VSD - current source or voltage source design

[jdsmith]

I'm looking for some guidance on current source vs. voltage source medium and low voltage drives. We have an application right now where we need to install two 400 HP 4160V drives and one 200HP 480V drive. All three motors are driving long vertical shaft pumps where across the line starting causes excess vibration and water hammer and has caused piping leaks in the past. The initial thought was to install soft starters to alleviate this, but from a process perspective using VSD's with pressure control on the output manifold solves the vibration issue and addresses other process concerns as well. The logistics of the area dictate 150-200 ft of cable between the drives and the motors, and none of this is a hazardous location. We've been looking at Rockwell current source drives for all three pumps because they claim to work with the existing non-inverter duty motors and because they say that cable length isn't a problem. These would be installed in a new climate controlled electrical equipment shelter (building) along with 480V switchgear and assorted other small electrical equipment.

Has anyone had problems with larger Rockwell drives of this design? Is there a better option we should look at?

 

[Besoeker]

I'm looking for some guidance on current source vs. voltage source medium and low voltage drives. We have an application right now where we need to install two 400 HP 4160V drives and one 200HP 480V drive. All three motors are driving long vertical shaft pumps where across the line starting causes excess vibration and water hammer and has caused piping leaks in the past. The initial thought was to install soft starters to alleviate this, but from a process perspective using VSD's with pressure control on the output manifold solves the vibration issue and addresses other process concerns as well. The logistics of the area dictate 150-200 ft of cable between the drives and the motors, and none of this is a hazardous location. We've been looking at Rockwell current source drives for all three pumps because they claim to work with the existing non-inverter duty motors and because they say that cable length isn't a problem. These would be installed in a new climate controlled electrical equipment shelter (building) along with 480V switchgear and assorted other small electrical equipment.

Has anyone had problems with larger Rockwell drives of this design? Is there a better option we should look at?

My tuppence worth....
Do you need variable speed?
Or just a starter to get the machines up to speed?
If the latter, a closed transition starter (Korndorfer) might be an option for the 4160V motors.
If you do need a VSD then stepping down from the 4160V to 480V and changing the motor might be more cost effective.

 

[jdsmith]

My tuppence worth....
Do you need variable speed?
Or just a starter to get the machines up to speed?
If the latter, a closed transition starter (Korndorfer) might be an option for the 4160V motors.

This electrical project is a small portion of a $5 million revamp of the non-potable water system in our plant. We (the sparkies) were initially asked to prepare a cost estimate for three soft starters which came to $65k each (if I remember correctly) for the 4160v motors. Apparently while we were doing that legwork the mechanical engineer decided it was worth getting a price on VSDs, which came back at maybe $225k each for the 4160 drives and $20k for the 200HP 480V drive. Apparently after seeing those numbers it's still worth spending the money because the underground piping is cast iron, which doesn't like water hammer or for the pressure to be any higher than necessary, or it leaks.

If you do need a VSD then stepping down from the 4160V to 480V and changing the motor might be more cost effective.

That's not something I've priced out yet and it might be worth looking at. It might save $150k each on two drives, with the additional cost of $50k or so for custom vertical motors, plus the cost of two more switchgear breakers to feed the drive, possibly higher bus rating and larger main/tie/main breakers in the switchgear, and a larger 4160/480 transformer.

Thanks!

 

[MJJBEE]

Rockwell drives

Rockwell drives

I have had no problem with Rockwell drives at 480. I Have no experience with their 4160 drives but I would probably consider them for the next drive replacement I'm doing at medium voltage along with all the other usual suspects.

 

[Jraef]

Current Source is an old technology that most manufacturers have moved away from because of the inherent problems, not sure why anyone would go back to it. Robicon, the world largest supplier of MV drives even before Siemens bought them a few years ago, was, for many years the champion of CSI drives because they were very robust compared to early PWM VSI drives. But VSI technology has changed dramatically and as of 2 years ago, Robicon dropped CSI technology altogether (except for supporting legacy systems). It tends to be less efficient and creates more harmonics. Its only advantage might be that it uses a big inductor in the circuit instead of a phase shifting transformer as 18 pulse and 24 pulse VSI drives do, and since AB does not make transformers, that may be what was behind their decision.

But another big problem that CSI always had (besides the harmonics and efficiency issues) is that the drive and motor had to be "tuned" together; a CSI drive has a capacitor on the load side and so if you want to change out a motor to install a spare, that involves determining and calculating the proper amount of capacitance. That capacitor also has a tendency to cause resonance with the motor and lead cable capacitance at certain speeds, so the overall installation tuning process can take upwards of a day as the motor is run through its paces with an engineer and a scope watching it to figure out where the bad spots are. Many people did not like that extra down time on change-outs. Something to consider.

To be fair, I now work for Siemens, albeit in a different division, but I had a lot of interaction with Robicon as a consultant in a previous life, along with ABB, Toshiba and GE, on large MV VFDs. I was a fan of CSI drives for a long time, especially after witnessing a set of Robicon CSI drives get flooded in a pump station and put back on line after the service tech removed the PC boards and ran them through a dishwasher! But I also saw the problems associated with CSI and was eventually sold on ABB's MV PWM drives after that. Now I happen to think that the Robicon Perfect Harmony MV drive is the cat's meow, but I have been accused of having a slightly biased opinion... :wink:

My advice, for what it's worth, is to stick with the now proven forward moving trend and go with a newer VSI drive from any one of the big worldwide manufacturers who can afford to stay in the game for the long haul. Going back to old technology is usually a result of a previously failed strategy. You are going to be making a major investment here and you want it to be in service for a long time with as little trouble as possible. While I did like the robust nature of CSI drives, I can't see it being available in the long run; upcoming regulations on harmonics and efficiency will be the likely death knell in a few years. That is what I think Siemens / Robicon and all of the other CSI drive manufacturers saw coming. I might be wrong, but as far as I know A-B is the only one left using it now, and for them it is a relatively new venture, so you might be one of their beta sites.