VFD OVERSIZING 2

[hooverdale]

Hello Everyone,

I know a there are a lot of vfd expert here. So here's my question.

We have a pumping station project for dewatering application that will be used for flood control.
Composed of two (2) units Teral Submersible Pump Model 600AAX-637 rated at 50HP, 400V, 3Phase, 60Hz. Controlled by Duplex VFD Type controller with Softstarter bypass.

However the VFD and Softstarter that was proposed by the supplier was one size higher, specifically 60HP, 380-460V, 3Phase VFD Model DANFOSS FC202.

Is it correct to oversize the VFD for this application? Or is it a marketing strategy?

 

[itsmoked]

With VFDs it's always about the current because the most expensive components in the VFD are "current rated". Yet, they label VFDs by HP. You can't lean too heavily on the HP ratings. Consider that the offered VFD is called a 60HP version yet the voltage can range from a measly 380V all the way up to 460V. To get 60HP at 380V you will certainly need more current than you'd need at 460V.

The makers "HP" their VFDs based on the highest voltage, in this case 460V. If you want to use a lesser voltage you're probably going to need a higher current capable VFD. Hence your 50HP pump using only 400V is going to need a 60HP VFD so its bigger power elements can dish out the added current needed to achieve 50HP with only 400V.



Keith Cress
kcress -

http://www.flaminsystems.com

 

[waross]

To put some numbers to Keith's information;
460V / 400V = 1.15 x Current
50 HP @ 460V x 1.15 = 57.5 HP @ 400V
Next available size; 60 HP

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[squeeky]

Dear Hooverdale, you gave some very exact figures in your question which is not often done, but you left off the enclosure type and the ambient temperature. As as been stated above, the limitation will be the current and thermal loading of your VFD. Having worked the load requirement and matched the next drive size up if there is not an exact match, you will have to consider the mounting. If it is in the open which is possible with some drives it will have maximum air flow and cooling but also maximum dust and contamination. Also in a pump station a risk of water. You may then consider installing it into an enclosure. Using IEC 61439 you will discover that the enclosure will be very big even if vented. Slightly small if forced cooled via fans and filters.This will cause a build up of heat. This may be the reason for derating the drive to a larger size.

I have seen pump stations in corrigated iron structures where the ambient gets to 55 degrees Celcius. The drive is rated for either 30 or 40 degrees. This could be another reason for the derating.

If your reticulation network is weak, there may be voltage variations which would result in a low voltage at the point of control. This again would increase the current and heat rise to a given HP. Along the same theme, if your supply transformer KVA rating is very near the total installed capacity, and / or you have long supply cables, your VFD will allow the starting but when fully loaded the transformer and cable voltdrop will cause the rise in current again.

We often select a drive to match an exsisting unit to reduce the spares requirements.

Did you ask your supplier? He may just be building in a margin of safety to ensure that the drive survives and his risk of warrenty claims and calls to sort out problems are less likely.

 

[DFordPE]

Manufacturers typically have tables showing the HP ratings at different voltages. However, current is the ultimate factor, as was said. Danfoss is a reputable company and they should be able to answer your question.

What I do find interesting is using an SSRV for a bypass. I've never seen that before.

 

[jraef]

I see it all the time in the pumping business. Being able to soft-stop the pump can help reduce or even eliminate the effects of water hammer. VFDs are perfect for that, but soft starters are a close second. So if your VFD fails and you must use the bypass device to keep the pump running while you wait for the VFD to be fixed, a soft starter can maintain that added protection of the piping system by being able to soft stop.

"Will work for (the memory of) salami"

 

[DFordPE]

Yes, I understand the theory. I just said I've never seen it. The SSRV only prevents water hammer if it has decel ramp - not all do.

 

[jraef]

True, but less so in the last 10 years. Once soft starter mfrs went digital, the ability to do Decel became software, not hardware.

"Will work for (the memory of) salami"

 

[hooverdale]

Hello Everyone,

Thank you all for your inputs. Yes we end up buying the 60HP VFD and Softstarter. What i don't understand with the supplier explanations was the reason why they oversize it because they said the application is for flood control and water can get slurry so pump will draw higher current and to compensate it VFD needs to be oversize. I don't understand it because pump was designed for such application and why would it draw much higher current?

Yes next time we would rate the VFD according to current. Lesson learned.

By the way, VFDs and SSRV will be installed in a NEMA 3R enclosure (vented and with exhaust fan).

