[js412000]

Not to hijack the previous post,but, can I simply use rectified 240 volts to run my 146 volt 20a servo motor? I am using a Mesa 8i20 brushless servo motor drive

that needs a separate DC power supply. The 8i20 is good for up to 400 volts. Can the machine be set to limit the top speeds and protect the motors?

This is a new build and I am using Fanuc 20s and 22-2000 motors and LINUX CNC.

This would eliminate an extra transformer.

[Al_The_Man]

If in the US why not use rectified 120v which will give you around 170vdc, otherwise 240 will come out to 335VDC.
The control will limit the maximum rpm through the feedback loop, but the 120v might be the wiser choice.
Al.

[js412000]

I am supplying my my machine with a rotary phase converter. The spindle VFD (7.5 HP) is the only thing that needs 3 phase. I have been told, on another forum,

that it is a bad idea to use the neutral supplied by the power company for 120 volt needs. Something about an indeterminate path for the neutral return can

cause problems with the delicate electronic parts.

I don't really see what the problem would be if I made sure I only used one leg of the 240 volts (edit- not the generated leg) and brought the white neutral wire in to the machine along with the three phase wires.

I have some transformers on hand that I can use for small loads, but nothing big enough for the servo drives.

[PCW_MESA]

Not to hijack the previous post,but, can I simply use rectified 240 volts to run my 146 volt 20a servo motor? I am using a Mesa 8i20 brushless servo motor drive

that needs a separate DC power supply. The 8i20 is good for up to 400 volts. Can the machine be set to limit the top speeds and protect the motors?

This is a new build and I am using Fanuc 20s and 22-2000 motors and LINUX CNC.

This would eliminate an extra transformer.

Its possible to do this but there are some disadvantages (and advantages)

Disadvantages: More current ripple in the motor (heating when idle),
More insulation stress in the motor (due to higher voltage)

Advantages: faster torque loop response, better torque near full speed

In any case the DC power supply for the 8I20 needs filtering so cannot be just rectified main power
(a large electrolytic output filter capacitor is required)

[cgo]

You are forgetting the most important reason that they use a power supply , isolation !
You will have a ground loop doing that, which will burn up your controllers and maybe even go as far back as the pc.

Very bad idea.

[PCW_MESA]

The 8I20 is designed to use rectified line voltage so this is not an issue

[js412000]

Thanks for all the responses. I have capacitors to smooth out the DC power.

I am going to power the control PC from machine electrical enclosure, not plug it into the wall outlet.

I am not understanding the whole isolated power supply concept.

I thought that code required that -

Every exposed metal part on the machine (and in the entire building) be connected to the green ground wire.

If I have a separately derived system, like a control transformer, I need to pick one wire on the secondary side, immediately after the transformer, and connect it with a jumper to the green ground wire.


I am not sure about the next point, Does every DC negative wire, on the entire machine, connect to the green ground wire (at the center of the star

grounding point)?

[PCW_MESA]

Thanks for all the responses. I have capacitors to smooth out the DC power.

I am not understanding the whole isolated power supply concept.

I thought that code required that -

Every exposed metal part on the machine (and in the entire building) be connected to the green ground wire.

If I have a separately derived system, like a control transformer, I need to pick one wire on the secondary side, immediately after the transformer, and connect it with a jumper to the green ground wire.


I am not sure about the next point, Does every DC negative wire, on the entire machine, connect to the green ground wire (at the center of the star

grounding point)?

No! A line derived DC power supply must be floating or you will have a line to ground short!
The 8I20 Frame must be grounded but not VBUS-

If and only if you use an isolation transformer, can you ground VBUS-

[cgo]

Most modern VFD and PLC products can also run directly because of opto-isolators but we still used an isolation transformer on the controllers just for extra safety, but I was dealing with 480 vac 3 phase on the power side , you dont get a second chance working with that kind of power if you put your hands in the wrong place.
Even if you can do it , doesnt mean that you should, sometimes.

[Al_The_Man]

QUOTE=js412000;1452978] I have been told, on another forum,

that it is a bad idea to use the neutral supplied by the power company for 120 volt needs. Something about an indeterminate path for the neutral return can

cause problems with the delicate electronic parts.

I don't really see what the problem would be if I made sure I only used one leg of the 240 volts and brought the white neutral wire in to the machine along with the three phase wires.
.[/QUOTE]

Doesn't make sense, only that when using one of the three phases is that the live conductor is one of the 240 conductors passed through and not the generated phase to pair up for 120vac.




