开云体育

David,
Would you share the details of that “very aggressive CMC” that you referenced? Sounds like a winner!
Gene – K1AVE


Sent from Mail<> for Windows 10

From: David Eckhardt<mailto:davearea51a@...>
Sent: Friday, January 22, 2021 7:27 PM
To: [email protected]<mailto:[email protected]>
Cc: NANO VNA<mailto:[email protected]>; [email protected]<mailto:[email protected]>
Subject: Re: [nanovna-users] [nanovna-f] EFFECT OF CMCs and/or BALUNS on RECEIVED NOISE

Remember, I have preamp #1 ON. Virtually all my local noise originates
from the SMPSs in our new appliances (2014). Our mountain home was burned
completely down in the High Park Fire of 2012 here in N. COlorado (we lost
everything except the two vehicles, a very few clothes, and our puppy). We
rebuilt down here. The mountain house was built before all the Chinese
SMPSs with no attention paid to EMC invaded everything (from China, with
Love). Today, ALL appliances have them and appliances are exempted from
FCC rules.

We recently had to replace a refrigerator. Old was Samsung - little RFI.
The new one is made in the USA, Hotpoint. Once installed and plugged in,
160 and 75 meters became useless. I built a very aggressive CMC and
installed it in the line cord to the new refrigerator. Solved the problem
completely, well, almost..... Every new appliance has multiple Chinese
SMPSs in them. All are RFI sources. It's been a 6-year battle on my part
to deal with the RFI. Not to mention another hobby is radio astronomy
which is even more sensitive to RFI.

With no preamps turned on in the Icom 7300 (or 7610), my noise floor is S-0
(-100 dBm) or lower most of the time during the day. On 160 and 75 toward
evening, it comes up a bit, but that is just cosmic and atmospheric noise.

Dave - W?LEV

On Sat, Jan 23, 2021 at 12:07 AM Joe WB9SBD <nss@...> wrote:

Hi Dave,

Why do you have such a HIGH noise floor at all?

Wow, It doesn't look low at all,

My noise floor if it is any higher than a S 2 or 3 I shut the radio off!

Joe WB9SBD

On 1/22/2021 5:54 PM, David Eckhardt wrote:

For those unbelievers in feedline chokes to properly treat CM energy on
feedlines, I made some quantitative tests using my HF antenna and Icom
7300. So,...... have a read of the short attachment.

I hope it leaves you with a belief in baluns (NOT transformers) and
CMCs.

Dave - W?LEV

PS: After some 3-years, I still have not removed the clear plastic
covering the surface of the touch screen.......

--
*Dave - W?LEV*
*Just Let Darwin Work*

I use CMCs on two antennas; a dipole and a 1/4wave 2 meter vertical with counterpoises. But I have a third antenna, an off center fed dipole with a balun. I have assumed that the balun will also function as a CMC. Do you agree?

On a slight off subject. I was surprised to see you state that the 7300 Smeter was 3dB per division. How did you determine that.

Thanks for the good information

Paul W8SBH

Remember, I have preamp #1 ON. Virtually all my local noise originates
from the SMPSs in our new appliances (2014). Our mountain home was burned
completely down in the High Park Fire of 2012 here in N. COlorado (we lost
everything except the two vehicles, a very few clothes, and our puppy). We
rebuilt down here. The mountain house was built before all the Chinese
SMPSs with no attention paid to EMC invaded everything (from China, with
Love). Today, ALL appliances have them and appliances are exempted from
FCC rules.

We recently had to replace a refrigerator. Old was Samsung - little RFI.
The new one is made in the USA, Hotpoint. Once installed and plugged in,
160 and 75 meters became useless. I built a very aggressive CMC and
installed it in the line cord to the new refrigerator. Solved the problem
completely, well, almost..... Every new appliance has multiple Chinese
SMPSs in them. All are RFI sources. It's been a 6-year battle on my part
to deal with the RFI. Not to mention another hobby is radio astronomy
which is even more sensitive to RFI.

With no preamps turned on in the Icom 7300 (or 7610), my noise floor is S-0
(-100 dBm) or lower most of the time during the day. On 160 and 75 toward
evening, it comes up a bit, but that is just cosmic and atmospheric noise.

