Dunki's sqrt(3)*rated voltage motor life rule of thumb 1

[brainsalad]

Greetings:

Can anyone familiar with the following rule of thumb,

"one hour of operation at sqrt(3) times rated voltage equates to about 700 h of operation at unity (rated) voltage" from "The Reality of High Resistance Grounding", Dunki-Jacobs, IEEE Trans. On Industrial Applications, Vol. IA-13, No. 5,1977

please offer some thoughts as to how this rule is figured? Thanks for your help.

 

[waross]

So when a motor is operated at the star voltage rather than the delta voltage the life expectancy is 1/700 or 0.14%.
Your question may be valid but please include enough context to identify the specific application.

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

 

[wareagle]

Quote waross "So when a motor is operated at the star voltage rather than the delta voltage the life expectancy is 1/700 or 0.14%."
I don't follow your math.. I would say the operation at the voltage stated would reduce the life time hours = 700/expected lifetime hrs.
I am not familiar with this rule. Reduction seems high.

 

[waross]

Possibly for high voltage motors. Definitely not 230/460/575 Volt class.
Did I slip a decimal place? Sorry. Still an unrealistic figure for most motors.

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

 

[KllrWolf]

That reduction does seem a bit extreme. According to that rule of thumb, a motor with a 10 year life expectancy would only run for 5.2 days. The operator would change things before that (so we hope) because the motor would have some odd power factor and current readings at best. We have actually tested a few motors like that to try to duplicate problems the customer was having. They ran fine at 8 hours a day for a month and did not appear to have any issues when we disassembled them.

 

[rcw retired EE]

The paragraph in the article says:

"It may be of interest to speculate on the probablity that (life) x V[sup]12[/sup] is a constant, which relates to the expectation that one hour of operation at (sqrt 3) times rated voltage equates to about 700 h of operation at unity (rated) voltage. This estimation becomes, however, invalid for low voltage."

Dunki-Jacobs was speculating on motor life being roughly proportional to the 12th power of voltage. Sqrt3[sup]12[/sup] = 729.

He did say it doesn't apply at low voltages. That portion of the article was discussing medium voltage systems and motors.

It would be interesting to know the background on the 12th power theory.

I checked the references but there were only articles by Fran Fox, Clyde Tipton & others discussing practical experience with HRG on the old 2400V delta systems at the Chevron refinery. (I started my career working for Clyde and installed some of Fran Fox's systems there.)

 

[electricpete]

Stone et al discuss similar principle to that discussed by rcwilson in their textbook "Electrical Insulation for Rotating Machines. Estimating this relationship can be important for evaluating voltage endurance tests which aim to quantify nominal-voltage service life through accelerated aging tests at elevated voltages. I believe the exponent 12 (or whatever is chosen) is based on empirical results.

If partial discharges are present, then the effect of the stress level (E in kV/mm) on the life of the insulation (L in hours) is most often represented by the inverse power model [2.5]:
L = c* E^n (2.2)
where c is a constant and n is called the power law constant. This model is based on work done by Eyring, among others [2.6]. As with thermal aging, below a threshold electrical stress, there is effectively no aging. This threshold is the discharge extinction voltage (DEV), where E is the DEV divided by the insulation thickness. Sometimes E in Equation 2.2 is replaced by (E – E0), where E0 is the threshold stress below which aging does not occur. If electric stress versus time to failure is plotted on log-log graph paper, the slope of the line according to Equation (2.2) would be n.
The power law “constant” is usually reported to range from 9 to 12 for machine insulation systems [2.7–2.9]. If one assumes n to be 10, then a two-times increase in electric stress will reduce the life by about 1000 times. Thus, the electric stress (voltage) has a very powerful influence on service life if PD is occurring.

2.10 IEC 1251, “Electrical Insulating Materials—AC Voltage Endurance Evaluation-Introduction,”
1993.
2.11 E. Persson, “Transient Effects in Applications of PWM Inverters to Induction Motors,” IEEE
Trans. IA, Sept. 1992, 1095–1102.
2.12 G. C. Stone, S. R. Campbell, and S. Tetreault, “Inverter Fed Drives: Which Motor Stators are at
Risk,” IEEE Trans. IA, Sept. 2000, 17–22.

=====================================
(2B)+(2B)' ?

 

[electricpete]

Sorry, I posted the references labeled 2.10 to 2.12 when I meant to post 2.6 to 2.9. Here are the relevant references:

2.6 H. S. Endicott, B. D. Hatch, and R. G. Sohmen, “Application of the Erying Model to Capacitor
Aging Data,” IEEE Trans. CP-12, March 1965, 34–40.
2.7 A. W. W. Cameron and M. Kurtz, “A Utility’s Functional Evaluation Tests for High Voltage Stator
Insulation,” AIEE Trans. Vol. 78, Part III, June 1959, 174–184.
2.8 Wichmann, “Accelerated Voltage Endurance Testing of Micaceous Insulation Systems for Large
Turbogenerators Under Combined Stresses,” IEEE Trans. PAS, Jan. 1977, 255–260.
2.9 D. R. Johnston et al., “Frequency Acceleration of Voltage Endurance,” IEEE Trans. EI, June
1979, 121–126.

=====================================
(2B)+(2B)' ?

 

[brainsalad]

waross: Apologies the statement is now corrected per rcwilson - thanks both. electricpete: how cool, thanks, I think that answers the question!