Topics Topics Edit Profile Profile Help/Instructions Help Member List Member List  
Search Last 1|3|7 Days Search Search Tree View Tree View  

EMC

:: EPE Chat Zone ≠:: ≠Radio Bygones Message Board :: » EPE Forum Archives 2010 - » Archive through 02 October, 2011 » EMC « Previous Next »

  Thread Last Poster Posts Pages Last Post
  ClosedClosed: New threads not accepted on this page        

Author Message
Top of pagePrevious messageNext messageBottom of page Link to this message

echase
Frequent Contributor
Username: echase

Post Number: 541
Registered: 07-2007

Rating: N/A
Votes: 0 (Vote!)

Posted on Thursday, 22 September, 2011 - 12:01 pm:   Edit Post Delete Post Print Post    Move Post (Moderator/Admin Only)

Anyone understand EMC? I built a high power phase controlled dimmer with a mostly resistive load that is injecting spikes 100 times per sec into the mains (conducted emissions) of 6V high upon load switch on. After rising very quickly to 6V it starts ringing at 1MHz and decays exponentially in amplitude and frequency with a time constant of about 5us (5 cycles of interference). I measured this on the neutral. Itís probably the same on the live or worse, but I can't discriminate between the 230VAC and the interference with my scope.

The relevant EMC limit EN50081-1 is 56dB relative to 1uV at 1Mhz as this is classed as continuous noise even though spiky. I am trying to work out if I need a filter to meet EN50081-1 e.g. how many dB is 6Vpp at 1MHz? dB = 20 * Log(6/1.41/2/0 .000001) gives 146dB and no simple filter I know will attenuate that to 56dB; 15 -30dB attenuation is more common. Or can I average the spikes out to calculate a lower dB? A simple capacitor might reduce the peak a lot.

My TV etc. is on the same ring main and does not notice the spikes but someone else with a similar unit claims his broadband router is affected by it.

(Message edited by echase on 22 September, 2011)
Top of pagePrevious messageNext messageBottom of page Link to this message

echase
Frequent Contributor
Username: echase

Post Number: 542
Registered: 07-2007

Rating: N/A
Votes: 0 (Vote!)

Posted on Thursday, 22 September, 2011 - 05:06 pm:   Edit Post Delete Post Print Post    Move Post (Moderator/Admin Only)

146dB should have said 126dB.
Top of pagePrevious messageNext messageBottom of page Link to this message

mee
Frequent Contributor
Username: mee

Post Number: 122
Registered: 05-2010

Rating: N/A
Votes: 0 (Vote!)

Posted on Thursday, 22 September, 2011 - 11:07 pm:   Edit Post Delete Post Print Post    Move Post (Moderator/Admin Only)

Hi echase. Haven't a clue, sorry. In my simple mind, for something to be proven to comply with 'ElectroMagnetically Compatible' regulations, it has to be compared against the specifications of another device that has a reference EMC specification.
Eh?
I wonder whether the term 'EMC' was created by an interested party in the EU government's loby to make sure that no small manufacturer would ever dare to make anything that could compete with the big guys without paying an inordinate sum to have it approved.
I'm a supporter of the EU in principle. But do I give a about the curve of my bananas? I could go on for a long time about what they've done badly in my view, but I've also got a list of what the EU could have done to help us. Top of the list is creating an EU competitor to Paypal. Fantastic amounts of money travel from here to the US via Paypal and the charges are daylight robery. I've nothing against the US either!
Rant over.
If you're still reading this echase, I still haven't a clue, but can you provide an outline diagram of the circuit connections and a sketch of the waveform that you're seeing?
As an aside, we have a shop that's near to residential properties. A neighbour came to me to complain that every time that we used our vacuum cleaner (made by a well know UK maker (is there more than one?)), his radio was drowned out.
"Is there anything I can do with my antenna'" he said. "No," I replied, "but you will probaly solve the problem by providing the radio with a good Earth connection". That fell on deaf ears so I provided him with enough information to do 'Listen again' on the internet.
What I know about electronics is entirely self taught but in my experience, I can imagine that the "simple capacitor" might well make the promlem worse. For me, EMI is a black art. My first option would have been series inductors (plural). Most capacitors (apologies if you already know this but it might be of use to someone else) start to drop off in performance (resistance for want of a better word at this time of night) as frequency increases.
If Murata's still around, have a look at the data sheets. Cheers, M

(Message edited by mee on 22 September, 2011)
Top of pagePrevious messageNext messageBottom of page Link to this message

echase
Frequent Contributor
Username: echase

Post Number: 543
Registered: 07-2007

Rating: N/A
Votes: 0 (Vote!)

