Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > sci.electronics.design > #464429 > unrolled thread

EMI problems

Back to article view | Back to sci.electronics.design

Contents

  EMI problems John Larkin <jjlarkin@highland_snip_technology.com> - 2017-06-14 16:09 -0700
    Re: EMI problems Martin Riddle <martin_ridd@verizon.net> - 2017-06-14 20:05 -0400
      Re: EMI problems John Larkin <jjlarkin@highland_snip_technology.com> - 2017-06-14 18:01 -0700
        Re: EMI problems o pere o <me@somewhere.net> - 2017-06-15 09:28 +0200
          Re: EMI problems Phil Hobbs <pcdhSpamMeSenseless@electrooptical.net> - 2017-06-15 10:59 -0400
            Re: EMI problems Don Kuenz <g@crcomp.net> - 2017-06-21 02:53 +0000
          Re: EMI problems John Larkin <jjlarkin@highlandtechnology.com> - 2017-06-15 07:59 -0700
            Re: EMI problems George Herold <gherold@teachspin.com> - 2017-06-15 08:07 -0700
              Re: EMI problems Phil Hobbs <pcdhSpamMeSenseless@electrooptical.net> - 2017-06-15 11:12 -0400
                Re: EMI problems George Herold <gherold@teachspin.com> - 2017-06-15 08:42 -0700
              Re: EMI problems John Larkin <jjlarkin@highlandtechnology.com> - 2017-06-15 09:38 -0700
            Re: EMI problems Robert Baer <robertbaer@localnet.com> - 2017-06-16 22:06 -0800
    Re: EMI problems Jon Elson <jmelson@wustl.edu> - 2017-06-15 14:14 -0500
      Re: EMI problems John Larkin <jjlarkin@highland_snip_technology.com> - 2017-06-15 13:55 -0700
        Re: EMI problems Lasse Langwadt Christensen <langwadt@fonz.dk> - 2017-06-15 14:02 -0700
        Re: EMI problems Phil Hobbs <pcdhSpamMeSenseless@electrooptical.net> - 2017-06-16 12:43 -0400
          Re: EMI problems John Larkin <jjlarkin@highland_snip_technology.com> - 2017-06-16 10:11 -0700
            Re: EMI problems Phil Hobbs <pcdhSpamMeSenseless@electrooptical.net> - 2017-06-16 13:24 -0400
        Re: EMI problems Jeff Liebermann <jeffl@cruzio.com> - 2017-06-16 10:37 -0700
        Re: EMI problems Jon Elson <jmelson@wustl.edu> - 2017-06-16 13:31 -0500
          Re: EMI problems Phil Hobbs <pcdhSpamMeSenseless@electrooptical.net> - 2017-06-16 14:46 -0400
            Re: EMI problems John Larkin <jjlarkin@highland_snip_technology.com> - 2017-06-16 12:25 -0700
            Re: EMI problems George Herold <gherold@teachspin.com> - 2017-06-16 13:07 -0700
          Re: EMI problems John Larkin <jjlarkin@highland_snip_technology.com> - 2017-06-16 12:22 -0700
      Re: EMI problems upsidedown@downunder.com - 2017-06-16 09:35 +0300
        Re: EMI problems John Larkin <jjlarkin@highlandtechnology.com> - 2017-06-16 06:40 -0700
    Re: EMI problems Joerg <news@analogconsultants.com> - 2017-06-15 12:27 -0700
      Re: EMI problems John Larkin <jjlarkin@highland_snip_technology.com> - 2017-06-15 12:47 -0700
        Re: EMI problems Lasse Langwadt Christensen <langwadt@fonz.dk> - 2017-06-15 12:57 -0700
          Re: EMI problems Joerg <news@analogconsultants.com> - 2017-06-15 13:12 -0700
        Re: EMI problems Joerg <news@analogconsultants.com> - 2017-06-15 13:03 -0700
      Re: EMI problems "Lund-Nielsen, Jorgen" <jorgen.lund-nielsen@xyz123desy.de> - 2017-06-16 08:58 +0200
    Re: EMI problems Jeff Liebermann <jeffl@cruzio.com> - 2017-06-16 10:03 -0700
      Re: EMI problems John Larkin <jjlarkin@highland_snip_technology.com> - 2017-06-16 11:09 -0700
        Re: EMI problems Jeff Liebermann <jeffl@cruzio.com> - 2017-06-16 12:06 -0700
          Re: EMI problems John Larkin <jjlarkin@highland_snip_technology.com> - 2017-06-16 12:34 -0700
            Re: EMI problems Jeff Liebermann <jeffl@cruzio.com> - 2017-06-17 09:22 -0700
              Re: EMI problems John Larkin <jjlarkin@highlandtechnology.com> - 2017-06-17 10:08 -0700
                Re: EMI problems John Larkin <jjlarkin@highlandtechnology.com> - 2017-06-17 11:07 -0700
                  Re: EMI problems Tauno Voipio <tauno.voipio@notused.fi.invalid> - 2017-06-17 21:30 +0300
                  Re: EMI problems pcdhobbs@gmail.com - 2017-06-17 16:34 -0700
                  Re: EMI problems krw@notreal.com - 2017-06-17 20:44 -0400
                  Re: EMI problems George Herold <gherold@teachspin.com> - 2017-06-19 06:03 -0700
                Re: EMI problems Jeff Liebermann <jeffl@cruzio.com> - 2017-06-18 12:49 -0700
                  Re: EMI problems John Larkin <jjlarkin@highlandtechnology.com> - 2017-06-18 13:42 -0700
                    Re: EMI problems Jeff Liebermann <jeffl@cruzio.com> - 2017-06-18 15:50 -0700
                      Re: EMI problems John Larkin <jjlarkin@highlandtechnology.com> - 2017-06-18 17:15 -0700
                        Re: EMI problems upsidedown@downunder.com - 2017-06-19 09:39 +0300
                          Re: EMI problems upsidedown@downunder.com - 2017-06-19 09:48 +0300
                            Re: EMI problems Phil Hobbs <pcdhSpamMeSenseless@electrooptical.net> - 2017-06-19 12:45 -0400
      Re: EMI problems "John Miles, KE5FX" <jmiles@gmail.com> - 2017-06-16 15:30 -0700
        Re: EMI problems John Larkin <jjlarkin@highland_snip_technology.com> - 2017-06-16 18:10 -0700
          Re: EMI problems Phil Hobbs <pcdhSpamMeSenseless@electrooptical.net> - 2017-06-17 11:08 -0400
        Re: EMI problems upsidedown@downunder.com - 2017-06-17 14:25 +0300
        Re: EMI problems Jeff Liebermann <jeffl@cruzio.com> - 2017-06-17 09:35 -0700
          Re: EMI problems John Larkin <jjlarkin@highlandtechnology.com> - 2017-06-17 10:42 -0700
    Re: EMI problems upsidedown@downunder.com - 2017-06-17 14:45 +0300
      Re: EMI problems John Larkin <jjlarkin@highlandtechnology.com> - 2017-06-17 08:28 -0700

Page 3 of 3 — ← Prev page 1 2 [3]

#464887

>Hmmm, how does one do a tracking generator on an FFT-based spectrum 
>analyzer? 

