Not using X2 rated capacitors in AC filter for offlien SMPS?

Hello,
You can find a million internet reports on X2 capacitors…eg...
**broken link removed**
..but none of them state if it’s an actual regulatory requirement for a capacitor from line to neutral upstream of the diode bridge to be an X2 rated one. (in an offline SMPS)
I mean, surely it could be a ceramic and non X2 rated?..so what if it fails short….there’s a fuse upstream of it.
So is ceramic OK for AC filters in offline SMPS (upstream of the diode bridge)?

You can get X2 rated ceramic capacitors for 230V lines.

As for X and Y capacitor rating, the paper is quoting EN 132400. What does it say?

Thanks
https://shop.bsigroup.com/ProductDetail/?pid=000000000030065685
All these standards, the many thousands of them, are about £100 each, we simply cant afford them.

Which non-X2 capacitor do you particularly plan to use?

There can be no doubt that mains connected devices must withstand the respective voltage levels, including surge voltages specified for the grid category. The requirement is reflected in specifications for cable insulation, creepance and clearance and e.g. capacitor voltage handling. It's most critical as far as protection against hazardous contact voltage is involved.

You can consider line-to-neutral shorts less critical and rely on fuses. But can you be sure that a massive short takes place, a device can also catch fire under circumstances. Using capacitors with insufficient voltage specification might be rated as negligence in this case.

Which non-X2 capacitor do you particularly plan to use?

Click to expand...

We are thinking of this 1.5kV ceramic non-X2 capacitor......
**broken link removed**
...... with us having it downstream of a fuse i dont see how there could be a fire risk?

We also don’t see why, if X2 capacitors are a regulatory requirement, then why is it only a regulatory requirement to have them upstream of the bridge rectifier?

The Kemet catalog has this application note:

X7R dielectric is not recommended for AC line filtering or pulse applications. These capacitors and/or the assembled circuit board containing these capacitors may require a protective surface coating to prevent external surface arcing.

Click to expand...

**broken link removed**

Murata has 250VAC capacitors, 0.1 µF according to Japanese standards (GA255DR7E2104MW01L) and up to 56 nF X2 according to IEC (GA355XR7GB563KW06L).

You're right, you could use an ordinary capacitor there. Functionally, SMPS L-N capacitor are there strictly for EMC.

BUT it has to be of type X2-rated to meet electrical safety standards. "Ordinary" disk/film capacitors are a hazard across mains.
You can't pass for example IEC 60750 or IEC 61010 using an "ordinary" capacitor there, as it will fail the impulse/over-voltage test.
Also, regulatory agencies really really like components that have UL/CSA/IEC/VDE approvals.

You can't pass for example IEC 60750 or IEC 61010 using an "ordinary" capacitor there, as it will fail the impulse/over-voltage test.

Click to expand...

As for ceramic capacitors, I disagree, because they have generally a good pulse withstanding. But the surface arcing point mentioned by Kemet must be considered. It won't happen in the test but possibly after some months or years of operation. In so far I agree with the conclusion to use X2 rated.

Also, regulatory agencies really really like components that have UL/CSA/IEC/VDE approvals.

Click to expand...

If you don't go for UL or CSA, nobody cares.

Thankyou FvM, that quote from Kemet was very interesting..
Quoting from Kemet….

X7R dielectric is not recommended for AC line filtering or pulse applications. These capacitors and/or the assembled circuit board containing these capacitors may require a protective surface coating to prevent external surface arcing.

Click to expand...

…The thing is , the 100n, X7R capacitor that I mentioned in post #5 above is rated to 1.5kV. So there is of course no chance of it ever experiencing “surface arcing” at any voltage up to 1.5kV. –And since we have a 275VAC MOV upstream of this capacitor, then this capacitor will never experience a voltage above 1.5kV……and so this ‘surface arcing’ will never happen…surely?

Also, if X2 rated capacitors really are mandatory upstream of the bridge rectifier, then they would also surely be mandatory downstream of the bridge rectifier? –Because after all, bridge rectifiers don’t prevent the passage of high voltage spikes.
-And we all know that non X2 rated capacitors get used downstream of the mains diode bridge…including X7R ceramic capacitors and electrolytic capacitors…..so are we now saying that this is wrong?....ie so are we now saying that all capacitors either across live_neutral or across the post rectifier DC bus, must be X2 rated?

You can't pass for example IEC 60750 or IEC 61010 using an "ordinary" capacitor there, as it will fail the impulse/over-voltage test.

Click to expand...

…are you sure about “IEC60750”?...i thought it was IEC60950 that was responsible for safety rating?
Either way, we simply can’t afford all these thousands of regulatory standards….they’re £100 each

treez said:

...……are you sure about “IEC60750”?...

Click to expand...

