Electrician Talk banner

Disconnect ahead of VFD?

55K views 68 replies 29 participants last post by  ace24wright 
#1 ·
Awhile ago I was working with a company that was doing a lot of cooling tower retrofit work, a cooling tower company would replace the motors and fans and anything else mechanical. We would install drives on all the motors, all the drives I was installing were ABB ACH drives. Most if not all cooling towers have a disconnect for the motor right at the fan or belt access area or somewhere near it. I was dealing with an ABB rep at the time and he told me that there can not be a disconnect between the motor and drive, so what we ended up doing was installing lockable stop switches wired into the drives E stop control circuit so the conductors from the motor to the drive were continuous. Personally, I do not think this is a safe way to do it since a failure in the controls portion of the drive COULD cause the motor to start even with the switch locked in the off position. This was all passed by the AHJ and an engineer involved with the project.

I still see lots of towers using drives that have disconnects between the motor and drive and I still think this is the safest way to keep things safe when someone is working on the motor or fan. I asked the ABB tech rep why you could not have a disconnect between the motor and drive and he said that turning the disconnect off without stopping the drive first could damage the drive.

Is this true? I find it hard to believe since I see it done all the time, can this really damage a drive? Or is this just something specific to ABB drives.
 
See less See more
#2 ·
Cdn.,

First of all if you really want a isolating or local disconnect switch you should add a interlock microswitch on as well so that way when you hit the local disconnect or isolating switch the microswitch or aux concats will shut the VSD off.

That what I done that in European side.

If you try to disconnect the VSD motour while it still running the voltage spikes may damage the VSD controller I have see it at least couple time when someone try that.

I do not know how strict in your location reguarding of E stop and I know the E stop is NOT a means of disconnecting means. so therefore for true disconnect means just follow as I mention above a isolating switch or local disconnect switch AVCE aux concats to trip the E stop at the VSD controller.

I know there are couple other guys in the fourm they will echo simauir comment as my I know two guys for sure they will say something about this.

BBQ are you in the game ?

Merci,
Marc
 
#4 ·
Since I'm mostly an Allen-Bradley guy... I'll respond from an A-B persepective.

Yes, you can absolutely have a disconnect on the load side of a VFD, if you keep certain things in mind. Like the French sparky rightly points out, you MUST have it interlocked with the VFD itself. If you simply disconnect the three line side conductors, you'll toast your VFD. You absolutely need a 4th contact to interlock with the VFD to disable it from trying to run. In Allen-Bradley terms, that would mean installing the "Drive Guard" module in the VFD (standard fare, lately).

I like to use Meneekes IEC-style motor disconnects, which feature a low-current 4th pole as a standard item. Very LOTO friendly also.
http://www.mennekes.com/product_motordisconnect_features.html
Hubbell makes a suspiciously similar product.
 
#6 · (Edited)
Thanks for the answers guys, I suppose you could use an IEC type disconnect with an aux contact installed and wired to the stop circuit. I was looking at a brand new tower with a few fans the other day, all danfoss drives and all square D disconnects at the tower which definitely did not have aux contacts.
 
#13 ·
I wish they would install the aux switch on those disconnects. We kill a drive every month or two around here on average. The maintenance and AC guys never secure the drive before opening/ closing the disconnects.

You get a spiked wave that travels back to the drive & motor. It can knock a pin hole in the varnish or blow the drive.
 
#15 · (Edited)
There are technicalities here that are being over-interpreted and under appreciated a bit.

There is technically no problem HAVING a disconnect device on the load side of the VFD. What IS a problem is opening a device on the load side WHILE UNDER LOAD. Opening it when the VFD is powered off is not at all a problem. So that means if you have a known procedure of turning the VFD output off BEFORE you open the load side disconnect, that's OK. What the aux. contacts do is to operate slightly ahead of the main contacts and are wired to a disable input on the VFD. That way, by the time the main contacts BEGIN to open on the disconnect, the output of the VFD is already turned off.