One of your Grekhov diode pulsers would be about right. Of course that torture-tests the assumptions of linear systems theory a bit... 

cheers

Phil Hobbs

[toc] | [prev] | [next] | [standalone]

#464893

On Sat, 17 Jun 2017 11:07:38 -0700, John Larkin
<jjlarkin@highlandtechnology.com> wrote:

>On Sat, 17 Jun 2017 10:08:39 -0700, John Larkin
><jjlarkin@highlandtechnology.com> wrote:
>
>>On Sat, 17 Jun 2017 09:22:20 -0700, Jeff Liebermann <jeffl@cruzio.com>
>>wrote:
>>
>>>On Fri, 16 Jun 2017 12:34:02 -0700, John Larkin
>>><jjlarkin@highland_snip_technology.com> wrote:
>>>
>>>>There's also a huge amount of lower-frequency noise here, probably
>>>>from AC-line powered things in the neighborhood.
>>>
>>>I see the same thing even in a residential neighborhood in the middle
>>>of nowhere.  If I unplug various AC powered gadgets around the house,
>>>the noise level goes down a little.  I can see some of it with a scope
>>>connected to the power line through an LISN (Line Impedance
>>>Stabilization Network).  The really nasty stuff disappears when the
>>>sun goes down, which suggests a grid connected solar power system. The
>>>problem is that I live in the deep dark forest where such systems are
>>>rare.  Nothing within a mile radius but I'll keep looking.  A line
>>>filter usually gets rid of most such noise.  If you're located in an
>>>industrial area, you might want to investigate what the local PV
>>>systems are putting back into the power lines.
>>>
>>>>This motor controller was making 20 volt spikes in the lab, until we
>>>>hacked it.
>>>>https://dl.dropboxusercontent.com/u/53724080/Otis/EMI/VFD.JPG
>>>
>>>Neatly done.  No capacitors to form a low pass filter?
>>
>>The box on the right is a commercial filter, on the input to the
>>controller. That helped some. I hacked the toroids in series with the
>>outputs, running to the blower on the roof, and that got us down to
>>the local noise floor. It's all in metal boxes and conduit!
>>
>>
>>>
>>>>Here's the general spectrum:
>>>>https://dl.dropboxusercontent.com/u/53724080/Otis/EMI/300MHz.JPG
>>>>-28 dBm with just a Pomona lead antenna.
>>>
>>>30 MHz/div.  TV Ch 2 on the left.  FM broadcast band 88-108MHz seems
>>>quite strong.  One aircraft beacon.  Starting at 150Mhz, quite a bit
>>>of VHF commercial and public safety signals which quit at about 190MHz
>>>with TV channel 7.  Methinks if you check the UHF TV and commercial
>>>spectrum, the signal levels will higher.
>>
>>I still want to light an LED from the ambient RF.
>>
>>>
>>>-28dBm is 9mv rms or 25mv p-p into 50 ohms.  You should see a 25mv
>>>"noise" level on any long wires that can act as an antenna.  If you're
>>>working with low impedances, not a problem.  If you have bandwidth in
>>>the FM/VHF region at high impedances, I suggest a screen room, screen
>>>box, or differential designs.
>>
>>We're moving soon, about the same distance from the tower, and we have
>>considered some sort of screen room in the basement. But the EMI does
>>automagically test our breadboards and products for us.
>>
>>>
>>>BTW, nice Marconi/IFR/Aeroflex/Cobham 2399c spectrum analyzer.  I'm
>>>jealous.
>>
>>It's horrible and buggy. It has +-120 Hz spurs, which is like finding
>>a roach in your salad.
>>
>>
>>
>>
>>>
>>>>>You might want to calculate the resonant frequency of the Royal Dansk
>>>>>cookie can:
>>>>><http://www.rfwireless-world.com/calculators/circular-cavity-resonator-calculator.html>
>>>>>If you see a noise peak in that area, it could be something
>>>>>oscillating or filtering using the can as a resonator.  
>>>
>>>>I'm OK at 100 MHz. It might make a resonant cavity filter or something
>>>>at some higher frequency.
>>>
>>>Maybe.  I've built and seen VHF (150MHz) cavity filters in much
>>>smaller cans.  Fruit juice cans, paint cans, beer cans, etc all can be
>>>made to work.  
>>><http://vk4zxi.blogspot.com/2014/08/homebrew-cavity-resonatorduplexer-for.html>
>>>All it takes is a center rod between the two ends with a tuning
>>>capacitor at one end.  A length of insulated wire, soldered to one end
>>>of the can, and leaning against the top, forming a capacitor, is all
>>>it takes.  Small cavities have much less Q, but they still will
>>>resonate.  I don't have any cookie cans available.  I'll get some
>>>dimensions at the market and see what happens.  
>>>
>>>Or, you could just frequency sweep the cookie can and see what you
>>>get.  Just sacrifice some more Danish butter cookies to a worthy
>>>cause.
>>
>>Do I have to?
>>
>>The SA has a tracking generator; that might be fun.
>
>Hmmm, how does one do a tracking generator on an FFT-based spectrum
>analyzer?

An impulse (a shitload of signal averaging) and a deconvolution.

[toc] | [prev] | [next] | [standalone]