Dave - W?LEV

--
*Dave - W?LEV*
*Just Let Darwin Work*

Mark,

Manfred (I like the way you think) mentioned,”Instead of calculating the
flux from the current, it's often more practical to calculate it directly from
the applied voltage, because doing so is independent from inductance. When
calculating flux from current, we need to know the actual core permeability,
which varies with frequency and with drive level. When calculating flux from
voltage, the effect of permeability cancels out, allowing us to calculate the
flux without knowing the actual permeability at the specific frequency and
drive level.”
I’m not sure I understand this. For a given voltage across a winding would
the current not be determined by the Reactance (Z), therefore making L, and u,
a very important variable in determining the flux?

Indeed for a given voltage the current will be determined by reactance, or more correctly by impedance, and the permeability of the core material is one of the factors determining it. So, if you compare two inductors that are identical except for the permeability of the core materials, then with the same voltage applied the inductor with the higher permeability material will have less current flowing in it.

But if you then calculate the flux densities from these currents, you will again have to factor in the permeabilities of the cores. The inductor with the higher permeability material will get more flux per current, and since the current in it is lower, it cancels out the difference caused by current, and the flux in both cores will be identical.

That's what I meant when I wrote that when calculating flux directly from voltage, the permeability cancels out, and so we don't need to know the permeability nor the inductance, reactance nor impedance. For a core of given dimensions, with a given number of turns, and a given voltage at a given frequency, the flux density is independent from the permeability.

Ferrites are the misunderstood red-headed step child of the hobby and I
feel obligated to lobby for their liberal and proper application. :) Your
thoughts?

They are just one among many misunderstood or poorly understood things in the hobby!

Other examples of misunderstood things are push-pull linear power amplifiers: Most people think that they operate in class AB, while their true operating mode is more often a sort of dynamic class A, which is easy to implement with BJTs but more difficult to get right with MOSFETs. It explains the typical 45% efficiency obtained, when true class AB should give something close to 70%. Those amplifiers that get better efficiency than 45% usually run in something close to current-switching class D, or a hybrid class between them.

At a much lower level, I keep hearing hams giving lessons about the level to which you need to set the carrier control of a radio for optimal SSB transmission, or hams reporting that they have a 59 level of noise... Oh well.

It's a hobby. Most hams aren't professionals in any radio-related matter. Each learns to the level he can, over time.

What's scary, though, is how many professionals don't fully understand what they are doing! But that's too part of how this world is. Humans just aren't perfect, and just can't know everything. Everybody pushes his limits and sometimes gets into terrain he doesn't fully master yet. Anyone who decides to only do things he perfectly understands down to the finest details, is condemned to a life of inactivity.

Excuse me for that bit of friday evening philosophy!

Manfred

For those unbelievers in feedline chokes to properly treat CM energy on
feedlines, I made some quantitative tests using my HF antenna and Icom
7300. So,...... have a read of the short attachment.

I hope it leaves you with a belief in baluns (NOT transformers) and CMCs.

Dave - W?LEV

PS: After some 3-years, I still have not removed the clear plastic
covering the surface of the touch screen.......

Found my problem. Screen protector was wadded under case. Removed the
front, cleared out the cover, boots fine.

Thanks!

On Fri, Jan 22, 2021, 4:55 PM Michael Brickell VE3TKI <ve3tki@...>
wrote:

I have 3 laptops, two running Windows 10, and one running Window 7. I was
able to install the v1.0.39 firmware for my H4 nanoVNA on the Windows 7
laptop but not the Windows 10 ones. When I ran the DFuse v3.0.6 program I
did not see anything in the Available DFU Devices List. The problem was
that the bootloader driver did not install with the DFuse Demo v3.06
program.

I then downloaded the STM Bootloader Driver package from
, and
installed the amd64.exe driver on the Windows 10 laptops. Having done
this I found that when the nanoVNA is placed in DFU mode and connected via
the USB cable, an “STM Device in DFU Mode” appears in the list of USB
Controllers in Device Manager.

When I run the DFuse Demo v3.0.6 program I now see “STM Device in DFU
Mode” in the Available DFU Devices List , and the new firmware installs
properly.

Michael Brickell VE3TKI

Does not help.

Try tapping the screen anywhere. I have no idea if this will work, but it
may be required for proper boot.

Dave - W?LEV

--
*Dave - W?LEV*
*Just Let Darwin Work*

I have 3 laptops, two running Windows 10, and one running Window 7. I was able to install the v1.0.39 firmware for my H4 nanoVNA on the Windows 7 laptop but not the Windows 10 ones. When I ran the DFuse v3.0.6 program I did not see anything in the Available DFU Devices List. The problem was that the bootloader driver did not install with the DFuse Demo v3.06 program.

I then downloaded the STM Bootloader Driver package from , and installed the amd64.exe driver on the Windows 10 laptops. Having done this I found that when the nanoVNA is placed in DFU mode and connected via the USB cable, an “STM Device in DFU Mode” appears in the list of USB Controllers in Device Manager.

When I run the DFuse Demo v3.0.6 program I now see “STM Device in DFU Mode” in the Available DFU Devices List , and the new firmware installs properly.

Michael Brickell VE3TKI

Mark pretty much covered why stacking cores of differing materials is a
really bad idea. I've never tried it from day one as I understand the
problems which Mark addressed. However, connecting CMC of differring
materials in series has can and is done in practice. Certainly I've found
over the last month of building and measuring CMCs for my applications that
the 31 material truly is better as an all-around core material for HF
applications. 42 material just does not have the ?r (or ?i) to properly
address 160-meters with few enough windings to address 75 through 10
meters.

Please note that I started (long ago) with RG-142 wound through large cores
of 43 material. RG-142 is the 'high power' coax made from Teflon and
silver coated conductive components and is generally light brown on the
outside. That approach is fine for coaxially fed antennas, but I use
parallel wire feeders for a multitude of reasons. I require a 'true' CMC
made of bifilar winding with no twists or core cross-overs on the toroids
to assure good choking impedance to CM currents AND assure proper
differential amplitudes and phases on the DM side of the choke. Therefore,
none of those I've previously put out on this group are of the coax
variety. All are bifilar windings on the cores as noted in the table. My
measured data (using the HP 8753C with associated s-parameter test set)
does NOT apply to chokes (current chokes) consisting of coax wound on the
cores.

I'll attach the latest 'Choke Table' as it stood last evening.

I just received my 4" OD 31 material cores from KF7P Metalworks in Utah.
More data coming........

Dave - W?LEV

On Fri, Jan 22, 2021 at 7:53 PM Max via groups.io <kg4pid=
[email protected]> wrote:

Could you increase the turns and re-measure? I'd like to see what happens
since I need a CMC choke for a milti band antenna.
These cores were stacked correct? I wonder about two in series, a 31 and a
43. Would there be a difference compared to stacking?
Max KG4PID
On Friday, January 22, 2021, 10:19:25 AM CST, Jeff Anderson <
jca1955@...> wrote:

On Thu, Jan 21, 2021 at 01:36 PM, Miro, N9LR wrote:

That would be asking for trouble, like having BJT (bipolar transistors)

in

parallel. On different frequencies, temperatures, flux one is destined

to take

more then the other, and that one heats even more and takes even more on
itself

I'm wondering if there is any empirical data or theoretical derivations
showing that paralleling ferrite cores of different materials is a bad idea.

From a heating perspective, it seems to me that if calculations show
either core, stand-alone, does not overheat in the spec'd application, then
there should not be a heating issue if the two cores are combined (unless
overall common-mode resistance *drops* below the common-mode resistance
presented by either core itself, but then what would be the point of
paralleling the two cores?)

From my own measurements, the common-mode resistance of paralleled 43 and
31 cores looks pretty good (see the attached chart). From this measured
data (taken on an HP 8753C) I don't see any reason why these two cores
should not be paralleled. If anyone has different data, I would very much
like to see it. After all, it's always possible my measurements are in
error, or that I'm looking at the issue incorrectly.

