Idea about measure EMI using oscilloscope

[neilhao]

Update (26 July 2013)

Experiment Report:
[1] Magnetic Field Probe http://uniteng.com/index.php/2013/05/30/measuring-emi-with-homemade-magnetic-field-probe/

[2] Electromagnetic Shielding http://uniteng.com/index.php/2013/06/21/electromagnetic-shielding/

[3] Channel power measurements (26 July 2013) http://uniteng.com/index.php/2013/07/26/channel-power-measurements/

[4] Preamplifier 30 MHz to 6 GHz (August 18, 2013) http://uniteng.com/index.php/2013/08/18/preamplifier-30mhz-to-6ghz/

[5] Philips Sonicare Electric Toothbrush Measurements without Teardown (January 25, 2014)http://uniteng.com/index.php/2014/01/25/philips-sonicare-electric-toothbrush-measurements-without-teardown/

Hi everyone,

I am new to EMI. I tried to design a simple add-on board for the oscilloscope  which could amplify the EMI using diff amp, the attachments are what I designed, the RF Sensor will be the near field probe.

Did anyone do this before? will this idea works?


Thanks,

Neal

[qno]

You can connect the sense coil directly to the scope.


Here is some other simple test equipment.

http://www.tmworld.com/electronics-blogs/4376432/4/The-EMC-Blog

[plesa]

It seems to be good idea and can be quite handy with some software like Agilent Spectrum Visualizer.
I will just recomend to split the probe and the amplifier and make the amplifier small as possible (e.g SMA connectors).

[fpliuzzi]

Hello Neil,
Your project looks very interesting. Please post some measurement results when you construct the device.

You may find the following interesting. On the other end of the complexity scale, Doug C. Smith has a video where he uses a simple loop probe made from a paperclip. This probe, along with his 500MHz scope, is used to investigate some switching power supply noise...



As Bored@Work suggested, Doug's site at http://emcesd.com is a very interesting place for anyone interested in EMI.

Additional info on the paperclip magnetic probe is at http://emcesd.com/tt080699.htm

It surprised me how useful this very simple probe can be. It fits right in with my collection of E-Field and H-Field Probes.

----------

Also, as noted by qno, Kenneth Wyatt has an interesting five part article about putting together "An EMC troubleshooting toolkit". lots of good info about fabricating some low cost E- and H-Field probes, plus some DIY high frequency current probes (the article has saved me a LOT of money).

www.tmworld.com/electronics-blogs/emc-emi-rfi-esd/4378152/An-EMC-Troubleshooting-Kit--Part-1a-Emissions-

Regards,
Frank

[neilhao]

Hi plesa,

Thank you for your suggestion!

Roy Ediss well stated the theories and the various designs of the Magnetic Field Probes in his article, Probing the Magnetic Field Probe( http://uniteng.com/neildocs/references/Probing_the_Magnetic_Field_Probe.htm ). I implemented the design 2(a) which mentioned in the Roy Ediss’s article.

I used RG405 (Semi-rigid Coax Cable with Copper Outer Conductor, Datasheet: http://uniteng.com/neildocs/datasheets/RG405-U.pdf ) and a SMA connector [male body (inside threads) with male inner pin] in my implementation. The probe looks very cool, I will post some picture once I get the amplifier board form the factory.

By the way, I have Rigol DS2202 and RF Explorer 3G on my hand. I plan to use the diff amp board for the DS2202 oscilloscope, because although it seems the oscilloscope with a probe is enough to measure the EMI caused by something likes switchable power supply, but  I think the gain of the oscilloscope may not enough to measure weak EMI signal, right? I think the probe could be used with the RF Explorer directly since RF Explorer is a high gain equipment, even in some case, we may need the attenuator to protect the spectrum Analyzer.

Neil

It seems to be good idea and can be quite handy with some software like Agilent Spectrum Visualizer.
I will just recomend to split the probe and the amplifier and make the amplifier small as possible (e.g SMA connectors).

[neilhao]

Hi Frank,

I will post the measurement results once I am sure the board works

Thanks for the information about EMI, especially I found the articles form www.tmworld.com are very useful. I plan to make an electric field probe after I get magnetic field probe works~

Hello Neil,
Your project looks very interesting. Please post some measurement results when you construct the device.

