"Unapproved" input filtering

[Pentium4]

Non safety approved input filtering seems to be becoming a lot more common. I have a KME power supply (Branded Pixxo) that has 3 Y caps, and 3 X caps. One of the X caps is directly connected to the AC receptacle, and this one has all the normal safety approved logos and what-not. The other 2 X caps and 3 Y caps are after the fuse and do not have the approved logos. Even the X caps just say "470K 200V" in huge letters and nothing else. Do these still do the normal "duty" as safety approved X caps? The Y caps look like ceramic caps with a blue sleeve, these serve no purpose? If they were to fail would they just blow the fuse?

[Pentium4]

Re: "Unapproved" input filtering

Here's a picture of the writing on one of the X caps:

[goodpsusearch]

Re: "Unapproved" input filtering

Oh no! If the yellow cap doesn't have anything else written on it except "104K 400V" then it is not a proper X cap and it should be considered unsafe for AC input filter.

[Pentium4]

Re: "Unapproved" input filtering

Thanks, I'll look for some replacements. A few more questions. How come some Y caps have a ferrite bead on one of the legs? (Usually the one that is grounded) And sometimes there's a Y cap (or fake one) next to the bulk capacitors connected to the ground of the corner of the PCB, does this Y cap have to always match the other 2 (or 4) before it?

[Pentium4]

Re: "Unapproved" input filtering

Also, how come some X caps have a resistor in parallel with them and some don't?

[c_hegge]

Re: "Unapproved" input filtering

^
The resistor is there to discharge it when power is disconnected. There really only needs to be a resistor for one of them, as the others are connected in parallel with it and will discharge through it too. If there is no resistor at all, though, then it's a corner cut.

[Pentium4]

Re: "Unapproved" input filtering

Cool, thanks for the info. Even most cheapo units have at least one. Any input on the other questions? Pun intended

[SIDMX]

Re: "Unapproved" input filtering

...the X caps just say "470K 200V" in huge letters and nothing else.The Y caps look like ceramic caps with a blue sleeve, these serve no purpose? If they were to fail would they just blow the fuse?

Those sure do their duty but in an undesired way, failure mode of ceramic caps tends to be short-circuit ITOH failure modes for X and Y capacitors are open-circuit mode (or high impedance), Y caps should have low leakeage current because they are used in line to ground applications, also X and Y rated caps are designed to withstand high voltage pulses (for a limited time) without going into failure mode, something that non safety approved caps can't do. I've attached some files about safety ratings caps and about EMI filters, hope it helps you as much as they helped me.

...How come some Y caps have a ferrite bead on one of the legs? (Usually the one that is grounded) And sometimes there's a Y cap (or fake one) next to the bulk capacitors connected to the ground of the corner of the PCB, does this Y cap have to always match the other 2 (or 4) before it?

The ferrite bead (or ring) is there just to avoid high frequencies or so I believe, because that is what a ferrite bead do when there's no winding, it acts as a very small inductor so it can only block high frequencies. I have noted that there is some sort of correlation with the use of those ferrite beads when there's a jumper between traces to the location of the component or the component has long legs, so maybe this has to do something with avoiding the pickup of RF by the jumper or leg (that is acting like an antenna) and nothing more than that, but I maybe totally wrong. About the Y cap close to bulk caps, I have no idea what noise is filtering, maybe someone else could tell us both what's the purpose of that cap.

...Even most cheapo units have at least one...

I wish even the most cheapo have one but at least in my country NONE of the real cheapos has it.

[Pentium4]

Re: "Unapproved" input filtering

Those sure do their duty but in an undesired way, failure mode of ceramic caps tends to be short-circuit ITOH failure modes for X and Y capacitors are open-circuit mode (or high impedance), Y caps should have low leakeage current because they are used in line to ground applications, also X and Y rated caps are designed to withstand high voltage pulses (for a limited time) without going into failure mode, something that non safety approved caps can't do. I've attached some files about safety ratings caps and about EMI filters, hope it helps you as much as they helped me.

Thank you very much for the info! So it isn't worth risking it. Fortunately, input filtering components don't seem to fail very often.

The ferrite bead (or ring) is there just to avoid high frequencies or so I believe, because that is what a ferrite bead do when there's no winding, it acts as a very small inductor so it can only block high frequencies. I have noted that there is some sort of correlation with the use of those ferrite beads when there's a jumper between traces to the location of the component or the component has long legs, so maybe this has to do something with avoiding the pickup of RF by the jumper or leg (that is acting like an antenna) and nothing more than that, but I maybe totally wrong.

Cool. I get in a lot of In Win units, and these usually use ferrite beads on the Y caps. They're usually so covered in tan/dark brown glue that I usually scrap them for parts. On this one with 3 "Y" caps, I think I'll just install a matching pair and just remove the third one. I'll post some more pictures of the unit on Monday. Thanks for attaching those documents!

