Hello,
I stumbled upon TNY256 chip: http://www.powerint.com/sites/default/files/product-docs/tny256.pdf .
The chip is apparently used in many of those small switching power supplies plugged directly into electrical outlet.
On page 7 of the datasheet I see "Figure 12. 5.5 W AC Adapter with Universal Input (85-265 VAC)." - typical application circuit. And on page 8 "Figure 13. Recommended PC Layout for TinySwitch without Under-Voltage Lock Out Resistor.".
I don't understand why/what for "C5 2200 pF Y1 Safety" on page 7 is needed. I.e. I don't understand why capacitive connection between primary and secondary network is needed, especially taking into account that there is already parasitic connection through windings of T1 transformer.
Or that's the purpose - to protect from T1 parasitic capacitance of from breakdown, i.e. from possible electrical discharge going through insulation between primary and secondary winding of T1 ?
I stumbled upon TNY256 chip: http://www.powerint.com/sites/default/files/product-docs/tny256.pdf .
The chip is apparently used in many of those small switching power supplies plugged directly into electrical outlet.
On page 7 of the datasheet I see "Figure 12. 5.5 W AC Adapter with Universal Input (85-265 VAC)." - typical application circuit. And on page 8 "Figure 13. Recommended PC Layout for TinySwitch without Under-Voltage Lock Out Resistor.".
I don't understand why/what for "C5 2200 pF Y1 Safety" on page 7 is needed. I.e. I don't understand why capacitive connection between primary and secondary network is needed, especially taking into account that there is already parasitic connection through windings of T1 transformer.
Or that's the purpose - to protect from T1 parasitic capacitance of from breakdown, i.e. from possible electrical discharge going through insulation between primary and secondary winding of T1 ?
Your reply doesn't explain the word "safety". The word is typically used in conjunction with implied danger, and what is that implied danger is the thing I want to find out.
The text says it is part of an EMI filter. Not sure how this works, but as the cap connects the output to live mains it has to be a safety cap.
Yes and no. The "no" part is that EMI most of the times is mentioned in conjunction with intentionally introduced frequency jitter and just one time in conjunction with the capacitor in question.
The cap has to be rated for safety - an ordinary disc cap is not OK, even if it's rated for high voltage. In this case, a Y1 rating is needed, meaning that it is certified by the various safety agencies for to safely bridge the primary-secondary barrier and is 100% tested for a high withstand voltage. You can find out more about safety and EMI issues by checking out AN-15 at the PI web site. And yes, iit's an EMI filter cap and is instrumental in reducing common mode EMI. The transfomer parasitic capacitance from primary to secondary actually is the cause of a lot of the common mode EMI, the transformer primary being a prinncipal source of noise. The cap forms a short return path from secondary return to primary. Proper transformer construction is also instrumental in reducing common mode EMI.
Last edited:
The jitter spreads out the interference, thus getting it under the required level set by the relevant EMC standard. This seems to be the modern way of 'reducing' interference - just spread it out so more spectrum users get a bit less interference each. I think this could be regarded as an example of malicious compliance. Spark transmitters are banned, yet jittered SMPS have a similar effect and are accepted.
Jitter only gives you about a 5-6dB reduction in EMI at best. It can help to reduce the size of some EMI components (assuming layout and everything else is good, but it does not render an EMI filter unnecessary.
The engineer(s) who designed this circuit will be the best judges on the value of said capacitor. Is it for health safety [connected between RTN and power line] and/or EMI reduction?.
Excuse me, I am one of the engineers...The capacitor is in the circuit for the purpose of EMI reduction, and needs to be proply safety rated (Y1) because it bridges the primary-to-secondary barrier. It has no safety benefit. It one of the first things to add for common mode noise reduction.
It is yet to be proven. Did you understand my point about protecting the transformer primary <-> secondary winding parasitic capacitor from electric breakdown ?
Don't talk to me about proof - If you don't want to listen to one of the PI design engineers (20 years at PI, 30 years in SMPS), what can I say? The capacitor is put there for EMI reduction. If the transformer is properly designed for safety, it will pass a 3kV AC hipot test.
In the case of common mode line surge, the Y capacitor can provide an alternnate path for surge energy, but that is not its main purpose. The word "safety" comes into the picture beacuse the power supply is unsafe if the capacitor is not properly rated to be used from primary to secondary. Period.
In the case of common mode line surge, the Y capacitor can provide an alternnate path for surge energy, but that is not its main purpose. The word "safety" comes into the picture beacuse the power supply is unsafe if the capacitor is not properly rated to be used from primary to secondary. Period.
I actually went through http://www.powerint.com/sites/default/files/product-docs/an15.pdf and I see some problems or, at least, questions.
For starters, of we are talking about common mode voltage/EMI, what is the reference point ? The question is not that stupid - power adapters with just two legs can be connected to electric outlet two ways, so if we even decide that the N electric outlet line is the reference potential, we do not know which of the two legs inside the power adapter should serve as reference potential.
Another question is related to what they call "quiet" and "noisy" and they too use quotes around the words.
And their considerations regarding power supplies assumes that "-" output is ground/common - why ?
They do mention the Y capacitor ("Y" is rating and assumption it won't have electric breakdown), and they state that one of the transformer inter-winding capacitances are good ones and they increase the good one using the Y one.
I am really surprised - if I were designing a power supply, I would try to make it as isolated from the mains (including parasitic capacitance) as possible. In fact, in another thread I mentioned insulated power converters with spec'ed capacitance between primary and secondary - just 63pF IIRC.
So, engineers in one company increase capacitance and engineers of another company decrease it ?
...
I myself had to design a measurement circuit which was supposed to measure less than 1mV differential on top of tens of volts of common - at 400Hz. We had to use valves/tubes - ambient temperature 200C and above.