Thanks,

- Jeff, k6jca

--
*Dave - W?LEV*
*Just Let Darwin Work*

On Fri, Jan 22, 2021 at 08:30 AM, Jim Lux wrote:

You've got two cores with different permittivities, so the flux will
distribute unequally. So the real question is whether there are "pathological"
frequencies and power levels that might get you into trouble with saturation
or loss.

Thanks very much for the reply, Jim.

Ideally, the question of what happens when a Mix 31 and a Mix 43 core are bound together and turns wrapped through them would be answered by first solving appropriate equations and then verifying with measurements.

But I’m not sure what the equations would be (although they are probably derivable by someone smarter than I). So I skip forward to making measurements.

And as my measurements show, there doesn’t seem to be any surprises in Common-mode Z if 4 turns of RG-8 are run through the combined cores, at least over the HF range.

Of course, take my measurements with a block of salt. Others should verify for themselves, if they are thinking of trying something like this.

- Jeff, k6jca

Could you increase the turns and re-measure? I'd like to see what happens since I need a CMC choke for a milti band antenna.
These cores were stacked correct? I wonder about two in series, a 31 and a 43. Would there be a difference compared to stacking?
Max KG4PID

Use the search function in the forum messages and you'll find my mesg with the link - I don't recall it offhand

Hi,
Where do I find the above referenced "NanoVNA-App", I do not see this in the FILES section. I assume that this software makes it easy for a non-computer person to do the Firmware Update.
Glen Jenkins, WB4KTF, Austin, TX

Mark, KA2QFX:

Thank you so much for your clarifications. Hopefully many will read it and
gain a bit of knowledge and apply it over snake oil and hearsay! I'll
confess I'm a bit jaded by the lack of technical knowledge among the
average amateur population (licensed for 61 years and a degree in
physics). And.....there is more snake oil in the subjects of antennas and
feedlines than any other subject applicable to the hobby. Our local 'Tech
Net' held weekly on our 70-cm repeater used to be technical. Over the last
year, nothing but digital this digital that and wow-woopy digital
everything, all of which could be learned by just googling.

To quote your email:
" Ferrites are the misunderstood red-headed step child of the hobby
and I feel obligated to lobby for their liberal and proper
application. :) Your thoughts?"
Here are my thoughts:

I strongly second that statement! Not only ferrites, but antennas and
transmission lines in general. You have more working knowledge than I do
regarding magnetics and flux in the cores (not that I once went there in
college). But I do understand what you are addressing. I believe if
anyone took the time and effort to read, re-read, and understand what DJ0IP
has so well written about baluns, feedlines, and antennas (generally with
no math to scare off others), these concepts would no longer be so
mysterious and magical. I would propose augmenting your statement to
include antennas and transmission lines of which ferrites should be an
intimate part.

At least with the advent of affordable VNAs, many are learning the Smith
Chart and the 'other half' of the impedance besides just resistance and SWR
(oh.....that's yet another totally misunderstood concept). Some are even
beginning to understand Ohm's Law! When these units were introduced, I
viewed them as a highly affordable alternative to the expensive test
equipment from HP/Agilent/Keysite and R&S. They indeed are an extremely
useful educational tool as well. That's the purpose of this group. I used
to work with one of the sponsors of these NANOVNA groups. Once again,
thank you. Also to the credit of those in China who made these NANOVNAs
available, a huge thank you!!!!

BTW, the site I referred to can be accessed at (I've previously referred to
it many time within this group and others) :


Dave - W?LEV

--
*Dave - W?LEV*
*Just Let Darwin Work*

On 1/22/21 8:30 AM, Jim Lux wrote:

I think you'd need to look at the magnetic circuits. You've got two cores with different permittivities, so the flux will distribute unequally. So the real question is whether there are "pathological" frequencies and power levels that might get you into trouble with saturation or loss.

This is sort of different than, say, a core made of a mixture of the two ferrites.

Imagine a rope and a piano wire in parallel, with the rope being very slightly shorter in the no load condition. As you load it, the pianowire will take more and more of the load, since it's got a higher spring constant. Now make it an oscillatory load and both the rope and the wire being lossy.? It's not obvious at first glance what the behavior will be.

For anyone really wanting to get into this, find a copy of E.C. Snelling's "Soft Ferrites", preferably the 2nd edition. He probably covers mixed cores somewhere in there.

The first edition is available at archive.org


but the second edition is really the one you want.

On 1/22/21 8:19 AM, Jeff Anderson wrote:

On Thu, Jan 21, 2021 at 01:36 PM, Miro, N9LR wrote:

That would be asking for trouble, like having BJT (bipolar transistors) in
parallel. On different frequencies, temperatures, flux one is destined to take
more then the other, and that one heats even more and takes even more on
itself

I'm wondering if there is any empirical data or theoretical derivations showing that paralleling ferrite cores of different materials is a bad idea.

From a heating perspective, it seems to me that if calculations show either core, stand-alone, does not overheat in the spec'd application, then there should not be a heating issue if the two cores are combined (unless overall common-mode resistance *drops* below the common-mode resistance presented by either core itself, but then what would be the point of paralleling the two cores?)

From my own measurements, the common-mode resistance of paralleled 43 and 31 cores looks pretty good (see the attached chart). From this measured data (taken on an HP 8753C) I don't see any reason why these two cores should not be paralleled. If anyone has different data, I would very much like to see it. After all, it's always possible my measurements are in error, or that I'm looking at the issue incorrectly.

Thanks,

- Jeff, k6jca

I think you'd need to look at the magnetic circuits. You've got two cores with different permittivities, so the flux will distribute unequally. So the real question is whether there are "pathological" frequencies and power levels that might get you into trouble with saturation or loss.

This is sort of different than, say, a core made of a mixture of the two ferrites.

Imagine a rope and a piano wire in parallel, with the rope being very slightly shorter in the no load condition. As you load it, the pianowire will take more and more of the load, since it's got a higher spring constant. Now make it an oscillatory load and both the rope and the wire being lossy.? It's not obvious at first glance what the behavior will be.

For anyone really wanting to get into this, find a copy of E.C. Snelling's "Soft Ferrites", preferably the 2nd edition. He probably covers mixed cores somewhere in there.

On Thu, Jan 21, 2021 at 01:36 PM, Miro, N9LR wrote:

That would be asking for trouble, like having BJT (bipolar transistors) in
parallel. On different frequencies, temperatures, flux one is destined to take
more then the other, and that one heats even more and takes even more on
itself

I'm wondering if there is any empirical data or theoretical derivations showing that paralleling ferrite cores of different materials is a bad idea.

From a heating perspective, it seems to me that if calculations show either core, stand-alone, does not overheat in the spec'd application, then there should not be a heating issue if the two cores are combined (unless overall common-mode resistance *drops* below the common-mode resistance presented by either core itself, but then what would be the point of paralleling the two cores?)

From my own measurements, the common-mode resistance of paralleled 43 and 31 cores looks pretty good (see the attached chart). From this measured data (taken on an HP 8753C) I don't see any reason why these two cores should not be paralleled. If anyone has different data, I would very much like to see it. After all, it's always possible my measurements are in error, or that I'm looking at the issue incorrectly.

Thanks,

- Jeff, k6jca

Some thoughts on Common Mode Chokes and Ferrites

In reading the preceding responses I noted some comments that are easily open to misinterpretation. I’d like to address these, as I feel the application of ferrite devices is valuable commodity in ham radio that too often gets bad press because of misapplication and misnomer.

First off, someone inferred that for a given magnetizing current flux density is fixed no matter what the permeability (mu) of the ferrite core. This is incorrect. The H field intensity, which is proportional to the ampere-turns of the inducing coil IS fixed for a given current and turns number. However, the flux density produced (B) by that H field varies with u and cross sectional area, and inversely with magnetic path length of the magnetic material within the field. Therefore, the inductive reactance will be greater for increased u, smaller cross section and shorter length cores. Detrimentally, this will also lead to lower saturation levels, core heating, etc. Although it sounds like W0LEV has employed more than sufficient core material to avoid pushing those limits.
But it is important to make that distinction that as u increases to provide more effective inductance, whether for coupling transformer windings or creating CMC reactance, the losses and frequency responses also vary in a compromising manner. This forces us to be very critical in selecting cores best suited to their power and frequency demands.

