VFD With Bypass Star/Delta Starter

Since your client wants to operate any one of this starting device in the event of failure,use both in parallel to 6 terminals coming from the motor. Use MCCB in series but did you asked the reason for frequent failure of starter or the reason behind this requirement?

Is you client going to vary the pump output? If not then all he need is a soft starter. Which in has just about replaced the use of Star/Delta Starter.

You will need a 6 pole change over switch with some fancy wiring in it. it can be done, but I've got to echo "why is the VFD so unreliable"?

Dear Mr. Tony,

In my view, the Electricity Regulation may not permit this. Other CR4 Members can share their views and experience.

Thanks,

RAJESWARI.

Not worth it, way too complex. Either use an RVAT or RVSS as the bypass, that way the motor connections match the VFD.

Thanks for your comments.

Hithuanand: How can I directly connect six wires from motor to VFD? I think I need to use a six pole isolator between the VFD and the motor, but Isolator is advisable after the VFD?

Ozzb: Yes client need to vary the pump output, star/delta only for emergency.

TonyS: Money is not an issue here, I think client want to try all the options to get zero down time.

Rajeswari & JRaef: Thanks for your input.

I am not familiar with VFD connections.Please get help from experts and Gurus here.

What application of the pump? It will cost to provide interlocks for these devices. And if controlling the pump rate with a VFD is necessary. Then the switching on of the Star/Delta starter may get someone hurt if the VFD fails. Pump will go to full speed unexpectedly. Hoses and lines that have been ran at a lower pressure and volume may immediately rupture if not tested on a regular bases. Manual over ride would be less costly. Everyone would then be aware went it started up.

And if down time an issue. Store a back-up VFD don't that that long to change one out.

Zero down time can be achieved by installing hot standby equipment.

It is extraordinary to provide star-delta as backup to a VSD. What planet does the Client normally inhabit?

You cannot connect the 6 motor leads to the VFD in the same way that you would connect them to a Star-Delta starter, and it is not just a matter of an isolation switch, although isolating the VFD from the Star-Delta is paramount. You would need almost the equivalent of ANOTHER Star-Delta starter arrangement on the output of the VFD to re-configure the output conductors as Delta Only when the VFD is in play. Like I said, it would get VERY complex and the likelihood of someone getting even one wire wrong is very high, and may result in the destruction of the VFD.

Again, WHY?

To even attempt this will cost a lot of money and time and effort, and VFD's are NOT unreliable, in fact the very opposite.

If you want to run the motor, and I take it it is a surface motor, you will need to bring ALL six wires back to the S/D starter, then if you want to run the same motor with a VFD you will need an additional contactor to create a star point to run the motor in star.

Here is the questions..

1. do you or your client understand how a VFD works?

2. Have you completed a cost analyse to find out if you will get any benefits from a VFD?

2. Do you need to vary the flow rate (already asked) of the pump?

3. Are you using the VFD as a soft starter?

4. Is the motor rated for VFD operation?

5. Have you thought about harmonic mitigation?

6. size of motor?

7. Pump maximum flow rate, barrel per day, cubes per day, gallon per minute...

8. duty time?

You just can't install a VFD and hope for the best..... and if you answer NO or don't know for Question 2, then forget the VFD and as already suggested, get a soft starter, but the medium is water, so what is the point in either, unless the pump/motor is extremely large.

Make it easy for your client and yourself.. one or the other

http://www05.abb.com/global/scot/scot209.nsf/veritydisplay/0fe1791c07b06416c12577d100428fe2/$file/1sfc132060m0201.pdf

This is how ABB recomends doing the circuitry for softstarters. For the VFD you can use the same layout and that is with one MCB for both the VFD and star/delta starter. Using separate MCBs won't help unless you use some sort of disconector switch for the input and output of the VFD, in this way you are able to completely isolate the VFD in case of failure and get almost zero downtime for the pump. Of course it would be a bit exagerated but it can be done.

Linking both starters outputs is easy, use a busbar - an electrical knot expanded in space . And also you can use a busbar for the input of the drives. If one drive fails just cut down the power, untie the faulty drive and power the circuitry again. That would be 30 minutes downtime, most, if you have an electrician arround. That is because the faulty drive can trip the MCB when reconected.