Sorry typo

The common safety standards I know that would require an X2-rated cap are:

IEC 60950 Information technology equipment - Safety
IEC 60730 Safety Standard for Household Appliances
IEC 61010 Safety requirements for electrical equipment for measurement, control and laboratory use

Don't forget your mains fuse is the only thing stopping the capacitor from becoming a bonfire - mains is a large fuse and so the energy available for a fire is huge. OK if you have a metal enclosure.

Some things that make an X2 capacitor unique:
Over-voltage withstand ability, tested.
Dielectric testing at applied voltage 24/7.
Harsher temperature cycling.
Flammability of plastics/epoxy used.
Lead spacing is larger with consideration of HV creepage/clearances and flux-wash.

They're just a tougher capacitor. Are we arguing to save a $1?


If you are using a (safety-approved) MOV on mains, that can allow you to reduce the over-voltage category for your product. I can be more specific if you tell me where the SMPS is used, mains ratings, etc.

The safety-tests actually hit your product with HV impulses. Depending on where your common-mode choke is, the DC bus may see the spike or not.

Depending on where your common-mode choke is, the DC bus may see the spike or not.

Click to expand...

Thanks, I believe you really meant “Differential “ mode choke here.
May I ask if you are you implying that the reason that capacitors downstream of the bridge rectifier are not mandated to be X2 rated is because the regulations assume that users will have a filter inductor upstream of the diode bridge?

Don't forget your mains fuse is the only thing stopping the capacitor from becoming a bonfire - mains is a large fuse and so the energy available for a fire is huge. OK if you have a metal enclosure.

Click to expand...

…this is our fuse…its Schurter 3403.0168.xx (250VAC, 1.6A)….
**broken link removed**

iSchurter 3403.0168.xx Fuse datasheet….
https://www.farnell.com/datasheets/...57.947372085.1495328258-1679746183.1489787856

…..its a 1.6A fuse, and with 3 Amps in it, it would blow inside 20 seconds…..with 10 Amps in it, it would blow in 0.2 seconds……I just cannot believe that any fire could happen with this fuse in our product….its right at the mains connector.

If you are using a (safety-approved) MOV on mains, that can allow you to reduce the over-voltage category for your product.

Click to expand...

Thanks yes we are using littlefuse MOV, and they are all safety rated.

I am just wondering, is there a sentence in EN60950 which says words to the effect of…”Any capacitor upstream of the mains diode bridge in an offline product must be X2/X1/X3 rated”

Our product is going to be a light up a post in a in a pub garden. [a bar (alcohol drink selling) with an outdoor section].

From all of your your kind posts above, it seems the concensus is that no one actually knows (or remembers) what the regulations actually say about this.....but most just assume it is a good idea to use an X2 rated capacitor upstream of the diode bridge?
....of course, there lies the other point......why should X2 capacitors only be mandated upstream of the diode bridge..why not downstream aswell?.....after all, diodes certainly dont block mains transients. (in fact , i doubt most filter inductors in most offline SMPS's will block them significantly)

- - - Updated - - -

IEC 60950 Information technology equipment - Safety
IEC 60730 Safety Standard for Household Appliances
IEC 61010 Safety requirements for electrical equipment for measurement, control and laboratory use

Click to expand...

Thanks, ours is outdoor lighting equipment...so presumably none of those standards have any information for us?

- - - Updated - - -

If we google the following…

“Which EN standard covers X2 rated capacitor use in outdoor lighting products?”

We got loads of links, none of which tell us which actual standard we need to actually buy to find this out…so we are stuffed

It would be good to know what product approvals you are seeking. Mains-connected products have safety and liability upfront, so a lot of extra learning for an engineer.

I did some research and find the safety-standards are calling for an approved X-capacitor:

IEC 60950;
1.5.6 Capacitors bridging insulation
A capacitor connected between two line conductors in a PRIMARY CIRCUIT, or between one line conductor and the neutral conductor or between the PRIMARY CIRCUIT and protective earth shall comply with one of the subclasses of IEC 60384-14 and shall be used in accordance with its rating.

{Table 1C shows Impulse test X1 4kv and X2 2.5kV}
-----------------------------------------------------
IEC 60730-1:
24 Components
24.1
... Capacitors used to provide radio interference suppression shall comply with the requirements of IEC 60384-14.
-----------------------------------------------------

IEC 60384 is the parent standard for X and Y caps.

IEC 60384-14: Fixed capacitors for use in electronic equipment– Part 14: Sectional specification: Fixed capacitors for electromagnetic interference suppression and connection to the supply mains
-----------------------------------------------------

But surprisingly another safety standard says:
IEC 62368-1: "No requirements for X capacitors. An X capacitor is not a safeguard against electric shock."


I can answer your other questions, if I know more about SMPS type (flyback?) and power output.

The thing is , the 100n, X7R capacitor that I mentioned in post #5 above is rated to 1.5kV. So there is of course no chance of it ever experiencing “surface arcing” at any voltage up to 1.5kV.

Click to expand...