Manfred (I like the way you think) mentioned,”Instead of calculating the flux from the current, it's often more practical to calculate it directly from the applied voltage, because doing so is independent from inductance. When calculating flux from current, we need to know the actual core permeability, which varies with frequency and with drive level. When calculating flux from voltage, the effect of permeability cancels out, allowing us to calculate the flux without knowing the actual permeability at the specific frequency and drive level.”
I’m not sure I understand this. For a given voltage across a winding would the current not be determined by the Reactance (Z), therefore making L, and u, a very important variable in determining the flux?

Dave, W0LEV, spoke pf de-Qing his cores with resistance across the windings to flatten and improve the analytical value of his measurements. This sounded less than straight-forward and piqued my curiosity. Here’s what I think may actually be going on. When designing RF current transformers for measurement devices I noted that small single turn primaries induced reasonable voltages in a high turn (10:1, 20:1,…) secondary. If we simply rectified that secondary voltage there were non-linearities well beyond what the rectifier drop would induce. We also noted that the cores were getting hot at higher power levels even though currents were obstensively very low. It turns out that if the secondaries are resistively loaded (20-~100 ohms) to produce more current than simply rectifying, that this secondary current produced sufficient loading in the core such that is was much more linear with power change and ran cold. This was with HF frequencies using 47, 31 AND 77 cores iirc. Perhaps this is what Dave is seeing. While the effective LOAD of a feedline outer shield is certainly unpredictable it may play a critical role in this regard. Certainly if the feedline appears as an open (shorted ? wave) at some frequency this could be a problem as the CMC is essentially unloaded.

Lastly, and this is a little off topic, but it has been my experience that ferrite devices (BALUNS, UNUNs, Transmission Line Transformers of all kinds) get way too much defamatory critique by folks who just don’t know how and why to use them. To wit, they really don’t like working in reactive systems or being employed outside of their designed impedance.
Using a 9:1 balun to feed 50 ohms to a 450 ohm line isn’t going to work well when that line is terminated in a 72 ohm dipole. Or a 1:1 balun feeding a dipole isn’t going to play nice when you try to operate that antenna somewhere where it’s highly reactive. If we apply these devices properly they work very well indeed. I use a 4:1 UNUN to feed an 80 meter mobile antenna. It’s an amazing, relatively broadband matching device requiring little to know matching over a 100KHz, at which point I move the tap on the antenna to shift it’s resonant (Low Z) point. I believe from it’s relative performance it is significantly more efficient that having a de-tuned (High Z) antenna using a lot of matching reactance at it’s feedpoint. But YMMV. Ferrites are the misunderstood red-headed step child of the hobby and I feel obligated to lobby for their liberal and proper application. :) Your thoughts?

On Thu, Jan 21, 2021 at 12:02 PM, Rud61 wrote:

latest model of the nanovna which one?

version F or V2 or ? max frequency ?

and what is the name of the version with the large lcd display ?

Since you don't say why you want one, or what you would use it for, here's my thoughts.

If you are going to use it outdoors, get the NanoVNA-F. (There is just 'one' to choose from.)
Frequency? Only you know that. (for example, I have no need or use for GHz tests.)
TWO versions have the large display, -H4 and -F.
My "OPINION" is - - get the NanoVNA-F, from BH5HNU's dealer Deepelec or from a U.S. ham store.

I have an early, cheap clone 2.8 inch, with no shielding, and a decent -F clone. I like and use them both, but the -F is useful outdoors, where the antennas are. And can store s1p files, although I don't need that (yet).

You did not mention price in your question, so the best thing to do is buy one of every model. Then tell us which is best! (And send me the leftovers.)

No matter which model you order, be SURE to also get connector adapters (both male and female) for every kind of RF connector you have. And patch cables. I have had to order adapters for UHF, BNC, F and MCX. And a binding posts adapter like the Pomona 1296 can be very useful.

ANY NanoVNA is better than none. It's hard to go wrong.
--
Doug, K8RFT