Re: Low pass filter advice #40m #80m #vfo

Steven Dick

I assume it was properly terminated at both the load and the source.  Otherwise you will not get the proper filter response curve. You shouldn't feed the output of the filter into a high impedance scope input. It should include a 50 ohm termination at the filter output before it goes into the scope. You can improve the source termination by using a 50 ohm 3dB or 6dB attenuator at the input, rated for the power used.   Depending on the LPF design, it is possible to get a peak.  A Butterworth design  should give a smooth fall-off with no peak. Other designs may have a slight peak near cutoff or a rising response in the bandpass.

-Steve K1RF

------ Original Message ------
From: "Peter GM0EUL" <gm0eul@...>
Sent: 9/13/2020 9:28:14 AM
Subject: Re: [QRPLabs] Low pass filter advice #40m #80m #vfo

Thanks Alan- that puzzled me, I wasn't expecting a peak and don't know what to make of it or what sort of artefact may have caused it.  I want to try and understand it before I do anything to change it.

Peter GM0EUL


Re: QCX-mini update

Tech Guy

Sounds wonderful Hans. Thank you for your efforts again
John N4HNO 

Re: Low pass filter advice #40m #80m #vfo



Find your way to the website of AA8V, use your search engine to find Digital VFO for Vintage Transmitters.

Of course the drive impedances are different for thermions. You might enjoy your visit to professor Latta's website.

73 Curt

SMT Zerobeat kit


Today I built the wb9kz SMT Zerobeat kit. Nice looking zerobeat LED display! Would be a nice visual addition to the QCX+

Video on imgur:

Re: Low pass filter advice #40m #80m #vfo

Peter GM0EUL

Thanks Evan
That seems to have done it.  With 47 Ohm across the scope it behaves exactly as I would expect i.e the 40m filter is flat to about 8.2 MHz then drops off quickly.  Testing the 80m filter now but I'm expecting that to behave itself too.

Noted about the need for a transformer when coupling the vfo to the transmitter.  It isn't expecting 50 Ohm I don't think, but not sure what it is expecting.  I'll see if I can figure it out.

Peter GM0EUL

Re: Low pass filter advice #40m #80m #vfo

Evan Hand

If I am reading the filter specification right, you might want to start with a 50ohm termination on the output of the filter before the input to your scope for your test.  Rerun the test to see what you get. 

You may also want to verify the impedance of the transmitter to be sure that it is expecting a 50ohm source.  If it is different, then either a resistor pad or a transformer may be needed.

Just a couple of suggestions.

Re: Low pass filter advice #40m #80m #vfo

Peter GM0EUL

Thanks Alan- that puzzled me, I wasn't expecting a peak and don't know what to make of it or what sort of artefact may have caused it.  I want to try and understand it before I do anything to change it.

Peter GM0EUL

Re: Low pass filter advice #40m #80m #vfo

Alan G4ZFQ

The amplitude doesn't vary much from about 4 to about 9 MHz then it shoots up and peaks at 9.2 MHz then rapidly falls off and is almost fully suppressed by about 10 to 11 MHz.

Your description makes me wonder about the technique you used.
A LPF is flat until it begins to reject HF. There is no peak.

73 Alan G4ZXF

Low pass filter advice #40m #80m #vfo

Peter GM0EUL

Hi All
Can I request a bit of a reality check before I take my next steps.  I've made the vfo kit to add to my one-valve transmitter (this is the one that's the same as Hans's first transmitter, so arguably an ancestor of the QCX!).  The vfo has the 40m and 80m low pass filters and I had a look at their characteristics because I wasn't getting as much output as I expected.  

I've measured the peak to peak voltage output from the 40m filter on a scan from about 4 MHz to 11 MHz (have a look at the 1 min video here  )  Ignore the actual numbers but look at the amplitude of the wave against the frequency (bottom left readout) The amplitude doesn't vary much from about 4 to about 9 MHz then it shoots up and peaks at 9.2 MHz then rapidly falls off and is almost fully suppressed by about 10 to 11 MHz. 

The 80m filter has similar characteristics and the peak output is at 5.3 MHz and its suppressed by just over 6 MHz 

I think I need to tweak the filters so that the peak output is at the bottom of the bands of interest (i.e pull it down by about 2 MHz by fiddling with the coils) is that right or am I interpreting the data wrong or misunderstanding what its telling me?

