Transformer K-Factor for CNC's ?

What K Factor is normally recommended, if any, for a transformer that will run Cnc's, possibly 2 at the same time. I never heard of it before, I kinda doubt anyone even looks at it when installing one on a new machine?

I ran into that option while pricing out features from a transformer manufacturer in Canada. Standard units are K-1. Then K-4($200),K-9($420),K-13($590),K-20($870) option(Cad$).
The only 2 mention I found of K-factor while trying to find an answer online called for a K-20 for VFD drives/CNC's.

The main issue with harmonics seemed to be transformer overheat. I'm currently looking at 45Kva units, expecting it to be eventually loaded to maybe 60-70% capacity when I get a 30hp PP and run 2 machines at once. How much does regen dump heat into a transformer? Perhaps that 30% left up top is not enough to keep it cool if only a K-1 ? I'll also have AC in this new shop, 20°C and air flow. I want to do this only once and don't mind paying a bit, just want to make sure I get the right thing, without pissing away too much $ for nothing either.

thanks

Don't know about recommended k-factor.

Regen is just power going the other way. It either subtracts from the total power coming through he unit (actually reducing the load) or it really puts net power backwards through the system.

In either case, it is not active for very long if it is just from braking, etc, so in the larger scheme of things it is probably nearly irrelevant unless the machine cycle is such that there is a lot of braking. And, of course, the machine has to actually do an active AC output of the regen energy, and not just dump excess power to a resistor.

There may be some effect if there are a lot of harmonics, but in general, the returned power is limited to a fairly low harmonic content, perhaps 5% or so max. So even the harmonics should not be too much of an issue.

In your setup I don't believe that a K factor unit will do you that much good, for the extra coin.
The primary differences are an oversized neutral and a wound core instead of a stamped one.

It's for building loads that have many harmonic producing devices that utilize the neutral connection. Your machines do not have a neutral load per say, but do put some harmonics on the grounding conductor, which is connected to the neutral point.

You need many loads in a large facility that produce harmonics to necessitate using them. Computers, switch mode power supplies, electronic ballasts, machine drives. If you have many loads that produce harmonics continuously. They add up to be a problem for the neutral wiring and supply transformer.

For a machine or two on a single transformer, it's not too critical normally. Its when there are lots of load that produces harmonics that you need them. There is also one level up from K-rated called harmonic mitigating type, that serves to cancel them instead of just dissipating them.

Have a look at the ACME catalog for a comparison and description.

ACM_CAT_7_913

SAF Ω

SND said:

option(Cad$).
The only 2 mention I found of K-factor while trying to find an answer online called for a K-20 for VFD drives/CNC's.

Click to expand...

VFD's are not the greatest of offenders. Rectification into a bank of capacitors ahead of the switching nastiness means that UNLESS fully-regen, the line side is not much abused.

IF, repeat IF, a VFD even needs "help" it would be a "true sine wave" 3-P filter on the load side. Few modern VFD - or their typical loads - still need even those.

The "offenders" if one has such at all, are direct-off-the-line DEE CEE SCR-class drives. No capacitor bank. Switching transients are ON the line for lack of any other place to BE.

I use Corcoms up to 20 A for my small DC Drives, Big fat full-isolation transformers for my larger ones.

The vault-pig here is shared with several other residences, and I don't need unhappy neighbours. Nor my own computers, routers, WiFi, security systems, multi-media and such being insulted when they need not be.

2CW

Thanks, good to know.
I'm checking into a single electrostatic shielded K-4, both options are not much more $, turns out even the "standard" units have the same lead time to be built anyhow so might as well add a few options at the same time while I can.

SAF is right, assuming all of your loads on the CNC are three phase input. Most will be, but it is possible that some of the smaller servos are fed from one phase and neutral. The thing is, many of those that are fed that way will not typically be the servos that are running continuously. The big loads on a CNC are the spindles, and those are typically 3 phase input. So as he said, on a 3 phase input, the K-factor is not going to buy you much.

I woke up thinking that since it'll only be on maybe 8-14hrs a day and cold/hot cycling every day vs always being On like most transformers are. I probably better go with copper wound, less thermal expansion and probably better life that way vs Aluminium.

I've been rather surprised at the number of transformer manufacturers still around in Canada. There's gotta be at least a dozen.