How worried should I be about "dirty" power for a CNC?

I'm working on installing a Trak 2op CNC mill, sort of a miniature VMC, into the shop. From what I've read, it has no option for single-phase power, so I have to feed it 3Ph.

Not having native 3-phase, I have a new American Rotary converter, supposedly "CNC rated". The chart that came with it basically showed the generated leg within a few percent of the native legs. I have not put a meter on them myself to see.

I believe I've wired the machine so that the computer gets the two native legs, and will double-check that before I light the thing up. (Still going through the manual checking off the service points- it's a well-used machine, and I've already found a few easter eggs. )

Short of springing for a Phase Perfect or something, what else can I do?

Along with that, I've read posts about how voltage-sensitive some machines are; if it says it needs 208 volts, you better feed it 208, and not 220v or 240v. Nothing in my documentation- that I've found, yet, suggests such a sensitivity. Anyone have a Southwest Industries machine and have experience with what it might or might not be tolerant of?

Basically, I obviously don't want to damage this thing by some bonehead move. No tech is going to come to Alaska to fix it, it's long out of warranty, and I don't want to have to buy a new $2,500 speed control or something just because I didn't ask a stupid question. \

Doc.

You could always call Southwest and ask.

Bill

I have 3 cnc machines hooked to an American Rotary converter. I just measured the phase to phase voltages to make sure they were within what the machine manufacturer requires and fired them up.

In my case the phase to phase voltage was about 250v. That's at the top end of what my Brothers want to see so I installed a stepdown transformer at each machine.

Just look at the manual for the cnc machine and if needed call the mfr and see what they say the voltage range is. Then go for it. Also check if phase rotation matters. Also if the machine has a transformer that needs the taps set, like on my Mazak Lathe, that they are set right.

Personally I don't think the phase perfect is a step up, based on what I've read here.

From their website, it don't look too picky to me


https://www.southwesternindustries....ite Prep Guides/F16467-56_VMC2.pdf?rev=173750

Electrical
• A separate 208 VAC (208 to 240 V is acceptable), 30
amp minimum, 60 Hz, 3 phase circuit is required or 480
VAC (440 to 520 V is acceptable) 15 amp minimum, 60
Hz 3 phase circuit.
• Machine can be run on single phase 208-volt but cutting
capacity is reduced by 40%.
• Machine comes standard with 208-volts. 480-volts is
optional and strongly recommended it be installed at
factory.
• This machine is meant to be moved around the shop so
the appropriate quick disconnect drops should be
installed in the shop locations where the machine may
be moved. See pictures below, Figure 5 shows the male
plugs we provide. Female equivalent plugs are to be
provided by the user.
Air
• ½” ID minimum air connection is required.
• Air pressure must be a minimum of 90 psi.
• Air – 2.5 CFM (at 90 psi),

Bill in PA said:

You could always call Southwest and ask.

Click to expand...

-I plan to. But I'm also trying to figure out what I don't know, so I know what to ask.

• Machine can be run on single phase 208-volt but cutting capacity is reduced by 40%.

Click to expand...

-Wild. I have that manual printed out and I missed that entirely. No wonder I never made it as a proofreader.

So, that would make it roughly a 2HP spindle. Not ideal, but not as bad as it sounds, either. This machine is intended for small runs (sub-200 parts, normally probably half that) of sub-2" square parts in aluminum. The Speedio guys will snigger (as will even the HAAS guys ) but it'll still run rings around the same part off a Bridgeport.

Also check if phase rotation matters.

Click to expand...

-That was one of the things I definitely wanted to talk to SWI about. In order to properly connect the controller to the two native legs, I had to swap a wire in the plug on the end of the cable it came with. (Again, used machine.)

That, of course, is how one reverses the rotation on a 3-phase motor, but that was also before whatever this thing uses for a VFD. That one might wind up simply being a "turn it on and see!"

Personally I don't think the phase perfect is a step up, based on what I've read here.

Click to expand...

-Kind of a moot point as I can't afford one at the moment, but just out of curiosity, why not? I was given to understand they produced very close to 'digitally perfect' 3-phase, kind of like a big VFD without the variable aspect.

Doc.

