Benefits of Active PFC

[QuantumFlux]

In so far as it concerns computer power supplies, what are the benefits of active PFC (power correction factor) versus non-PFC? Does it just save the power company money or would it result in the power supplied to the computer's components having more stable voltages and cleaner power? Similarly, would other electronic components (like my amplifier and speakers) that are plugged into the same electrical line benefit from decreased electrical noise and harmonics?

 

[Matthias99]

Originally posted by: QuantumFlux
In so far as it concerns computer power supplies, what are the benefits of active PFC (power correction factor) versus non-PFC? Does it just save the power company money or would it result in the power supplied to the computer's components having more stable voltages and cleaner power? Similarly, would other electronic components (like my amplifier and speakers) that are plugged into the same electrical line benefit from decreased electrical noise and harmonics?

It increases the overall efficiency of your power delivery if your line power is noisy, which, as you pointed out, won't save you anything on your electric bill in most places (since it doesn't lower the wattage that it draws). However, you are still conserving electricity (just on the distribution side, not the usage side). Apparently in some places, that *does* lower your bill directly, and it might indirectly decrease the 'distribution charge' that is tacked onto the cost of your electricity over time if enough electronic devices use it.

Assuming the power regulation in your PSU/system is doing its job (and your line power is not *exceptionally* bad - like so bad that it's on the verge of not working), it shouldn't make a noticeable difference in terms of voltage stability. I don't think it would help other things plugged into the same line, but then again, I didn't major in EE.

 

[JSSheridan]

Originally posted by: Matthias99
It increases the overall efficiency of your power delivery if your line power is noisy, which, as you pointed out, won't save you anything on your electric bill in most places (since it doesn't lower the wattage that it draws). However, you are still conserving electricity (just on the distribution side, not the usage side). Apparently in some places, that *does* lower your bill directly, and it might indirectly decrease the 'distribution charge' that is tacked onto the cost of your electricity over time if enough electronic devices use it.

Assuming the power regulation in your PSU/system is doing its job (and your line power is not *exceptionally* bad - like so bad that it's on the verge of not working), it shouldn't make a noticeable difference in terms of voltage stability. I don't think it would help other things plugged into the same line, but then again, I didn't major in EE.

I did major in EE and this is more or less correct. It saves the power company money and the power on the distributer-side is cleaner. The reason being, the voltage a PSU draws is a sine wave, but the current isn't. That creates some harmonics on the distributer's system, and they have to deal with those. Peace.

 

[cyr0nk0r]

correct me if I am wrong, but somewhere along the line I have heard that Active PFC's are good for keeping the power supply cooler.

with an active PFC the PSU doesnt heat up as fast, and doesnt put out as much heat as a PSU without an active PFC.. is this true?

 

[Mark R]

Originally posted by: cyr0nk0r
correct me if I am wrong, but somewhere along the line I have heard that Active PFC's are good for keeping the power supply cooler.

with an active PFC the PSU doesnt heat up as fast, and doesnt put out as much heat as a PSU without an active PFC.. is this true?

Active PFC PSUs tend to produce a bit more heat. The active PFC circuit itself produces heat, and the rest of the PSU is essentially identical to a standard PSU.

 

[imported_rustybx]

Originally posted by: Mark R
Active PFC PSUs tend to produce a bit more heat. The active PFC circuit itself produces heat, and the rest of the PSU is essentially identical to a standard PSU.

Generally, this is incorrect.

Although the boost PFC stage does have loss it's minor. 98% efficiency or better is not uncommon for the PFC stage. Futher, the output of the PFC is a regulated voltage. This allows the main transformer and topology selection to be optimized for a very limited voltage range (say 10%) instead of the typical 4:1 range. This brings the main output stage to the 95% efficiency range. This yields an overall efficiency of well over 90%. Also, since models with PFC are "premium" and can fetch a higher price the designs usually incorporate higher performance devices.

I'm not aware of any active PFC supplies with lower efficiency (and thus more heat) than a non-PFC design. There are passive techniques for PFC which are less efficient overall but these methods are unlikely to be used at these power levels due to the size of the laminated steel inductors. Passive topologies are generally used when cost is the main factor and the power is less than 150W or so.