#465011

On Saturday, June 17, 2017 at 2:07:44 PM UTC-4, John Larkin wrote:
> On Sat, 17 Jun 2017 10:08:39 -0700, John Larkin
> <jjlarkin@highlandtechnology.com> wrote:
> 
> >On Sat, 17 Jun 2017 09:22:20 -0700, Jeff Liebermann <jeffl@cruzio.com>
> >wrote:
> >
> >>On Fri, 16 Jun 2017 12:34:02 -0700, John Larkin
> >><jjlarkin@highland_snip_technology.com> wrote:
> >>
> >>>There's also a huge amount of lower-frequency noise here, probably
> >>>from AC-line powered things in the neighborhood.
> >>
> >>I see the same thing even in a residential neighborhood in the middle
> >>of nowhere.  If I unplug various AC powered gadgets around the house,
> >>the noise level goes down a little.  I can see some of it with a scope
> >>connected to the power line through an LISN (Line Impedance
> >>Stabilization Network).  The really nasty stuff disappears when the
> >>sun goes down, which suggests a grid connected solar power system. The
> >>problem is that I live in the deep dark forest where such systems are
> >>rare.  Nothing within a mile radius but I'll keep looking.  A line
> >>filter usually gets rid of most such noise.  If you're located in an
> >>industrial area, you might want to investigate what the local PV
> >>systems are putting back into the power lines.
> >>
> >>>This motor controller was making 20 volt spikes in the lab, until we
> >>>hacked it.
> >>>https://dl.dropboxusercontent.com/u/53724080/Otis/EMI/VFD.JPG
> >>
> >>Neatly done.  No capacitors to form a low pass filter?
> >
> >The box on the right is a commercial filter, on the input to the
> >controller. That helped some. I hacked the toroids in series with the
> >outputs, running to the blower on the roof, and that got us down to
> >the local noise floor. It's all in metal boxes and conduit!
> >
> >
> >>
> >>>Here's the general spectrum:
> >>>https://dl.dropboxusercontent.com/u/53724080/Otis/EMI/300MHz.JPG
> >>>-28 dBm with just a Pomona lead antenna.
> >>
> >>30 MHz/div.  TV Ch 2 on the left.  FM broadcast band 88-108MHz seems
> >>quite strong.  One aircraft beacon.  Starting at 150Mhz, quite a bit
> >>of VHF commercial and public safety signals which quit at about 190MHz
> >>with TV channel 7.  Methinks if you check the UHF TV and commercial
> >>spectrum, the signal levels will higher.
> >
> >I still want to light an LED from the ambient RF.
> >
> >>
> >>-28dBm is 9mv rms or 25mv p-p into 50 ohms.  You should see a 25mv
> >>"noise" level on any long wires that can act as an antenna.  If you're
> >>working with low impedances, not a problem.  If you have bandwidth in
> >>the FM/VHF region at high impedances, I suggest a screen room, screen
> >>box, or differential designs.
> >
> >We're moving soon, about the same distance from the tower, and we have
> >considered some sort of screen room in the basement. But the EMI does
> >automagically test our breadboards and products for us.
> >
> >>
> >>BTW, nice Marconi/IFR/Aeroflex/Cobham 2399c spectrum analyzer.  I'm
> >>jealous.
> >
> >It's horrible and buggy. It has +-120 Hz spurs, which is like finding
> >a roach in your salad.
> >
> >
> >
> >
> >>
> >>>>You might want to calculate the resonant frequency of the Royal Dansk
> >>>>cookie can:
> >>>><http://www.rfwireless-world.com/calculators/circular-cavity-resonator-calculator.html>
> >>>>If you see a noise peak in that area, it could be something
> >>>>oscillating or filtering using the can as a resonator.  
> >>
> >>>I'm OK at 100 MHz. It might make a resonant cavity filter or something
> >>>at some higher frequency.
> >>
> >>Maybe.  I've built and seen VHF (150MHz) cavity filters in much
> >>smaller cans.  Fruit juice cans, paint cans, beer cans, etc all can be
> >>made to work.  
> >><http://vk4zxi.blogspot.com/2014/08/homebrew-cavity-resonatorduplexer-for.html>
> >>All it takes is a center rod between the two ends with a tuning
> >>capacitor at one end.  A length of insulated wire, soldered to one end
> >>of the can, and leaning against the top, forming a capacitor, is all
> >>it takes.  Small cavities have much less Q, but they still will
> >>resonate.  I don't have any cookie cans available.  I'll get some
> >>dimensions at the market and see what happens.  
> >>
> >>Or, you could just frequency sweep the cookie can and see what you
> >>get.  Just sacrifice some more Danish butter cookies to a worthy
> >>cause.
> >
> >Do I have to?
> >
> >The SA has a tracking generator; that might be fun.
> 
> Hmmm, how does one do a tracking generator on an FFT-based spectrum
> analyzer?
> 
Our SRS SA has a chirp output, which does a sine wave centered in each of 
the 512* bins of the FFT, and diddles the phases in some way so that the 
amplitude doesn't get out of control.  

George H. 
*there might only be 400 bins?     

> 
> -- 
> 
> John Larkin         Highland Technology, Inc
> 
> lunatic fringe electronics

[toc] | [prev] | [next] | [standalone]

#464973

On Sat, 17 Jun 2017 10:08:39 -0700, John Larkin
<jjlarkin@highlandtechnology.com> wrote:

>I still want to light an LED from the ambient RF.

>>-28dBm is 9mv rms or 25mv p-p into 50 ohms.

Well, your spectrum analyzer shows 9 mv rms into 50 ohms, which is:
  P = E^2/R
  P = (9*10^-3)^2 / 50 = 
  P = 1.62 microwatts
Ummm... I think you'll need a bigger antenna.

>We're moving soon, about the same distance from the tower, and we have
>considered some sort of screen room in the basement. But the EMI does
>automagically test our breadboards and products for us.

That's fine as long as you don't require the EMI or RFI for the
circuit to operate.  

>>BTW, nice Marconi/IFR/Aeroflex/Cobham 2399c spectrum analyzer.  I'm
>>jealous.
>
>It's horrible and buggy. It has +-120 Hz spurs, which is like finding
>a roach in your salad.

Sigh.  All my various spectrum analyzers are ancient.  None of them
are FFT type.  None of them have any kind of storage.
<http://802.11junk.com/jeffl/pics/home/slides/test-equip-mess.html>
I'm looking for something better, but have some other priority
purchases to deal with first.  Incidentally, the IFR-1500 in the
picture has a really awful power supply.

>>Or, you could just frequency sweep the cookie can and see what you
>>get.  Just sacrifice some more Danish butter cookies to a worthy
>>cause.
>
>Do I have to?

Yes.  Cookies must be sacrificed in order for the can to resonate.  If
you left the cookies inside the can while sweeping, it will lower the
cavity Q making it more difficult to see the dip or peak in the trace.

>What's the point of living if you can't waste time on fun stuff?

I have a confession to make.  I spend far too much time on fun
diversions, academic exercises, and pursuing oddities, than I should.
It's a bad case of arrested growth, where I've never really outgrown
my childhood curiosity.  While this had provided considerable
entertainment and education values, it tended to greatly irritate my
former employers.  The standard line was "Am I paying you to do this"?

>>The SA has a tracking generator; that might be fun.

>Hmmm, how does one do a tracking generator on an FFT-based 
>spectrum analyzer?

Notice that tracking generators produce sine waves.  As I recall, you
don't do sine waves.

The tracking generator appears to be a hardware option on the IFR
2399c.  I couldn't find block diagrams or theory of operation for the
2399C or other FFT spectrum analyzer to see how it's done.  The Rigol
DSA815-TG, 832, and 875 all are FFT SA's with built in tracking
generators.  No block diagrams for those found.  Well, I did find this
photo of the DSA815-TG.  Tracking generator is at bottom.
<http://www.eevblog.com/forum/blog-specific/eevblog-391-rigol-dsa815-spectrum-anayser-teardown/?action=dlattach;attach=33922;image>
<https://www.flickr.com/photos/eevblog/sets/72157632115219155/>
but that doesn't count as a block diagram. 

My guess(tm) is that the tracking generator starts at some frequency.
The SA saves the highest value in the FFT bin that corresponds to that
frequency.  The bin values are saved and the frequency is incremented
to land in the next higher bin.  Repeat about 1000 times, and you
should have a continuous trace.

However, if you don't like sine waves, there are other ways to
generate a frequency response curve.  A noise generator works quite
well:
<https://www.youtube.com/watch?v=bbdTRX4_2DE>
Record an FFT trace, average over the previous 1000 or so
measurements, and you get a nice looking line instead of a noisy mess.
I didn't want to invest in a tracking generator, so I went for a cheap
eBay noise generator:
<https://www.youtube.com/watch?v=QMRxTSAjjLs>
<http://www.ebay.com/itm/252980159450>   

I can also get a frequency sweep without a spectrum analyzer using a
sweep generator, RF detector, DC amp, and an oscilloscope.  
<http://802.11junk.com/jeffl/pics/home/slides/BL-shop5.html>
The pile of HP8620, and the white Wavetek on the shelf are example.