In my limited experience the only thing phase rotation mattered for was coolant pumps.

Eh, the phase perfect uses the same technology as a VFD. It adds some things that are less often, or never, found on a VFD, but the conversion method is the same.

The difference is that two lines are not affected at all, they are a pass-through Like an RPC, the PP only powers the third wire. VFDs make all three "phases" ("both" if they are single phase output).

The PP does regulate the voltage on the third wire, so that the voltage are all maintained within 1%, even if the input voltage rises or falls. So balance is much better than an RPC. VFDs inherently have almost perfect balance.

What is added is a bidirectional Power Factor Compensating front end (PFC). That makes the PP draw power as if it were a resistor, at a very high power factor. It also allows power being returned from the load to be passed back to the power line, just as if the motor were directly connected to 3 phase. While there are some VFDs incorporating a PFC "front end", and a few that include the bidirectional power flow, the vast majority do not have either. An RPC will not have a high power factor, since it is a motor.

I believe, but do not know for certain, that the output also has some sort of sine wave filter on it to clean up the waveform and prevent electronic noise (EMI). The input may also have some filtering, at least. The PP should have this, to comply with FCC regulations, but almost no VFDs have anything like that.. And an RPC really does not need it.

JST said:

What is added is a bidirectional Power Factor Compensating front end (PFC). That makes the PP draw power as if it were a resistor, at a very high power factor. It also allows power being returned from the load to be passed back to the power line, just as if the motor were directly connected to 3 phase.

Click to expand...

-As I was looking around for what eventually became this machine, I happened across a late-model 10EE, mid 70s as I recall, and in what appeared to be very good shape. I was in the market for a CNC mill, not a manual lathe, but I certainly wouldn't mind a 10EE if one fell into my lap.

I mentioned it to some of my regulars, and one of them noted it could only be run on an PP or native 3-phase because it was... what a 'regenerative' drive? Something like that. And would try to... if I have it right, backfeed while braking, etc. And thus don't play well with VFDs and rotaries.

Idle speculation as I wasn't going to buy it anyway, but I have to admit that if I had a Phase Perfect already installed, I'd have been a lot more tempted.

As for this machine, I've been going over the manual- with maybe a little more attention - getting ready to light it up for the first time. I think I have all my waterfowl into relative colinearity, but want to make sure.

If it has no trouble running off the rotary, I'll leave it at that. If it turns out to be easier to go single-phase, well, that's not the ideal choice but it's at least a workable option.

Doc.

I think I have all my waterfowl into relative colinearity


This is great im stealing this lol.
I have people i like to tell them to get their ducks in a row and become offended. This will let me walk away before it dawns on them.


When I find it I don’t need it
When I need it I can’t find it!

Pete Deal said:

In my limited experience the only thing phase rotation mattered for was coolant pumps.

Click to expand...

You do not own a surface grinder or B-port?
Not sure what dirty power is phase or shape.
There is the machines AC and DC side. Often a transformer or two or three in between. They do things.
Bob

CarbideBob said:

Not sure what dirty power is phase or shape.

Click to expand...

-In my limited experience, and in this context, I'm given to understand that "dirty" generally means a low or fluctuating voltage on the generated leg off the rotary. Which is why- among other reasons? -that it's recommended that the actual computer be powered off the two 'native' legs.

The rotary I got was supposed to be more or less "CNC ready", although not their top of the line version.

There is the machines AC and DC side. Often a transformer or two or three in between. They do things.

Click to expand...

-Again, partly why I'm here asking what are probably stupid questions. I read a few things over in the CNC section, and occasionally somebody will have to put a step-down transformer on their machine, because it wants 208V and their wall-supplied power is 220V.

I'd rather run this thing at reduced power off single phase than pop something expensive with poor three phase.

This is great im stealing this lol.

Click to expand...

-You don't even have to file the serial numbers off. I stole it myself.

Doc.

CarbideBob said:

........................
Not sure what dirty power is phase or shape.

Click to expand...

it can be low or unstable voltage, possibly surges of high voltage, as Doc says. It can have spikes or a lot of electrical noise on it. It may have a bunch of distortion /harmonics/etc on it. It can have bad balance between phases.

