Contactor before or after the VFD, or ?

[JesperMP]

Hi,

I would like to hear some of you experts on which method of disengaging a VFD is the best, in particular with respect to E-stop.

1. A contactor in front of the VFD.
Problem is that communication to the VFD will be lost when the contactor is opened (when the VFD control is powered from the mains supply - most are these days).
Also, I have heard that a long waiting time is required before reapplying power (is that really true, does it apply to some or all VFDs ?).

2. A contactor after the VFD.
Problem is that the VFD will experience high voltages on the output terminals if the contactor is opened while the VFD powers the motor.
Some counter this by sending an "inhibit" command to the VFD (from an aux contact on the same contactor), the "inhibit" input is then setup to let the VFD enter cost-to-stop mode thereby avoiding overvoltage alarms. Will this work with any VFD ?

3. No contactor at all.
No problem if no E-stop or safety stop is required (a completely enclosed pump or fan for example).
Some use the VFD to bring the motor to a controlled stop at E-stop.
Problem is that this approach is not allowed in the EU (is it really OK to do so in North America ?).

So, what are you guys doing ?
Are there regulations that has to be followed ?
Thanks for any info

 

[ScottyUK]

Hi JesperME,

There was a thread running about a year ago on this very subject and it had a lot of opinions expressed and some good justifications for both methods. Have a search through the old threads.

Personally I would put it upstream of the VFD, provided you are using it as a true ESD, not a convenient way for lazy operators to stop the machine. To me a true ESD should only be operated when an unsafe or abnormal condition is developing, and in this situation the slight disadvantages of losing comms to the drive are fairly trivial. The downtime caused is unlikely to be as long as that caused if a downstream contactor causes an output stage failure.

At least one post in the old thread I referred to earlier indicated that some modern drives are reasonably tolerant of contactors on the output, but I'm still a sceptic.



------------------------------

If we learn from our mistakes,
I'm getting a great education!

 

[skills]

I have yet to see a drive installation quide or manual that doesn't explicitly depict 'downstream' output side contactors as anything but a no-no. Generally a picture with a red diagonal conveys the manufacturers intent. I'm sure some drives might be OK with this under some circumstances; but, the physics of what's happening to the transisters is immutable. This isn't good unless you're viewing the drive as a sacrifice in these events.

I've used output contactors; but, I informed everyone that they shouldn't be surprised if we blow it off the wall and every 9-10 months, we would!

 

[buzzp]

As far as waiting a long time to reapply power, this is dependant on the application and should have nothing to do with the drive (ie downhold pump or something where the water can back flow,spinning the motor backwards then when reapply power shafts get broken- just one example).

 

[JesperMP]

Re 1. A contactor in front of the VFD:
There is a CAUTION in the manual for Telemecanique Altivar 11 to
"..[avoid cycling power] in less than 5 minutes. .. to avoid damaging the precharge resistor and capacitor."

Re 2. A contactor after the VFD:
Off course the normal starting AND stopping is via the VFD. The contactor is first disengaged after the drive has come to a complete stop. The dangerous scenario will only happen when E-stopping.

 

[buzzp]

Is that cycling power to the drive or cycling the output power? Sounds like the input power to me.
Ooops! I shoud read more carefully. The contactor was put in front of the vfd on item #1, ignore me.

 

[ScottyUK]

Case 1 applies to frequent cycling of the of the power supply to the drive every few minutes. This (hopefully) doesn't apply to a contactor only used to provide an ESD function, and the manufacturer's caution can be disregarded if the contactor is only operated under initial power-on and ESD conditions.

Case 2 does not apply to most ESD situations because there will not be time for the drive to ramp down to an orderly stop prior to opening the contactor. There may be certain specific applications where an abrupt stop gives rise to even greater danger, and a controlled stop is therefore necessary even under ESD situations. Only you know whether your application is one of these situations, but the majority are not.



------------------------------

If we learn from our mistakes,
I'm getting a great education!

 

[JesperMP]

Re 2. A contactor after the VFD:
Here's an extract from the AB Powerflex manual:
"Allen-Bradley Drives can be used with an output contactor between the drive and motor. This contactor can be opened under load without damage to the drive. It is recommended, however, that the drive have a programmed “Enable” input and that this input be opened at the same time as the output
contactor."

