isolated Tx - no service neutral

[GITRDUN]

All 3 phase equipment has a ground wire ran with the 3 phase wires and or is grounded through the conduit only. We do have ground rods for almost all machines but right now they are not connected to the equipment. Thats sort of a catch all for the CNC gurus is to put ground rods in. I simply cant see the reasoning myself and seems like backwards thinking to my feeble mind not to mention the fact that it surely creates another path for nasty currents to get into the equipment. No power station within a few miles from us.

I spoke with a well trusted CNC repair company the other day and told them to schedule me in to come down and go through our power setup with a fine tooth comb. I have known these guys for probably 20 years and they know more about CNC machines than anyone ive ever met. Not sure where else to turn to right now.

 

[gar]

140809-1523 EDT

GITRDUN:

I can not directly address the Okuma machines, but I can the HAAS machines.

Do not use a secondary grounding device (ground rod) at a CNC machine. Use only the EGC to ground the machine.

HAAS has a well designed front end. The incoming 3 phase source can be either delta or wye. HAAS only makes delta connections to the source. However, do not apply more than 285 V RMS sine wave to any hot input wire relative to ground, or between any two hot legs.

The nominal 230 V 3 phase input connects to a 3 phase combination circuit breaker and disconnect switch. Under extremely rare conditions this breaker might trip on over current. Always shut this disconnect off when the machine is not to be used for a while, for example at night. Also the disconnect should be turned off if a lightning storm is expected.

At the output of the disconnect are two major connections.

One connection following the disconnect is the input 3 phase hot lines to the CRM relay contacts.

The other connection is to a small power supply and logic board for control of the CRM. This is where power to operate the CRM originates. Immeadiately at the input are three small fast blow fuses. Following the fuses are 6 Transorbs (possibly some other brand) that bridge between the hot lines and to ground (EGC). A Transorb is a pair of series connected Zener diodes back-to-back. Thus, a very well defined voltage threshold. In this case 285/.707 = 404 volts. Thus, any line-to-line or line-to-ground voltage greater than 400 V will blow at least one of the three fuses.

Loss of two fuses will cause CRM to drop out. I will have to go back and look at the circuit to see if a single open fuse will cause drop out This circuit will respond to a moderately short overvoltage condition, probably one cycle including the drop out response time of the CRM relay. How short does a very large voltage spike have to be to not trip one of the fuses, I don't know.

For most over voltage conditions a blown fuse in this circuit is an indicator that over voltage occurred. If none of these fuses blew in association with one or more of your failures, then my first guess is that the problem is not over voltage, unless the voltage spike was extremely short in duration.

The 3 phase delta output from CRM goes to selected taps on the primaries of a 3 phase transformer. The transformer is wye to wye with both mid points (neutral) floating. So fundamentally the primary and secondary side are delta from an external circuit perspective.

The primary side is used as an autotransformer to provide a nominal 230 V to the spindle power supply. The secondary side serves other loads.

There are some small circuit breakers for the coolant and chip conveyor motors. In the newer machines one DC supply powers both the spindle motor and the servos.

Do any of the fuses in the CRM circuit blow? Do any of the small motor circuit breakers blow?

Can you absolutely say that failures occur on power off? If so, then is this when just CRM is turned off, or if CRM is left on and the main disconnect is turned off? On our 6 machines over a 10 to 20 year period we have never had a fuse or breaker trip on power off or power down which drops CRM. Our only tripping has been on abnormal source line conditions.

.​

 

[GITRDUN]

Specifically on the Haas machines - no fuses blow - ever. The only cuircut breaker that trips is the main transformer breaker no matter if it is a new machine with power supply onboard in the vector drive or the older non vector drive style which have a couple seperate power supplys in the machine. As of right now i cannot say that the failures occur precisely when the machine is powered off. But i will know this shortly as every employee has been instructed to open the machine cabinet and switch off every cuircut breaker inside of a machine after they have powered it off and disconnected the main disconnect. My maintenance man has been assigned the task of double checking every machine at the end of shift to make sure breakers are switched off. If they are tripping when the machine is powered off then it is only a matter of time until someone finds one already tripped after they turn the machine off. The main disconnects are NEVER left on after a machine is powered off.

