Run 575 volt motors on 480 ?

[Besoeker]

I didn't see besoeker's post until I was done. And I don't have access to that computer generated motor model. I am supprised the efficiency went up at all - I would have expedcted it to go down.

Bes - makes me wonder if your example isn't pushing the limits of your model. I'm sure you are right - just surprised.

Hi CF.
Yes, it is a computer model but it's a model of real motors for a pumping station project. As with many projects, we had to provide guaranteed performance data at the bid stage. After installation that performance data gets tested. Getting it wrong isn't an option. Big $$$$. We haven't yet in decades of doing this. Still time, of course. I'm old but still not dead. Still time to screw it up right royally before I shuffle off my mortal coils.:grin:

OK. Slightly more seriously....
Boiled down to its basic elements, motor torque is essentially a product of stator flux and rotor current.
Reduce the supply voltage and you reduce the stator flux. And the losses associated with that. But, to get the same output torque, the rotor current has to increase. So you increase the rotor losses.
There is no one solution that fits all.

As a footnote, I have checked a couple of other motors where I have detailed calculations and the results are very similar to those I gave a couple of posts back.

 

[PowerQualityDoctor]

It is based on the standard simple motor model. This concept is known for more than 30 years.

 

[Besoeker]

It is based on the standard simple motor model. This concept is known for more than 30 years.

My calculations use the Steinmetz exact equivalent circuit which has been known for just a wee bit longerthan that......


I understand what they are trying to sell.

 

[PowerQualityDoctor]

It is the same model. The 30 years is when this idea was first invented. It is well understood concept, which can show exactly how the 575v motor would operate under 480v voltage. There are few companies using the same concept (with some differences - this company is the only one who does it without harmonics), so the concept works. How do you explain your completely different calculations?

 

[Besoeker]

There are few companies using the same concept (with some differences - this company is the only one who does it without harmonics), so the concept works. How do you explain your completely different calculations?

My calculations were for a real world motor - one for which I had to provide guaranteed performance figures at various duty points before we were awarded the contract. The guaranteed figures were written into the contract - and subsequently tested for compliance. Just to make matters a little more interesting, it was a variable speed drive application so the calculations had to carried for different voltages, different frequencies
and different loads.

By contrast, the article you linked gave claims, not calculations.

 

[PowerQualityDoctor]

When variable speed drives get into calculation, the results are completely different. This is not the discussion here!

 

[Besoeker]

When variable speed drives get into calculation, the results are completely different. This is not the discussion here!

Yes, I am well aware of that. My point was that, if you are required to produce guaranteed figures for all the variables involved, you need to have a pretty good handle on the motor characteristics and the means of making accurate calculations from that data.

See some experimental data at xxxxxxxxxxx

I had looked this up before I posted my previous comment.
The granulator case study is....interesting.
It is claimed that the current drops by 46.1% and the kVA also drops by 46.1%.
That infers no change in voltage.
So what does?

 

[PowerQualityDoctor]

The product has normal voltage on its input and reduced voltage on the output. If the measurements were done on the input (which is the correct way - include the product losses in the calculation), the change in current and total power should be exactly the same.

 

[Besoeker]

The product has normal voltage on its input and reduced voltage on the output. If the measurements were done on the input (which is the correct way - include the product losses in the calculation), the change in current and total power should be exactly the same.

So isn't the input kVA to the device the same as the output kVA from it?

 

[PowerQualityDoctor]

Every device has its efficiency, which means the input kVA is not the output kVA (assuming no change in PF). This product has 99.5% efficiency, or 0.5% losses (similar figure to soft starters, much better than inverters with 95% efficiency).

 

[Besoeker]

Every device has its efficiency, which means the input kVA is not the output kVA (assuming no change in PF). This product has 99.5% efficiency, or 0.5% losses (similar figure to soft starters, much better than inverters with 95% efficiency).