 

[JesperMP]

Re 1. A contactor in front of the VFD:
Here's some more AB PowerFlex manual text.
"!ATTENTION: A contactor or other device that routinely
disconnects and reapplies the AC line to the drive to start and stop the motor can cause drive hardware damage. The drive is designed to use control input signals that will start and stop the motor. If an input device is used occasionally, an auxiliary contact on that device should also be wired to a digital input programmed as a “Stop” function."

The two AB Powerflex quotes seems to indicate that AB prefer a contactor on the output side of the VFD over a contactor on the supply side.

 

[jraef]

As I recall, I chimed in on the old thread as an opponent of output contactors. However I have since been involved in several projects where VFDs have Soft Starters as a redundant backup, and of course there were load isolation contactors on the output of both devices. In every case, operators went through several changeover routines under load during commissioning. In no case was the VFD ever harmed. I don't know if it is relevant, but all of these were Robicon VFDs and I was told they have been doing this for years with no trouble. I also heard that Yaskawa and several other VFD manufacturers offer this as a routine option without qualifiers as to duty cycling or advanced base blocking of the transistors. I think it may have become a remnant issue from older VFD technology that many manufacturers are reluctant to take a chance on changing, but those who did are finding it is no longer an issue. Not being one of the latter I cannot say with absolute surety that it is OK for every drive design however.

"Venditori de oleum-vipera non vigere excordis populi"

 

[ozmosis]

If you have a look at thread237-81749 then there was a number of discussions about the 'Safety-Off' function now available on a number of VSD's these days.

 

[DickDV]

At the risk of repeating a recent thread on this same E-stop subject, a little engineering judgement goes a long way to advance real safety during E-stop conditions.

First, if you put an E-stop contactor in the input leads of an inverter, from the time you open the contactor until the DC bus capacitors discharge, you will get normal motor at-speed operation. Following that you get coast-to-stop operation of the motor and coupled machine. If it's a fast stop beginning as soon as you open the contactor that you are looking for, this is most definitely not what you want.

Second, if you put an E-stop contactor in the drive output leads (motor leads), from the time you open the contactor you will get coast-to-stop operation of the motor and coupled machine. If it's a fast stop you want, this isn't it either. In fact, with this arrangement, it is all-together likely that routine stops will be faster and more predictable than E-stops especially if the drive has snubber or regen braking capabilities.

The optimum way to get rapid, predictable stopping on E-stop, in my opinion, is to use a control input on the drive to stop the machine taking advantage of whatever braking capacity the system has. You may need to arrange the E-stop signal to command a steeper decel ramp to stop than a normal stop uses. Upon motor stop, for safety purposes, it may be desireable to open a contactor in the drive input or output leads or both for isolation.

My point is that the drive has the capability to stop the machine more quickly than coasting and, in many cases, this rapid stop represents real enhanced E-stop safety.
In extreme cases, it will be possible to force the input/output contactors open after a short period of time to guarantee at least a coast-to-stop if the drive somehow isn't responding to the E-stop control signals. That would provide the best of both worlds for those applications that really need it.

The case of E-stopping DC drives is somewhat easier due to the ability to de-energize the drive and have an M contactor (DC contactor in the motor armature leads) drop out on removal of power. This contactor would be configured to throw the motor armature leads across a fixed braking resistor and utilize the internal generator feature of DC motors to dynamically brake the motor quickly to a stop. This will not work properly, however, unless, on drive power-down, the power supply for the motor field remains energized. A shunt wound DC motor will not regenerate with a dead field so coast-to-stop is all you would get. At least, with DC motors, this option is available for fast stopping with proper system configuration where, with AC, it takes extra drive equipment and an energized functioning drive to get it to work.

Again, it doesn't hurt to think thru the options before setting up the E-stop system.

 

[oceanamber]

Although I can understand the merits of an emergency stop system using the electronics of the VSD to brake the motor in my opinion we need to keep the emergency stop system as simple as possible. A simple device operating an upstream contactor should shut the system down whether the smoke is coming from the motor, from the VSD or for some mechanical reason when the motor just needs to be stopped. In the unlikely event of the contactor sticking you can still stop the drive via the VSD usefully eliminating a point of common mode failure on what is a critical action.

 

[ruggedscot]

Id shy away from an output contactor of any description, best have the motor directly connected to the drive. We use 30kW motors hooked to danfloss drives and have not had any problem. The motors being fluid pumps. E stop can be carried out by linking to the drive and stopping the motor that way. Id say that a contactor up stream is okay but you need to hoom this in so that the motor is stopped or a stop command generated at the drive when the isolator is switched off.

Rugged

 

[DickDV]

A couple of further considerations regarding my earlier post.

