Continue to Site

Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

EMC of non switch mode power supply equipment?

Status
Not open for further replies.
T

treez

Guest
Hello,
Our non SMPS equipment that connects to the mains draws the following attached input current waveform from the mains. What do you think will be the mains harminics, and the EMC connducted emissions? (pass or fail)

There are only 20 switching operations of transistors over each mains cycle, so surely there will be very little conducted EMC problem?
 

Attachments

  • Input current.jpg
    Input current.jpg
    76 KB · Views: 64

Hi,

There are high frequency components included. One can see this in the high dV/dt in the signal.

To analyse it I'd use an FFT on the signal.

Klaus
 
  • Like
Reactions: treez

    T

    Points: 2
    Helpful Answer Positive Rating
Thanks, but sorry That is the mains input current that i show in post one. Also, we would need a LISN to make meaningful high frequency spectrum measurments, and we dont have a LISN. We dont know if its worth buying one because apparently it has to be very tightly bonded to earth ground, and we are working on how we would assure this. Ie , where in our building do we make the tight earth ground connection?...presumably the eaerthing rod point, but anyway, i am digressing to this post....
https://www.edaboard.com/threads/366369/#post1569208
 

The time base on that input current scan is approx 1ms/div. The “craggy” bits that you see have di/dt’s far less than a normal SMPS's switching transitions. So why is it failing as if it were an SMPS?

Also, a normal SMPS switches very noisily and with high di/dt about 100000 times per second……our power supply switches just 1000 times every second, and each of those switching transitions is far less in di/dt than an individual switching transition of a typical SMPS. So why is it failing EMC?, ...when it has 100 times less switching transitions than a typical SMPS?


Since our switching transitions occur every 1ms, then do you think we stand a good chance of failing the mains emissions at that mains harmonic?, ie the 20th mains harmonic frequency will be a harmonic failure for us?
 
Last edited by a moderator:

Hello, this concerns the above luminaire..

Hello,
Our 150W nonisolated, linear regulator based offline LED light has failed EMC conducted emissions from 150khz to 6MHz.
This light is not switch mode power supply based, but does have a 1W high voltage buck converter based on LNK302 in order to provide the bias power to the control circuit.

This buck converter has a pi filter upstream of it, consisting of a 470uH inductor (datasheet below) and two 1812 size 220nF ceramic capacitors.

Do you think the EMC failure could be due to conducted emissions flowing through the interwinding capacitance of the 470uH inductor?

744045471 Inductor datasheet (470uH)
https://katalog.we-online.com/pbs/datasheet/744045471.pdf
 

The datasheet specifies its self resonant frequency at 5MHz, so that's unlikely. A plot of the spectrum might help.

None of your above analysis includes any mention of input filters. Most offline SMPS will have one, do you?
 
  • Like
Reactions: treez

    T

    Points: 2
    Helpful Answer Positive Rating
Thanks , but no its linear regulator based and has no filter whatsoever....only the 1w buck bias supply has a filter, but that is downstream of the mains rectifier, and also downstream of a diode.
We dont want a filter. Filters have capacitors, and capacitors go short after suffering high mains transients. The pi filter i spoke of is downstream of a diode so it doesnt hold the voltage up on the DC post rectifier bus....so its ok........
 

This relates to posts #1-4 above.
The EMC failure is bad for us. Our product has no filter at all, not even a decoupling capacitor, ...do you think that the EMC test house saw this, and then secretly but deliberately activated a noisy_mains_supply_injector to try and upset our equipment, and interfere with it, such that it started malfuncationing and then failed EMC?

Apparently they (the EMC test houses) secretly, without telling you, put an injection circuit at the mains input to your equipment. That typically consists of an inductor in series with each line that can carry the equipment current and has a high inductive impedance at the frequencies of interest. Then they use capacitors to inject the noise into the lines between the inductors and the equipment under test.

I am sure the EMC test house secretly do this sort of thing, especially when they get a product which has no filter...do you agree?
 
Last edited by a moderator:

A test house does nothing secretly. They perform tests according to specifications, fully documenting the test conditions.

If you have any doubts about how they interpret the specifications, join the tests or discuss the procedure in advance.
 