 

[gar]

140810-1345 EDT

GITRDUN:

Our 1994 HAAS has only 3 Transorbs while an about 2000 machine has 6 Transorbs. The 400 V threshold of the Transorb on our 2000 machine would blow at least one of the fuses if any input exceeded a peak of 400 V, line-to-line or line-to-ground. Thus, if your newer machines are not blowing any of these fuses, then I do not believe there is an input overvoltage problem.

HAAS makes no use of an external or internal neutral. Internal to the main power transformer the primaries are connected as a wye and thus synthesize a neutral which is connected to nothing. Same for the secondary of this transformer. There is no need for an external neutral, and if it was brought to the machine there is no place to connect it.

There is no good reason for the main breaker to ever trip except from some severe internal overload. For example a shorted diode in the 325 V power supply.

You have too many individual machines showing problems that there is some piece of information that is missing. On the surface your power distribution, and any neutral or grounding of it does not seem to be a likely origin of the problems. The HAAS machines should work fine on a moderately balanced floating delta or corner grounded source. To repeat there is no need in HAAS for a neutral, and no way to use it.

There may be some very large very short duration voltage transients that do not cause the Transorbs to blow a fuse, yet that might damage some electronic components that are tied to the input lines.

What is the ambient air temperature in the HAAS cabinet with the door closed, and a hot day (105 F.).

.

 

[starbolin]

The most important piece of troubleshooting information has not been provided and no one has asked. When did this behavior start and what was the last change to the plant before it started???

I'm thinking bad TX somewhere. Behavior is different when system is unloaded from when it is loaded. Over-voltage trips could indicate shorted turns in a TX primary which might not show when the secondaries are loaded.

 

[starbolin]

...

I spoke with a well trusted CNC repair company the other day and told them to schedule me in to come down and go through our power setup with a fine tooth comb. I have known these guys for probably 20 years and they know more about CNC machines than anyone ive ever met. Not sure where else to turn to right now.

That is probably a good course of action. Someone needs to spend some hours on site. The EC you had out only saw dollars and had no interest in you as a customer. The minimum would be a amp-clamp check on every machine power conductor, every TX conductor and on every ground line plus carefully checking each machine transformer plate and jumper configuration. Following that I would want a ground impedance check on each machine and TX ground. In you case I would want to measure TX leg voltages with one leg at a time loaded to see any imbalances.

Good hunting.

 

[SceneryDriver]

Might this be a case of the POCO switching power factor correction caps in and out during off-peak hours? The switching can cause spikes. The OP has stated he is several miles from a substation. How far is far enough to not see an effect from cap switching events? (I realize there's no hard and fast rule, though).


SceneryDriver

 

[gar]

140811-0814 EDT

starbolin has a good question. What changed from a time when there was no problem to the time when failures began?

Some minor corrections to my previous comments. The transient or overvoltage detection circuit in our 1994 HAAS machine has 6 MOVs instead of 3 as shown on the circuit diagram in the manual. The shape of the voltage detector is that of an MOV rather than a Transorb.

The note in the manual says that a blown fuse will drop CRM. This is not correct. A blown FU3 will not drop CRM. A single phase transformer is supplied from FU1 and FU2 to power CRM. Thus, a blown FU1 or FU2 or both will drop CRM, but not FU3.

GITRDUN has said that no fuses have blown. Thus, there has been no major overvoltage problem, other than the possibility of a large short duration pulse that was insufficient to blow one of the fuses.

The type of failures mentioned do imply a large voltage spike.

.

 

[gar]

140811-0859 EDT

We have had one or more of FU1-3 blow in a lightning storm.

.

 

[JoeTroubleShoot]

Have the power company record the voltage at the transformer or service entrance. Like others have mentioned, utility switching could be causing problems for a fraction of a second that you will never be lucky enough to catch with a hand held meter. It's most likely free and will narrow down your problem to the inside at least.