I use A-M-C drives extensively and the option on most models is a built in supply, this is direct off of the main AC supply, they have rectifier and smoothing included.
In this case the DC cannot be referenced to Earth ground.
If a transformer is used and you want to reference to ground, it is best to do it on the DC common, not the secondary, and definitely not on both otherwise you blow the rectifier.
On the majority of my PC based systems, I bond or reference all power to the central earth ground point.
Your PC internal power supply is already connected to earth ground.
Al.

[js412000]

PCW- So you are saying that with a line derived DC power supply. #1 and #9 chassis gnd connectors and the printed circuit board mounting holes would be

grounded to the green ground wire that eventually leads to the power company supplied electric meter outside the building.

#4 motor power + and #5 motor power - would not be grounded.



I have a regulated 24 DC power supply ,with a transformer inside it, I want to use to supply limit switches and logic power. Does the negative side of that need to be grounded?



Do I need a separate supply each for logic power and limit switches?

I also have a couple of plain control transformers that output 24V AC, laying around, if they would be better for this purpose.

Al_The_Man-

So you are saying that you like to earth ground every DC negative ,on the entire machine, except for the kind that

you described where it can not be done.

[Al_The_Man]

Do I need a separate supply each for logic power and limit switches?

I also have a couple of plain control transformers that output 24V AC, laying around, if they would be better for this purpose.

Al_The_Man-

So you are saying that you like to earth ground every DC negative ,on the entire machine, except for the kind that

you described where it can not be done.

I like to use 24vdc where possible for control out on the machine, just one supply for all.
I generally ground all the commons where possible.
In conjunction with machine bonding.
http://www.automation.siemens.com/do.../emv_r.pdf?p=1
I have installed many PC based systems in some pretty harsh environments and never had a problem.
Al.

[js412000]

Thanks for the input. I am hooking up one of the encoders now and want to test a motor on reduced voltage (for safety) maybe tomorrow.

[neilw20]

Slightly different take on things, but still the safety issue.
I have beautiful 3 phase hybrid stepper drivers, supplied direct from 220vAC (wish they were rated for 240!!)
Internally, rectified mains. All opto isolated control inputs.
They can be run without a transformer, but manufacturer recommends isolation transformer.
Suits me, as I can step 240 down to 220.
Main reason for transformer, is that if a motor winding breaks down to ground it can destroy things rapidly.

If the machine is not earthed solidly, it can become a death trap!
You can't add safety detection, for seeing if the earth is good, as it, by definition, this breaks the isolation rules of many electrical standards which specify solid bonded earth, with no current flowing through it.
How can you measure earth continuity without having some (albeit) small current flow?
Then the measurement circuit becomes a safety issue... and so on.

Would be nice to just use an auto transformer to get from 240 to 220 but the cost saving is just not worth it when it comes to SAFETY.
This is $400 worth of safety insurance.
Transient spikes within inductive devices, like motors, can break down the insulation.
That transformer gets rid of the transient problem.

USE A TRANSFORMER !!, unless it is running on mars, away from humans, transients and safety regulations.

[js412000]

If a motor winding breaks down to the frame, won't that blow a fuse ?

If every exposed metal part is bonded together and connected to earth ground, how would the operator be shocked?

I have plenty of other machines in my shop that are connected directly to the lines.

[neilw20]

Yes, if securely bonded to ground, and permanently wired that is fine.
It is equipment that is plugged into a power outlet where the security of the earth is a problem.

[Al_The_Man]

Would be nice to just use an auto transformer to get from 240 to 220 but the cost saving is just not worth it when it comes to SAFETY.
.

A buck-boost transformer could also be used, either across the mains supply or transformer secondary.
Al.

[Larken]

You should be able to use both lives of 220ac with a half bridge rectifier to get the positive, then use the neutral as the negative. (Since the neutral is really a center-tap on the 220V.) It will give 170vdc, the same as rectified 120, but have the negative at gnd potential .

The motor frame should be grounded , but neither power rail of the electronics should ever be grounded, Imo.

Al, on big AMC drives, is the tach or encoder inputs ground, ever at the same potential as the main power neg ?

[js412000]

That is an interesting thought, I will have to do some reading up on the subject,

Right now I have a 2kva step down transformer followed by a variable autotransformer, for testing.

I did have one small hiccup and blew the input fuses, but the transformer shows no sign of overheating with just one servo motor running.