-- 
Jeff Liebermann     jeffl@cruzio.com
150 Felker St #D    http://www.LearnByDestroying.com
Santa Cruz CA 95060 http://802.11junk.com
Skype: JeffLiebermann     AE6KS    831-336-2558

[toc] | [prev] | [next] | [standalone]

#464976

On Sun, 18 Jun 2017 12:49:49 -0700, Jeff Liebermann <jeffl@cruzio.com>
wrote:

>On Sat, 17 Jun 2017 10:08:39 -0700, John Larkin
><jjlarkin@highlandtechnology.com> wrote:
>
>>I still want to light an LED from the ambient RF.
>
>>>-28dBm is 9mv rms or 25mv p-p into 50 ohms.
>
>Well, your spectrum analyzer shows 9 mv rms into 50 ohms, which is:
>  P = E^2/R
>  P = (9*10^-3)^2 / 50 = 
>  P = 1.62 microwatts
>Ummm... I think you'll need a bigger antenna.

Certainly. But a high-Q resonant antenna can do some serious coupling.

A good LED is visible in office light at 1 uA, which is around 2 uW. 

>
>>We're moving soon, about the same distance from the tower, and we have
>>considered some sort of screen room in the basement. But the EMI does
>>automagically test our breadboards and products for us.
>
>That's fine as long as you don't require the EMI or RFI for the
>circuit to operate.  
>
>>>BTW, nice Marconi/IFR/Aeroflex/Cobham 2399c spectrum analyzer.  I'm
>>>jealous.
>>
>>It's horrible and buggy. It has +-120 Hz spurs, which is like finding
>>a roach in your salad.
>
>Sigh.  All my various spectrum analyzers are ancient.  None of them
>are FFT type.  None of them have any kind of storage.
><http://802.11junk.com/jeffl/pics/home/slides/test-equip-mess.html>
>I'm looking for something better, but have some other priority
>purchases to deal with first.  Incidentally, the IFR-1500 in the
>picture has a really awful power supply.
>
>>>Or, you could just frequency sweep the cookie can and see what you
>>>get.  Just sacrifice some more Danish butter cookies to a worthy
>>>cause.
>>
>>Do I have to?
>
>Yes.  Cookies must be sacrificed in order for the can to resonate.  If
>you left the cookies inside the can while sweeping, it will lower the
>cavity Q making it more difficult to see the dip or peak in the trace.
>
>>What's the point of living if you can't waste time on fun stuff?
>
>I have a confession to make.  I spend far too much time on fun
>diversions, academic exercises, and pursuing oddities, than I should.
>It's a bad case of arrested growth, where I've never really outgrown
>my childhood curiosity.  While this had provided considerable
>entertainment and education values, it tended to greatly irritate my
>former employers.  The standard line was "Am I paying you to do this"?

It's impressive how many apparent product failures, or goofy poinless
experiments, turn out to be valuable later on.

>
>>>The SA has a tracking generator; that might be fun.
>
>>Hmmm, how does one do a tracking generator on an FFT-based 
>>spectrum analyzer?
>
>Notice that tracking generators produce sine waves.  As I recall, you
>don't do sine waves.

But my customer does, in his specs, so I have to humor them.

We do make a few arbitrary waveform generators that can be dumbed down
to make sine waves. This one makes 1 MHz with about -50 dB THD, not
bad for a signal generator.

http://www.highlandtechnology.com/DSS/T346DS.shtml

A custom lowpass filter improves that by about 80 dB, so we're about
-130 on harmonics to test our photodiode amp.


>
>The tracking generator appears to be a hardware option on the IFR
>2399c. 

We have it.

 I couldn't find block diagrams or theory of operation for the
>2399C or other FFT spectrum analyzer to see how it's done.  The Rigol
>DSA815-TG, 832, and 875 all are FFT SA's with built in tracking
>generators.  No block diagrams for those found.  Well, I did find this
>photo of the DSA815-TG.  Tracking generator is at bottom.
><http://www.eevblog.com/forum/blog-specific/eevblog-391-rigol-dsa815-spectrum-anayser-teardown/?action=dlattach;attach=33922;image>
><https://www.flickr.com/photos/eevblog/sets/72157632115219155/>
>but that doesn't count as a block diagram. 
>
>My guess(tm) is that the tracking generator starts at some frequency.
>The SA saves the highest value in the FFT bin that corresponds to that
>frequency.  The bin values are saved and the frequency is incremented
>to land in the next higher bin.  Repeat about 1000 times, and you
>should have a continuous trace.

Sweeping the tracking gen must slow down the scan, unless it sweeps in
big jumps. I could try that and see if it's so.

>
>However, if you don't like sine waves, there are other ways to
>generate a frequency response curve.  A noise generator works quite
>well:
><https://www.youtube.com/watch?v=bbdTRX4_2DE>
>Record an FFT trace, average over the previous 1000 or so
>measurements, and you get a nice looking line instead of a noisy mess.
>I didn't want to invest in a tracking generator, so I went for a cheap
>eBay noise generator:
><https://www.youtube.com/watch?v=QMRxTSAjjLs>
><http://www.ebay.com/itm/252980159450>   
>
>I can also get a frequency sweep without a spectrum analyzer using a
>sweep generator, RF detector, DC amp, and an oscilloscope.  
><http://802.11junk.com/jeffl/pics/home/slides/BL-shop5.html>
>The pile of HP8620, and the white Wavetek on the shelf are example.

-- 

John Larkin         Highland Technology, Inc

lunatic fringe electronics 

[toc] | [prev] | [next] | [standalone]

#464988

On Sun, 18 Jun 2017 13:42:52 -0700, John Larkin
<jjlarkin@highlandtechnology.com> wrote:

>On Sun, 18 Jun 2017 12:49:49 -0700, Jeff Liebermann <jeffl@cruzio.com>
>wrote:
>
>>On Sat, 17 Jun 2017 10:08:39 -0700, John Larkin
>><jjlarkin@highlandtechnology.com> wrote:
>>
>>>I still want to light an LED from the ambient RF.
>>
>>>>-28dBm is 9mv rms or 25mv p-p into 50 ohms.
>>
>>Well, your spectrum analyzer shows 9 mv rms into 50 ohms, which is:
>>  P = E^2/R
>>  P = (9*10^-3)^2 / 50 = 
>>  P = 1.62 microwatts
>>Ummm... I think you'll need a bigger antenna.

>Certainly. But a high-Q resonant antenna can do some serious coupling.

Sure, but it doesn't need to resonant.  Aim a big dish antenna, with a
broadband feed (waveguide beyond cutoff, LPDA, bow tie, etc) at Mt
Sutro and see what comes out the coax.  I have the Mt Sutro RF
exposure data buried somewhere and can probably calculate the field
strength from some of the transmitters at your location.  However,
it's Sunday, I'm overheated, and quite sober, which means I don't have
a handy excuse if I mangle the calculations.  How far away from Mt
Sutro are your old and new offices?