Thanks, Hooverdale

 

[jraef]

To understand their response, you have to understand the marketing of VFDs and the terms "Variable Torque" and "Constant Torque" as it applies to how they are sold.

The primary job of a motor is to supply torque, at a specific speed, which we relate to as "power". But still, the first part of that is the torque, because that is what the LOAD requires both to accomplish the work, and to accelerate the load from a standstill. But not all loads are created equal and with some, the amount of work required by the load varies greatly depending on the speed, we call those "Variable Torque" (VT) or "Quadratic" loads, because the power to speed relationship is a quadratic equation. As speed increases, the load requirement increases by the CUBE of the speed change. This also works vice versa of course, so as speed LOWERS, the load on the motor is lowered at the cube of the speed drop. Normally, meaning using full speed motors, we don't really care, it's just an oddity. But once we introduce a speed control method for the motor, such as a VFD, it makes a HUGE difference. So if I select a motor for full speed power, but run it at 50% speed, now the LOAD on that motor is reduced to 12.5% of what the original sizing was.

How that plays into VFD sizing is that from a marketing and pricing standpoint, VFDs are sized based upon the ability for the internal components to handle the maximum load, PLUS the ability to handle any temporary OVERLOAD conditions, including the extra torque that may be necessary to overcome inertia and accelerate it from a stop. But in that VT machine, like a centrifugal pump, the load inertia is very very low, plus the need for being able to handle an overload condition is usually very very slight, knowing that the pump will be sized for MAXIMUM flow at full speed. So there is no NEED for added overload capability of the VFD, it should never need it. Therefore the VFD industry began, years ago, to offer "down-sized" versions of their drives that are expected to ONLY be used on VT machines like centrifugal pumps and fans. The difference is, they CANNOT handle any overload conditions, but should never have to. So what you might call a 50HP VFD in all other applications can actually be used as a 60HP VFD if the load it strictly "Variable Torque".

Now enter your viscosity issue. The reality of the "never going to overload" issue is that, it only applies when all things are equal to the lab test conditions. When viscosity increases, meaning the specific gravity of the fluid increases, the "cube law" no longer applies as equally as it does with clean water. So that is the likely basis of there concern. The "60HP" rating of the drive is likely a VT rating, but they are concerned that with a higher specific gravity in the fluid, the actual power requirement curve of the motor may end up looking flatter than a simplistic centrifugal machine.

All that said however, your point about the pump motor having ALREADY been selected based on it being a slurry is likely valid as well. My guess is that they just do not want to "ASSume" that and the difference in price is not worth the risk of failure.


"Will work for (the memory of) salami"

 

[djs]

I once did a project where we had transfer/re-circulating pumps on some caustic soda tanks. During the plant start-up the tanks were filled, but the re-circulating pumps where not run on a schedule. When the pumps were started (after a month or so) all of them would trip out due to motor overload. The caustic soda in the solution had settled and was more viscous at the bottom which caused the motor overload.

As an aside I have found that many pump suppliers will under size the electric motors supplied with their pumps. This means that the unit will operate at the normal operating point, but will be overloaded if operated off of that point. Kind of like having a ticking time bomb you don't know about.

 

[Compositepro]

This is a good discussion but I must correct a few points on the fluids side. Viscosity and density are two completely separate fluid properties. It is the density that has the most important effect on a centrifugal pump. Doubling the fluid density will double the power used. Viscosity has a relatively minor effect within the viscosity range that a centrifugal pump is suited for. Take as an analogy powering a car. Density has the effect of a more massive car, whereas viscosity has the effect of greater rolling resistance due to low tire pressure. However, a pump is constantly accelerating fluid, unlike a car which accelerates and then runs at constant speed.
Also, it is probably a myth or misconception that turning-off a centrifugal pump will cause water-hammer. Water will still flow through a undriven centrifugal pump. It is just like taking your foot off the accelerator of your car.

 

[jraef]

Thanks for the primmer on the fluid terminology issues, I didn't do well in fluid dynamics, hence my becoming an EE...