OK. I stand corrected. The output kVA is substantially the same as the input kVA.
The transformation to a lower voltage thus means that the motor current isn't reduced to the extent that the supply current is.
Looking again at the granulator case study, you have a 37kW motor running at less than 15% of its rating which is a pretty inefficient thing to anyway, and it from the data given, it doesn't seem to be a very efficient motor in the first place.

Perhaps a more effective solution would be to size the motor according to the required load.

 

[PowerQualityDoctor]

This is the typical load of recycling granulators - they are designed to be able to work during peak load. I did measurements on tens of such granulators in different sizes, both inline (part of the manufacturing line) and central, and the maximum average load was 40%. The peak load, however, reaches 100%. Usually the load was around 25% on average and the no-load loading is 10-15%.

 

[Besoeker]

This is the typical load of recycling granulators - they are designed to be able to work during peak load. I did measurements on tens of such granulators in different sizes, both inline (part of the manufacturing line) and central, and the maximum average load was 40%. The peak load, however, reaches 100%. Usually the load was around 25% on average and the no-load loading is 10-15%.

But the figures relate to just the no load condition.
Selective data to make the case?


My previous post was a little tongue in cheek. I do understand the need for motors to deal with full power and idling conditions.
As it happens we have just delivered a couple of 75kW variable frequency drives for a Sumitomo plastics injection moulding machine.
We got the order on Monday and delivered the systems in IP43 panels yesterday. Cables were delivered to site today. Installation by my guys will be over the weekend.
The drives are to power the hydraulic pumps and the duty cycle is results in off load running some of the time. Energy saving is effected by reduced speed.

There's more than one way to skin a cat.

 

[roger]

To all who have been or will be reading this thread, you can see I have removed some of PowerQualityDoctors links, after talking to him via PM it has become apparent that his intent was not advertising but to provide technical support to his posts and with that said I offer my apologies to PowerQualityDoctor.

PQD, if you'd like to repost the article be my guest.

Roger

 

[mivey]

Thanks Roger. It is harder to follow the arguments without the link or the referenced data being posted.

 

[Besoeker]

To all who have been or will be reading this thread, you can see I have removed some of PowerQualityDoctors links, after talking to him via PM it has become apparent that his intent was not advertising but to provide technical support to his posts and with that said I offer my apologies to PowerQualityDoctor.

PQD, if you'd like to repost the article be my guest.

Roger

This is one of those that I made reference to.
http://www.powersines.com/aarticles/c2787.php

 

[PowerQualityDoctor]

The links

The links

The article which explains the concept is at at http://www.energycentral.com/enduse...cle-Costs-using-Sinusoidal-Motor-Controllers/.

The case studies are at http://www.powersines.com/aarticles/c2787.php

I agree with Roger not to link to the products themselves. If you will be interested I am sure you'll find them...

Using VFD to save energy is great solution where possible. However, many applications do not allow it. In addition, it is important to check the overall performance - VFD has 5% or higher losses and create huge amount of harmonics. One of my customers installed VFD on plastic injection. He got 16% energy saving, but 1.6% increase in the overall facility voltage harmonics !!! This means that he added at least 1.6% losses facility wide !!! This is great deal - he was happy (think he save money), the supplier was happy, the utility was happy (increased revenues) and I was happy (consultancy...). My rule of thumb for VFD saving is minimum 20% speed reduction. Note that you also have bad affect on the motor life time.

But the figures relate to just the no load condition.
Selective data to make the case?

To compare performance with granulators you must have equal condition, so no load conditions are very accurate for comparison. The actual results are better although shows worse values: Saving 15% of 15% load is 2.25% of full load while saving 10% of 40% load is 4% of full load. This means more kW saving. In granulators, the results are around 10% on average operation (less accurate comparison, though). You can also see the other granulator case study.

