Discussion:
Quantifying power quality, 50 mf capacitor
(too old to reply)
Chris Schaefer
2014-10-13 13:57:35 UTC
Permalink
-----Original Message-----
From: RE-wrenches [mailto:re-wrenches-bounces at lists.re-wrenches.org] On
Behalf Of Ray Walters
Sent: Sunday, October 12, 2014 16:21
To: RE-wrenches
Subject: Re: [RE-wrenches] Quantifying power quality

Hi Dan;

I, like Larry, must be amazingly lucky then, because we've solved power
quality issues for many years with 50mf caps. Usually, its been front
loading washers not running on Trace SW inverters, but also Grundfos CP
pumps, too. Never had a problem, and some of these have been running that
way for over 15 years.
I never had a power quality issue with an Exeltech, so I've never used this
trick with your inverters.

R.Ray Walters
CTO, Solarray, Inc
Nabcep Certified PV Installer,
Licensed Master Electrician
Solar Design Engineer
303 505-8760
Wrenches,
Be careful about connecting *any* capacitors across an inverter's AC
output. It's not a smart thing to do.
While a capacitor may help in some cases when a load has a substantially
poor power factor due to significant inductance, there are two serious
1. The power factor seen by the inverter will be equally bad in the
opposite polarity when the inductive load turns off, and IF the capacitor is
still connected across the lines. Larry implied the capacitor he connected
was disconnected when the load turned off.
2. Larry "shot-gunned" a solution. He guessed, and got lucky. We don't
know how much actual capacitance was needed in this specific situation.
Capacitors connected across an inverter's AC output can destabilize its
voltage control loops, leading to erratic AC voltage, resulting in possible
damage to the inverter, the load(s) .. or both.
Power factor is one of the most misunderstood aspects of electricity.
Dan
(Professional inverter design engineer .. among other things...)
--------------------------------------------
On Sun, 10/12/14, Starlight Solar Power Systems <larry at starlightsolar.com>
Hi William,
If you recall a couple years ago we had some discussion here about a
problem with a specific make appliance being powered by a Magnum inverter.
The problem, it was discovered, was the appliance input PF was creating
current demands that the transformer based inverter could not keep up with.
When I powered the load with a low cost, high frequency inverter, it worked
fine. Apparently the HF inverter can keep up with the current changes faster
than the Magnum.
To fix the problem, a 50mf run type capacitor was installed in the
appliance so that it was applied across the AC input when powered on.
Everything worked fine then because the cap was taming the wild current
created by the low PF.
Try looking at the current and voltage waveform together (AC shunt, dual
trace scope) at the UPS input. If the current is displaying large spikes
and/or is out of phase, you may have found the problem is with a non linear
load from the UPS. It may be the SI, like the Magnum, can not keep up and so
the UPS input circuits are throwing a fit.
Or, something else you can try, connect a high frequency inverter and try
to power the UPS. This can be a portable 12 volt battery and inverter. If
the problem disappears, well, you take it from there.
Larry
Hi William,

Power quality, in your case, refers to how close to 60 Hz (frequency
component), 120/240 VAC RMS (amplitude component) your voltage waveform is.
Every waveform is the superposition of different frequencies, amplitudes
etc. Power quality requirements for a devices input or output are usually
defined in total harmonic distortion (THD %) and in voltage ranges. THD is
an attempt to quantify harmonics on a waveform. It is the ratio of the sum
of all harmonics to the fundamental frequency (60Hz in your case).
A Sunny Island has THD and amplitude tolerances similar to that of the grid
(3%, 105-132VAC). I have found that most generators aren't as close and the
distortion in a waveform is easily discernible.

My approach to your problem might be to ask for the power quality
requirements for the UPS not to beep. Document that the Sunny
Islands/Generator system is delivering these requirements during beep events
using a power quality analyzer. Put problem back on UPS guy if you can
prove that your system is delivering adequate power quality. This is how
our utility deals with alleged power quality problems.

If the SI system has enough reserve, run it totally off-grid with the grid
as backup. Then they will never see any transitions.