Any thoughts or definitive wisdom appreciated.

Peter GM0EUL

Re: QCX-mini update

Bill Cromwell


Would that require a fine tip Sharpie Marker or can we get by using the standard size Sharpie?


Bill KU8H

bark less - wag more

On 9/13/20 4:53 AM, Lex PH2LB wrote:

from a employer point of few, the cutting of the SMD part strips from the reels and writing the values on them (not all 0603 components have values) would be a punishment for Hans's co-workers.

Re: #40m #40m

Don Blachura

Hello Bob,
Thanks for the reply. I have checked amazon and found the usb to ttl  interface. 
also found a usb to ttl cable with a mini plug on it, but the tip is tx not rx. Can you explain 
how it should be hooked up to the rig?I am not very good at all this new  tech. See the interface,
now I am lost.


Re: BPF 80m

Alan G4ZFQ

it is the negative sign that I was referring to.

Did you look up what Korinthenkacker means?

Razvan knows he used a double negative but what does it matter?
Everyone knew what he meant.

73 Alan G4ZFQ

Re: BPF 80m

geoff M0ORE

Hi Razvan,

I am not saying that insertion loss is negative, it is the negative sign that I was referring to.


On 13/09/2020 00:14, DL2ARL wrote:
 geoff M0ORE

I humbly excuse my english as not being breed at Oxford.
As for the insertion loss being negative, I know a word that fits  marvelously: in german they say to it Korinthenkacker.
Your fine remark has helped a lot to  clear the problem described above, thank you.
Enjoy your digestion.

Yours Razvan DL2ARL

Re: Compressing or expanding turns on a toroid.

geoff M0ORE

The late W4ZCB described a simple L & C oscillator to which you added the un-known component and then measure the frequency and by using a simple bit of maths or a spreadsheet calculate the un-known parameter. You needed a good reference capacitor to do the initial calibration, I was lucky to be able to borrow a 0.1% tolerance to make my own standard. Not perhaps  as good as a lab standard but I am aware of the limitations.

On 12/09/2020 23:53, wb8yyy via wrote:
Yes nice explanation that physical arrangement of the coil adjusts inductance.

One can measure inductance by resonating with a capacitor, measuring frequency,  and calculating inductance from that. Here in US the njqrp once offered a little kit that measured inductance and capacitance, in era before cheap lcd. Read them out in morse. I don't know what's out there now, but an LC meter would be a nifty offering,  yeah take advantage of a market here too focused on their coils.

73 curt

Re: QCX+ Resistor hot

K3YV <eeb3568@...>

Hi Curt. Thank you. The cover plaque was fun and really appreciated. I am changing the resistor value to adjust the brightness and solve the dissipation problem. Lots of room for two resistors. 73

Re: QCX-mini update


On Sat, Sep 12, 2020 at 11:29 PM, Hans Summers wrote:
The hard-core sadistic fanatic (not suggesting necessarily that this includes you Janez OM) could always use hot air to remove all SMD components from the assembled boards, then have fun identifying all the tiny grains and cleaning everything up and then re-soldering them to his heart's content...


I wouldn't mind soldering 0603 components (have the equipment for that) but I fully understand Hans reason not to sell a "have a fine masochistic SMD Hamradio kit building experience" version of the QCX mini and from a employer point of few, the cutting of the SMD part strips from the reels and writing the values on them (not all 0603 components have values) would be a punishment for Hans's co-workers. 

Re: QCX-mini update


On Sat, Sep 12, 2020 at 06:30:29AM -0700, Glenn ODonnell wrote:
now ... about that QCX Pico ... hmmmm, intriguing if my old eyes were better 😊
Naturally Hans would include a special mode which "displays" all the pertinent data by encoding it as CW and playing it to the user... no LCD screen necessary!

No pesky knobs or buttons either, just "key" in your commands ;)

Re: transmitter/transceiver idea

Jens Groh

Thank you for the Teensy mention, Ron.

Maybe an even better idea: 
Could the QSX PCB get jumpers to optionally disconnect the Si5351A interface, the ADC/DAC interface (assumed to be a serial one) and the LCD interface?
Then the QSX could become an ideal platform for modifications in this spirit, but with the added value that it is a transceiver not only a receiver.


