mm. NB ten turns so pitch per turn is 19.6 mm.

 ( TEF notes sorry , minor cock up on the Catering Front here ..the use angle in the VK example is 200 degrees not 220. This has 
but a minor effect so I have not redone the calculations below) ; 
 
The HGT of the core is 4.3mm.  
 
The nominal pitch angle of the wiring  is therefore  Atan (2/ 4.3) mm or  24.9 degrees.  That's the nominal pitch 
down angle of the  B vector from the ctr line of the mag racetrack when the core is viewed al Horizonto... 
 
The cosine of that is 0.91 which is the fraction of flux aligned with the racetrack. However that wire spiral is 
three dimensional so therefore the correct figure for the aligned flux is  Cos squared or 0.82.
0.4 uH original L for a close wound coil   (as per the diag provided 0.39u) and as per The Toroid king calc. 
 
(which can be desk checked)  ,
 
this becomes therefore  0.4 * 0.82 = 0.32 approx uH. Which is rather close to the data given for the sparse wound 
coil.  And of course this makes no allowance for end to end capacitance which seems to be a bit of a black art 
subject. 
 
So there is no black magic there but the coil winding technique has a big influence on the final L and therefore 
the detail filter performance. BTW the sparse coil will have higher Q  (Xl at frequency / R ) . There is  hard  data on this in the Amidon 
technical section on their website and ditto in the RF engineering publications you can find on the the 
Micrometals website.
 
Unfort. there is no real corresp between the Micrometals core materials and the Amidon materials but the MicroM data 
is better presented than the Amidon.  Amidon materials are used in most amateur projects because they are easier 
to get hold of in small qty - eg The Toroid king or in Australia, MiniKits of Adelaide.

So what you get with one hand is taken away with another. Ain't RF physics a bitch !
 
 
If you buy a Tcvr made by say ICOM to more or less military standards you have one level of engineering and test. 
If you buy a LPF board on a dirt cheap tcvr out of Shenzen well you might get something else in terms of filter 
performance and spurious emission.
 
 
BTW the other non aligned  flux does not disappear - Faraday says it cannot - it becomes flux that loops back 
 
around the creating wire and of course it becomes leakage flux too - more serious if the part is against 
 
ferromagnetic casings and such like. Therefore toroids do leak - but the wiring configuration defines how much 
 
leakage.  Even 50 Hz toroid mains transformers with near radial wiring technique do have a small external field.


Re: QCX PTT Mod

Shane Justice <justshane@...>
 

Paul,

I wonder what value Hans meant?- Value could also point to the wattage capacity of the resistor, or the tolerance. I doubt he would specify a resistance value and then comment it doesn't matter. He may have meant any value in the ranger of 100k- say 90k to 120k.

Probably best for Hans to say what value was noncritical, and provide some additional detail.

Best regards,
Shane


On Sep 15, 2020 at 07:46, w2eck <eckerpw@...> wrote:

In the paper Hans put out on the PTT mod to the QCX for use with the 50W PA, he describes a hardware mod
 by adding a 10K resistor (value entirely uncritical) between the RX signal and +5V. My question is this:
- my junk box has 100, 100k and 1k resistors.
 - so Hans says value uncritical- so would any of these I have work in this application?

73 Paul
w2eck


Re: Product Suggestions? #15m #80m

Kelly Jack
 

Hans,

Is there any chance that a band option for 160M could be added to the firmware for those wishing to experiment with a lower reference crystal for the si5351? I have a QCX+ on order and am hoping to experiment with getting it running on 160m. ie lower frequency clocks with the lower reference crystal, not a 4X frequency option with divider.

73


Simon
VK3ELH


QCX PTT Mod

w2eck
 

In the paper Hans put out on the PTT mod to the QCX for use with the 50W PA, he describes a hardware mod
 by adding a 10K resistor (value entirely uncritical) between the RX signal and +5V. My question is this:
- my junk box has 100, 100k and 1k resistors.
 - so Hans says value uncritical- so would any of these I have work in this application?