Wire the Circuit breaker, then a three pole changeover switch, then the VFD, then the star delta in series. For operation always switch on the star delta and let it time out to delta running. Then switch on the VFD and ramp up the speed. In the event of VFD failure, The changeover switch routes power from the VFD input to the star-delta input, shorting out the VFD. Then the VFD (helpful if it is connected by short bus bar links top and bottom) can be removed for repair.

VFDs, especially large ones are much more unreliable than star delta starters. If our questioner has a plant that would be put out of action by a VFD failure and the company will lose $??K per hour, the precaution to install a backup star delta is perfectly reasonable. I have several clients with this arrangement fitted.

if you know or use VFD's that fail frequently, have you thought about getting better ones that don't fail? Just a thought!

how often do you clients have to use the S/D starter as the VFD has failed?

Once that I know of, about four years after the star-delta backups were installed, but they would not always tell me. The one occasion a VFD went down was on a 132Kw shear mixer. Down time on that particular plant costs the company about £8K ($12.5K)per hour. They have four pumps/mixers backed up in this way. That one failure saved them about twenty times the total installation costs because they do not keep four different sizes of spare VFDs ranging from 75 to 200Kw in their own stores. It was that incident that persuaded other clients to adopt the practice. Worst case is another company, who would lose £31K ($49K) for every hour of down time.

Correct me if i am wrong.. if I follow all you have said, you will use the VFD to start the motor AFTER the S/D has timed out, so the motor will be ramped up by the VFD in DELTA.. Yes?

Yes, they keep upstream valves closed during run up so they don't need much torque.

that is the most ridiculous thing to do, shut the valves, run up the pump then open the valves.

The downthrust (valve closed) and then throwing the pump into up thrust (valve open) will give a very small MTBF for the pump.

Have they not heard of pump curves?

If you short out the VFD you will put mains onto the VFD output damaging the devices. so, you also need to isolate the VFD output unless it is removed every time!

BTW, the VFD's I work with are very reliable up to 700kW at LV, but they need to be installed and operated correctly and periodically checked for minor maintenance. If you have suffered worse experience then you can greatly improve your installations should you want to. Maybe you are not getting sufficient guidance and information re this?

Dear Mr.kra101,

While the driven equipment is a PUMP, AND STUDYING THE RESPONSE from CR4 members, which informs lot of complications.

The simple solution will be "to put one more pump with seperate drive" as a stand by unit. Solution itself, should not be a greater problem than the original problem.

Good Luck.

RAJESWARI.

If the motor is not very big. I think less than 50A current draw. I would make a connection on the cable and one each on the VFD and the star/delta starter. In the event of faliure simply connect the plug to the other output.

It will cost 1 male connector, 2 female and additional 1 hour to fit when installing the units.

Yes! Possible. But, the circuitry will be very complex.

Thank you all.

I will try to convince the client. If he still insists for it then I think I will use jhhassociates proposal and also will consult with VFD manufacturer for their recommendation.

Once again thank you all for your time and effort.

OMG, have any of you even LOOKED at the 6 wires to the motor necessary for Star-Delta starting? That solution from jhhassociates is NOT going to be as simplistic as it sounds!

Oh well, I give up. Have fun at the fireworks show.

GA - I would like to add that a disconnect in front of the VFD could be added or the MCCB could have separate circuit breakers for the VFD & star-delta. This would allow for the VFD to be serviced without taking the pump off line.

But this decision all depends upon the operation of the plant. If it is a 24/7 plant, add the LOTO (lock out tag out) disconnect/breaker, if there is nightly scheduled maintenance, skip this additional expense.

It's a bit hard to see at that size, but I think you show a CO device, which I would interpret as a "Change Cover" set of contactors. So now you have 5 contactors to accomplish this, 4 of which meed perfectly functioning mechanical interlocks to avoid a disaster. OK, well enough, the design accomplishes the task as initially described, but maybe not the intent.