I've seen similar problems where the voltage was dropping from 480 to 180 between two legs and causing three phase breakers to trip. It happened every Friday at this particular place and after a few visits we installed a recorded and caught it right away. It ended up being a faulty utility transformer. Each event lasted less than a second. Might be a good direction to go before you spend more money, although they may want the recorder on for a few weeks.

 

[GITRDUN]

We had our power setup completely changed about 4 years ago and i dont honestly recall having this much trouble before that time. But we also have about three times as many CNC's now as we used to. Our origional setup was a 240V Delta from the POCO. Our setup now is utility connecting to our 480Wye Tx feeding into a 208Wye which supplys our single phase, dual phase and three phase to most machines. The 480 feeds three phase to only 4 machines.

I am withholding any more guesses on my part until we find out for sure exactly when the failures are occuring, either when the disconnect is switched off, or during the night when the disconnect is off. I know for sure they are not occurring during switch on from catching 5 machines with breakers tripped before they were even switched on. This will be a large piece of the puzzle.

 

[gar]

140813-0915

GITRDUN:

A HAAS or other CNC machine is a big chunck of cast iron, varoius steels, and sheet metal. There is some wiring that is not totally enclosed within this metal structure, but it is very close. Thus, I classify the wiring and circuitry to be within a Faraday shield. Some of the circuitry is electrically connected to the shield. Other circuitry is isolated.

External connections to the machine are:

(1) AC main power ---

3 phase 208 delta (CNC is only a delta load), grounded by the neutral point of a 3 phase wye source.

The EGC is always connected to the CNC machine chassis, and originates at the source 3 phase neutral and grounding electrode. This EGC wire is permanently attached at the CNC, and its circuit is never opened at the CNC.

(2) An air line, could be rigid steel pipe. All our machines are non-conductive hose.

(3) Communication ports like RS232 or Ethernet.

After the main disconnect/circuit breaker is turned off there is a fairly large air gap between the three incoming 208 lines and anything inside the cabinet.

With a high confidence I do not believe any failures like you have seen can occur after the main disconnect is put into the off state, other than a direct lightning strike to the machine.

More later.

.

 

[paul]

Get a somebody out there who has a power quality analyzer and hook it up to your system for about a week or two. That should give them enough data to find out what is going on. For example, we like to use an independent company called Western Electrical Services. They do all of our breaker testing, transformer testing and power quality testing. You should be able to find a company in your area who supplies the same type of services.

 

[GITRDUN]

140813-0915

With a high confidence I do not believe any failures like you have seen can occur after the main disconnect is put into the off state, other than a direct lightning strike to the machine.

More later.

.

I agree totally, anyone with common sense would. But the fact is that the breakers inside the machines are not getting tripped off when the machines are switched on, I KNOW THIS AS A FACT. Therefore it is either happening at the moment when the machines main disconnect is switched off, or during the night while the main disconnect is off. Common sense tells me its happening when the disconnect is switched off, but until we catch one in the act i cant answer that question positively.

 

[iceworm]

I agree totally, anyone with common sense would. But the fact is that the breakers inside the machines are not getting tripped off when the machines are switched on, I KNOW THIS AS A FACT. Therefore it is either happening at the moment when the machines main disconnect is switched off, or during the night while the main disconnect is off. Common sense tells me its happening when the disconnect is switched off, but until we catch one in the act i cant answer that question positively.

I read through all the post and I must have missed this:

Are the CBs tripped? Is the handle in the TRIPPED position?
or
Are the CBs turned OFF? Is the handle in the OFF position
​

I did not see a clear distinction in the posts (and may have missed it)

... But the fact is that the breakers inside the machines are not getting tripped off when the machines are switched on, I KNOW THIS AS A FACT. Therefore it is either happening at the moment when the machines main disconnect is switched off, or during the night while the main disconnect is off. Common sense tells me its happening when the disconnect is switched off, but until we catch one in the act i cant answer that question positively.

This puzzles me completely.
There are no AFCI, No GFCIs, no GFEs, no shunt trips, no UVs, no motor operators? This is a question not a statement. But it appears to be true reading the previous posts.