 

[Besoeker]

Using VFD to save energy is great solution where possible. However, many applications do not allow it. In addition, it is important to check the overall performance - VFD has 5% or higher losses

I've yet to come across any that were as bad as that. Most of those I have dealt with have been 2.5% or less. In any case, the energy savings result from reducing the power taken by the load. That's something just varying the voltage can't do in any meaningful way.

and create huge amount of harmonics.

Yes, variable speed drives do result in harmonics. When we install a VSD, we always consider the harmonics. We take compliance with the Electricity Association's G5/4 as a minimum and, if mitigating measures are required they are taken.
On the larger drives, typically above 1,000 kW, we sometimes use a wound rotor machine and control the rotor rather than the stator. This gives a very efficient, low harmonics arrangement, particularly for pumps fans etc..

Note that you also have bad affect on the motor life time.

Can have. I have seen a few with problems -the result of poor installation practice.

To compare performance with granulators you must have equal condition, so no load conditions are very accurate for comparison.

Yes. For the no load condition.

 

[PowerQualityDoctor]

You are right, many modern VFDs has rated high efficiency and rated low harmonic distortion. However, it is marketing camouflage. Manufacturers measure the efficiency at different conversion frequencies and states the best values. The problem is that the maximum efficiency is usually at the lowest frequency (minimum switching losses) and the lowest harmonic pollution is at the maximum frequency. IEC plans to define a standard which force manufacturers to state at what frequency the data was measured and to provide values at different frequencies.

Another issue is the efficiency at low load. There is a new draft for IEC 60034-31 standard (Rotating electrical machines ? Part 31: Guide for the selection and application of energy-efficient motors including variable-speed applications) expected to be released soon that states: "Frequency converters generally have a high level of energy-efficiency. As with motors, their efficiency drops at partial load (see Figure 7)". The graph shows that the efficiency of 50% speed goes below 90%.

The last issue about frequency inverter efficiency is the possibility of the motor to use the energy. When measuring efficiency, the RMS of the power is measured. However, the motor uses mainly the fundamental frequency.

Since that, the recommendation is to use inverters only where it has significant benefits. See Martin Doppelbauer's lecture at last EEMODS conference:
"A variable frequency converter can be used to reduce the speed to a lower level. But a converter is a costly device and will also add further losses to the system. For these reasons, frequency converters should not be used in full-speed, full-load applications just for the purpose of constant speed
reduction alone."

To this end, in constant speed applications, such as escalators, conveyors, bucket elevators, granulators, vacuum pumps, grinders, crusher, mixers etc., the most applicable energy controller is the SinuMEC.

 

[Besoeker]

You are right, many modern VFDs has rated high efficiency and rated low harmonic distortion. However, it is marketing camouflage. Manufacturers measure the efficiency at different conversion frequencies and states the best values. The problem is that the maximum efficiency is usually at the lowest frequency (minimum switching losses) and the lowest harmonic pollution is at the maximum frequency.

Switching losses do affect losses of course. But not input harmonics except on active input inverters which have reduced magnitudes of lower order harmonics. And lower efficiency. For the majority of inverters which have B6U front ends, the switching frequency of the carrier is decoupled from the supply side.

As I have pointed out previously, we have to commit to efficiency and harmonic figures at the bid stage of a project. Fudging the numbers, as you suggest happens in your second sentence, is not an option. We don't do it.

To this end, in constant speed applications, such as escalators, conveyors, bucket elevators, granulators, vacuum pumps, grinders, crusher, mixers etc., the most applicable energy controller is the SinuMEC.

(Not sure you should have mentioned the product....but I'm not a Mod.)
Some of those applications do use variable speed drives. We have supplied a fair number for use vacuum pumps for cryogenic plants. They usually have two operating speeds - operating speed and slow roll mode.
And some fixed speed applications need a variable speed drive. We make drives and motors for machine tool applications. They are usually not more than 56kW but have to be small to get the dynamic performance required. Normal operating speed is 20,000 rpm which needs a frequency converter. Stopping has to be within ? 0.5 degrees and within 0.5s.