¿ªÔÆÌåÓý

ctrl + shift + ? for shortcuts
© 2025 Groups.io
  1. QRPLabs
  2. Topics

Low Power fix on QCX 80 and a suggestion for a possible addition to QCX test gear to measure inductance.

 

That would be fantastic. Speculations and possibly too obvious stuff follows:
With the LPF unpopulated you could place the resonant LC under test between the C28 and C27 points and use the already present signal injection to sweep and detect the peak. BPF would need to be disabled, like disconnecting C1 or similar.
If you want to have the entire LPF fitted it would be possible to redirect R34 to get the signal off board via the LC under test and fed into the BNC, or perhaps after C29.
The rest is firmware (: Auto/manual sweeping for finding the peak and perhaps tuning the coils etc.

On Thu, Nov 30, 2017 at 4:38 AM, Hans Summers <hans.summers@...> wrote:

Yes, adding an inductance meter would be a nice feature... or some way of measuring/checking the toroids. I am thinking about how that could be done.?
 

If you only have wire with one colour, you can use coloured felt markers to make each piece of wire distinct.

George VY1GP
Arv Evans
 

Nick

Should work.? Use very small (2 to 3 pf) capacitors to couple into and out of the DUT.
I do this with an HP Signal generator and use an Oscilloscope to detect peak voltage.
No reason the same thing could not be done using the Si5351a as a signal source
with an RF detector and digital voltmeter (ADC input to AVR micro-controller).? Could
be enhanced by using an Arduino and software to detect peak and specific off-peak
signal levels so you could determine resonance and Q-factor.? If using an Arduino with
ttl/USB PC interface you could even grab the readings and convert that to graphical
representation.? This is starting to look a lot like a SNA (Scalar Network Analyzer).

Arv
_._


On Wed, Nov 29, 2017 at 9:44 PM, <nickpullen@...> wrote:
Could you build the whole kit except the LPF sections, and use thiose know capacitors and the Gig gen and the RF meter to build a simple test circuit to test the inductors, and a new menu item in the firmware?
 

Could you build the whole kit except the LPF sections, and use those know capacitors and the Sig gen and the RF meter to build a simple test circuit to test the inductors, and a new menu item in the firmware?
Arv Evans
 

Interesting...sweeping the LO while monitoring the detected RF for peak output could
provide a way to generate the response curve for your LPF or, with a known capacitor
value it could be used to find resonance point, and from that calculate the effective
inductance of an individual winding.

Arv? K7HKL
_._


On Wed, Nov 29, 2017 at 8:38 PM, Hans Summers <hans.summers@...> wrote:
Hi Des, Keith

Yes, adding an inductance meter would be a nice feature... or some way of measuring/checking the toroids. I am thinking about how that could be done.?

Note that yes, it is true, the quadrature LO will only work down to 3.2MHz - this does not affect transmit. The quadrature LO is used only on Receive.?

73 Hans G0UPL

On Wed, Nov 29, 2017 at 10:30 PM, QRP_nut <raspiham1973@...> wrote:
Hi Keith,

Interesting idea regarding the swept source, it would also be a useful additional feature for the QCX test equipment.

While my QCX was delivering lower than expected output I did try winding the frequency up/down to get an idea of what was happening in the filter section of the PA but unfortunately the QCX VFO (while in quadrature mode) drops out below 3.2 MHz but it was clear the power was increasing as the frequency was reduced which in turn helped me identify the problem with the number of turns on the cores.

Good luck with the build Keith and a couple of tips regarding the toroids when you come to them. Solder the PA and filter toroids in "lightly" or semi permanently until your sure the turns are correct. I was "silly" and trusted to luck Hi. I followed the manul to the letter but on re-checking the toroids inductance I found they all needed two or three turns removing in order to bring them to the correct value. Wish now I had checked them prior to fitting to the PCB. I also made the mistake of soldering them in "fully" which made removal/replacement harder than it needed to be but with care and patience they came out OK and the PCB was none the worse for the hacking about.

If you have good eyesight then winding and fitting T1 will probably not be an issue for you but if you have to work under a magnifier as I do then this sometimes reduces the depth-of-field and makes identification of the 8 connections hard. I found the job becomes a "breeze" if you wind secondary-1 and secondary-2 with different colour coil/ magnet wire if you have it. This means each alternate winding is a different colour on the core and makes identification of the windings and connections much easier.