Am 19.08.2020 um 23:11 schrieb Ron Carr <rcarr@...>:

I like this idea.   A Teensy 4.x and audio board could also be used.  There are some hams on the Teensy forums who have built SDR receivers, one by DD4WH he calls a convolution SDR ( uses FFT and inverse FFT instead of a phasing approach ).

Re: Squeezing more power out of an IRF510 based RF amp with a Silicon change out (not being RD16HH).

Timothy Fidler

FQP13N10D part versus the old  IRF510 cockroach :

I knocked up the table with the two compared.  As per Han's comments at higher HF the  FQP13 will dog down. However at up to about 10 Mhz it
will basically keep pace with the IRF510 part and if equipped with a gate ballasting inductor  of relatively small value (as shown)  should perform well at 14 Mhz as well.

Does anyone have a true LTspice model for the FQP13N10D before I butcher something up for simulation ?

I will update the table in due course as results come in .  I'd love to hear back from 30m and 20 m users who test this part.  The part is particularly to get in two off qty in Australia, from Mini kits.

The Gfs  (transconductance) of the part is much higher than the IRF 510 part  so the bias may be very sensitive to set up. Beware.

As per all devel work , run tests at lower Vcc initially and have all power feeds fused with low value fuses. 

And the point is ? More of those little black electrons up the wire.. why make do with 40 Wrms when you can have 55 or sixty ? Maybe. Always Maybe.

Comment - the power gain of a common source MOS amplifier is a closed form solution with  Y parameters,  but you show me someone who uses Y parametrs to get an accurate solution for an RF HF Power amp and I'll show you a worse liar than Bill Clinton.

Re: QCX+ RF Power #qcx


I am reminded of a time a friend retuned a 25W 2m transmitter and eventually got almost 60W output, as measured on a Bird Wattmeter. Unfortunately the fundamental was still only 25W, with the remaining power contained in spurs and harmonics that earned the wrath of the regulators.

So all I am saying is have a way to check for second and particularly third harmonic levels when you start raising the cut off frequency of the low pass filter.

Yes, it is entirely possible that the inductors are higher value than the design goal. And reducing turns to achieve the design value is a very good thing. Just be careful out there!


On Sep 12, 2020, at 18:53, Arv Evans <arvid.evans@...> wrote:


My idea was to determine the tuning range of compressing versus expanding windings.  Once known this could predict whether you need to add or remove turns or just adjust turn spacing.

Adjustment of turn spacing is more about distributed capacitance than about changing the inductance, so making a predication formula or software tool may be a bit difficult.


On Sat, Sep 12, 2020, 4:05 PM Al Sines <sasines@...> wrote:

Not sure if this is exactly what you are thinking but so far with my new QCX PLUS 20 meter these are the RF output results:

Input voltage 14.5 
Reverse polarity diode is installed

After winding coils L1, L2 & L3 as specified in the manual:

1 watt out measured on bird wattmeter 

After spreading the windings on all three coils as much as possible:

3 watts out on the bird wattmeter 

After removing one turn each from L1, L2 & L3:

4.5 watts out on the bird wattmeter. 

I wound my coils very tight and close together at the start. 

I am thinking about removing one more turn each from L1, L2 & L3 to see the effect on the RF output.

In a way I am hesitant to do so cause it all works so well but sometimes my OCD makes me want to get the whole 5 watts out at about 13.2 volts (13.2 is the voltage of my bioenno when charged)

What say you?  Thanks

Alan, W3AL

On Sep 12, 2020, at 13:14, Arv Evans <arvid.evans@...> wrote:

Might be interesting and informative if some ambitious person 
would do some tests and make chart or on-line calculator that 
shows the effect (tuning range) of spreading and compressing 
turns on a toroid.  This could possibly show ends of tuning ranges 
and indicate at which point one would need to add or subtract 
a turn versus where compressing or expanding turns could be 


On Sat, Sep 12, 2020 at 11:09 AM ajparent1/KB1GMX <kb1gmx@...> wrote:
Try one turn from L4 and spreading the turns for best output.

Please reply on list so we can share.
No private email, it goes to a bit bucket due address harvesting

14241 - 14260 of 66503