I have a 450W power supply design for a telecom product going into production in Q4. Active PFC allowed me to get full load efficiency over 95%...or about 22W total dissipation. A more traditional design at 70% would have more than 70W of dissipation. Higher efficiency devices were more expensive but the savings in fan cost makes the differential essentially nil.

 

[Mark R]

Originally posted by: rustybx

Although the boost PFC stage does have loss it's minor. 98% efficiency or better is not uncommon for the PFC stage. Futher, the output of the PFC is a regulated voltage. This allows the main transformer and topology selection to be optimized for a very limited voltage range (say 10%) instead of the typical 4:1 range. This brings the main output stage to the 95% efficiency range. This yields an overall efficiency of well over 90%. Also, since models with PFC are "premium" and can fetch a higher price the designs usually incorporate higher performance devices.

Very interesting, thank you.

I was perhaps somewhat swayed by the information I'd seen in various PSU datasheets - One thing that struck me about some of the earlier Active PFC PSUs was that even the claimed efficiency was lower. The same manufacturer quoted '70% typ.' for non-PFC models and '65% typ.' for their PFC models!

I had attempted to calculate the power losses in an active PFC stage a while ago when I tried to design a PSU (I never got very far due to lack of time). I think I got about 95%, but that was for readily available off-the-shelf parts. You're right that premium parts would make a significant difference.

I also hadn't considered the improved efficiency & performance of the regulated DC voltage from the boost stage. But then, I don't design PSUs for a living

Measurements, possibly of limited accuracy, by various enthusaists have, for the first active PFC PC PSUs suggested very little difference in efficiency - typically 68-75%. Very similar to the older non-PFC supplies.

I'm impressed at your design's efficiency - I hope we can see similar products for PCs soon. Just wondering what techniques you use, if you can say: Full bridge? Synchronous rectifiers?

 

[imported_rustybx]

Originally posted by: Mark RI was perhaps somewhat swayed by the information I'd seen in various PSU datasheets - One thing that struck me about some of the earlier Active PFC PSUs was that even the claimed efficiency was lower. The same manufacturer quoted '70% typ.' for non-PFC models and '65% typ.' for their PFC models!

This may be due to early desings simply adding a PFC front end without changing the rest of the supply...like you said in an earlier post.

Originally posted by: Mark RMeasurements, possibly of limited accuracy, by various enthusaists have, for the first active PFC PC PSUs suggested very little difference in efficiency - typically 68-75%. Very similar to the older non-PFC supplies.

I did a little reseach and I think you're right. For silver box designs the PFC takes up space otherwise used for the main supply. Shrinking the main supply negatively effects efficiency. Or possibly the manufacturer reduced the performance of the main supply to offset the cost of the PFC. Interesting.

Originally posted by: Mark RI'm impressed at your design's efficiency - I hope we can see similar products for PCs soon. Just wondering what techniques you use, if you can say: Full bridge? Synchronous rectifiers?

The PFC is peak current mode with line modulated fixed off time (LM-FOT). This allows continuous inductor current at the sine peaks with a simpler control scheme than average mode control. The main supply took a more brute force approach to reduce development time. It's a multi-output active clamp forward converter modified for resonant switching of the primary FETs (i.e. lossless switching and recovery of parasitic energy). All outputs (except aux) use self driven synchronous rectifiers. I doubt this would be practical for PCs right now since I estimate part cost about 2x higher than a more traditional design. Probably not much interest in $200 power supplies.

 

[jagec]

As far as the end user is concerned, active PFC has NO benefit.

 

[NeoPTLD]

If you're a commercial user and your facility is charged based on KVA or you're assessed a poor PF penalty you might save money by using an active PFC in all computers.

Active PFCs draw less current for the power by keeping the VA as close as possible to actual power so it will allow more computers per circuit.

Unless you have a lot of computers you wish to power off of one circuit and you're challenging the circuit capacity, residential customers gains nothing from active PFC. It makes the power company happy though.

 

[Tab]

Anyone have any good links as to how electricity and power supplies work in general? Whats the difference between active and passive PFCs?