>A good LED is visible in office light at 1 uA, which is around 2 uW. 

I didn't know they were that good.  Impedance of the LED would be
around:
   P = I^2 * R
   2uW =1uA^2 * R
   R = 2/1 = 2 ohms
So the fun part will be transforming the inpedance from a 75 ohm
antenna to 2 ohms.  37.5:1 transformations are possible but not fun.

Incidentally, we've been here before in 2012, except that you were
planning to do it using the AM broadcast band:
<https://groups.google.com/d/msg/sci.electronics.design/zqpg1dz56HU/5k2ofruUPqoJ>

>>I have a confession to make.  I spend far too much time on fun
>>diversions, academic exercises, and pursuing oddities, than I should.
>>It's a bad case of arrested growth, where I've never really outgrown
>>my childhood curiosity.  While this had provided considerable
>>entertainment and education values, it tended to greatly irritate my
>>former employers.  The standard line was "Am I paying you to do this"?

>It's impressive how many apparent product failures, or goofy poinless
>experiments, turn out to be valuable later on.

Yep.  I've designed 3 products, at 3 different companies, where
everyone within criticism range announced that it was a dumb idea,
nobody would pay for it, it's a waste of time, and so on.  I didn't
keep track of the reasons they should have failed.  All of them
eventually had a very successful product run and lifetime.  Nobody
ever thanked me.  

Lesson learned:  Don't tread on managements toes.

>>My guess(tm) is that the tracking generator starts at some frequency.
>>The SA saves the highest value in the FFT bin that corresponds to that
>>frequency.  The bin values are saved and the frequency is incremented
>>to land in the next higher bin.  Repeat about 1000 times, and you
>>should have a continuous trace.

>Sweeping the tracking gen must slow down the scan, unless it sweeps in
>big jumps. I could try that and see if it's so.

I haven't seen one in action or found the block diagram so I don't
know how it works.  If it's anything like using an RF sweeper and
spectrum analyzer connected with no sweep synchronization, it's going
to be really slow.  Since the two sweeps are at different rates, the
only time anything appears above the baseline is when the generator
and SA frequencies just happen to coincide.  With a short persistence
SA screen, it's hard to see anything.  Doing the same with a storage
scope, DSO display, or an HP8750A "storage normalizer" should produce
usable sweeps.


-- 
Jeff Liebermann     jeffl@cruzio.com
150 Felker St #D    http://www.LearnByDestroying.com
Santa Cruz CA 95060 http://802.11junk.com
Skype: JeffLiebermann     AE6KS    831-336-2558

[toc] | [prev] | [next] | [standalone]

#464991

On Sun, 18 Jun 2017 15:50:08 -0700, Jeff Liebermann <jeffl@cruzio.com>
wrote:

>On Sun, 18 Jun 2017 13:42:52 -0700, John Larkin
><jjlarkin@highlandtechnology.com> wrote:
>
>>On Sun, 18 Jun 2017 12:49:49 -0700, Jeff Liebermann <jeffl@cruzio.com>
>>wrote:
>>
>>>On Sat, 17 Jun 2017 10:08:39 -0700, John Larkin
>>><jjlarkin@highlandtechnology.com> wrote:
>>>
>>>>I still want to light an LED from the ambient RF.
>>>
>>>>>-28dBm is 9mv rms or 25mv p-p into 50 ohms.
>>>
>>>Well, your spectrum analyzer shows 9 mv rms into 50 ohms, which is:
>>>  P = E^2/R
>>>  P = (9*10^-3)^2 / 50 = 
>>>  P = 1.62 microwatts
>>>Ummm... I think you'll need a bigger antenna.

>
>>Certainly. But a high-Q resonant antenna can do some serious coupling.
>
>Sure, but it doesn't need to resonant.

A resonant antenna can be a lot smaller than a plain one, but in the
FM band I can do a 1/4 or 1/2 wave, and make it broader band to slurp
up more stations.

  Aim a big dish antenna, with a
>broadband feed (waveguide beyond cutoff, LPDA, bow tie, etc) at Mt
>Sutro and see what comes out the coax. 

I had in mind a bit of wire, not a giant structure!



 I have the Mt Sutro RF
>exposure data buried somewhere and can probably calculate the field
>strength from some of the transmitters at your location.  However,
>it's Sunday, I'm overheated, and quite sober, which means I don't have
>a handy excuse if I mangle the calculations.  How far away from Mt
>Sutro are your old and new offices?

Current, 2.2 miles. New will be 2.5.

>
>>A good LED is visible in office light at 1 uA, which is around 2 uW. 
>
>I didn't know they were that good.  Impedance of the LED would be
>around:
>   P = I^2 * R
>   2uW =1uA^2 * R
>   R = 2/1 = 2 ohms
>So the fun part will be transforming the inpedance from a 75 ohm
>antenna to 2 ohms.  37.5:1 transformations are possible but not fun.

I'd need a rectifier that can deliver a few uA into maybe 2 volts. The
gross load impedance is ballpark 1 meg. So I need a step-up, not down.


>
>Incidentally, we've been here before in 2012, except that you were
>planning to do it using the AM broadcast band:
><https://groups.google.com/d/msg/sci.electronics.design/zqpg1dz56HU/5k2ofruUPqoJ>


No rush.


>
>>>I have a confession to make.  I spend far too much time on fun
>>>diversions, academic exercises, and pursuing oddities, than I should.
>>>It's a bad case of arrested growth, where I've never really outgrown
>>>my childhood curiosity.  While this had provided considerable
>>>entertainment and education values, it tended to greatly irritate my
>>>former employers.  The standard line was "Am I paying you to do this"?
>
>>It's impressive how many apparent product failures, or goofy poinless
>>experiments, turn out to be valuable later on.
>
>Yep.  I've designed 3 products, at 3 different companies, where
>everyone within criticism range announced that it was a dumb idea,
>nobody would pay for it, it's a waste of time, and so on.  I didn't
>keep track of the reasons they should have failed.  All of them
>eventually had a very successful product run and lifetime.  Nobody
>ever thanked me.  
>
>Lesson learned:  Don't tread on managements toes.
>
>>>My guess(tm) is that the tracking generator starts at some frequency.
>>>The SA saves the highest value in the FFT bin that corresponds to that
>>>frequency.  The bin values are saved and the frequency is incremented
>>>to land in the next higher bin.  Repeat about 1000 times, and you
>>>should have a continuous trace.
>
>>Sweeping the tracking gen must slow down the scan, unless it sweeps in
>>big jumps. I could try that and see if it's so.
>
>I haven't seen one in action or found the block diagram so I don't
>know how it works.  If it's anything like using an RF sweeper and
>spectrum analyzer connected with no sweep synchronization, it's going
>to be really slow.  Since the two sweeps are at different rates, the
>only time anything appears above the baseline is when the generator
>and SA frequencies just happen to coincide.  With a short persistence
>SA screen, it's hard to see anything.  Doing the same with a storage
>scope, DSO display, or an HP8750A "storage normalizer" should produce
>usable sweeps.

I suppose a smooth or stepped sweep through each of successive FFT
windows would work.