Anyway, have fun building the QCX.
73,
Des (M0AYF)



 

Hi Des, Keith

Yes, adding an inductance meter would be a nice feature... or some way of measuring/checking the toroids. I am thinking about how that could be done.?

Note that yes, it is true, the quadrature LO will only work down to 3.2MHz - this does not affect transmit. The quadrature LO is used only on Receive.?

73 Hans G0UPL

On Wed, Nov 29, 2017 at 10:30 PM, QRP_nut <raspiham1973@...> wrote:
Hi Keith,

Interesting idea regarding the swept source, it would also be a useful additional feature for the QCX test equipment.

While my QCX was delivering lower than expected output I did try winding the frequency up/down to get an idea of what was happening in the filter section of the PA but unfortunately the QCX VFO (while in quadrature mode) drops out below 3.2 MHz but it was clear the power was increasing as the frequency was reduced which in turn helped me identify the problem with the number of turns on the cores.

Good luck with the build Keith and a couple of tips regarding the toroids when you come to them. Solder the PA and filter toroids in "lightly" or semi permanently until your sure the turns are correct. I was "silly" and trusted to luck Hi. I followed the manul to the letter but on re-checking the toroids inductance I found they all needed two or three turns removing in order to bring them to the correct value. Wish now I had checked them prior to fitting to the PCB. I also made the mistake of soldering them in "fully" which made removal/replacement harder than it needed to be but with care and patience they came out OK and the PCB was none the worse for the hacking about.

If you have good eyesight then winding and fitting T1 will probably not be an issue for you but if you have to work under a magnifier as I do then this sometimes reduces the depth-of-field and makes identification of the 8 connections hard. I found the job becomes a "breeze" if you wind secondary-1 and secondary-2 with different colour coil/ magnet wire if you have it. This means each alternate winding is a different colour on the core and makes identification of the windings and connections much easier.

Anyway, have fun building the QCX.
73,
Des (M0AYF)


 

Hi Keith,

Interesting idea regarding the swept source, it would also be a useful additional feature for the QCX test equipment.

While my QCX was delivering lower than expected output I did try winding the frequency up/down to get an idea of what was happening in the filter section of the PA but unfortunately the QCX VFO (while in quadrature mode) drops out below 3.2 MHz but it was clear the power was increasing as the frequency was reduced which in turn helped me identify the problem with the number of turns on the cores.

Good luck with the build Keith and a couple of tips regarding the toroids when you come to them. Solder the PA and filter toroids in "lightly" or semi permanently until your sure the turns are correct. I was "silly" and trusted to luck Hi. I followed the manul to the letter but on re-checking the toroids inductance I found they all needed two or three turns removing in order to bring them to the correct value. Wish now I had checked them prior to fitting to the PCB. I also made the mistake of soldering them in "fully" which made removal/replacement harder than it needed to be but with care and patience they came out OK and the PCB was none the worse for the hacking about.

If you have good eyesight then winding and fitting T1 will probably not be an issue for you but if you have to work under a magnifier as I do then this sometimes reduces the depth-of-field and makes identification of the 8 connections hard. I found the job becomes a "breeze" if you wind secondary-1 and secondary-2 with different colour coil/ magnet wire if you have it. This means each alternate winding is a different colour on the core and makes identification of the windings and connections much easier.

Anyway, have fun building the QCX.
73,
Des (M0AYF)
 

Des M0AYF wrote...

it would only take a small number of additional
components (and software) to make some sort "peak
detector" such that the resonance of each PA stage
inductor can be checked while building.
While thinking about your suggestion, I had another
thought. How about connecting a dummy load and sweeping
a range of frequencies to see where max power was and
using the results to determine what needed to be removed?
Of course, this does mean completing the build first.

I have my first (80 M) QCX nearly built. I still have to
wind the toroids, but I've left the BNC connector, volume
control, rotary encoder, push buttons and jacks off as my
intention is to put it in a QRP Labs aluminum enclosure.

--
73 Keith VE7GDH

1 - 8 of 8