Also, if X2 rated capacitors really are mandatory upstream of the bridge rectifier, then they would also surely be mandatory downstream of the bridge rectifier?

Click to expand...

1. There is, though. 2. No. You obviously don't understand the specific requirements for capacitors operated with large AC voltage. You need to consider effects like partial discharge that slowly destroy an insulator if it's not designed for AC load.

Thankyou

You obviously don't understand the specific requirements for capacitors operated with large AC voltage. You need to consider effects like partial discharge that slowly destroy an insulator if it's not designed for AC load.

Click to expand...

Do you have a reference we can read up on about this? We can find loads on Partial Discharge occurring in capacitors when subjected to sudden high dv/dt…however, these documents don’t tell us whether the partial discharge still occurs if the Vpk stays below the capacitor’s rated voltage?

Ill now come to the point if i may…

We make small lighting units for bars in a certain restaurant chain in Northern Ireland.This restaurant chain is actually our owner. We make Lighting over outdoor bar tables etc for them.
We have just lost our entire EU grant money due to brexit…..our product doesn’t have enough room for a filter because the designer didn’t include enough room. We now don’t have enough money to re-do the product to make it big enough for X2 rated capacitors. We now have to add capacitance and can’t fit an X2 rated capacitor in there (not enough room). If the non X2 ceramic capacitor is going to be dangerous, then we obviously wont use it…..the thing is, since we have a 1.6A fuse in the product, surely there will be no danger?
We just need to get the products through EMC testing then install the products…then when we find more investment, we can go back and replace the lamps when we have the money to make the products with enough room for x2 rated capacitors. (if its even necessary for us to do that).

the specific requirements for capacitors operated with large AC voltage

Click to expand...

Thanks for this…..considering the voltage downstream of the mains rectifier bridge, since we have power factor correction, our DC Bus voltage has a high AC content…..as you know its highly fluctuating DC. Surely if a ceramic capacitor is not allowed upstream of the diode bridge due to the AC…..then it is also not allowed downstream of the diode bridge due to the high AC content of the voltage waveform there?

You need to consider effects like partial discharge that slowly destroy an insulator if it's not designed for AC load.

Click to expand...

How long would it take the 100nF, X7R, 1.5kV ceramic SMD capacitor of post #5 above to be destroyed by partial discharge if used…
1…..Upstream of the mains diode bridge?
2….Downstream of the mains diode bridge?

- - - Updated - - -

No. You obviously don't understand the specific requirements for capacitors operated with large AC voltage. You need to consider effects like partial discharge that slowly destroy an insulator if it's not designed for AC load.

Click to expand...

Thanks…ceramic capacitors, as you know, get used in RC snubber circuits across secondary diodes in SMPS…..there they experience high AC voltages..but I don’t remember anyobody ever saying that ceramic capacitors should not be used for that?

I fear, the question can't be answered. It's not said that the non-X2 capacitors will be destroyed. I just wanted to emphasize that they are according to manufacturer information not designed for permanent large AC load and may at worst case suffer breakdown.

The X2 or similarly specified AC capacitors mentioned in post #6 are same form factor as your 1.5 kV capacitor.

Thanks, indeed they are the right profile....there is just the fact that as you know it says they are not suitable for AC line filtering applications...what they dont say, and what we need to know, is what exactly happens if you do use them in AC line filtering applcaitions?

Hello, this is related to the above and shows how X1 or X2 rated capacitors are not necessary in filters of offline power supplies.....do you agree?

Capacitors C1 and C2 of the following offline LED driver schematic (page 6)…..
http://www.mouser.com/ds/2/328/rdr378-269473.pdf
(BOM on page 10)

….these capacitors (C1 and C2) are simply just film capacitors and have no X1 or X2 rating….this shows that it is not a regulatory requirement to use X1 or X2 rated capacitors in offline power supplies….

C1 MEXXD31004JJ1:
http://www.dura-elec.com/pdf/MEX Series.pdf

C2 ECW-F2W334JAQ:
http://www.mouser.com/ds/2/315/ABD0000C183-947669.pdf

The post seems to confuse the original question. The shown capacitors C1 and C2 are DC filters. Noone suggested X2 for DC filters.

Thanks yes, but the point is that that CLC filter could equally have been placed upstream of the rectifier bridge instead…
…Is it so that if placed upstream of the rectifier bridge then the capacitors would have needed to have been X1 or X2 rated?
….and if that is the case…then why?......why does the position of the mains filter upstream or downstream of the rectifier bridge have any impact on whether or not the filter capacitors should be X1/X2 rated or not?

- - - Updated - - -

…or are we saying that if you want to use a mains input filter, then it is better to put it all downstream of the rectifier bridge because then you don’t have so many constraints on what type of capacitor you have to use?

There's a fundamental difference between AC and DC rating of capacitors. You should know if you read the datasheets and application notes. Shifting the filter capacitor to the other side of the rectifier might be no great change EMI wise, but it surely is related to capacitor rating.