But there are a lot of "gotchas" in that, the main one being that many novices do not understand the concept of "off" on a VFD. For example if you have a Decel function enabled, and you put the disconnect aux. contact in the regular start-stop circuit, opening the aux contact will NOT disable the output of the VFD, because it will be trying to DECELERATE the motor.

So for this to work properly, what you have to do is find and understand the input commands on the VFD that will completely turn OFF the output transistors. The problem with this however, is that there are many different VFD mfrs and many different wys of them describing this functionality, which means there are lots of ways to screw it up. Some call is "E-Stop", some call it "Base Block", some call it "Emergency Power Off", etc. etc. etc. Sometimes it's as simple as using a function that would be there for an external overload or thermistor device (because if something is overloaded, it also must kill output power immediately). It therefore requires a THOROUGH knowledge of the VFD programming for make sure you are doing it correctly. Then add into the mix that I have seen several cheap brands of VFD that do NOT provide any such function!

So the SAFE BET is to not install a disconnect (or contactor) on the output side unless you are ABSOLUTELY SURE you know what you are doing in programming the VFD.

How it potentially damages the VFD by opening a load side disconnect or contactor when under load is like this:

When you interrupt the flow of current with an air gap device like a disconnect or contactor, there is an arc formed that lasts until there is enough separation between the contacts to extinguish it. While the arc is still active, there is a very rapid increase in the voltage potential across the gap because of what is essentially a capacitor effect. That rapid change in voltage (called dV/dt for delta (change) in Voltage over delta in time) can cause unpredictable effects in the operation of semiconductor devices, such as the transistors on the output side of the VFD. So if some of those transistors fire at the wrong instant, they can cause a short circuit and become damaged. In addition, the dV/dt can cause tiny little "holes" to be punched through the silicon layers of the transistor material. The effects of that are cumulative. So even if you got lucky and the unpredictable effects are not bad, every time you open a switch on the output you are incrementally damaging the transistors and it's only a matter of time before they fail.
 
#16 ·
There are technicalities here that are being over-interpreted and under appreciated a bit.

There is technically no problem HAVING a disconnect device on the load side of the VFD. What IS a problem is opening a device on the load side WHILE UNDER LOAD. Opening it when the VFD is powered off is not at all a problem. So that means if you have a known procedure of turning the VFD output off BEFORE you open the load side disconnect, that's OK. What the aux. contacts do is to operate slightly ahead of the main contacts and are wired to a disable input on the VFD. That way, by the time the main contacts BEGIN to open on the disconnect, the output of the VFD is already turned off.

But there are a lot of "gotchas" in that, the main one being that many novices do not understand the concept of "off" on a VFD. For example if you have a Decel function enabled, and you put the disconnect aux. contact in the regular start-stop circuit, opening the aux contact will NOT disable the output of the VFD, because it will be trying to DECELERATE the motor.

So for this to work properly, what you have to do is find and understand the input commands on the VFD that will completely turn OFF the output transistors. The problem with this however, is that there are many different VFD mfrs and many different wys of them describing this functionality, which means there are lots of ways to screw it up. Some call is "E-Stop", some call it "Base Block", some call it "Emergency Power Off", etc. etc. etc. Sometimes it's as simple as using a function that would be there for an external overload or thermistor device (because if something is overloaded, it also must kill output power immediately). It therefore requires a THOROUGH knowledge of the VFD programming for make sure you are doing it correctly. Then add into the mix that I have seen several cheap brands of VFD that do NOT provide any such function!

So the SAFE BET is to not install a disconnect (or contactor) on the output side unless you are ABSOLUTELY SURE you know what you are doing in programming the VFD.