Luckily, it was narrow band 400Hz, so we used transformers. Interestingly enough, ferrite core transformers were superior exactly because with them (ferrite is non-conductive) inter-winding capacitance was much smaller, so common mode rejection was much stronger.
So from my experience increasing capacitance between primary and secondary is complete fallacy.
Hmmm......
Someone asks a question, gets an entirely reasonable answer, given
the capacitor is not for "safety", it must be a "safety" type given its
circuit position, and then they start arguing about the answer given.
Its not cricket. If you don't know what its for, and get a reasonable
answer, don't change the subject to whether your convinced or not
about the answer, go off and find out if the answers right or wrong.
Only argue if you have found a reasonable alternative purpose .....
rgds, sreten.
Someone asks a question, gets an entirely reasonable answer, given
the capacitor is not for "safety", it must be a "safety" type given its
circuit position, and then they start arguing about the answer given.
Its not cricket. If you don't know what its for, and get a reasonable
answer, don't change the subject to whether your convinced or not
about the answer, go off and find out if the answers right or wrong.
Only argue if you have found a reasonable alternative purpose .....
rgds, sreten.
:O)
Thank you, wrenchone, for the reply.
Only other chatter I'll add , not to contradict Mr. W, but to further enlighten anyone interested, there are line or mains-rated capacitors in the X and Y 'class' for different types of EMI filtering, typically line to line and line to ground. IIRC, X are usually across the line and Y are from line to ground. I can't remember if there are '1' and '2' 'modifiers' to the X and Y categories. What I recall about the 'safety' terminology is that these capacitors have been designed to endure a component's lifetime in an electrically stressful location...all manner of electrical spikes, etc. cause more problems that the 'nominal' line voltage. These types of capacitors are acknowledged to be 'safe' in such applications in accordance with standards in many countries...the various safety agency logos are often found on the capacitor body...SEMKO, NEMKO and DEMKO come to mind because they rhyme (I think it's Sweden, Norway and Denmark). You do not want to install capacitors subject to the stress of life on a power line without such ratings...this is where the safety comes in. A capacitor that lacks the appropriate design methods can actually catch fire...transients, corona, etc. will eventually damage some types of film caps...beyond any self-healing capabilities.
What you would or wouldn't do for filtering isn't relevant...commercial products ofter have to meet conducted and radiated emissions testing, and there are filter designs that solve common and differential mode EMI challenges. You have no choice in some cases for exporting, say, to Europe, but to meet the EU standards.
Thank you, wrenchone, for the reply.
Only other chatter I'll add , not to contradict Mr. W, but to further enlighten anyone interested, there are line or mains-rated capacitors in the X and Y 'class' for different types of EMI filtering, typically line to line and line to ground. IIRC, X are usually across the line and Y are from line to ground. I can't remember if there are '1' and '2' 'modifiers' to the X and Y categories. What I recall about the 'safety' terminology is that these capacitors have been designed to endure a component's lifetime in an electrically stressful location...all manner of electrical spikes, etc. cause more problems that the 'nominal' line voltage. These types of capacitors are acknowledged to be 'safe' in such applications in accordance with standards in many countries...the various safety agency logos are often found on the capacitor body...SEMKO, NEMKO and DEMKO come to mind because they rhyme (I think it's Sweden, Norway and Denmark). You do not want to install capacitors subject to the stress of life on a power line without such ratings...this is where the safety comes in. A capacitor that lacks the appropriate design methods can actually catch fire...transients, corona, etc. will eventually damage some types of film caps...beyond any self-healing capabilities.
What you would or wouldn't do for filtering isn't relevant...commercial products ofter have to meet conducted and radiated emissions testing, and there are filter designs that solve common and differential mode EMI challenges. You have no choice in some cases for exporting, say, to Europe, but to meet the EU standards.
The answer is not reasonable.
If one sees "safety goggles" ( Goggles - Wikipedia, the free encyclopedia ), he/she can read: "Goggles or safety glasses are forms of protective eyewear that usually enclose or protect the area surrounding the eye in order to prevent particulates, water or chemicals from striking the eyes." (emphasis in red is mine).
So, similarly, if I see "safety capacitor", I expect an explanation in the style "the capacitor protects FOO entity from BAR bad things happening to FOO".
If you native English/American speakers don't understand that simple things about your own language, it's very sad.
Because the answer was not reasonable, I had to look for info myself, and now I see even more crap in the "answers" from the datasheet - I am still and definAtely waiting for the definition of reference point for common mode EMI measurement.
...
Thank you again for supercalifragilisticexpialidocious ( https://en.wikipedia.org/wiki/Supercalifragilisticexpialidocious ) and ghoti ( https://en.wikipedia.org/wiki/Ghoti ).
May I suggest you seek answers at/from the source: www.powerint.com. Additional customer service phone numbers are also found on the last page of the application note. Peace!
Hi,
Your just being dull and not reading the answers you've been given properly.
To repeat because your being lazy the capacitor must be a "safety" type,
that does not mean in the circuit it has anything to do with "protection".
Anyone who thinks that a thread is validly only about how and what they
think because they started it, is mainly doing themselves a disservice.
The answer is reasonable, what you think about it is your problem.
rgds, sreten.
I think Wrenchone was very insightful in his response, as well as giving you additional information to reference.
The capacitor has to be of Y safety standard because if it were to fail it can present a danger.
This capacitor is not intended to parallel the existing primary - secondary transformer capacitance but rather to close the loop of this p-s transformer capacitance back to the converter source. In doing so reducing currents which would otherwise take the chassis return back to the source.
Hope this helps, even though I think Wrenchone is much more qualified here.
-antonio
- Status
- Not open for further replies.