[momaka]

Re: "Unapproved" input filtering

As SIDMX mentioned, the difference between safety-rated X/Y caps and regular caps is their failure mode. Safety Y caps usually fail open, not short like normal ceramic capacitors do. With X caps, I think it is something similar, though I am not exactly sure.

If these non-safety caps not sitting behind a fuse, and they do fail shorted, you can get a hard short between Live and Ground (via non-safety Y caps) or Live and Neutral (via non-safety X caps). If you have a bad braker or bad house wiring, that makes things quite risky (but luckily, most cheapo PSUs also have very thin input wires so they may burn before the house wiring goes, if the breaker doesn't trip).

That said, even if the non-safety caps are sitting behind a fuse, you still have the risk of a live case or partially live case if grounding on the PSU is not good. As SIDMX mentioned, safety Y caps have low leackage, so touching said energized case is not as dangerous.

As far as functionality, there isn't much difference between the regular and safety X/Y caps. Both will supress RFI and EMI equally well.

[SIDMX]

Re: "Unapproved" input filtering

...Safety Y caps usually fail open, not short like normal ceramic capacitors do. With X caps, I think it is something similar, though I am not exactly sure.

To be honest I'm not 100% sure either because I couldn't find a contundent argument to this, going by the citation of IEC 384-14 (first pic) that is commonly found in X2 datasheets I do believe that there is no restriction for a short circuit failure mode even if Wikipedia says the opposite, if there are not restrictions why some PSU manufacturers put X2 caps before the fuse? that is a potential fire hazard in case of a bad breaker or bad wiring as you mentioned, however there are some X2 caps that don't tend to go short circuit, these are the ones with self-healing feature, reading from Wima excerpt (second pic) note how they don't put short circuit as a disadvantage for their metallized construction, one more example from Faratronic (third pic) also implies that they should not go short circuit so at first I think, ok, so as long as they are metallized film they shouldn't short, but a quick look at datasheets from other metallized caps (see attached Okaya LE Series ds) made me realize that there's no sure thing so maybe the answer for what is the failure mode of an X cap? should be: Look at the datasheet!

Also we need to be careful about what types are used in PSUs (or used for repairs) specially the ones before the fuse and if using a non self-healing one better to put it after the fuse don't you think?

[momaka]

Re: "Unapproved" input filtering

Very interesting post, SIDMX... but now I am more confused .

I always thought metalized poly caps (i.e. X caps) are bigger than their regular film cousins for a given capacitance. Yet, the article by WIMA states the opposite? Or am I way off here?

From what I understand from the text excerpt from the 3rd picture, when a metalized poly cap encounters a high voltage spike, the energy from the arc can take out the metal coating that connects the affected layers, thus removing the short. The result is loss of capacity, but the capacitor "heals itself". So if that is true, then I guess a badly abused metalized poly cap eventually looses so much capacitance that it is as good an an open circuit. Does that sound right?

In any case, I don't like when I see X caps before the fuse. If the board has a spot or more than enough X caps, I do consider removing any "excess" X caps from the receptacle (actually, I did do that on two PSUs of mine). The nice thing about that is, I have extra caps to put in other PSUs that might need them .

[Pentium4]

Re: "Unapproved" input filtering

Okay here are some pictures of the new input filtering and an "overall" look of the PSU. Looks decent. Replaced the 20A ultra fast on the 12V with a 30A ultra fast, and replaced all the caps. I love power supplies

To be honest I'm not 100% sure either because I couldn't find a contundent argument to this, going by the citation of IEC 384-14 (first pic) that is commonly found in X2 datasheets I do believe that there is no restriction for a short circuit failure mode even if Wikipedia says the opposite, if there are not restrictions why some PSU manufacturers put X2 caps before the fuse

KME obviously wanted to be cheap by getting non safety approved input filtering but they must have had a reason to put a safety approved X cap before the fuse?

That said, even if the non-safety caps are sitting behind a fuse, you still have the risk of a live case or partially live case if grounding on the PSU is not good.

How can they get away with this? Sounds very dangerous...

[SIDMX]

Re: "Unapproved" input filtering

I always thought metalized poly caps (i.e. X caps) are bigger than their regular film cousins for a given capacitance. Yet, the article by WIMA states the opposite? Or am I way off here?

Well, I'm not sure but since the text excerpt came from a Wima catalog, that may be true only for Wima caps.

From what I understand from the text excerpt from the 3rd picture, when a metalized poly cap encounters a high voltage spike, the energy from the arc can take out the metal coating that connects the affected layers, thus removing the short. The result is loss of capacity, but the capacitor "heals itself". So if that is true, then I guess a badly abused metalized poly cap eventually looses so much capacitance that it is as good an an open circuit. Does that sound right?

