Power Supply Quality

It is likely that the upstream cabling is no longer adequate and undersized for current conditions. Moan like billy-o to the utility supplier and get them to sort it out.

Power factor correction will only be of value if the utility supplier charges for loads where the power factor is away from unity. The forum cannot see the tariff charges.

How much power are we talking here.....You could go with a UPS on more sensitive equipment maybe....

http://www.schneider-electric.com/en/product-category/8000-uninterruptible-power-supply-%28ups%29/

10~40kVA

Two good suggestions already from other posts.

Are line voltage drops caused in your own installation or by nearby users whose motors are too big for supply capacity?

You may have to tolerate the dips and feed sensitive control gear of your plant from UPS.

Are dips long duration or only about 20 milliseconds long? Most good computers or control gear will stand 20 ms loss by design. Usually, because the PSU is not fully loaded, the "hold-up" time is twice that. Many AC voltage regulators cannot deal with dips of a few cycles.

One "dodge" might be to double-up the power supply of the most sensitive control gear - this will add reservoir capacitance reserve to bridge dips. Using a PSU of double current rating but same "hold-up" time specification @ rated load would have same effect. I have added considerable electrolytic capacitors to output of DC PSU where the DC voltage is not critical (relay supply).

Some monitoring of supply to establish nature of dips is needed.

Using from supply line, a transformer & resistor divider to the sound input of a laptop PC & some sound recording software can be a way of recording what you are getting. Even an undervoltage relay(s) with different voltage settings, dropout delays can give you a measure of the magnitude & length of dip.

Personally, I would use a "Bitscope" 2 channel USB oscilloscope triggered "falling edge" from one channel fed by a bridge rectifier with limited hold-up time, simulating your control power supplies. It has pre-trigger buffer & 0.1 second/cm timebase.

Gut: Give us more information about what kind of load center (breaker box, MCC panel, etc.) that you have, what loads (the kinds and magnitude thereof), and how far you are from your local transformer, and how far distant (with how many houses, businesses, etc. hooked up from the nearest substation?

Elevators, James. Some owners just don't have the care of electrical building contractor they hire. All they see is connect and connect.

10-40 kVA seems a lot for elevators on a small island. Are we talking product elevators, like grain elevators, or personnel elevators?

You have 18 questions left to answer. LOL

DATAQ has data logger you can attach to laptop via USB for $29 buck, and it has four +/- 10 V channels, and other connections you could use, such as a trigger if needed.

You would need to use a Hall effect current sensor, or other, such as CT, and use a potential divider to see the correct voltage. Then you set up each channel in the appropriate engineering units, and press the "GO" button. It writes the data to a file on your laptop, and you can analyze it later with a variety of programs.

Not bad for $29. The stand alone units cost much more.

So now you are into counting question, buddy? Not bad for the treatment =) elevators here only range that much. Thanks for the dataQ.

Who in hell would get on and actually ride a $29 elevator? Get real.

Who is responsible for this job?

If you are paying someone to provide good grid quality, call them and ask them to solve that problem.

If you are the one supposed to solve it, you might need to have a better knowledge of Electrical Engineering and Power Electronics to solve it by yourself.

My guess is that you are using a digital multimeter to measure the voltage. Power quality that can disturb your equipment would last only a few milliseconds at most, and you cannot measure that with the multimeter. Therefore, your waveform must not be stable. I think you will need a power converter similar to a 3 phase UPS, but you might not need batteries or even electrolytic capacitors. Be careful not to buy something you do not need.

If you cannot find a commercial product for your application and you need a custom made converter you can give us a call. For power levels above a few hundreds of kW, the engineering cost for custom made converters can be lower than over-sizing your installation with products on the market.

Ivanov is correct in his assertion that you may not see everything required in the solution by merely using a DVM.

You need a datalogger attached to see everything that goes by, for a minimum time including the events that "trip" your equipment on low volts.

Oh, Datal loggers aren't cheap. I only have this elevator service tool logs to chew.

Gutmonarch,

You told us you get +10% to -10% variation of voltage and that you have a muti-purpose service tool that is telling you this and which, maybe, does a degree of logging.

1/ But were these dips/surges a cycle or two, maybe with a nasty spike causing malfunction of electronics or seconds due to severe supply dips or elevator loading??

2/ You wrote "failure", but did that mean someone stuck in lift until it was all reset or burnt-out parts??

3/ And what parts failed or "locked-out"??

4/ Is this a new problem on an installation that worked OK for years??

If the volt drop is due to lift load or supply, a separate feed to controls from another circuit direct from incomer might avoid most of the dips and prove a point with a temporary extension lead.

If it is an interference spike problem, it might come from worn contactors in the lift control or deteriorated ground bonding.

A 10% change in line voltage is normal. For 115V supplies, low line is 105V, and high line is 125V. That kind of change shouldn't damage anything. You may be seeing transients from motors (back EMF) and so forth that go much higher. I agree with Tornado that power factor correction is irrelevant. What has failed? Elevator electronics?, Motors? Circuit breakers?

Elevators typically (if personnel type) have a lot of stops and starts, surely he is looking up the wrong skirt here, has poor motor starter electronics, bad connections, and a host of other (worn out junk) problems.

Gutmonarch,

I hope you have seen that LOTS of information is needed if we are to give you a good answer or a couple good options. Otherwise, we are just walking blind.

The answers suggest you are working with elevator(s), but you have not said. Sometimes I have found that the problem lies in the power quality going to the controls and not the main or motor loads. A constant voltage transformer can give very good results and will filter out most transients and also correct most peaks and sags in the incoming line, but these are far from cheap. They are available in 1-phase and 3-phase designs. If you are seeing peaks or sags from a large motor starting it may be that the motor is still going to work but the controls are having problems. If the peaks or sags are independent of when the motor starts or stops then the problem is coming at you from the source (which may or may not be something you can work on).

• What are the loads on the equipment you are working with (HP/kW, duty cycle, number of starts or stops per hour, etc.)?
• What are the controls (relay, PLC, other)?
• Describe the supply circuit--gauge, length, etc., along with what the source is.
• How many and what use for the elevator(s)?
• How long has this problem been here?
• Any changes/repairs to the equipment?
• What has already been done and the results of this?
• What happens if you do nothing?
• Do you have a budget for any repairs?

JMM

Let's say the actual pf now is 0.85 and the premise power is 44 kVA at 480 V

the elevator motor 40 hp 460 V 0.85 PF 0.8 EFF. 55 A rated and the supply cable length 1000 ft.The elevator may present a start current of 6 times the rated-if no VFD is involved. So the start current could be elevated and the voltage drop also. According to NEC[Tb. 310.15(B)(16)] you need 6 awg copper- the resistance is 0.49 and reactance 0.064 ohm/1000ft. The voltage drop will be 8.6% at steady state and 22.34 % at motor start. So you have to increase the conductor cross section to 3 awg and then the voltage drop will be 4.65 % at steady state and 14.32% at start.

If you'll rise the pf to 1 the current will drop to 45 A -since only active power will stay unchanged [37.4 kw=37.4 kVA] the steady state voltage drop will be 4.05% but the voltage drop will be still 14.32% at start.

You have to replace the cable or to provide VFD for elevator.

If cable is 1/0 awg- for instance- then the voltage drop at steady state will be 2% and at at motor start 9.77%.