Posted on Friday, 23 September, 2011 - 10:45 am:   Edit Post Delete Post Print Post    Move Post (Moderator/Admin Only)

mee think you are right about the capacitor. The classical approach to reducing this sort of common mode interference (think it is common mode) is a common mode inductor in series which I guess effectively provides a high resistance to the 1 MHz so blocks it from getting into the mains.

I need to put my waveform through a FFT programme I guess. Would this do an RMS average of each frequency so effectively average the short 5us spike out over the full 10ms repetition period? I can transfer the Fluke scope screen shots to PC but not sure they would be in a format to enter into say PSpice to analyse. Or can I hand draw it into PSpice? Any ideas folks?
Top of pagePrevious messageNext messageBottom of page Link to this message

741
Frequent Contributor
Username: 741

Post Number: 341
Registered: 08-2005


Rating: N/A
Votes: 0 (Vote!)

Posted on Sunday, 25 September, 2011 - 01:43 pm:   Edit Post Delete Post Print Post    Move Post (Moderator/Admin Only)

Are you using zero-crossing detection?

Can you post a scope screenshot?

If your scope has the ability to save screenshot to file, maybe it can also save it as CSV data?

Maybe you can gain clues by measuring the frequency of the ring.

For an undamped LC, w0 = sqrt[1/(CL)]

In your case, it is damped, so the frequency you measure, w1 is given by

w1 = sqrt[w0^2 - y^2]

Where
y = R/(2L) = damping factor
Top of pagePrevious messageNext messageBottom of page Link to this message

rob_guyer
Member
Username: rob_guyer

Post Number: 6
Registered: 08-2011

Rating: N/A
Votes: 0 (Vote!)

Posted on Sunday, 25 September, 2011 - 11:13 pm:   Edit Post Delete Post Print Post    Move Post (Moderator/Admin Only)

I have performed EMC testing to the ANSII (USA) standard and I reasonably certain that the IEC standard would specify the test apparatus used to measure conducted interference for compliance to the voltage limit. In outline, the ANSII standard required the item under test to be inside a small Faraday cage with all leads brought through small holes and connected to their intended source/load.--An electrostatically shielded current transformer is then clamped to each lead in turn and connected through a step attenuator to a calibrated heterodyne receiver. The receiver had a gaussian IF passband and known second detector response. The output voltage was calculated by an empirically derived formula

Any other set-up might give a misleading result so your equipment could possibly be compliant. I used a very small HF radio glued to one end of a ABS plastic pipe to probe the innards of my device. Not at all satisfactory, so prevention is easier than the cure.
Top of pagePrevious messageNext messageBottom of page Link to this message

echase
Frequent Contributor
Username: echase

Post Number: 546
Registered: 07-2007

Rating: N/A
Votes: 0 (Vote!)

Posted on Sunday, 25 September, 2011 - 11:46 pm:   Edit Post Delete Post Print Post    Move Post (Moderator/Admin Only)

Will try a screenshot but need to install some software first to do it. It is a 1Mhz ring at first and then switches to around a tenth of that at <1/4 the amplitude.

It canít be zero crossing so at worst case on peak of sine wave it is suddenly connecting 300V to a 13A RMS load so is trying to instantaneously draw around 20A. Hence the spikes and ringing.
Top of pagePrevious messageNext messageBottom of page Link to this message

dselec
Frequent Contributor
Username: dselec

Post Number: 187
Registered: 11-2010

Rating: N/A
Votes: 0 (Vote!)

Posted on Monday, 26 September, 2011 - 08:08 pm:   Edit Post Delete Post Print Post    Move Post (Moderator/Admin Only)

Buy a simple line filter on ebay it has double t filter made from 4 coils and 3 capacitors

Administration Administration Log Out Log Out   Previous Page Previous Page Next Page Next Page