High starting torque, low running power 1

[VermontPE]

I have a process driving many screw conveyors in a mining facility. The material being conveyed is a fine powder. Generally we try to empty the conveyors at the end of a production run, but sometimed residual material is left in the conveyors. When this happens we need high torque to get them started again. Most of these conveyors have 10HP 460V 3ph induction motors on them. History has shown that a 7.5HP motor is not enough to start the screws, but when they are running 3HP is enough to keep them going. The PF of the 10HP motor is very low (.5) when running so lightly loaded. What methods have been used to solve this problem of the requirement for high starting torque but low running power? As you can tell we are oversizing the motor to avoid the problem but in the new super-energy-efficient world this option is becoming unacceptable. Could I use a part-winding motor in the opposite manner to that which it is usually applied? That is, disconnect some of the winding once the motor is started.

Tom Gilmartin, P.E.
Rutland, VT USA

 

[jraef]

That means that the slip at any given torque will increase to about 300% of the full-voltage value.

Please explain this. Do you mean to say that if you have 4% slip at full voltage, you will have 12% in Wye? How so? The number of poles has not changed, the winding resistance has not changed, the frequency has not changed. What am I missing?

"Our virtues and our failings are inseparable, like force and matter. When they separate, man is no more." Nikola Tesla
Read the Eng-Tips Site Policies at faq731-376
Member, [blue]P³[/blue]

 

[CJCPE]

jraef

Reducing the phase winding voltage reduces all of the torque values on the torque-speed curve without changing the speed values much. Between no-load and full-load, torque is pretty much proportional to slip. If the 10 Hp motor has 50 RPM slip at 30 lbs-ft, full load, it will have 15 RPM slip at 9 lbs-ft, 3 Hp load. For the Y connection, the torque for 50 RPM slip will drop to 1/3 or 10 lbs-ft. Slip for 9 lbs-ft will increase to from 15 RPM to 45 RPM. If you draw the curves, you can see what happens.

 

[jraef]

Huh. When you put it that way, I see your point. If you lower the entire torque values on the curve without changing speed, the angle of the curve from breakdown torque to synchronous is lower, which would increase the slip RPM percentage of the new curve, so in that case it would be additional slip losses. I was thinking of it in terms of what creates slip in the first place. It makes sense when I think about it in terms of another old motor trick, that of oversizing a motor in order to reduce slip speed loss. Thanks for the enlightenment. Have a Star.

"Our virtues and our failings are inseparable, like force and matter. When they separate, man is no more." Nikola Tesla
Read the Eng-Tips Site Policies at faq731-376
Member, [blue]P³[/blue]

 

[aolalde]

VermontPE:

See the following paper by Reliance Electric about Using AC PWM Drives on Non-Standard Voltages and Frequencies;

http://www.reliance.com/prodserv/standriv/appnotes/d7726.pdf

 

[aolalde]

Marke & Jraef:

According to NEMA MG1 part 12.58.1 the following losses shall be included in determining an induction SC motor efficiency:

Stator I^2*R
Rotor I^2R
Core Loss
Stray Load Loss
Friction and Windage Loss

My question is; what is the meaning of the “extra slip losses” or “slip speed loss” that you mentioned?

 

[Marke]

Hello aolalde

The slip losses are dissipated in the rotor resistance, so they are covered in the list that you included.

Best regards,

Mark Empson

http://www.lmphotonics.com

 

[CJCPE]

Reliant Energy of Houston has some information that discusses energy savings by motor voltage reduction.

http://www.reliant.com/business/1,,CID433945,00.html?2id=220037&1id=220172&3id=433947

They provide a set of curves showing motor losses vs applied voltage for various load levels. I used data from the curves to calculate the efficiencies for the minimum loss points on the curves:

Full Load, 88.5% at 430V vs. 88.3% at 460V
3/4 Load, 89.1% at 408V vs. 88.4% at 460V
1/2 Load, 89.6% at 359V vs. 86.7% at 460V
1/4 Load, 89.2% at 279V vs. 80.4% at 460V

Note that, in this case, the optimum voltage at 30% load would be more than 57.7%.

If the existing motors are delta connected with terminals available for Y connection, the Y delta idea might provide good results. Although it would introduce additional losses and harmonics, a solid state voltage reduction device might provide significant savings also.

If the motors are replaced with new motors of the highest obtainable efficiency at the actual load point, adding a Y delta or solid state voltage reduction arrangement may not add so much efficiency improvement.

 

[aolalde]

Good work CJCPE I agree. We should consider that those are typical figures and could change for different type and motor manufacturer.

Thanks for the clarification Marke. In reality the slip will be proportional to the shaft load ( very reduced for this aplication) and the magnetic strength of the motor.The induced voltage in the rotor is proportional to the slip and the rotor current proportional to the induced voltage. The resultant loss in the rotor winding (cage) is I^2*R.

Loading a 10 HP motor at reduced voltage but taking only around 3 HP load, will probably not increase the slip as compared to that of 10 HP load (nominal slip) at full voltage.

 

[electricpete]

A brief discussion to explain the behavior discussed above. (nothing new to most of the particpants here).

Consider two major losses I^2*R and core losses. As we decrease voltage (with constant load), they go in opposite directions (I^2*R increases and core losses decrease).

Which one wins? Depends on load. I^2*R is dominant increases with load while core loss doesn’t. I^2*R effect wins at high load (perhaps above nameplate in the example above) and higher voltage is more efficient. Core effect wins at low load and lower voltage is more efficient at low load. You could draw a curve of most efficient voltage vs load and program the voltage accordingly... I guess that’s sort of what the Nola devices do.

==============================

I agree delta-wye start is an innovative approach. To play the devil’s advocate you might think about the followiong questions: Will the energy savings would be worth the cost of installation/material PLUS the increase in complexity and possible reduction in reliability. In my mind increased complexity almost always means reduced reliability... if nothing else there are just more components to fail. In this particular case we might think about possible trips occuring during switching if not set up exactly right... possible degradation of motor windings due to switching surges.
We don’t have any wye-delta starters so I am not familiar with them and just speaking in generalities. Others are welcome to comment from experience.

Also in my mind there is an element of uncertainty about the design assumptions – in the past you have only experienced problems during starting. But perhaps there are other temporary power overloads during conveyer operation which you can ride thru in the existing configuration but which would cause problems in the wye configuration.

One other possibility to examine would be a special high-starting torque motor of a lower horsepower rating. Still subject to the uncertainty mentioned above regarding momentary overloads during operation.

=====================================
Eng-tips forums: The best place on the web for engineering discussions.

 

[itsmoked]

I suggest VermontPE gets a delta-wye starter and a regular power meter and runs one of his of many conveyors for 30 days. Then block the starter in delta for a month.(or alternate every week) This is chump change for a plant of his size and could give a good handle on everything discussed in these great posts, including the delta-wye feasibility.