73 Paul
w2eck


Re: Compressing or expanding turns on a toroid.

Fred Piering
 

DK:
You are absolutely correct. Most inductance meters measure the "Apparent" inductance which includes the self capacitance of the windings. 
There are simple methods to measure the true inductance and self capacitance.
However, from the simple circuit standpoint, I would think the only thing it cares about is what it looks like, not why.
I have performed measurements that have shown the change in self capacitance with windings from full distribution to 270 and 180 degrees.
73
Fred, WD9HNU


Re: QCX+ mechanical problem

Jerry DiPietro
 

I used a hammer and it fit fine.

On Sep 12, 2020, at 10:02 AM, Larry Taylor <W7ldt.az@...> wrote:

Everybody building the + has dealt with this. Grad a piece of sand paper, place it on the bench, grad the nut with fingers (or pliers) and sand away on in 2 of the 6 sides to the nut. It works just great. 


VFO/Signal Generator - current consumption?

Steve Rawlings
 

I understand that the supply voltage is 5 volts but I'd like to know the current consumption.  

I've looked through the web resources, but I must have missed it.  I'd appreciate a pointer to where it's specified.

Many thanks,
Steve


Re: 7.7 Band Pass receiver input filter characteristics

Hans Summers
 

Hello Pierre
 
New to this group and interested by this nice QCX+ my last experience soldering components was with the late Heathkits and I need to refresh my knowledge of HamRadio!
Could you give the vertical scale of these graphs? 

The graphs in section 7.7 are spectrum analyzer screenshots from my Advantest R3361C spectrum analyzer, via GPIB to USB adapter. The top of the grid represents an insertion loss of 0dB. The vertical scale is decibels with 10dB per division. So the vertical scale goes from 0dB insertion loss at the top, to 80dB insertion loss at the bottom.
 
May I understand that a QCX+ 20m can receive with a good sensitivity from 9.69MHz to 22.54MHz, possibly by resetting C1?

It is optimized for 20m; the range 9.69MHz to 22.54MHz is what I measured as the tuning range using the C1 trimmer. But basically yes, you can tune it to measure in this range. 
 
May I also understand that it can transmit some power on the 30 and 17m bands?

Yes BUT... On 30m the harmonic attenuation will not be so good, and perhaps not good enough to satisfy regulatory requirements. On 17m since 18MHz is likely above the 20m filter cut-off (3dB) frequency, the output power will be attenuated. 
 
One last question, I plan to buy one kit with the Japanese firmware but could also wait for the QCX-mini, will this Japanese firmware be usable with the mini?

QCX, QCX+ and QCX-mini all use the same firmware chip. HOWEVER, there is going to need to be a minor change to the firmware, to make the LED backlight on the QCX-mini controllable by a menu item. If you use current firmware 1.05 then the QCX-mini LED backlight will be OFF. You could hardwire it to be on, of course, via a simple hardware modification. 

All the Japanese firmware version does is prevent operation outside the Japanese band limits. This is a requirement for the Japanese license, strictly speaking, though many Japanese amateurs are using the normal QCX firmware. There is nothing else about the firmware version that is Japanese. I.e. the menus are still in English and the Morse is still international Morse etc. Only the frequency restrictions are implemented. The penalty for this is that the WSPR functionality is disabled - this is necessary in order to make space for the frequency check. 

At the moment the allowed operating frequencies, with the Japanese firmware version, are:

3500000 - 3575000
3599000 - 3612000
3680000 - 3687000
3702000 - 3716000
3745000 - 3770000
3791000 - 3805000
7000000 - 7200000
10100000 - 10150000
14000000 - 14350000
18068000 - 18168000
21000000 - 21450000
24890000 - 24990000
28000000 - 29700000

I have in mind some dim recollection of someone telling me the 80m allocations have been changed recently; if anyone can confirm this and what the new band limits are, I can update that in the firmware. 

