Mains conducted EMC test with picoscope?

[The Electrician]

The EPSMA web site says a new standard is coming. The one you referenced (**broken link removed**) is a bit old.

I don't see how your power factor can be .99+ with the amount of distortion a 5 step waveform has. See "distortion power factor" at: https://en.wikipedia.org/wiki/Power_factor

 

[Flyback]

I don't see how your power factor can be .99+ with the amount of distortion a 5 step waveform has. See "distortion power factor" at:

Thanks, the 0.99 power factor also surprised me but on the simulator we took the Active Power and the Apparent power, and went [Active Power]/[Apparent Power] and it comes out as 0.99

We took Apparent power by VRMS*IRMS (input)

And the Active power is gotten by ALT-RIGHT-CLICKING the source whereby it takes the cumulative sum of the instantaneous v*i values over one mains cycle....so it effectively integrates the v*i product at each instant with respect to time, and then divides by the period.

 

[kubeek]

As I guess you already know, simulation is usually not even close to the real deal. What power factor does your actual device have?

 

[Flyback]

The actual LED driver product is 0.99 in power factor according to the Hameg Power analyser. I took that reading when there was no input filter attached to it whatsoever.

 

[Flyback]

This relates to the Mains input current of post #3 above.
Our offline product with this current waveform, has failed EMC
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?

 

[kubeek]

Now that is absolute nonsense. Especially on a test that is very easily reproduced in a lab.

 

[Flyback]

Thanks, but it is not at all easy to reproduce the test in a lab......for a start, you have to have a grounding place in your lab which is very tightly connected with earth.....and if your lab is on insulative clay soil, that will be difficult......you have seen a conducted emc lab...there is all that grounded metal sheeting all round the room......thats not cheap to make or find space for.

Then you have to pay 2.5 £grand for a LISN and a spectrum analyser

 

[JimB]

Apparently they (the EMC test houses) secretly, without telling you, put an injection circuit at the mains input to your equipment.

There are some classes of equipment which are tested for susceptibility to noise (voltage spikes) on the mains supply.
Why would the do it secretly without telling you?

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...

Unless the test house is somewhat corrupt, why would they do this.
The equipment does not "have to be fitted with a filter" it just has to meet the standards. If it can meet them without a filter - job done.

do you agree?

Seems to be your favourite expression when launching into frequent polemic rants.
No, I do not agree.
I think that you are completely wrong.

Thanks, but it is not at all easy to reproduce the test in a lab......for a start, you have to have a grounding place in your lab which is very tightly connected with earth.....and if your lab is on insulative clay soil, that will be difficult......you have seen a conducted emc lab...there is all that grounded metal sheeting all round the room

Yada yada yada....
Those are the test conditions REQUIRED by the STANDARDS.

It is not necessary for you to have all that to do your own "Pre-Compliance" tests in-house, which you can compare with the test house results.
If the test house results are say 10dB over the limit, then you need to reduce your own measure values by >10dB to be within limits.

Then you have to pay 2.5 £grand for a LISN and a spectrum analyser

Just part of the cost of doing business, a bit like hiring a competent engineer.

Just put on your big boys trousers, suck it up and stop moaning.

JimB

 

[Flyback]

There are some classes of equipment which are tested for susceptibility to noise (voltage spikes) on the mains supply.
Why would the do it secretly without telling you?

Because surely a offline product without an input filter is like a red rag to a bull for them...and they will ahve suspicions about it, that it might get disturbed by nearby switching equipment and start being noisy, so they will do their darnedest to fail the product.

 

[JimB]

Because surely a offline product without an input filter is like a red rag to a bull for them...and they will ahve suspicions about it, that it might get disturbed by nearby switching equipment and start being noisy, so they will do their darnedest to fail the product.

Just because you are completely paranoid it does not mean that they are out to get you.

JimB

 

[Flyback]

..If they pass a product which subsequently fails, they are in big doo doo...if they see a product with no filter whatsoever, they fear the possibility of subsequent failure..no matter how remote, so they just try and sink it out of sight....its all about possible litigation, and their reputation.

 

[JonSea]

Maybe it's time for a little (long) vacation? Perhaps far away where everyone isn't in the conspiracy against you?

 

[Flyback]

Thanks, but its the only way to explain why our non SMPS LED driver has failed EMC.....You can see in post #3 how EMC_benign the mains input current is...and the "switchings" of the linear regulator happen only once every ms, and are far less in di/dt and dv/dt than the switchings of an SMPS.

 

[JimB]

Thanks, but its the only way to explain why our non SMPS LED driver has failed EMC.....You can see in post #3 how EMC_benign the mains input current is...and the "switchings" of the linear regulator happen only once every ms, and are far less in di/dt and dv/dt than the switchings of an SMPS.

What you mean is that the whole world is wrong and that your LED driver cannot possibly be wrong.
Not only that, you are unwilling the do your own tests to prove that your LED driver is right or wrong.

JimB

 

[Flyback]

Id love to do the tests, the company cant afford an emc conducted emissions set up....with all that grounding etc, and the spec analy and the LISN £2500....no can do

 

[ronsimpson]

US ebay.com "lisn"
$250 to $500usd. There are several in that range.
Every thing you need is on ebay. (not calibrated)

there is all that grounded metal sheeting all round the room

I did not do that.
I used a sheet of aluminum to cover a work table. 3x6 feet??
Connect all grounds to the desk top, then connected to earth ground but that did not make any difference. It is important to connect everything to the same ground. (does not have to be real earth ground)

 

[kubeek]

You can make your own LISN, just use the correct components. You can borrow a spectrum analyzer.
Stop finding lame excuese and start solving the real problem - your device producing emissions above the required limit.

And stop with all that "my dI/dt is lower than in a smps therefore I should pass" crap. A smps has good low esr caps on the input which drastically lower the current spikes upstream of the psu, usually have a PFC too, and most have a common mode choke and X and Y filter caps. You said you have none of that, so even though you have lower dI/dt you are letting all of that current go straight to mains, which is exactly why you are not passing the emc test.

 

[ronsimpson]

On ebay there is one spectrum analyzer like mine $250(usd). Probably should get something better.
There are some nice unites in the 500 to 1,000 range. I have used most of them.

Here is a new one. I do not know anything about it.
**broken link removed**
/122370793306?hash=item1c7dde1f5a:g:UGEAAOSwSlBYsOiF

There are several "USB-spectrum analyzer". Again; I do not know anything about it.

There is one for 110,000.00. lol

 

[Mosaic]

Flyback, even a Rigol DSA can do quasipeak EMC etc...gets u in the ballpark.

 

[ci139]

get 100mH

(1-st -- i ** don't do HV nor commercial stuff) i see it as large filter inductor for below ** 20V 5A DC power (attempt to confirm the value for 150W revealed) http://www.compliance-club.com/archive/keitharmstrong/design_techniques3.html (unlike many others this guy has an email , wow ???)