Why have Active PFC in Audio SMPS?

[grigson]

Hi,

I have just seen the datasheet for a 180W (peak) offline SMPS which comprises a single stage active power factor corrector...

here it is, its the SMPS180 by Hypex......

http://www.hypex.nl/docs/SMPS180_datasheet.pdf

I cannot , for the life of me, understand why Hypex have made a Power Factor Corrected SMPS for Audio usage.

There is not a single country in the world that requests Power Factor Correction for Audio usage.

Since this (single stage) SMPS is power factor corrected , it will comprise a dreadfully slow feedback bandwidth, (~10Hz)which is totally unwanted in audio applications, where good transient response is required.

In the datasheet, Hypex extol the virtues of them having the bulk storage capacitance on the secondary side...........

.....i can't think why they believe that this is virtuous, the entire switch-mode industry knows only too well that capacitive storage banks are best placed where you have the highest voltage...........generally at the mains side.
Capacitive energy storage quadruples with doubling capacitor voltage because of the square law of capacitive energy storage.

The lack of capacitance on the primary side of the SMPS180 means that it is harder for the SMPS180 to filter the high frequency switching harmonics from the mains....i'm not saying it cant be done, but your hampered by the lack of primary side capacitance.

The SMPS180 will, however, reduce mains harmonic current levels, in comparison to a non-PFC design.........however, what on earth is the point of doing this when the regulatory bodies have no requirement of it.?

A PFC design will be more expensive and require more engineering effort than a non-PFC design, and since there are no advantages in the Audio world of using PFC designs, why have Hypex chosen to do a PFC design.?

Perhaps i am being too cynical there, the inrush current will be less with a single stage PFC design....however, inrush is easily circumvented with NTC's.

One point about Single stage PFC design is that the peak FET currents will be higher, and the transformer will need to be bigger.
The FET RMS current will also be higher, and a bigger FET heatsink, or more expensive low RDS(on) FET wil be required.

Can any reader think of a reason for using the SMPS180 in an Audio application?

Here are Audio SMPS's of several 100W's power level, which have no PFC stage..........................

https://www.coldamp.com/store/sps80.html

http://www.abletec.com/index.php?id=33

The deleterious point about Hypex SMPS180 is that it is a *single stage* PFC design............if it had comprised a PFC Boost converter, followed by a downstream SMPS, then that would have made sense, since the high voltage bus provided by the Boost PFC stage, would mean a convenient high input voltage for the downstream SMPS stage, which would allow it a very good transient response.

 

[ScottyUK]

Well, most desktop PCs with a 400W PSU spend the majority of their time idling at about 50 - 60W. Occasionally one might get up to a couple of hundred if it is working hard, but that is unusual. Not a radically different scenario really is it?

Anyway, an active front end is now a commodity item because it is a necessity rather than an expensive option. Are margins really that tight?


----------------------------------

If we learn from our mistakes I'm getting a great education!

 

[grigson]

zzzjim:
So are you saying that really we should use PFC because otherwise we will have big input electrolytics which will eventually tire of the repeated huge inrush currents and go bang?

ScottyUK:
Margins are tight indeed......The market for guitar amplifiers is not large, and also, our competitors are not using PFC , even with their 300-400W products.


ScottyUK: I take your point about the PC which idles at 60W.........but in all truith, a guitar amplifier would not be giving 60W for any length of time at all.....even if it was 300W peak.
-seriously, your average in-the-home guitar amp user is going to get really bored of playing if they play it loudly for long....thats not to say that the majority of in-the-home guitar amp users can't play the guitar properly.......but most of us aren't exactly Hank Marvin.

Anyway, as mentioned a guitar amplififer is a transient load........it is never continuous...it simply goes from no-load to full-laod and vice versa.....the lingering sound in between strums is extremely low power.

Such a load is useless for PFC......
-This is because PFC stages have limits on the excusion of V(out).......usually, V(out) is not allowed to go outside 380 - 440VDC (for a 230VAC use)

.....so when the Vout trys to go above 440VDC, the PFC contoller immediatley stops doing power factor correction and siomply shuts off the fet.

-similarly at 380VDC, the PFC controller stops doing PFC and slams the FET on for long duty cycles to bring the output up.

-With a guitar amoplifier load, the PFC would simply always be hitting the end-stops of Vout, and would never be acting as a power factor corrector....due to the transient load.

The only reasonably useful type of PFC for guitar amplififer would be a voltage mode boost converter with a bandwidth around 500Hz.

10Hz is just too slow and you'll constantly ping up and down from 380VDC to 440VDC.

-though in truth, even this wouldnt be that worthwhile,,,,since its losses would be higher than any potential savings in the supply system due to harmonics of a non-PFC supply.

 

[Skogsgurra]

"its losses would be higher than any potential savings in the supply system due to harmonics of a non-PFC supply"

Please understand that it is not about what you think is needed. It is about what you and everyone else *must* do. Like it or not - obviously the latter.

Gunnar Englund

http://www.gke.org

--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.

 

[grigson]

thanks, i understand..