This relates to the above. (it concerns the same LED driver)
Please help us to design the offline EMC filter to allow us to pass EMC.....
Hello,
We have a 150W linear current regulator based offline LED driver which has failed conducted EMC, as in the following EMC scans (attached).
The product normally has absolutely no AC mains filter, and no post rectifier bus capacitance whatsoever.
For one of the scans, a 220nF X2 capacitor was added to the AC input of the product, and this brought the ‘quasi peak’ line down by 30dB. However, the ‘Average’ line was only brought down by about 8dB.
This confirms to us, that when designing an conducted EMC filter for a product that’s failed EMC, one should mainly look at the “average” scan line and not the quasi peak line. Do you agree?
 

Attachments

  • Conducted emissions _220nF input capacitor.jpg
    Conducted emissions _220nF input capacitor.jpg
    83.8 KB · Views: 92
  • Conducted emissions _no input capacitor.jpg
    Conducted emissions _no input capacitor.jpg
    83.3 KB · Views: 102

Conducted EMC problem is common mode problem

Hello,
Our 30W, Offline, Power factor corrected LED driver has failed Conducted EMC in the Live connection. (it does NOT use any Switch Mode Power supplies, just controlled linear current regulators.)
It has actually passed conducted EMC in the Neutral connection.
(the EMC scans are attached.)
We take it that this means that the problem is a common mode conducted problem, and that we need a common mode choke?
The strange thing is that we have a similar product that operates at 150W. –Its Live and Neutral conducted EMC plots are the same.
I am wondering if their has been a breach in the insulation sheet that separates the PCB from the earthed heatsink....perhaps when we programmed it the programming pins went right through the sheet.
 

Attachments

  • Conducted EMC _Live.jpg
    Conducted EMC _Live.jpg
    81.4 KB · Views: 102
  • Conducted EMC _Neutral.jpg
    Conducted EMC _Neutral.jpg
    80.4 KB · Views: 104
Last edited by a moderator:

Re: Conducted EMC problem is common mode problem

To handle EMI generated by your product, you should have at least an idea which part of the circuit is the interference source.
 

Re: Conducted EMC problem is common mode problem

Banks of leds are kind of swiched in and out by the IC of the product as the mains half sines go up and down at the HV DC bus....so i reckon the IC that does this 'handing over' from bank to bank of leds is causing some di/dt in the current which is creating EMC...what confuses us is why its not the same in the live and neutral?.......the product has no filter whatsoever, but then again, it is not a switch mode power supply.

So do you agree we have a common mode noise problem?
 

Ref the conducted EMC plots in post #10 above, the live current is obviously not equal to the neutral current....so the "balance" must have flowed in the earth conductor.....so do you think that if many of these products are plugged in to the same mains circuit, then the RCD could get tripped?
 

Thanks,
For the Failed EMC scans in post #11 above, do you think it needs a common mode filter , or a differential mode filter in order for it to pass? (one scan is conducted emissions in the live line of the product, the other in the neutral line of the product...they are differently shaped plots)
 

Hello , this is related to the above in a way, so please may i place it here?...

Our product actually resides at the end of 8 metres of mains cable.
This obviously provides a convenient few microhenries of line inductance which would help us in passing EMC.
Do you think the regulations would allow us to get conducted EMC certification with the 8 metres of cable connected to the product? Or would we have to have just say a meter of mains cable during the conducted EMC scan?
 

Hello,
The attached is the mains input current and voltage to a 115W Offline LED driver (not Switch Mode Power Supply based). It has an AC Mains input filter comprising a 33mH inductor and a 220nF X2 capacitor downstream of the inductor.
The efficiency is above 86%.
The Power factor is 99%
Do you think there would be difficulty getting this through the conducted EMC test and the mains input harmonics level test?

Its hard for us to say with all the windowing functions, frequency window sizing etc that the ECM test houses use in their measurements.
 

Attachments

  • Mains input i  and v waveform.jpg
    Mains input i and v waveform.jpg
    56.6 KB · Views: 95

I really don't know what answer you could possibly expect by posting a screen capture of a jagged waveform.

Sounds like your first step should be to access the detailed test specifications. Until then you're shooting blindly. If you're hoping that someone here would hand them over for free, I think you're out of luck.
 

thanks, the waveform in post #18 above looks jagged on the high timebase, but actually its very rounded when you close in and it looks like the maximum frequency in there is about 33khz, which is well below the limit start point of 150khz....so surely i believe we should be ok?
 

Status
Not open for further replies.

Similar threads

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top