But sine waves are boring.



-- 

John Larkin         Highland Technology, Inc

lunatic fringe electronics 

[toc] | [prev] | [next] | [standalone]

#465006

On Sun, 18 Jun 2017 17:15:45 -0700, John Larkin
<jjlarkin@highlandtechnology.com> wrote:

>n Sun, 18 Jun 2017 15:50:08 -0700, Jeff Liebermann <jeffl@cruzio.com>
>wrote:
>
>>On Sun, 18 Jun 2017 13:42:52 -0700, John Larkin
>><jjlarkin@highlandtechnology.com> wrote:
>>
>>>On Sun, 18 Jun 2017 12:49:49 -0700, Jeff Liebermann <jeffl@cruzio.com>
>>>wrote:
>>>
>>>>On Sat, 17 Jun 2017 10:08:39 -0700, John Larkin
>>>><jjlarkin@highlandtechnology.com> wrote:
>>>>
>>>>>I still want to light an LED from the ambient RF.
>>>>
>>>>>>-28dBm is 9mv rms or 25mv p-p into 50 ohms.
>>>>
>>>>Well, your spectrum analyzer shows 9 mv rms into 50 ohms, which is:
>>>>  P = E^2/R
>>>>  P = (9*10^-3)^2 / 50 = 
>>>>  P = 1.62 microwatts
>>>>Ummm... I think you'll need a bigger antenna.
>
>>
>>>Certainly. But a high-Q resonant antenna can do some serious coupling.
>>
>>Sure, but it doesn't need to resonant.
>
>A resonant antenna can be a lot smaller than a plain one, but in the
>FM band I can do a 1/4 or 1/2 wave, and make it broader band to slurp
>up more stations.

The receiver antenna capture area is proportional to the square of the
wavelength. The capture area is sometimes illustrated as an ellipse
about 1/2 x 1/4 wavelengths, with a half wave dipole in the middle.

The captured power [W] thus increases rapidly at longer wavelengths as
long as the power density  [W/m²] remains the same. Thus, you would
expect to get much larger power levels from an AM transmitter than an
FM transmitter with some distance and power.

The capture area of an electrically short (much less than 1/2
wavelengths) dipole is nearly as good as full size dipole. The problem
with very short dipoles is that it is reactive (huge capacitive
reactance). You can only extract power from the resistive part of the
impedance, not the reactive part. 

Loading coils can be used to generate inductive reactance and hence
cancel the antenna capacitive reactance (conjugate match). Thus, for
best power extraction, the antenna elements do not have to be
resonant, but the antenna _system_ (antenna+loading coil) must be
resonant. The problem is that if you need a large resistance or
reactive change, the network is quite narrow band.

[toc] | [prev] | [next] | [standalone]

#465007

On Mon, 19 Jun 2017 09:39:47 +0300, upsidedown@downunder.com wrote:

>On Sun, 18 Jun 2017 17:15:45 -0700, John Larkin
><jjlarkin@highlandtechnology.com> wrote:
>
>>n Sun, 18 Jun 2017 15:50:08 -0700, Jeff Liebermann <jeffl@cruzio.com>
>>wrote:
>>
>>>On Sun, 18 Jun 2017 13:42:52 -0700, John Larkin
>>><jjlarkin@highlandtechnology.com> wrote:
>>>
>>>>On Sun, 18 Jun 2017 12:49:49 -0700, Jeff Liebermann <jeffl@cruzio.com>
>>>>wrote:
>>>>
>>>>>On Sat, 17 Jun 2017 10:08:39 -0700, John Larkin
>>>>><jjlarkin@highlandtechnology.com> wrote:
>>>>>
>>>>>>I still want to light an LED from the ambient RF.
>>>>>
>>>>>>>-28dBm is 9mv rms or 25mv p-p into 50 ohms.
>>>>>
>>>>>Well, your spectrum analyzer shows 9 mv rms into 50 ohms, which is:
>>>>>  P = E^2/R
>>>>>  P = (9*10^-3)^2 / 50 = 
>>>>>  P = 1.62 microwatts
>>>>>Ummm... I think you'll need a bigger antenna.
>>
>>>
>>>>Certainly. But a high-Q resonant antenna can do some serious coupling.
>>>
>>>Sure, but it doesn't need to resonant.
>>
>>A resonant antenna can be a lot smaller than a plain one, but in the
>>FM band I can do a 1/4 or 1/2 wave, and make it broader band to slurp
>>up more stations.
>
>The receiver antenna capture area is proportional to the square of the
>wavelength. The capture area is sometimes illustrated as an ellipse
>about 1/2 x 1/4 wavelengths, with a half wave dipole in the middle.
>
>The captured power [W] thus increases rapidly at longer wavelengths as
>long as the power density  [W/m²] remains the same. Thus, you would
>expect to get much larger power levels from an AM transmitter than an
>FM transmitter with some distance and power.
>
>The capture area of an electrically short (much less than 1/2
>wavelengths) dipole is nearly as good as full size dipole. The problem
>with very short dipoles is that it is reactive (huge capacitive
>reactance). You can only extract power from the resistive part of the
>impedance, not the reactive part. 
>
>Loading coils can be used to generate inductive reactance and hence
>cancel the antenna capacitive reactance (conjugate match). Thus, for
>best power extraction, the antenna elements do not have to be
>resonant, but the antenna _system_ (antenna+loading coil) must be
>resonant. The problem is that if you need a large resistance or
>reactive change, the network is quite narrow band.
>

In addition, a random wire longer than 1/2 wavelength doesn't increase
extracted power, since any voltage induced into half wave sections
tend to cancel, so on FM band a random wire longer than 150 cm is not
of much use. 

Of course yagis etc. can be used, but they tend to be directional. The
directivity is good thing, if there is only a single transmit tower at
line of sight, but a bad thing, if you want to capture power from
multiple towers or from (indoor) reflections.

[toc] | [prev] | [next] | [standalone]

#465019

On 06/19/2017 02:48 AM, upsidedown@downunder.com wrote:
> On Mon, 19 Jun 2017 09:39:47 +0300, upsidedown@downunder.com wrote:
> 
>> On Sun, 18 Jun 2017 17:15:45 -0700, John Larkin
>> <jjlarkin@highlandtechnology.com> wrote:
>>
>>> n Sun, 18 Jun 2017 15:50:08 -0700, Jeff Liebermann <jeffl@cruzio.com>
>>> wrote:
>>>
>>>> On Sun, 18 Jun 2017 13:42:52 -0700, John Larkin
>>>> <jjlarkin@highlandtechnology.com> wrote:
>>>>
>>>>> On Sun, 18 Jun 2017 12:49:49 -0700, Jeff Liebermann <jeffl@cruzio.com>
>>>>> wrote:
>>>>>
>>>>>> On Sat, 17 Jun 2017 10:08:39 -0700, John Larkin
>>>>>> <jjlarkin@highlandtechnology.com> wrote:
>>>>>>
>>>>>>> I still want to light an LED from the ambient RF.
>>>>>>
>>>>>>>> -28dBm is 9mv rms or 25mv p-p into 50 ohms.
>>>>>>
>>>>>> Well, your spectrum analyzer shows 9 mv rms into 50 ohms, which is:
>>>>>>  P = E^2/R
>>>>>>  P = (9*10^-3)^2 / 50 = 
>>>>>>  P = 1.62 microwatts
>>>>>> Ummm... I think you'll need a bigger antenna.
>>>
>>>>
>>>>> Certainly. But a high-Q resonant antenna can do some serious coupling.
>>>>
>>>> Sure, but it doesn't need to resonant.
>>>
>>> A resonant antenna can be a lot smaller than a plain one, but in the
>>> FM band I can do a 1/4 or 1/2 wave, and make it broader band to slurp
>>> up more stations.
>>
>> The receiver antenna capture area is proportional to the square of the
>> wavelength. The capture area is sometimes illustrated as an ellipse
>> about 1/2 x 1/4 wavelengths, with a half wave dipole in the middle.
>>
>> The captured power [W] thus increases rapidly at longer wavelengths as
>> long as the power density  [W/m²] remains the same. Thus, you would
>> expect to get much larger power levels from an AM transmitter than an
>> FM transmitter with some distance and power.
>>
>> The capture area of an electrically short (much less than 1/2
>> wavelengths) dipole is nearly as good as full size dipole. The problem
>> with very short dipoles is that it is reactive (huge capacitive
>> reactance). You can only extract power from the resistive part of the
>> impedance, not the reactive part. 
>>
>> Loading coils can be used to generate inductive reactance and hence
>> cancel the antenna capacitive reactance (conjugate match). Thus, for
>> best power extraction, the antenna elements do not have to be
>> resonant, but the antenna _system_ (antenna+loading coil) must be
>> resonant. The problem is that if you need a large resistance or
>> reactive change, the network is quite narrow band.
>>
> 
> In addition, a random wire longer than 1/2 wavelength doesn't increase
> extracted power, since any voltage induced into half wave sections
> tend to cancel, so on FM band a random wire longer than 150 cm is not
> of much use. 

They generally have strongly lobed patterns, though, so you'll get more
signal if you wiggle them around to maximize it.

Cheers

Phil Hobbs

-- 
Dr Philip C D Hobbs
Principal Consultant
ElectroOptical Innovations LLC
Optics, Electro-optics, Photonics, Analog Electronics

160 North State Road #203
Briarcliff Manor NY 10510

hobbs at electrooptical dot net
http://electrooptical.net

[toc] | [prev] | [next] | [standalone]

#464755

On Friday, June 16, 2017 at 10:03:56 AM UTC-7, Jeff Liebermann wrote:
> You can eliminate a ground loop if you only ground one end of a
> coaxial cable.  This may seem counter-intuitive.  The idea is to keep
> the coax cable from acting as a one turn transformer.

You can leave the ground intact and get a loop antenna, or you can
lift the ground at one end and get a dipole antenna.  Choose your poison.

Lifting a coax ground is almost always a terrible, terrible idea.

-- john, KE5FX

[toc] | [prev] | [next] | [standalone]

#464770

On Fri, 16 Jun 2017 15:30:48 -0700 (PDT), "John Miles, KE5FX"
<jmiles@gmail.com> wrote:

>On Friday, June 16, 2017 at 10:03:56 AM UTC-7, Jeff Liebermann wrote:
>> You can eliminate a ground loop if you only ground one end of a
>> coaxial cable.  This may seem counter-intuitive.  The idea is to keep
>> the coax cable from acting as a one turn transformer.
>
>You can leave the ground intact and get a loop antenna, or you can
>lift the ground at one end and get a dipole antenna.  Choose your poison.

As far as the signal is concerned, if there are ground potential
differences between the source box and the load, opening a shield will
make them worse.

>
>Lifting a coax ground is almost always a terrible, terrible idea.

Right.


-- 

John Larkin         Highland Technology, Inc
picosecond timing   precision measurement 

jlarkin att highlandtechnology dott com
http://www.highlandtechnology.com

[toc] | [prev] | [next] | [standalone]

#464822

On 06/16/2017 09:10 PM, John Larkin wrote:
> On Fri, 16 Jun 2017 15:30:48 -0700 (PDT), "John Miles, KE5FX"
> <jmiles@gmail.com> wrote:
> 
>> On Friday, June 16, 2017 at 10:03:56 AM UTC-7, Jeff Liebermann wrote:
>>> You can eliminate a ground loop if you only ground one end of a
>>> coaxial cable.  This may seem counter-intuitive.  The idea is to keep
>>> the coax cable from acting as a one turn transformer.
>>
>> You can leave the ground intact and get a loop antenna, or you can
>> lift the ground at one end and get a dipole antenna.  Choose your poison.
> 
> As far as the signal is concerned, if there are ground potential
> differences between the source box and the load, opening a shield will
> make them worse.
> 
>>
>> Lifting a coax ground is almost always a terrible, terrible idea.
> 
> Right.

Ground loops are very low impedance at low frequency, and so can do
weird things.  We don't usually expect a 5-mV signal with a circulating
current of an amp.

A parallel combination of a 1-ohm resistor and a bunch of capacitance in
series with the shield, combined with a diff amp, can be good medicine
for that problem.

Cheers

Phil Hobbs

-- 
Dr Philip C D Hobbs
Principal Consultant
ElectroOptical Innovations LLC
Optics, Electro-optics, Photonics, Analog Electronics

160 North State Road #203
Briarcliff Manor NY 10510

hobbs at electrooptical dot net
http://electrooptical.net

[toc] | [prev] | [next] | [standalone]

#464810

On Fri, 16 Jun 2017 15:30:48 -0700 (PDT), "John Miles, KE5FX"
<jmiles@gmail.com> wrote:

>On Friday, June 16, 2017 at 10:03:56 AM UTC-7, Jeff Liebermann wrote:
>> You can eliminate a ground loop if you only ground one end of a
>> coaxial cable.  This may seem counter-intuitive.  The idea is to keep
>> the coax cable from acting as a one turn transformer.
>
>You can leave the ground intact and get a loop antenna, or you can
>lift the ground at one end and get a dipole antenna.  Choose your poison.
>
>Lifting a coax ground is almost always a terrible, terrible idea.
>
>-- john, KE5FX

The RF signal in a coaxial cable propagates in the insulation material
between the outside of the inner conductor and inside of the outher
conductor. Disconnecting the shield at one end, the coax acts more
like a G-line.

One way around would be disconnecting the shield at one end, but
bridge it with a small capacitance good quality capacitor. It will
effectively block any mains and LF related ground loop currents, but
the coaxial line works as intended on HF and low-VHF frequencies.

[toc] | [prev] | [next] | [standalone]

#464841

On Fri, 16 Jun 2017 15:30:48 -0700 (PDT), "John Miles, KE5FX"
<jmiles@gmail.com> wrote:

>On Friday, June 16, 2017 at 10:03:56 AM UTC-7, Jeff Liebermann wrote:
>> You can eliminate a ground loop if you only ground one end of a
>> coaxial cable.  This may seem counter-intuitive.  The idea is to keep
>> the coax cable from acting as a one turn transformer.