..but the Regulatory bodies are ultimately governed by Professional Engineers......and we believe that these are not the kind of people who will dogmatically enforce the regulations if they think that there is cause for discretion...which with guitar amplifiers, there obviously is.

We do not beleive that they will close our company down if we do non-PFC guitar amplifiers with 300W peak.

The very worst they might do is simply give us 6 months to change the PSU to a power factor corrected type.

We know that Engineers, such as on this forum, are reasonable , thinking type people, who ultimatley do what's logical and proper , and follows good engineering sense.

the dogmatic types who enforce laws without thinking simply dont understand engineering well enough to have a decision at the regulatory body's HQ.

So we must go ahead and do non-PFC......because when a guitar is not strummed, it pretty much consumes no power........and the guitar owning population simply dont strum away like mad-men for hours, or even minutes on end, whilst on full power.

..even if they did, they still, on average, wouldnt get above 1/3rd maximum power, on average, during there frenzied strumming.

 

[Skogsgurra]

Thanks for your kind estimation of my 'the dogmatic types who enforce laws without thinking'.

Not so much appreciated.

Gunnar Englund

http://www.gke.org

--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.

 

[itsmoked]

Thanks for your insights grigson. Your arguments make sense to me.

Sadly the drones who would be doing the enforcing won't have any problem tagging you as they won't care about any logic. This is demonstrated countless times by enforcement of the many asinine regulations bureaucrats have come up with.

You should have a debugged, tested, ready for production PFC design in your drawer.

Keith Cress
kcress -

http://www.flaminsystems.com

 

[grigson]

Hi,

The following two documents, confirm that PFC is not a requirement in the world of Audio for <600W.

EN61000-3-2 states “75-1000W” needs PFC but does not state how that is measured….EN61000-3-2 states that you must draw >75W “continuously” in order to need a PFC stage….guitar amps with nominal power of ~400W definitely don’t draw >75W continuously.

DOCUMENT 1:-

http://www.olfer.com/articulos_electronica/normaEN61000_gb.html

“2. Is power consumption continuously in the range of 75 W to 1000W?
Note: consumption is only considered (for all phases) during normal operation. Operation under extreme conditions or the rated power of the equipment are not considered. In other words, we only consider the real consumption of the equipment.”

“10. Equipment with a rated power input of up to 600W are really at the limit of requiring PFC.”

DOCUMENT 2:-

http://www.powerphysics.com/lab-notes-journal/2010/8/10/what-is-power-factor-correction-pfc.html

“PFC is required by the European Union CE mark for amplifiers greater than 600 Watts”

 

[MacGyverS2000]

While those documents provide some possible insight, no document other than IEC 61000-3-2 can confim that PFC is not a requirement... keep that in mind, as you have said on several occasions in this thread that you know PFC is unnecessary (usually attached to "no one else does it, so it must be okay").

The second link is laughable as "proof". The first link is better as far as explanations go.

You may be under the limit, but I do not know the exact testing methodology. Running a 300W amp at 1W to show it beats the 75W limit may be acceptable from a governmental testing standpoint, but it seems to defeat the heart of the requirement.

Dan - Owner

http://www.Hi-TecDesigns.com

 

[ScottyUK]

You really need to read the IEC standard, and in particular Appendix C.3 "Test conditions for audio amplifiers". It details how the tests will be performed. It does not take account of whether the amplifer is being used by Mark Knopfler or Lemmy from Motorhead. It doesn't care that it is a guitar amplifier at all.


----------------------------------

If we learn from our mistakes I'm getting a great education!

 

[grigson]

Hi,

I have just received confirmation from a major , quality , European smps design and manufacturing company, with volumes in the 100’s of thousands for their audio smps products……..

..they told me that PFC is not needed for audio smps below 600W.

..they also told me that Audio SMPS are tested for approvals at 1/8th maximum power.

-they said that no EN regulations dont really specify this, but it is just an accepted fact.

 

[ScottyUK]

It's all in the standard. Buy the damned standard.


----------------------------------

If we learn from our mistakes I'm getting a great education!

 

[MacGyverS2000]

I have just received confirmation from a major , quality , European smps design and manufacturing company, with volumes in the 100's of thousands for their audio smps products........

..they told me that PFC is not needed for audio smps below 600W.

..they also told me that Audio SMPS are tested for approvals at 1/8th maximum power.

-they said that no EN regulations dont really specify this, but it is just an accepted fact.

Because obviously the wise thing to do here is to take the word of some random guy at a potential competitor as to what the standards mean without purchasing/reading them yourself in detail.

He could be right... or he could be looking at an ultra-easy way to make sure someone doesn't take any market share (you spend millions in capitol making units that can't be sold). You don't know. But considering your other posts, you seem quite content with rushing to judgement with little to no useful data, and that judgement is always in favor of what you want it to be. Talk about fitting the data to fit the hypothesis...

I think I'm done with this one...

Dan - Owner

http://www.Hi-TecDesigns.com

 

[FreddyNurk]

Its not clear as to who you're trying to convince here, but arguing your case against eng-tips won't get you a waiver from whatever the standards may say. ScottyUK is right, buy the standard.

Nothing said here is going to make any difference to the content of the standard.