So NO, the CBs are not "tripping" when the Disconnect is off

Disconnect OFF = no current flow.
No current flow = no tripping

Minor teasing to follow:
If the CBs are OFF (as opposed to tripped) You likely have the same issue as the poor guy with the residential panel with the mystery trips (or switched off) I'd be looking at the kids or secret military mind control radiation:roll:

Seriously - Paul has an excellent suggestion. Get power monitoring equipment.

ice

 

[gar]

140813-1441 EDT

GITRDUN:

Late yesterday I recorded some information from several of our HAAS machines, but did not have time to correlate with the circuit diagrams in the HAAS manuals.


Oldest machine is a VF-2, 6-93, serial range 2000

Main disconnect/breaker is 30 A.
Small main transformer breaker 25 A, 115 V, 3 phase (obviously secondary side)
Coolant pump 3 A, 230 V, 2 poles
Solenoids 3 A. 115 V, 2 poles


VF-3, 3-06, serial range 7000

Main disconnect/breaker is 40 A.
Small main transformer breaker 25 A, 115 V, 3 phase (obviously secondary side)
Coolant pump 5 A, 230 V, 3 poles
Solenoids 5 A. 115 V, 2 poles
Fuse FRN-R15, 230 V, 3 poles


VF-3, 3-96, serial range 14000

Main disconnect/breaker is 40 A.
Small main transformer breaker 10 A, 115 V, 1 pole
(obviously secondary side)
Coolant pump 10 A, 230 V, 1 poles
User 15 A. 115 V, 1 poles
Another main 10 A, 115 V, 1 pole


VF-0E, 6-00, serial range 21000

Main disconnect/breaker is 40 A.
Small main transformer breaker 10 A, 115 V, 1 pole
(obviously secondary side)
Coolant pump 10 A, 230 V, 1 pole


SL-20T, 4-00, serial range 62000

Main disconnect/breaker is 40 A.
Small main transformer breaker 10 A, 115 V, 1 pole
(obviously secondary side)
Coolant pump 10 A, 230 V, 1 pole
User 15 A. 115 V, 1 pole

A separate board with 3 phase breaker.



The combinations for these different time frames are all slightly different.

From your descriptions the machines are shut down with the power down panel button. Thus, power is removed from all active circuits, except the power supply for CRM, by the power down sequence dropping out CRM.

Thus, when the main disconnect is opened there is virtually no load. no current, and no switching transient.

.

 

[gar]

140813-1525 EDT

iceworm:

The small breakers, in contrast to the large main disconnect/breaker, I do not believe have a tripped state different than the off state. In otherwords similar to a Potter and Brumfield breaker I use in a power supply. And I don't know about the main disconnect/breaker relative to whether it has a tripped state..

Our new 2014 machine a has changed structure for the main disconnect. Now the disconnect lever directly protudes thru the cabinet door.

.

 

[iceworm]

Question still stands:
For any CBs that are acting erratically: Are they OFF or are they tripped? I'd recommend not guessing, but state with certainty. It's okay to look up the part number and get a cut sheet for operation.

As I said: Breakers tripping or flipping with the power off are unlikely to the 3rd water. So, it matters - Flipped or Tripped?

ice

 

[GITRDUN]

Question still stands:
For any CBs that are acting erratically: Are they OFF or are they tripped?

ice

Im pretty sure most if not all of the machines breakers dont have a tripped state, they are either flipped on or off. I will double check to make sure thats correct however and post back later.

The large main disconnects do not have a breaker, they are just a simple disconnect switch.

We shut machines off by pushing the control off button, then going to the back side of the machines and flipping the main disconnect switch to the off position. Power up is reverse. This is standard for any CNC machine ive ever seen no matter what model or year it was. The main disconnects are built into the machines.

 

[gar]

140813-2138 EDT

GITRDUN:

The HAAS large main disconnect is also a breaker. Look at the disconnect's handle and you will see the ampere rating.

Look at the information in any of your HAAS manuals and it will indicate the breaker rating of the main disconnect.

HAAS has on line a mill manual titled --- The Mill Operator's Manual, June 2007. . The filename is 96-8000.pfd . I copied this several years and on page 193 it has electrical requirements listed. Here it clearly indicates the rating of the disconnect/breaker for several size machines.

.