Yeah that's what I understood too, although neither Wima nor Faratronic use the phrase "open circuit", in fact I could only find one manufacturer that openly write "failure mode is open circuit" see here at the middle of the page.

Okay here are some pictures of the new input filtering and an "overall" look of the PSU. Looks decent. Replaced the 20A ultra fast on the 12V with a 30A ultra fast, and replaced all the caps. I love power supplies

Nice job on that Pixxo/KME! ...there's a way to put another cap for +12V rail before the pi coil in these KME units

[budm]

Re: "Unapproved" input filtering


http://www.capacitorguide.com/filter-capacitor/
http://www.tecategroup.com/capacitor...capacitors.php

[Pentium4]

Re: "Unapproved" input filtering

Nice job on that Pixxo/KME! ...there's a way to put another cap for +12V rail before the pi coil in these KME units

Nice! How'd you get that thing in there? Yours looks to have a bigger core transformer.

http://www.capacitorguide.com/filter-capacitor/
http://www.tecategroup.com/capacitor...capacitors.php

Good read, thanks budm!

[SIDMX]

Re: "Unapproved" input filtering

http://www.capacitorguide.com/filter-capacitor/
http://www.tecategroup.com/capacitor...capacitors.php

Thanks budm, if I get it right Kemet says that their X caps or at least the ones mentioned in that file don't short, Tecate (there's a beer brand in MX by the same name ) says that theirs usually go short circuit and capacitorguide implies that X caps can fail in both ways. So this supports the argument to check the datasheet to really know how a particular X cap will act in failure mode.

Nice! How'd you get that thing in there? Yours looks to have a bigger core transformer.

It's really easy, you'll need to relocate one jumper that is used to reinforce the current capability of the ground path, that jumper is placed just below the group regulation coil and you can reinstall it in the solder side, soldering it only on the surface so that one of the holes used by the jumper gets free to use for the cap, then clear an unused hole that is covered by solder in the point where the +12V winding coming from the regulation coil is soldered to the +12V output coil and that's it, just place a cap in there.The first pic shows in red the place where the jumper used to be and marked in green the spots that are now used for the cap, and in third pic the jumper has been reinstalled in the solder side.
About the size of core, I'm not so sure, I don't trust KME stickers but it is posible.

[Pentium4]

Re: "Unapproved" input filtering

SWEET! Thanks SIDMX for sharing, that's awesome I'll have to do that to mine as well. BTW, nice custom RS cap

I'm going to post this in power supply build quality pictoral but figured I'd post it here too, another classic example of "unapproved" or no input filtering turned into a useful filter (there is a 560kΩ resistor under the small X cap). Scrapped input parts from the following dead PSU's: Bestec, CWT ISO, Allied, Antec SP500. I also added a bridge rectifier heatsink from a dead piece of trash Leadman

BEFORE:


AFTER:

[kaboom]

Re: "Unapproved" input filtering

Looks like you had fun- I like this!

Beware of the possibility of instabilities if you shift the dominant pole around. Depending on the margins in that supply's compensation, you might not be able to get away with putting a cap before that PI-choke.

I did the same in a supply I upgraded without any trouble. It, too, was a half-bridge, 5V-based design, so less emphasis on the +12 regulation may work in your favor.

Hope you can find something here that helps:
https://www.badcaps.net/forum/showthread.php?t=14405

Edit- Looks like I added the cap after a PI choke...

[SIDMX]

Re: "Unapproved" input filtering

SWEET! Thanks SIDMX for sharing, that's awesome I'll have to do that to mine as well. BTW, nice custom RS cap

You are welcome! Yeah I really suggest you to try this mod because even if this units have been designed to use only one cap for 12V rail IMHO one cap is not enough specially if the power supply is pushed hard (ie- a somewhat power-hungry system), I don't have a load tester nor a scope but this mod resolved the issues a friend was having in his custom-built entry-level gaming computer (FX-8120, HD 7770, N68-VS3 FX, 8GB DDR3, Hard Drive and DVD burner) that system was presenting sudden shutdowns and random BSODs, the system was used only for six months before the issues appeared, he was reluctant to change the PSU (Pixxo) so I offered him a recap and he agreed. Long story short the recap only fixed the BSODs but shutdowns persisted to a lesser extent until that second cap in 12V rail was added and it's been working perfect for a year now.

Looks like you had fun- I like this!

Beware of the possibility of instabilities if you shift the dominant pole around. Depending on the margins in that supply's compensation, you might not be able to get away with putting a cap before that PI-choke.

I did the same in a supply I upgraded without any trouble. It, too, was a half-bridge, 5V-based design, so less emphasis on the +12 regulation may work in your favor.

Hope you can find something here that helps:
https://www.badcaps.net/forum/showthread.php?t=14405

Edit- Looks like I added the cap after a PI choke...

Thanks for the info kaboom, it is very interesting and very helpful. I see now that I haven't been as careful about compensation as I should, will do some research about it.