A Sunny Island that switches to and from grid power probably does not
transfer smoothly enough to protect sensitive equipment, but if it is online
all the time, using solar to charge the batteries, I would not bother with a
UPS.

On Sun, Oct 12, 2014 at 11:27 AM, <toddcory at finestplanet.com> wrote:
perhaps i am not understanding the set-up, but why feed a ups with a ups?
doesnt the sunny island already provide ups capabilities for the loads
connected to it?

todd

On Sun, Oct 12, 2014 at 10:40 AM, Tump <tump at swnl.net> wrote:
Thats the best fix I've heard yet!
Document then DUMP IT! With the power logger you would also be able to
capture the transition from utility to SI for your home auto gusto see if he
is able to accept the power quality too.

On Oct 12, 2014, at 10:24 AM, Chris Mason wrote:

I recently worked on a Sunny Island/Sunny Boy system in a nearby island,
that powered and protected a complete Home Automation system, including the
airco system keeping it cool. I had to restart the Sunny Islands to sort out
a COMS problem, which revealed some IT issues. It turns out the Sunny Island
system had kept it all running smoothly for about four years without an
issue so no-one had realized some of the firmware was crunched. I would get
rid of the UPS altogether and that ends the problem.

Good luck.

On Sun, Oct 12, 2014 at 6:01 AM, Tump <tump at swnl.net> wrote:
I have in the past have used the Fluke power logger 1735 , It will monitor
1/3 phases voltages spikes/dips in V, A, Hz , +++ AND one is able to print
out a time line of all the "occurrences" that may trip the UPS. I rent this
from some folks in Texas quite reasonable ~ $230/month or weekly too. NO
MORE finger pointing!
I found that when the generator was charging the batteries, the UPS's
acceptable voltage window was below the limits and began beeping, due to the
poor voltage regulation of the gen under hi load/charging conditions.
Are you able to log the Hz as well as the voltages? The ups sometimes have
very close voltage windows that they work within. The SI/SB's power quality
should NOT be the issue. With this power logger one is able to see a time
line and figure out what it operating at the time of "fault/beep" utility
power or generator power. Contact me off list for rental contact info or if
you would like to see what the graphs look like. Tool is easy to use and
VERY informative.
On Oct 12, 2014, at 1:21 AM, William Miller wrote:


Friends:

I am still having a pi$$ing match with the company that installed the
expensive home entertainment system for a client over the issue of power
quality. The UPS beeps frequently and the AV system crashes now and then.
The lead installer has copped an attitude and is blaming the Sunny Island /
Sunny Boy / Generac system for the AV system problems. The UPS emails
reports in flurries that indicate a "power failure" has occurred.

I have a Dataq data acquisition system sampling at a rate that allows me to
see the individual cycles on the system output on one of three phases.
During the time that a power failure is reported to have occurred each and
every cycle looks great. I know this is not a definitive test of power
quality so I upped my game. I bought a Fluke VR1710 power monitor and I am
recording power quality from the same receptacle that feeds the UPS.

I have some log files from the Fluke Power Log software but the instructions
are not that great and I don't know how to interpret the data. Do any of
you have any experience with this software and could possibly advise me?

In a more general sense, how good is power to be called "good"? I have done
some research and the industry material is a bit overwhelming. Does anyone
have a short answer?

And lastly, even though the UPS is beeping, if it is doing what it is
designed to do, the power quality should not be an issue because the UPS
corrects for any problems. (Actually the UPS was beeping-the best move in
this situation was to mute the annunciator so at least the homeowner is not
hearing the beeper.)

I know others of you are dealing with the same problems because it came up
on this forum. I am interested in an approach that verifies the Sunny
Islands are producing adequately good power, or not, so I can either address
the problem or get the AV guy off my rear end.

The system is off grid with:
3 SI6048s in 3 phase
2 SB5000s
1SB4000
1 Generac 40kw generator
A Savant home automation system with
Furman 1500 UPS.

Yes, when the SIs are running into the Furman we have cascading UPSs, but
when the generator is powering loads it transfers straight through the SIS
and then it is not cascading UPSs.

I would hope the Furman could "play nice" with the SIs and maybe it can. If
indeed power quality from the SIs is acceptable, maybe we can program the
Furman to be more accepting, or just shut off the beeper and email alarms
and let it performed its intended function in silence.

I will evaluate if alarms are more prevalent during gen run or non-gen run.
I will also contact Furman.

Thanks in advance.

William millersolar.com
805-438-5600

Friends:

The system is off grid with:
3 SI6048s in 3 phase
2 SB5000s
1SB4000
1 Generac 40kw generator
A Savant home automation system with
Furman 1500 UPS.

Yes, when the SIs are running into the Furman we have cascading UPSs, but
when the generator is powering loads it transfers straight through the SIS
and then it is not cascading UPSs.

I would hope the Furman could "play nice" with the SIs and maybe it can. If
indeed power quality from the SIs is acceptable, maybe we can program the
Furman to be more accepting, or just shut off the beeper and email alarms
and let it performed its intended function in silence.

I will evaluate if alarms are more prevalent during gen run or non-gen run.
I will also contact Furman.

Thanks for the feedback.

William
Exeltech
2014-10-13 14:27:47 UTC
Permalink
Hello Ray,

The "capacitor across the load" trick works in some cases where mod-square-waves are involved, and where high harmonic content is present. You're effectively creating something of a "low pass filter", and taking a bit of the edge off of the square waves. A capacitor isn't needed for a true sine wave inverter because the harmonic content is very low already.

As Larry mentioned in his post, he wired the capacitor across the load in a manner that when the load was disconnected, so was the capacitor. If you're going to do it .. that's the best way, but be careful just the same. Leaving a capacitor of that size connected at all times is a bad idea. I won't go into detail as to the "why?", but suffice to say such explanation is complicated. In simple terms, if an inductive load causes a lagging power factor of X, and a capacitor of appropriate value connected in parallel with the inductive load results in an [essentially] unity power factor, then when the inductive load is disconnected, and the capacitor isn't, you end up with a leading power factor of X .. which is just as bad as the inductive load by itself. That said, a capacitor connected across a mod-square wave inverter will tend to "soften" the square edges somewhat, and may make certain loads "happier" .. but the effect on the inverter isn't
predictable, particularly where the AC output voltage control circuits are concerned.

To say you've been amazingly lucky would be appropriate.


Dan




-----Original Message-----
From: RE-wrenches [mailto:re-wrenches-bounces at lists.re-wrenches.org] On Behalf Of Ray Walters
Sent: Sunday, October 12, 2014 16:21
To: RE-wrenches
Subject: Re: [RE-wrenches] Quantifying power quality

Hi Dan;
Post by Chris Schaefer
I, like Larry, must be amazingly lucky then, because we've solved power
quality issues for many years with 50mf caps. Usually, its been front
loading washers not running on Trace SW inverters, but also Grundfos CP
pumps, too. Never had a problem, and some of these have been running that
way for over 15 years.
I never had a power quality issue with an Exeltech, so I've never used this
trick with your inverters.
R.Ray Walters
CTO, Solarray, Inc
Nabcep Certified PV Installer,
Licensed Master Electrician
Solar Design Engineer
303 505-8760
Wrenches,
Be careful about connecting *any* capacitors across an inverter's AC output.
It's not a smart thing to do.
While a capacitor may help in some cases when a load has a substantially poor
1. The power factor seen by the inverter will be equally bad in the opposite polarity
when the inductive load turns off, and IF the capacitor is still connected across the
lines. Larry implied the capacitor he connected was disconnected when the load
turned off.
2. Larry "shot-gunned" a solution. He guessed, and got lucky. We don't know how
much actual capacitance was needed in this specific situation. Capacitors connected
across an inverter's AC output can destabilize its voltage control loops, leading to
erratic AC voltage, resulting in possible damage to the inverter, the load(s) .. or both.
Power factor is one of the most misunderstood aspects of electricity.
Dan
(Professional inverter design engineer .. among other things...)
Continue reading on narkive:
Loading...