¿ªÔÆÌåÓý

ctrl + shift + ? for shortcuts
© 2025 Groups.io
  1. Nanovna-Users
  2. Topics

Plate RF Choke Measurements #measurement

 

Re visit this topic as I am looking at plate chokes for a tube amplifier.

I used 1010 points from 100 kHz to 30 MHz. There are 3 chokes shown here. The older National R175 shown in a prior post with its problems of 2 in band resonant conditions, a 230 uH homebrew choke for 160 - to - 40 meter. Fine but horrible beyond 15 MHz and finally a fairly nice choke at 100 uH on a ceramic core complete with a 500 pF/20 kV bypass capacitor. This choke shows a single resonance but is out of band.

It appears unnecessary to provide a short across the choke to pick up these loops if a shunt S11 measurement is used. In all cases the parallel resonate point crosses the open circuit side of the chart. A short across the choke simply rotates the reflection towards the short circuit side of the chart however the series loops stay in tact.

Alan

National_R175.png
Homebrew_230uH.png
RFC_GA1404C_500pF_20kV.png

 

Dear Alan,

thank you for sharing the measurements. I assume you used NanoVNA-saver.
Can you please share the s1p or s2p files if saved?

TNX, 73 Ady YO2NAA

On Fri, Dec 27, 2019 at 5:14 AM alan victor <avictor73@...> wrote:

Re visit this topic as I am looking at plate chokes for a tube amplifier.

I used 1010 points from 100 kHz to 30 MHz. There are 3 chokes shown here.
The older National R175 shown in a prior post with its problems of 2 in
band resonant conditions, a 230 uH homebrew choke for 160 - to - 40 meter.
Fine but horrible beyond 15 MHz and finally a fairly nice choke at 100 uH
on a ceramic core complete with a 500 pF/20 kV bypass capacitor. This choke
shows a single resonance but is out of band.

It appears unnecessary to provide a short across the choke to pick up
these loops if a shunt S11 measurement is used. In all cases the parallel
resonate point crosses the open circuit side of the chart. A short across
the choke simply rotates the reflection towards the short circuit side of
the chart however the series loops stay in tact.

Alan
 

Hi Ady.

I only did a screen capture. The measurement and cal are one port only. So I would have only S11 data available.

The main thrust of the post is you need to capture a reasonable number of data points, 10 bands for 100 kHz to 30 MHz was used. You can probably get by with much less. Perhaps 401. The series loops appear with no added short applied. The choke is measured in place on the chassis. Take care with this placement if you are concerned about a problematic loop near an operating band!

A one port measurement appears sufficient.

I can capture a S data file. Might be of interest if you desire some other form of display or post processing. Although this is all very nice, I am not parting with my GDO!

Alan
 

Hi Alan,

thank you for the info. I think it is a good thing to save the measurement
files, they are taking very little space and can be re-analyzed in the
future with various software tools.
Also can be used as reference for future measurements of the same DUT, to
see any modifications.
Actually I wanted to look at other graphs, for example S11 R and X, more
comfortable for me to see resonances.
I totally agree the choke should be measured in place.
For S11 measurement of course no shorting is needed, both ends are
connected to CH0. For GDO like measurement, shorting is needed to close the
circuit.
It would be interesting to compare S11 measurements with GDO and also with
NanoVNA in GDO mode (small coupling coil on CH0).

73 Ady

On Fri, Dec 27, 2019 at 12:08 PM alan victor <avictor73@...> wrote:

Hi Ady.

I only did a screen capture. The measurement and cal are one port only. So
I would have only S11 data available.

The main thrust of the post is you need to capture a reasonable number of
data points, 10 bands for 100 kHz to 30 MHz was used. You can probably get
by with much less. Perhaps 401. The series loops appear with no added short
applied. The choke is measured in place on the chassis. Take care with this
placement if you are concerned about a problematic loop near an operating
band!

A one port measurement appears sufficient.

I can capture a S data file. Might be of interest if you desire some other
form of display or post processing. Although this is all very nice, I am
not parting with my GDO!

Alan
 

Yes, agree.
I will re visit this and with another choke as I need one in the input, grid side of the amp.

73' Alan
 

I think the Rp resistance of the choke will need to be plotted vs frequency. This will give an insight into how much power will be dissipated in the choke. A good plate choke suitable for a 1kW amplifier might show an Rp >400k ohm across most of the HF band.
 

That is a nice inductor. Consider that a 100 uH L is used at that power level, the unloaded Q is near 100.

I'll return to look at some as I recall they certainly hug the chart edge so the Qul prior to looping is pretty good.

Alan

1 - 7 of 7