You may find the following interesting. On the other end of the complexity scale, Doug C. Smith has a video where he uses a simple loop probe made from a paperclip. This probe, along with his 500MHz scope, is used to investigate some switching power supply noise...



As Bored@Work suggested, Doug's site at http://emcesd.com is a very interesting place for anyone interested in EMI.

Additional info on the paperclip magnetic probe is at http://emcesd.com/tt080699.htm

It surprised me how useful this very simple probe can be. It fits right in with my collection of E-Field and H-Field Probes.

----------

Also, as noted by qno, Kenneth Wyatt has an interesting five part article about putting together "An EMC troubleshooting toolkit". lots of good info about fabricating some low cost E- and H-Field probes, plus some DIY high frequency current probes (the article has saved me a LOT of money).

www.tmworld.com/electronics-blogs/emc-emi-rfi-esd/4378152/An-EMC-Troubleshooting-Kit--Part-1a-Emissions-

Regards,
Frank

[fpliuzzi]

By the way, I have Rigol DS2202 and RF Explorer 3G on my hand. I plan to use the diff amp board for the DS2202 oscilloscope, because although it seems the oscilloscope with a probe is enough to measure the EMI caused by something likes switchable power supply, but  I think the gain of the oscilloscope may not enough to measure weak EMI signal, right? I think the probe could be used with the RF Explorer directly since RF Explorer is a high gain equipment, even in some case, we may need the attenuator to protect the spectrum Analyzer.

I constructed a collection of probes using the info in Kenneth Wyatt's article above and Doug Smith's emc site. I used surplus semi-rigid cables from ebay that already had sma male connectors on them to form many of the probes. This allowed me to put together my collection of probes and attenuators for under $100 total (photo below).

All of my various E-field, H-field, and current probes work very well with my RF Explorer 3G. This tiny spectrum analyzer is pretty sensitive. As far as overload protection goes; the right hand side connector on the RF Explorer 3G can handle up to 1 Watt (+30dBm) without being damaged (the left side connector can handle only about 3 mW (+5dm), unless you attach the new $40 RF limiter that Ariel is selling as an accessory).

Frank

[Smokey]

One more thing to put on your Christmas list...
http://www.home.agilent.com/en/pd-2058001-pn-N9311X/close-field-probe-set-30-mhz-to-3-ghz?nid=-35189.996175.00&cc=US&lc=eng

[w2aew]

One more thing to put on your Christmas list...
http://www.home.agilent.com/en/pd-2058001-pn-N9311X/close-field-probe-set-30-mhz-to-3-ghz?nid=-35189.996175.00&cc=US&lc=eng

If you're going to purchase something, I much prefer the near field probes from Beehive Electronics
http://www.beehive-electronics.com
...they're more than an order of magnitude cheaper than the Agilent set, and work very well.  I used one of them while making this video on transient EMI debug:

[neilhao]

Hi Frank,

Your kit looks very cool! I was inspired by your AC current probe. I planed to design a DC current front end using the hall effect sensor for the oscilloscope, but I never realized that the coil which usually used as AC current probe for DMM, it can also be used by oscilloscope and spectrum analyzer.  I am bounded set thinking I added the current probe to my to do list.

For the power limiter for RF explorer, I do not want to buy it right now, since it's frequency range is DC ~ 3.0 GHz. I am waiting for the 5.8G edition

I have a set of attenuators form mini circuit which could cover the frequency range DC ~ 6.0 GHz. I think, to a certain extent, using attenuators is enough for measurement after carefully calculation 

Ps. The attachment is the RG405 cable I used. It also contains a SMA male connector.   Theoretically, I think all semi-rigid coax cables should work.

[neilhao]

Hi Frank,

I took a picture of my Magnetic Field Probe, I will apply the clear resin coat to the probe later for isolating.

I compared mine with other people's probe, it seems my probe has relative larger gap at top. Hope it will work~


Neil

[fpliuzzi]

Hi Frank,

I took a picture of my Magnetic Field Probe, I will apply the clear resin coat to the probe later for isolating.