How it potentially damages the VFD by opening a load side disconnect or contactor when under load is like this:

When you interrupt the flow of current with an air gap device like a disconnect or contactor, there is an arc formed that lasts until there is enough separation between the contacts to extinguish it. While the arc is still active, there is a very rapid increase in the voltage potential across the gap because of what is essentially a capacitor effect. That rapid change in voltage (called dV/dt for delta (change) in Voltage over delta in time) can cause unpredictable effects in the operation of semiconductor devices, such as the transistors on the output side of the VFD. So if some of those transistors fire at the wrong instant, they can cause a short circuit and become damaged. In addition, the dV/dt can cause tiny little "holes" to be punched through the silicon layers of the transistor material. The effects of that are cumulative. So even if you got lucky and the unpredictable effects are not bad, every time you open a switch on the output you are incrementally damaging the transistors and it's only a matter of time before they fail.

Very well said:thumbsup:
 
#22 ·
I don't like using the exception for VFD's. To many of my co-workers walk in and just turn things back on. They don't use LOTO and don't seem to know any better.
But I can't get management to add the aux switches. They claim a cost factor. But yet we replace drives non-stop.
 
#35 ·
Is there anything code about using the breaker as the disconnect if it's qualified persons servicing the gear?

Seems to me that have VFD's fail from opening a disconnect at the motor is a operational, training, procedural problem. If I was in charge of a place I'd have the disconnects labeled with a warning about securing the VFD first. Smoke test a VFD and you'd be on the short list for the next round of lay offs.
 
#39 ·
There are situations in MANY companies where the issue of an air-gap disconnecting means is a non-negotiable requirement; period. Ford Motor Co. comes to mind, they require disconnects within sight of all motors. In addition, the machine Safety level certifications that are already in force everywhere but North America require not one, but TWO contactors in series on ANY motor that needs the highest level of Safety category implementation, regardless of where the power comes from, including VFDs. these rules are coming to a theater (of operation) near you in the not so distant future. So there is a need for these practices to be understood and implemented, WHERE NECESSARY, in the proper way.
 
#42 ·
spoke to an inspector recently about this issue as we were planning on installing VFD's and had been informed that a disconnect opening when drive running caused damage he said there was no need of a disconnect beside the motor as long as there was a notice at the motor clearly stating where the disconnect was located and again a notice on the disconnect clearly identifying the location of the motor
 
#45 · (Edited)
Sir, good day! If i have a three water cooled chillers which each are having power capacity is 251kw, FLA is 421.14Amp and starting current is 1086 Amp. What should be the size of circuit breaker that I'm going to use each machine?

Thank you in advance and More Power.
You should start another new thread, this is not germaine to the original thread that you posted in. That said, this is a forum for North American electrical engineering issue for the most part. What you are describing is something that would relate to whatever the electrical codes and standards are in your country. What we would tell you that applies to North America may not be correct in your country. You should hire a qualified electrical engineer.
 
#48 ·
JRaef said:
You should start another new thread, this is not germaine to the original thread that you posted in.


The goddamn germans got nothing to do with it, don't you hear good?
Not German, GERMAINE!


As in Germaine Jackson, Michael's brother. :rolleyes:
 
#50 ·
Alot of the places I work at have a kaleidoscope of drives from Cutler hammer to AB. We have disconnects at each motor or (some of you are going to get excited) receptacles. The operators constantly turn those disconnects on and off and sometimes the drive trips but they just reset it. You can have almost all the faults talk to your plc so the operator can see it on the HMI and reset it. *we log that information and review it* thats for 5 hp or less. Over that hp, either run a feed all the way to the motor disconnect first and then back to the drive. OR we install a custom box where the operator has to key off the drive before its possable for him to physically get at the disconnect. Of course there are lots of variables so its hard to say this will work or will not we almost exclusively use coast to stop.
 
#51 ·
430.102 B 2 Says that the controller disconnecting means Shall be permitted to serve as the disconnecting means for the motor and exception 2 makes the allowance for a locking means when located out of sight of the motor, also FPN 1 seems to make a pretty obvious clarification for vfd's as being subject to additional hazards to property. It seems to me that a lockable disconnect anywhere up stream of the drive including the circuit breaker would satisfiy the NEC. What am I missing?
 
#52 ·
IIRC, it has to be a facility with an established LOTO safety procedure that is reviewed and monitored, something to that effect. So it doesn't universally apply to every installation.
 
This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.
Top