Electronics > Power/Renewable Energy/EV's

Power Factor Correction unit - different capacitor sizes?

(1/3) > >>

Thermoelectric:
Howdy,

Question for those who may design/know more about PFC units than me. Got a 137.5kVAR power factor correction unit, 415v, 207a rated, all that jazz. It's old and a fuse has blown on the smallest (most used) capacitor "bank". Put my hand up to have a look at it and try to fix it.

It's made up of a few separate sets of capacitors, 2x50kVAR, 1x 25kVAR and 1x 12.5kVAR section. My plan was to just ditch the caps from the 12.5kVAR section and swap in some of the less used caps from the larger sections (and reduce total capacity, building has LED lighting now and inverter airconditioners so much less need for PFC, having smaller stages is beneficial).

However, in my travels I noticed that they haven't used multiples of the same cap for each section, instead being as follows

* 50kVAR = 5x 10kVAR caps
* 25kVAR = 3x 8.25kVAR caps
* 12.5kVAR = 2x 6.6kVAR caps
Here's what I assumed it would have been

* 50kVAR = 4x 12.5kVAR caps
* 25kVAR = 2x 12.5kVAR caps
* 12.5kVAR = 1x 12.5kVAR cap
I think I can still make it work as I intended it to, however, I'm curious if anyone knows why you would use different size caps for each stage, rather than a multiple of the same? Is it just convenience or is there something more to it, like in electronics where you'd have a small cap for decoupling and a large cap for bulk capacity? I can't quite figure it out.

Cheers!

note: capacitor values provided above don't match the photos - kVAR scales with voltage, 12.5kVAR cap rated at 525v ends up being a 10kVAR cap on 415v

f4eru:
Wow, impressive stuff.
I wonder that they don't have inrush protection when switching ON, how does this not create crazy transients ?
How are the contactors not welded yet ?


Who needs microfarads, when you can specify your caps in VAR :))
https://search.abb.com/library/Download.aspx?DocumentID=9AKK107046A1971&LanguageCode=en&DocumentPartId=&Action=Launch

Alti:

--- Quote from: f4eru on January 28, 2022, 09:54:16 am ---I wonder that they don't have inrush protection when switching ON, how does this not create crazy transients ?
--- End quote ---
Indeed, maybe the ESR does the job of current limitation.
Cu 10x30 bar for 207A, kind of overkill.


--- Quote from: Thermoelectric on January 28, 2022, 09:02:45 am ---It's old and a fuse has blown on the smallest (most used) capacitor "bank".
--- End quote ---

With every unfortunate turn-on it is going to strain the fuses. So, I would not be concerned about the tripped fuse in such case, maybe this design wears out fuses by design. Check the fuses, if these are M-effect fuses then every time you exceed rated current they become weaker and weaker (by design).
If this is the case then perhaps manufacturer mentioned that, as a standard maintenance procedure, you should replace all M-effect fuses periodically instead of waiting for this to happen.

Did you verify the caps are the actual problem?


Circlotron:
If you were building one from scratch it would be nice to use back to back SCRs for switching the caps in and out, turning on when there is zero volts across the SCR. Not zero volts at the supply but zero volts across the SCR.

NiHaoMike:
Does the control module save stats as to when the different capacitors have been used? I would say to take a 10kVA module from one of the 50kVA banks (turning it into a 40kVA bank) and install it where the 12.5kVA bank goes. And since it's rated to fail safely with as much as 125A fuses, you could upsize the wires, contactor, and fuses to prevent fuse degradation from inrush.

Or if both modules in the 12.5kVA bank test fine, split them up with separate contactors and fuses? Should give you more fine grained control.

--- Quote from: Circlotron on January 28, 2022, 11:49:25 am ---If you were building one from scratch it would be nice to use back to back SCRs for switching the caps in and out, turning on when there is zero volts across the SCR. Not zero volts at the supply but zero volts across the SCR.

--- End quote ---
Could install zero crossing SSRs in parallel with the contactors along with a timer to coordinate the switching. That'll give near zero inrush without the substantial amount of heat a SSR would give off.

Navigation

[0] Message Index

[#] Next page

There was an error while thanking
Thanking...
Go to full version