EEVblog Electronics Community Forum

Electronics => Projects, Designs, and Technical Stuff => Topic started by: mike buba on October 13, 2021, 05:50:24 pm

Title: Active Power Factor Correction; off-the-shelf or custom made, need advice please
Post by: mike buba on October 13, 2021, 05:50:24 pm
Hi, I am looking into Active Power Factor Correction devices for my application (230 Vac to 3x400 Vac onverter with a DC/DC boost converter in-between). The current model has a diode rectifier to get DC voltage and inductor at the input AC line to limit the max. current. The inductor is big, heavy and does not limit the max. input current much (50 mV = 50 A at 1 kW output?! [attach=2].

I was looking into Active Power Factor Correction devices, mainly reading this document (https://toshiba.semicon-storage.com/info/docget.jsp?did=68570) and running Matlab Simulink simulation (https://uk.mathworks.com/matlabcentral/fileexchange/69653-active-power-factor-correction?s_tid=srchtitle).
[attach=1]

I have now two options for which I could use a bit of advice on what would be the faster and the easiest solution:

a) Are there any off-the-shelf Active Power Factor Correction device for 3kW, input 230 Vac, 50 Hz and output 350 Vdc?

b) The second option would be to build one Active Power Factor Correction device. Based on the simulation and document I need:
Output power is 3 kW at 350 Vdc so the input sinusoidal current is max. 20 A and inductor max. current 20 A.

If I go with the second option, mounting semiconductor devices and passive elements is not a problem, but are there already available control and signal acquisition interface board, ideally with transducers and control algorithm? I've just finished creating the main PCB board and drivers for that 1-phase to 3 phase converter and it was custom made, exhausting, tiring and $$$, so I would like to avoid that :)
Title: Re: Active Power Factor Correction; off-the-shelf or custom made, need advice please
Post by: jonpaul on October 13, 2021, 07:21:58 pm
Bonjour your post is not clear mentioning the PFC and one><3 ph conv.

A system block diagram will clarify.

There are no "off the shelf" PFC at 3 KW.

Three phase PFC can be nixed  if a transformer and 6 or 12 phase rectifier are used.

PDF design is very well known, see the many books, IEEE, PELS, PESC papers and seminars.

Especially my colleague Dr Richard REDL has done pioneering work on these issues.

Finally the magnetics are critical and will be custom.

How many are building, one? 100 pcs 1K

Bon Chance

Jon



Title: Re: Active Power Factor Correction; off-the-shelf or custom made, need advice please
Post by: bdunham7 on October 13, 2021, 07:39:33 pm
These can be paralleled, so that covers easy and fast.

https://www.artesyn.com/power-supplies/websheet/42/aif04zpfc-series (https://www.artesyn.com/power-supplies/websheet/42/aif04zpfc-series)
Title: Re: Active Power Factor Correction; off-the-shelf or custom made, need advice please
Post by: jbb on October 13, 2021, 09:29:05 pm
I may be misunderstanding, but… do you currently have the following arrangement?
- single phase input
- diode rectifier to ~300V DC (and big line inductor)
- DC link cap L
- DC-DC boost converter from 300V to ~500V
- DC link cap H
- three phase inverter

If so, maybe you could modify what you’ve got. Normally a single phase PFC spits out 380V ((240V + 10% surge) * 1.41 + margin).  But you certainly could make a PFC stage spit out >=500V to supply your three phase inverter directly.

Maybe you can remove DC link cap A and modify the existing DC-DC boost to do PFC boost? (May change inductor rating and required current sense range.)

Note: if you pursue this approach then the DC link cap B will need to filter out the single phase power ripple (ie 100Hz ripple), and may need increased capacitance and/or ripple current rating. This ripple will also mix into the three phase inverter output so your three phase inverter stage would then benefit from line voltage feed forward when converting from voltage command to switch duty cycles.
Title: Re: Active Power Factor Correction; off-the-shelf or custom made, need advice please
Post by: jonpaul on October 14, 2021, 05:22:56 am
Rebonjour:

Mike a moment please....

1/ PFC are usually boost so any desired DC bus can be designed, as long as it is > peak V of input line. On a 120/240V line, 360 VDC  is normal but 500 or more is possible.

2/A  Toroid is the worst for for transformers at 140 kHz and will be difficult to wind, terminate or insulate or gap. Pot cores are not intended for such powers or voltages. Use ETD, EER, or other round center leg modified EE or EI shapes, see TDK, EPCOS and other ferrite suppliers.

Foil, litz or bunched wire may be needed depending on skin and proximity effect and tolerable losses.

3/ Power electronics and power magnetics design is a specialty and you could start to learn more but years of experience are needed to make a safe and effective design.

Bon Chance

Jon

Title: Re: Active Power Factor Correction; off-the-shelf or custom made, need advice please
Post by: mike buba on October 15, 2021, 08:51:44 am
Quote from: jbb on October 13, 2021, 09:29:05 pm
Note: if you pursue this approach then the DC link cap B will need to filter out the single phase power ripple (ie 100Hz ripple), and may need increased capacitance and/or ripple current rating. This ripple will also mix into the three phase inverter output so your three phase inverter stage would then benefit from line voltage feed forward when converting from voltage command to switch duty cycles.

Would Electrolytic capacitors be ok for the output? For DC/DC converter I used film, but I cannot find any at mF values
Title: Re: Active Power Factor Correction; off-the-shelf or custom made, need advice please
Post by: mike buba on October 15, 2021, 09:32:20 am
Quote from: jonpaul on October 14, 2021, 05:22:56 am
Rebonjour:

2/A  Toroid is the worst for for transformers at 140 kHz and will be difficult to wind, terminate or insulate or gap. Pot cores are not intended for such powers or voltages. Use ETD, EER, or other round center leg modified EE or EI shapes, see TDK, EPCOS and other ferrite suppliers.


I have a pot core inductor used in 20 kHz switching DC/DC boost converter. Can this be used for PFC application (at 20 kHz switching)?

Pot-core inductor data: