Using a Power Quality Analyzer to rule out "dirty power"

[11bgrunt]

A large facility builds one new building with 32 roof HVAC units. 10 ton compressors begin failing and each trade is pointing fingers at each other. To date, 13 scroll compressors have failed. The HVAC Co. believes somehow all units were run at length in reverse rotation.
Someone else says out loud that dirty power is the problem.
I have a Fluke 435 and an Amprobe DM-lll.
For several reasons the Amprobe is the go to meter. Easy to use and a good viewable display outdoors.
We agreed to do PQ test to remove "dirty power' from the list of options.
Service entrance to this new building is served by a POCO 1500kVA padmount, 480/277VAC. Six conductors per phase. This PQA does not have Flex current CTs, they are clamp types with jaws for cables up to 1.5". I clamped one conductor from each phase. All four voltage leads were connected. I would like to measure the bundle amps but not with this meter.
The PQA shows 90 amps with 1.4% ThdV, 10% ThdI. I say that since these conductors are in parallel and the run is less than 100' from POCO the service entrance, the difference in impedance for each cable will be very small. We have tested this concept in the past with conductor bundles that were too large for the Flex CTs that we owned at the time, and found the results acceptable.
I use 90 amps X 6 = approx. +or- 540 amps. The Thd is a percentage of the E or I and is fixed across the total
I report no problems based on this testing method and saw no reason to install recorders.

What are your thoughts.

 

[ron]

The IEEE 519 limits at the PCC are on page 2 and 3 of the PDF here

https://ewh.ieee.org/r3/atlanta/ias/IEEE_519.pdf

Keep in mind total demand distortion vs. total harmonic distortion when looking at the second chart since you measured THD

 

[zbang]

The HVAC Co. believes somehow all units were run at length in reverse rotation.

I'd be asking the HVAC company how that can happen. Unless there's either a mis-wired ATS or lots of mis-configured VFDs, it would be really hard for the phase rotation to occasionally change. (Maybe the rotation was wrong at one time, but if it's correct now, it's probably going to stay correct.)

Are the compressors (motors) failing or are the controls failing? Electronic controls are usually much more sensitive to many PQ issues than the motors.

How's the voltage balance? Could be that one phase is a bit "soft".

 

[tkb]

It’s funny that the other trades always blame all problems on electrical without any substantiation, and we have to prove them wrong.

I have gone down this rabbit hole more times than I would like to count, only to find it is usually a mechanical issue.

 

[ActionDave]

It’s funny that the other trades always blame all problems on electrical without any substantiation, and we have to prove them wrong.

I have gone down this rabbit hole more times than I would like to count, only to find it is usually a mechanical issue.

:thumbsup: Been there.

 

[11bgrunt]

I'd be asking the HVAC company how that can happen. Unless there's either a mis-wired ATS or lots of mis-configured VFDs, it would be really hard for the phase rotation to occasionally change. (Maybe the rotation was wrong at one time, but if it's correct now, it's probably going to stay correct.)

Are the compressors (motors) failing or are the controls failing? Electronic controls are usually much more sensitive to many PQ issues than the motors.

How's the voltage balance? Could be that one phase is a bit "soft".

The HVAC thinks EC closed main and sent power to roof unit disconnects that were closed before HVAC startup checks.
The 13 compressors have failed and 13 compressors were replaced.
Voltage balance is very good.

 

[__dan]

Did the motors fail or the compressor connected to them, ie, do the motors pass a megger test and run fine when power is applied, even if the compressor is toast. Meg the motors and power one up on a test bench, get the FLA (if it doesn't smoke or otherwise complain).

Seen two compressors fail on startup, both while attended to by refrigde techs. Both slogged liquid at the suction side and they were done. One while filling and the other because it ran without the DX coil fan on and liquid some passed through back to the compressor.

Saved two others by running up to the roof and throwing the disconnect (to off). One was put in service without loosening the vibration isolation mounts (from the shipping position, noticed the vibration). The other was going bang bang bang. Sheaves were not set and the belt set dropped into the flat, stuck there and tore in half, then whipped around beating on the nearest thing which was the DX coil of the RTU.

If the motors are fine but the compressors are done, it's mechanical. But it could be because the fans on the coil were not on properly.

If it's actual motor damage, looking at the supply is indicated.

 

[Jraef]

The HVAC thinks EC closed main and sent power to roof unit disconnects that were closed before HVAC startup checks.
The 13 compressors have failed and 13 compressors were replaced.
Voltage balance is very good.

So in my opinion, that would be the fault of whomever installed the electrical /disconnects for the Air Handkers. I would NEVER install a piece of equipment and leave the disconnects closed BEFORE a rotation check could be performed. But OF COURSE the EC would, at some point, energize the system, he can’t get paid until he can show that it’s good. So who installed the the equipment and left the disconnects closed?

 

[Cow]

So in my opinion, that would be the fault of whomever installed the electrical /disconnects for the Air Handkers. I would NEVER install a piece of equipment and leave the disconnects closed BEFORE a rotation check could be performed. But OF COURSE the EC would, at some point, energize the system, he can’t get paid until he can show that it’s good. So who installed the the equipment and left the disconnects closed?

That's the million dollar question.

 

[zbang]

So who installed the the equipment and left the disconnects closed?

Probably the HVAC company, of course :lol:.

(That, or the Disconnect Fairies came and played with all the handles.)

 

[Jraef]

Probably the HVAC company, of course :lol:.

(That, or the Disconnect Fairies came and played with all the handles.)

That was my thought (the HVAC contractor, not the fairies...). I've had a problem with HVAC contractors doing electrical work for a long time, ever since I had one blow up my 250HP VFD by connecting it backward.

He saw the word "Load" on the feeder breaker and "Load" on the VFD terminals, then "Line" on the motor connection box and "Line" on the VFD, so he connected it, in his words, "Line to Line, Load to Load. What's the problem?".
:rant::dunce:

 

[Besoeker]

First thing I'd do is to hook my digital scope to the supply and check the voltage waveform. A quick eyeball of it will give you an indication. A dirty sine wave is easy to spot.

 

[Jraef]

As _dan said earlier, if the “failure” is not in the motor itself, no point in looking at the power. They said the compressors failed. We don’t yet know what that means but if it was not the motors, then power quality has no involvement.

And feeding it with reversed power on energization is not a “power quality” problem, it’s a procedural problem.

 

[Besoeker]

As _dan said earlier, if the “failure” is not in the motor itself, no point in looking at the power. They said the compressors failed. We don’t yet know what that means but if it was not the motors, then power quality has no involvement.

And feeding it with reversed power on energization is not a “power quality” problem, it’s a procedural problem.

Yes, but if he is being asked to prove/eliminate it as an issue it is a quick and simple means of doing it.

 

[RumRunner]

Scroll compressors can be run in either direction. It is inconceivable to have these things running in reverse and subsequently destroy the motors as others had theorized.

Perhaps the HVAC guy needs a brush up on scroll compressors when he said it was running in reverse. Not only that it's impossible to check rotation of a hermetic AC motor, the two spiral scroll mechanisms don't rotate.

The movable half of the two spiral compressor element wobbles, that generate the compressing action caused by the eccentic cam that is coupled to the motor shaft.

All of these things are enveloped in one " seamless" casing and welded throughout--no serviceable part, just the refrigerant ports ( low and hi pressure) and the motor picker head.

If these are capacitor- run motors, using the wrong capacitance value would cause the motor to run hot and noisy-it may also vibrate.
.

 

[GoldDigger]

You can often identify the rotation direction of a hermetically sealed compressor by looking at the reaction torque "kick" of the whole unit when power is applied. Then you would only have to know the right direction.

Sent from my XT1585 using Tapatalk

 

[RumRunner]

You can often identify the rotation direction of a hermetically sealed compressor by looking at the reaction torque "kick" of the whole unit when power is applied. Then you would only have to know the right direction.

Sent from my XT1585 using Tapatalk

What reaction do you expect from a scroll compressor?
Most likely from reciprocating pump not scroll pump

 

[GoldDigger]

What reaction do you expect from a scroll compressor?
Most likely from reciprocating pump not scroll pump

Regardless of the nature of the pump mechanism, any motor with a rotating component will have more angular momentum while spinning that while stationary. That means that a torque must be applied to the entire pump/motor assembly to get the rotor spinning. \
The only exception would be a motor assembly with two counter rotating rotors or with a reversing gear between the motor rotor and an equally massive pump rotor.

 

[markebenson]

People can debate all day. If they think thats the problem just use a recent model oscilliscope and provide the facts.

 

[Besoeker]

People can debate all day. If they think thats the problem just use a recent model oscilliscope and provide the facts.

Yes, that was my point a few posts back. Quick and simple. Or even simple and quick.....