>You can leave the ground intact and get a loop antenna, or you can
>lift the ground at one end and get a dipole antenna.  Choose your poison.

I think you mean monopole antenna, but I get the idea.  Please note
that I'm not suggesting that lifting the shield be a part of design.
Not only can it cause other problems, but production would probably
shoot it down as difficult to build and easy to do wrong.  What it's
good for is testing for ground loops.  On some connectors (phono,
PL259), I can simply retract the plug until the shield disconnects,
eliminating the need for a special cable or adapter.

I have a collection of 1:1 RF transformers, which is another way to
eliminate or test for ground loops.

I sorta cheated with my examples of web articles indicating that
grounding one end of the shield was proper practice.  Those were
mostly about CAT5, which is a bundle of twisted pairs, surrounded by a
shield.  Grounding one end there works because the shield is NOT part
of the signal path.  In ordinary coax cable, the shield is part of the
conduction path, and needs to be intact.

>Lifting a coax ground is almost always a terrible, terrible idea.
>-- john, KE5FX

I would say it's only a bad idea and that it does have its uses.

-- 
Jeff Liebermann     jeffl@cruzio.com
150 Felker St #D    http://www.LearnByDestroying.com
Santa Cruz CA 95060 http://802.11junk.com
Skype: JeffLiebermann     AE6KS    831-336-2558

[toc] | [prev] | [next] | [standalone]

#464848

On Sat, 17 Jun 2017 09:35:35 -0700, Jeff Liebermann <jeffl@cruzio.com>
wrote:

>On Fri, 16 Jun 2017 15:30:48 -0700 (PDT), "John Miles, KE5FX"
><jmiles@gmail.com> wrote:
>
>>On Friday, June 16, 2017 at 10:03:56 AM UTC-7, Jeff Liebermann wrote:
>>> You can eliminate a ground loop if you only ground one end of a
>>> coaxial cable.  This may seem counter-intuitive.  The idea is to keep
>>> the coax cable from acting as a one turn transformer.
>
>>You can leave the ground intact and get a loop antenna, or you can
>>lift the ground at one end and get a dipole antenna.  Choose your poison.
>
>I think you mean monopole antenna, but I get the idea.  Please note
>that I'm not suggesting that lifting the shield be a part of design.
>Not only can it cause other problems, but production would probably
>shoot it down as difficult to build and easy to do wrong.  What it's
>good for is testing for ground loops.  On some connectors (phono,
>PL259), I can simply retract the plug until the shield disconnects,
>eliminating the need for a special cable or adapter.
>
>I have a collection of 1:1 RF transformers, which is another way to
>eliminate or test for ground loops.
>

Use a clamp-on ammeter for 60 Hz, or some sort of CT for higher
frequencies.




>I sorta cheated with my examples of web articles indicating that
>grounding one end of the shield was proper practice.  Those were
>mostly about CAT5, which is a bundle of twisted pairs, surrounded by a
>shield.  Grounding one end there works because the shield is NOT part
>of the signal path.  In ordinary coax cable, the shield is part of the
>conduction path, and needs to be intact.
>
>>Lifting a coax ground is almost always a terrible, terrible idea.
>>-- john, KE5FX
>
>I would say it's only a bad idea and that it does have its uses.

I vote for "always terrible."


-- 

John Larkin         Highland Technology, Inc

lunatic fringe electronics 

[toc] | [prev] | [next] | [standalone]

#464811

On Wed, 14 Jun 2017 16:09:43 -0700, John Larkin
<jjlarkin@highland_snip_technology.com> wrote:

>
>I designed this little photodiode amp and I'm trying to measure its
>output noise. All we can see on a spectrum analyser is radio and TV
>stations.

Is this a one off proof of concept design ?

In that case, why do you insist doing noise measurements in hellish RF
environment ? Move the test equipment and DUT into a properly shielded
room or better into a deep cave or mine.

If the final product is intended to be working in such hellish RF
environment, the  shielding should be designed from the beginning. Put
everything (including the photodiode) inside a die cast aluminum or
copper box. Since the photodiode needs to detect some outside light,
make a small hole in the box. 

To prevent RF penetrating into the box through the hole, put a small
metallic tube around the hole. It will then act as a waveguide filter,
cutting out any low frequency RF noise. As the maximum frequency
required is only 100 MHz, quite large diameter tubes can be used, so
it doesn't limit the photodiode field-of-view too much. If only a
narrow f-o-v is required, a 1 cm diameter metal tube would be
sufficient, having cut off at several GHz.

[toc] | [prev] | [next] | [standalone]

#464826

On Sat, 17 Jun 2017 14:45:34 +0300, upsidedown@downunder.com wrote:

>On Wed, 14 Jun 2017 16:09:43 -0700, John Larkin
><jjlarkin@highland_snip_technology.com> wrote:
>
>>
>>I designed this little photodiode amp and I'm trying to measure its
>>output noise. All we can see on a spectrum analyser is radio and TV
>>stations.
>
>Is this a one off proof of concept design ?

No, we just shipped the first batch of 8 units. I'm still working on
the test report (there are 97 specification line item that I'm
supposed to comply with... there are only 50 parts on the board) and
I'm still doing some tests on a couple of engineering units. We *had*
to ship last week for some systems test or something.

>
>In that case, why do you insist doing noise measurements in hellish RF
>environment ? Move the test equipment and DUT into a properly shielded
>room or better into a deep cave or mine.

This would be ideal, but I'd need a mile or two of extension cord to
power the gear.

https://dl.dropboxusercontent.com/u/53724080/Truckee/CW_inside.jpg

and I'd have to hire some grunts to backpack the equipment in.

I was thinking about getting some EMI test equiupment for the cabin in
Truckee, a generator and a spectrum analyzer and one of those
surfboard wideband antennas. It's ideal for open-field testing...
surrounded by mountains and cell phones barely work. But we go there
to get away from work.


>
>If the final product is intended to be working in such hellish RF
>environment, the  shielding should be designed from the beginning. Put
>everything (including the photodiode) inside a die cast aluminum or
>copper box. Since the photodiode needs to detect some outside light,
>make a small hole in the box. 

The actual environment will be inside a wafer scanner in a
semiconductor fab. The photodiode will be inside a stainless steel
vacuum vessel. There won't be much EMI. The guy who wrote the noise
spec just went wild. 

>
>To prevent RF penetrating into the box through the hole, put a small
>metallic tube around the hole. It will then act as a waveguide filter,
>cutting out any low frequency RF noise. As the maximum frequency
>required is only 100 MHz, quite large diameter tubes can be used, so
>it doesn't limit the photodiode field-of-view too much. If only a
>narrow f-o-v is required, a 1 cm diameter metal tube would be
>sufficient, having cut off at several GHz.

The only EMI problem that I have is the gigantic Sutro Tower
interfering with my wideband noise measurement. The cookie can fixed
that.


-- 

John Larkin         Highland Technology, Inc

lunatic fringe electronics 

[toc] | [prev] | [standalone]

Page 3 of 3 — ← Prev page 1 2 [3]

Back to top | Article view | sci.electronics.design

csiph-web