I suppose it could have phase errors, where the three are not, or not always, at the correct 120 degree phase difference, although that is less likely to bother most machines. Might lead to more heating, and less power, from a motor.

Anything that is not a nice sine wave of a constant voltage would be "dirty power". The question is always "how dirty is it", and "how much can the equipment it powers take".

CarbideBob said:

There is the machines AC and DC side. Often a transformer or two or three in between. They do things.
Bob

Click to expand...

Machines that rectify the AC are often more tolerant of some things. Most phase errors would not even be noticed (unless extreme). But since they are also often CNC, electrical noise, spikes, etc, may cause trouble with the internal electronics unless the machine has good filters on the AC.

Also bad balance on the phases voltages may lead to "ripple" on the DC, which can cause heating, stress on some rectifiers, stress on the bus capacitors, etc. That can be true of even plain VFDs.

Transformers usually do nothing much but change voltage. They can reduce some forms of noise, but usually just pass through disturbances like voltage surges or dips, phase errors, or bad balance.

JST said:

Transformers usually do nothing much but change voltage. They can reduce some forms of noise, but usually just pass through disturbances like voltage surges or dips, phase errors, or bad balance.

Click to expand...

I see and and certainly would always bow in this world to you.
My comments come from experience in nc and early cnc when simply adding a big oversized transformer up front even if 1:1 would eliminate problems.
Mu boss at the time would say "There is nothing like a big chunk of iron to absorb transients from other stuff around.". Welders near cnc and robots always a problem.
Is this sort of a like a choke being added to VFD with nasty output? I know it not the same but do you get the effect since it can not saturate as fast and steps on rise time?\
Is it a very simple and pooped low pass?
Bob

They can act as a low pass.

Wye delta can actually block some forms of distortion. Voltage changes, those usually go right through.

EDIT: Large power transformers are usually not "shielded" to block noise from being directly capacitively coupled from primary to secondary. So "spike noise" may be able to go right through them, even though they can have some effect as a low pass filter.

A proper "shield" is a sheet of metal that is grounded, located between the windings. Some "shields" are simply a layer of wire wound into the transformer, with an end out for grounding. Those are less effective, because the inductance of the wound wire "coil" means they are not very effective against high frequency noise.

Home - Smith Electric Motorworks
I have used this brand for 20 years, both in CNC mill and CNC EDM, No issues except bearings and a burnt wire

rkdeckel said:

Home - Smith Electric Motorworks
I have used this brand for 20 years, both in CNC mill and CNC EDM, No issues except bearings and a burnt wire

Click to expand...

They have far better information on their site compared to phase technologies. Thank you for sharing that.

JST said:

EDIT: Large power transformers are usually not "shielded" to block noise from being directly capacitively coupled from primary to secondary. So "spike noise" may be able to go right through them, even though they can have some effect as a low pass filter.

Click to expand...

All of my older stuff, from when this was more of a concern, had large isolation transformers between incoming power and everything else. No idea what type they were but according to the manuals, intended to act as a filter as well as having taps to adjust for various input voltages. Any ideas what they might have been ?

EmanuelGoldstein said:

All of my older stuff, from when this was more of a concern, had large isolation transformers between incoming power and everything else. No idea what type they were but according to the manuals, intended to act as a filter as well as having taps to adjust for various input voltages. Any ideas what they might have been ?

Click to expand...

They MAY have had some form of a shield. The shield cuts the noise by shunting noise to ground because the shield is grounded.

There are two basic paths through the transformer.... One is linked by the magnetic field, i.e. "normal transformer action". The other is capacitive coupling between primary and secondary, which is very direct.

Any transformer also has some "leakage inductance", which plays no part in coupling of the primary and secondary. That inductance tends to filter out high frequencies, so that they are more-or-less blocked from the normal transformer action path. Losses in the steel also do some filtering. So all transformers, shield or not, have some blocking capability.

For the capacitive coupling, the leakage inductance is less effective, for several reasons, the details of which are not important here. The capacitive coupling is generally always present, and most ways to reduce it also make the transformer less efficient.

BUT, a good shield between primary and secondary can block that direct path.