As to the water hammer issue though, the point is not just that there is a pump, but that there is a pump and a check valve. It is the slamming of the check valve that causes water hammer because in closing the valve rapidly; the kinetic energy of the moving fluid is trapped, can't compress and becomes a shock wave. That's what Pump Control Valves do, they leave the pump at full output and slowly choke off flow, then shut down the pump. But their main function is usually flow control and when the VFD is used for flow control instead, you eliminate the PCV. By soft-stopping the pump with the VFD, you lower the kinetic energy remaining in the fluid movement and allow the check valve to close in a near neutral flow environment. No kinetic energy left to be trapped in the pipe, no water hammer. Then as I said, a soft starter can closely emulate that function if the VFD is not working.

"Will work for (the memory of) salami"

 

[DFordPE]

The affinity laws (cube/variable torque) are a nice approximation of load on fans and pumps, but they only work in a perfect world, not what we have to deal with. However, if they did always work a 10% speed increase would only yield 10% more flow but would require 33% more current. Assuming that the motor is fully loaded, as is the case with many pumps and fans, the motor won't tolerate it so the VFD doesn't need to provide it.

Constant torque loads, such as conveyors, often see instantaneous overloads. That is the primary difference between the two types of drives, the ability to handle those situations.

We must also understand that most drives and now labelled a "normal" and "heavy" duty. Normal duty being variable torque and heavy duty being constant torque.

 

[hooverdale]

Hello Everyone,

The difference in price is not worth the risk of failure.

This was exactly what my boss told me. Its convincing.

The difference is, they CANNOT handle any overload conditions, but should never have to.

I also pointed this out to the VFD supplier that why allow the motor to run in an overload condition for a long period of time. Why not oversize the motor instead? Unfortunately, i got no valid answer from them.

As an aside I have found that many pump suppliers will under size the electric motors supplied with their pumps. This means that the unit will operate at the normal operating point, but will be overloaded if operated off of that point. Kind of like having a ticking time bomb you don't know about.

This usually happens when designers specify horsepower rating of the pump and to comply with it pumps supplier push pump selection to the limits.

With regards to water hammering i'm also interested in this topic because we had problems with it recently. I'll create another post with regards to it.

Thanks everyone,

Hooverdale

 

[jraef]

I also pointed this out to the VFD supplier that why allow the motor to run in an overload condition for a long period of time. Why not oversize the motor instead? Unfortunately, i got no valid answer from them.

If the VFD supplier is not the motor supplier, that would explain their inability to answer that question. Conditions outside of their ability to influence.

"Will work for (the memory of) salami"

 

[squeeky]

We pump water from rivers and dams. The mechnical guys remove the non-return valves so the when the pump turns off, the reverse flow "blows" away the mud building up at the suction pipe. This can cause the motor to run at super synchronous speeds in reverse. A high speed hum is heard. If the operator wants a quite night he stops the pump and then starts it again once is is "singing". It trips the breaker and now he can sleep. Replicating this there is also a laarge bang which I assume is coming from the coupling and bearings. I wish we had the money to install Soft Starters on the 280kW and 315kW motors.

I look forward to Hooverdales new posting.

 

[DFordPE]

It's common to see compressors operate into the service factor of the motor for short periods of time. I have also seen motors relabeled to 15% higher HP with a 1.0 SF.

There will always be unusual cases or poor operating practices that require special care.

 

[rotw]

quote
So if I select a motor for full speed power, but run it at 50% speed, now the LOAD on that motor is reduced to 12.5% of what the original sizing was.
unquote

I think this is a misconception. Let me give the example of a centrifugal compressor. When a centrifugal compressor is started with a fixed speed motor, the load vs. speed is indeed quadratic (ignoring the breakaway torque effect)- in this case yes, the statement was correct in sense that it takes few seconds to accelerate to the rated speed and relationship load vs. speed is generally based on a certain assumption/simplification (like affinity laws which has been mentioned).

If you operate on variable speed mode, the operating map of compressor needs some attention.

The iso-power lines vs. the performance curves at different speeds are lines nearly parallel and their slope is depending on several things where behavior of compressed gas plays a role. You could encounter a situation where the iso-power lines are very steep so same iso-power line crosses both the rated speed and the minimum operating speed at a quite low value (say 60 or 70%). So motor has to cope with a situation of high power at both low speed and high speed and I don't have to explain which is worse case for the torque.

Often what you specify is that train will operate on specific (discrete) operating points, these points are then plotted on a graph used by vendor to size the VSD and motor, but you might have de facto limited the potential operating envelop of the machine where operating points of concerns do not cover worse case.


"If you want to acquire a knowledge or skill, read a book and practice the skill".