I compared mine with other people's probe, it seems my probe has relative larger gap at top. Hope it will work~

Neil

Neil, your probe looks fine. I never came across a specified dimension for that gap at the top of the loop, so I just used a thin (0.025") cutoff wheel in my dremel tool to carefully cut the slot in the outer jacket of the semi-rigid cable after the loops were formed. I coated the probes with a few coats of epoxy to insulate them.

By the way, if you decide to make one of Ken Wyatt's toroidal current probes, like the ones shown in figure 4 of his article "An EMC troubleshooting kit - part 1a (emissions)", a copper foil shield must be added to them.

The caption below figure 4 states "The current probes pictured were photographed prior to the installation of a simple e-field shield constructed from copper tape. Wrap the tape around the windings, leaving a narrow slot around the inside of the toroid to allow h-field coupling."

I used the following photo of a very fine commercially available RF current probe as a reference for how this narrow slot in the copper foil shield was to be implemented. As a final step I just wrapped black electrical tape over my completed current probe to insulate it.

Regards,
Frank

[neilhao]

Hi Frank,

Thank you for your tips about the current probe, I will update my work progress here

I planed to implement a DC/AC current to voltage add-on board (Based on hall current sensor ACS712) for oscilloscope, after I finish the Magnetic Field Probe. And then I will start to work on the AC current probe.

I aim to design a hall current probe for the oscilloscope with less than 5 bucks cost. The hall current probe will have the BW from DC to 80KHz, and 30A measurement range. If I finally get a hall current sensor and a AC coil current sensor, then I think I should able to measure the current in a relative wide frequency range. 

Thank you for your help! You taught me a lot of things.



Neil

[neilhao]

My first experiment about the probe.

I connected the probe to the RF explorer directly, and I measured a 100Mhz oscillator. It seems the output wave of the oscillator is not sine wave or square-wave...


Neil

[neilhao]

Experiment 2: Using oscilloscope to measure a 100Mhz oscillator, directly. It seems  the oscilloscope is good enough to measure the EMI directly.

[neilhao]

Experiment 3: Using oscilloscope to measure a 100Mhz oscillator with preamplifier (LMH6552, gain =10). The preamplifier is working, but according to the result in the Experiment 2, I think the preamplifier is not necessary in the most cases, since it will make the measurement circuit to be more noisy.

[neilhao]

I did a little research about how to select the proper shielding materials, the report is here:

http://uniteng.com/index.php/2013/06/21/electromagnetic-shielding/

The experiment data about this topic will be updated later.


Neil

[neilhao]

Topic about how to measure the Channel Power using the RF explorer:

Channel power measurements (26 July 2013) http://uniteng.com/index.php/2013/07/26/channel-power-measurements/


Neil

[neilhao]

Topic about Low cost Preamplifier:

Preamplifier 30 MHz to 6 GHz (August 18, 2013) http://uniteng.com/index.php/2013/08/18/preamplifier-30mhz-to-6ghz/




Note: agilent has the similar thing for US$ 2,421 (http://www.home.agilent.com/en/pd-146774-pn-87405B/preamplifier-10-mhz-to-4-ghz?&cc=US&lc=eng), but we can DIY a preamplifier for less than 50 bucks.

Neil

[neilhao]

Topic about Philips Sonicare Electric Toothbrush Measurements without Teardown:

Conclusion

The Philips Sonicare Electric Toothbrush is designed to brush the teeth 262 times per second, which means the teeth were supposed to be brushed 31440 times in exactly 2 minutes. This toothbrush is definitely more efficient than human hand.

However, the wireless charging design is still not perfect, because it is continually generating a 68.34 kHz alternating magnetic field which is not power efficiency. Comparing with the Qi Wireless Charging Standard [2], Qi Charger has better power efficiency since it will not continually generating an alternating magnetic field when there is no receiver on the charging pad or the receiver is fully charged. Hope Philips will improve the wireless charger design in future.

More details:
Philips Sonicare Electric Toothbrush Measurements without Teardown (January 25, 2014) http://uniteng.com/index.php/2014/01/25/philips-sonicare-electric-toothbrush-measurements-without-teardown/




Neil

[SArepairman]

Does anyone have backups of this website? 

revive this topic!

i want to make oscilloscope accessories!!!!!!!!!!!