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


#qsoparty Fun with 40m in QRO QSO Party, report #qsoparty

Hans Summers
 

Hi all

Since the thread on the QRO QSO party somehow got hijacked, here's a new one. Apologies if it is not of interest... 

I was a bit unlucky at 1900Z, I had a reply to my first CQ on 40m immediately (which was not unlucky), from UT2LA/QRP Sergey. I sent his report etc. But then during his over, I had a family interruption... for 45 minutes! I came back at 1951Z and worked HA6VH Geza and E74LZ Meho. After Meho, (and after 2000Z) people kept calling and calling... there were several at once... I QSOed 4L6QL Mike then LZ2GA Georgi, there were still people calling, but I was so so tired and planning to get up early for the 0300Z session so reluctantly I switched off the radio. 

This morning at 0300Z it was 6am local time and everyone was asleep. Even, miraculously and highly uncharacteristically, The Baby. 40m was beautifully quiet at my QTH and very little QSB. I had a long stream of QSOs, most of them with very good signals:

RA3LK Anatol
UA6YH Alex
UA6YX Yuri
UR5EOH Toly
DJ0YI Ben
UA4CDV Vlad
R4PCE Alex
DH2FW Frank
HA0HW Laci (starting 0404Z so just outside the hour). 

No QCXs in any of them :-/   But at least Ian G4GIR made up for that in the 1300Z session yesterday, one of my five 20m QSOs that session. 

Sorry to go on... but this was my first time on air for 1 month since the last QRP Labs QRO QSO party! The QSO parties help get me on air, and remind me what fun I *should* be having if I wasn't so busy :-/   Now I have good enthusiasm for the day! I should do this every morning!

Another question... in case anyone has any ideas... I have a 0-30V, 0-20A linear power supply type RXN-3020D. Purchased new last year. Same ubiquitous design thing, as made by many manufacturers. I power it on, and set it for 20V. Current draw on key-down is somewhere between 4.5A for around 50-60W RF output (including the current consumption of the QCX). *IF* I turned the knobs all the way clockwise the display reads 31.2V. HOWEVER... over the course of an hour of operating, the output voltage drifts downward. I have to keep making adjustments, turning the voltage knob clockwise a bit, to keep the output reading 20V. By the end of an hour's operating, the voltage knob is almost all the way clockwise and the thing can just barely make it past 20V. Any ideas anyone? I ought to open it up probably. But time is so short... 

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


Re: QCX+ 20M Lost sensitivity

Alan G4ZFQ
 

My guess is you lost a solder connection somewhere.
A bad connection on T1 is possible. I'd resolder that first.

73 Alan G4ZFQ


Re: QRO QSO Party

Timothy Fidler
 

 

Aha.. it is the old third shoe in the pocket trick...  a bit like the poisoned needle in the  Phonograph trick as Maxwell Smart  would say.  I hear head scratching .. Phongraph.. wht a the hell is that??.. was that before the Titanic ?  Perhaps that is Hans' NEXT BIG THING.. retro phonographs.
BTW taking a shoe off and banging it on the table was the sort of thing you did a a  Communist meeting in Pskov in the 1920s.. many many upper crust Russkies were hellishly embarrassed when Nikita did that at the UN .  It made the Russians look like a bunch of oafs.. well they did steam roller half of Europe with their ideology post WWII .. that's a bit oafish.. but all better now.. we kissed and made up.. sort of.     TEF


Sidetone level

Jim Painter
 

I have finished my QCX+ 40M and is working well with the following exception.

The sidetone level on receive is fine but on transmit it is very week (barely audible). I have tried adjusting the sidetone level but it makes no difference to Rx or Tx.

I checked the sidetones on my QCX 20M. Both Rx and Tx are fine and pretty much the same level. But I find that the sidetone level adjust has no effect on this unit either.

Thanks...Jim...KQ3S


Re: QCX+ 20M Lost sensitivity

@CurtisM
 

Bruce

My guess is you lost a solder connection somewhere. If you check your alignment and its no longer nulled, most likely its within audio, okay easy guess since nearly all of receiver is audio.