This design leaves out a significant issue, that of the HORRENDOUS spike that can happen when a Star-Delta starter transitions from star to delta. In this design, that spike will hit the front end of the VFD. In a "normal" bypass scenario we don't typically worry too much about isolating the front end rectifier unless, as was mentioned, we want a service isolation switch. In that case a manual disconnect device is more than adequate. But in THIS case, because the bypass is Star Delta, you would be unwise to expose the diode bridge to that transition spike that will likely happen on the Star- Delta starting. So now you must add a 6th contactor and control circuitry so that the isolation does not depend on an operator remembering to turn the manual isolation switch.

So in order to avoid the "extra" cost of a solid state bypass starter, which could use as few as 2 contactors (the CO in that diagram) and would have OL protection built-in, you are now going to add 4 more contactors, another mechanical interlock, a thermal OL relay and some relatively complex control circuitry, all to "increase the reliability".

Makes zero sense to me...

Thanks for the comments. I agree with you about the extras needed and the control circuit for the S-Delta etc... But the OP is interested in this scenario and thinks that it may be cheaper (?).

In the case of the spikes affecting the vfd input stages: Most of the VFD do come with choke and filters either in-built or to be added to conform with norms. Also, since the S-D is for when the VFD needs replacement or else, then all is needed is to disconnect the short wires from the MCCB to the VFD. This does not affect the downtime... and maintenance can be done on the unit etc.

On the submitted picture/drawing small characters: I am sorry for that but I did not think it would be that small. You can copy/save and open as a picture etc.

Thanks

The sixth contactor could be eliminated if the isolation switch has an auxiliary contact which inhibits operation of the star-delta when the isolation switch is not open.

Agreed - zero sense ...

VFD's today are pretty reliable but I always specify them with an integral bypass contactor, so that if if fails, you can do an across-the-line start. If the motor is very large, this may not be an option, though. Let us know the HP-that really drives the answer to your question. Using two different starters to one motor is pretty complicated, as has been pointed out and has the potential to be hazardous if not handled right. Consider buying a 2nd VFD that can be swapped out if the first one fails. That will be less expensive than a VFD/Y-Delta conglomeration.

with every NEW comment, I read and read again your statement, and that of others, and for the life of me I cannot understand why.

Who told your client that VFD's are unreliable?

I install and operate large submersible pumps for the oil industry all over the world WITH VSD/VFD.. (same animal different name).... and apart for the over temp trips due to some VSD's being in the desert, I have used the same VSD's on the same wells for 5 years or more with NO failures, in fact the surface unit proves to be way more resilient, durable and reliable than the downhole motors and pumps.

So why make work and increase your costs trying to resolve a problem that does not exist?

I agree, a VFD is now very reliable. If 100% up time is that big of a deal, add a second pump & VFD in parallel to provide a live standby. Alternate between the VFD & pump sets to keep both in good working order and continually tested. Adding complex high powered switching logic to bypass the VFD may decrease reliability and contribute more to downtime than than the VFD. The complexity will decrease the MTBF (mean time between failure) and increase the MTTR (meant time to repair). Make sure that your customer is aware that back up systems require routine maintenance and testing.

Memory lane: When VFDs were relatively new (1970's) and expensive, we were the VFD manufacturer, we had clients in the sewage lift and municipal water districts that had to bring multiple motors on and off line depending upon demand. The VFD was used to ramp a motor up and then with contactors, the motor was transferred to line and then the VFD was free to pick up another motor to run at variable speed. The procedure was revered to bring motors off line. Sometimes we did open transfers and sometimes we did closed transfers (shorting out the VFD). The largest installation was eleven 100 hp pumps on a single VFD. The matrix of motor contactors was a nightmare. It was all controlled with relay logic.

Dear

I am new member CR4 working as a electrical engineer I read your mail and want to share circuit diagram of VFD bypass with star delta but couldnot attached drawings

kindly help how I attached and also your mailing adress

Awaiting your response

Naveed Aziz

CR4 ID: SNA

If you are a new EE, stop now before you begin to think this is a viable solution. It is not. The previous diagram was a good exercise in proving the theoretical possibility, but also proved that the solution is so complex and prone to failure that it should never be considered from the standpoint of practicality. If you have to provide a bypass for a VFD, use a solid state soft starter. Star-Delta starting is bad for motors anyway, and now you add the risk of also destroying a very expensive VFD.

So, what happens when the pump needs maintenance?

Would it be better to install two pumps and operate duty/standby?

12 years later? I doubt anyone cares now.