First, in the "coast-to-stop" situations cited, if, for whatever reason, that is the only acceptable stopping method, a mechanical safety brake could always be applied to get a faster fail-safe stop. A word of caution tho, stopping so hard that drive shafts break, gear teeth strip off, belts slip, or drive chains break, is hardly an improvement over coasting. Worm gears that lock up when driven backward (ratios over about 25/1) are also unsuitable for passing braking torques. Again, a little thought is in order.

Second, under "coast-to-stop" conditions, it may work out that guarding the machine for safety purposes is a better solution that a fast stop.

Third, there are situations where a slower stop is actually safer. For example, if you are transporting a ladle of molten metal across a room, a sudden e-stop that spills molten metal on the floor and splatters it all around is hardly a "safe" stop. Similarly, E-stop when driving your car is not always locking up all four wheels in an uncontrolled slide!

In summary, I am most comfortable with E-stop solutions when operators, engineers, and inspection authorities have had a chance to think thru their options and pick a solution that is acceptable to all. Trying to do this negotiation on the shop floor with the machine controls already half built is ugly at best.

 

[JesperMP]

rugged,
I think that in your case, the E-stop is essentially there to have a handy save-the-pump button, rather than a person-safety stop button.

A pump "normally" doesnt have any dangerous aspects (unless extreme high pressures are created, or the pump is used to drive hydraulic actuators f.ex.)

If that is the case, then stopping the pumps via the VFDs is quite allright.

 

[ruggedscot]

I had trouble some time ago with a soft start, and I guess this needs to be looked at here as an electromechanical brake has been talked about. Worked with a conveyor that has a brake that sometimes would not release, this was due to the fact that it was connected to the motor star point and phase, this would have worked if it had been a DOL starter but as a soft start was used we had the brake coil not getting a clean enough jag to release properly leading to the motor not running as the drive would hit the current limit.

Any brake devices ensure that they are off of the supply side and switched as required.

E stopping has now been affected by newer installations that utilise modern drive control. Official stance is that an E stop opens the supply and ensures that the system stops. Now due to drives this has to be looked at in a manner that does not put drive units at risk. Simply opening the supply to a drive many times puts undue stress and strain on the input stage - drives tend to rectify the AC to an intermediate stage and then convert back to AC for the motor. The output stage if being driven and the circuit is opened this can overstress the output stage leading to failure of the semiconductors. Id be avoiding that as much as I can as those are pretty darn expensive.

Running a drive under normal operational conditions should mean that the operator does not require to use the e stop, if the e stop is being used then one has to look into this and identify why, if it is being used as a means to switch off the motor then a way should be looked into to provide a local point of operation for the user operator to remove the operation of the e stop. This can be undertaken now with out much expense with the way that drives operate now. No heavy wiring required. If I was doing this I would look into doing it this way and Id incorporate an e stop with a flip cover and have this open the mains supply contactor. The reset button for this would not be in the locale so as to prevent the operator relying on this as a switch off point. A local isolator switch on the motor for maintenance purposes would also link through this to ensure that the mains was dropped when isolator was opened.

Rugged

 

[ScottyUK]

Ruggedscot,

You've summarises my opinion very well, especially about operators taking the easiest route to do their job - a well-designed system makes the right way easy and the wrong ways harder, so the operators natuarlly favour the correct method.

The key factor is to discriminate between an E-stop and a normal stop. The E-stop should occur very infrequently in a well-designed process, so the detrimental effect on the drive link capcitors and precharge resistor caused by disconnecting the supply should be insignificant.



------------------------------

If we learn from our mistakes,
I'm getting a great education!

 

[advidana]

Be smart, check with the manfactures recommended method of using the vfd. I would get a contactor that is built in with the vfd if required, that way if you have problems the vfd the manufacture would have to fix it if it fails.

 

[ruggedscot]

All very well getting one with an inbuilt contactor but this increases cost and also you may find that the set up is still configured not to switch the contactor on and off to suit the motor when the drive can do this electronically.

We are getting sidelined here with start / stop functions and e stop functions which are two different scenarios. If you do have a genuine e stop condition the last thing on the persons mind is the drive or any other equipment.

With modern electronics and drive controls sometimes its better to do a controlled stop through the drive. Or have some sort of electro mechanical brake. Cutting the supply like old systems used to do may increase the danger. Some rotational equipment may have a long run down time compared to what could be achieved with a controlled stop commnad.

Reading the drive manuals and research on the subject is required.

Rugged