Just turn on the builtin test signal. Check peaking of  the trimmer capacitor. Leaving the test source on, poke around with your finger and see if your audio recovers when you touch the now open connection. This is how I found the issue with my first qcx. Oh I did this once on my valve receiver, except using a stick.

73 curt


Re: 7.7 Band Pass receiver input filter characteristics

@CurtisM
 

Pierre

bon jour.  sitting listening to my 80m QCX+ this rig is amazing.  I highly recommend its assembly - wonderful manual and thoughtful design.  the companion 25/50 watt amplifier sits completed - ready for alignment and careful check out. 

the rig and its control firmware are not really intended for multi-band use - but there are examples out there.  the input receive bandpass filter is essentially a high Q single-tuned filter - it needs nothing more.  now the important part - the special quadrature cancellation network combining the unique mixer and dual audio chains only works on one band.  choose your favorite band for your first QCX+

Please find the video tour of QRP-Labs ... and download the manual to see what you think.  You will be thrilled by your build and operating experience - glad to have you join us here.  I am merely a several-times builder of these fine products. 

73 Curt wb8yyy


7.7 Band Pass receiver input filter characteristics

Pierre Guillot <jc_glt@...>
 

New to this group and interested by this nice QCX+ my last experience soldering components was with the late Heathkits and I need to refresh my knowledge of HamRadio!
Could you give the vertical scale of these graphs? 
May I understand that a QCX+ 20m can receive with a good sensitivity from 9.69MHz to 22.54MHz, possibly by resetting C1?
May I also understand that it can transmit some power on the 30 and 17m bands?
One last question, I plan to buy one kit with the Japanese firmware but could also wait for the QCX-mini, will this Japanese firmware be usable with the mini?
73 to all - FK8IH - Pierre


QCX+ 20M Lost sensitivity

Bruce KS4V
 

I was listening to my QCX+ yesterday on 20 meters and all of a sudden I noticed my received signals dropped off considerably.  I can only copy a few of the loudest signals, which are very weak on my QCX+.
The receiver portion had been working quite well up to that point.  I  went through the alignment procedure again and noticed that as I turn C1 to peak the bandpass filter, I do not see but a one or two bar difference in the entire range of C1.  Transmitter is putting about 2.4 watts from a 12 v power source.  I will try removing a turn off L3 to boost the power after I get the receiver straightened out.
I am looking for ideas where to start with very limited test equipment.  Thanks, Bruce  KS4V


Re: Compressing or expanding turns on a toroid.

N3MNT
 

Had some time today so I pulled all the LPF out of my U3S and stuck them on my VNA and measured insertion loss while adjusting the windings.  The results  showed that there was no one size fits all and the changes were not huge.  On some filters compressing L1 and L3 while expanding L2 produced best results. On some compressing all was best, while on one compressing the windings produced the worse results.   For all six filters, the max insertion loss was 0.4db. at the operating freq. 


Re: LightAPRS-W (+WSPR) tracker question

BrianB
 

Dwight,

There needs to be a script running somewhere to pick the WSPR spots from the database, see if it's 15-min or more newer than the APRS spot, reformat it, and finally write it to APRS.fi

Most of us have been running the scripts on our own computers. Recently Pedro of LU7AA.org has been helping out with some but I'm not sure if he supports the LightAPRS-W.

Once there is a high altitude spot in APRS, and it's formatted correctly, HabHub and other systems will pick it up from there.

Hope this helps.

73,
BrianB
N6CVO


Re: QRO QSO Party

Peter GM0EUL
 

I was on 40m at 1900 using my qcx and Juma PA100-D running about 75 watts to a dipole.  Very difficult conditions, lots of activity on the band but terrible qsb so people were fading in and out, tripping over each other and causing qrm.  Managed to work Carl, IK1YRA, who was running 100 watts to a vertical (non-qcx).  

73
Peter GM0EUL 

8681 - 8700 of 60996