Mysterious vibration in long conduit between solar inverters and main service panel

Status
Not open for further replies.
180612-1702 EDT

Redwood Infrared:

Get a good scope, amplifier, and well shielded (magnetically and electrostatically) microphone, plot the sound waveform, and measure its fundamental frequency and harmonics. Expect a relation to 60 Hz line. When synced to AC line does the sound waveform drift in time? Are there other frequencies unrelated and not synchronized with the power line.

Also look at line current.

Sound level increasing with load current (generated current) means the sound originates from magnetic field forces. Is the sound level about the same anywhere along the conduit?

Is the frequency much higher than 60*3*2 = 360 Hz?

.
 
Redwood Infrared said:
ATSman, forgive me but I keep coming back to this and I'm having trouble parsing your second situation. I'm not sure what you mean by mismatch when you say the 120v hot conductors were connected to the ground conductors - that sounds like a straight up short circuit to me? Can you please clarify?
Click to expand...

Yeah, my age is showing :weeping: What I meant to say was the NEUTRAL and GROUND conductors (120v hot, neutral, ground 3-wire) were crossed at the plug/socket connection causing the neutral current to return on the ground paths back to the supply subpanel. These net currents generated the EMI like a transmitter antenna that cased the screen jitter. Not sure if this is possible with the newer type LCD/ LED monitors. Can anyone comment on that issue?
 
180612-2141 EDT

To make acoustic measurements you may need a non-magnetic, non-conductive, acoustic transducer (microphone) coupled thru a plastic tube to a remote electrical microphone to avoid errors from high magnetic and electric fields near the noisy cable.

Note: a hairpin loop of wire with the same magnitude and waveform of current flowing in both conductors, but in opposite directions has a high opposing force between the two wires. Without mechanical restraint this will tend to open toward a circular shape.

With an AC current this force will vary from zero to a maximum twice per cycle. Thus, the potential for vibration is at double the AC frequency. Three wires with three phase power and the vibration fundamental should be 6 times the fundamental.

.
 
gar said:
180612-2141 EDT



With an AC current this force will vary from zero to a maximum twice per cycle. Thus, the potential for vibration is at double the AC frequency.

.
Click to expand...

Not to confuse OP.

Never found anything you have ever said to disagree with, but perhaps should rethink that sentence.
Apply 100Hz ac to loudspeaker, you hear 100 Hz, not 200 Hz:
Or put another way - apply a dc bias and you hear the same, not shifted. etc. Note: full wave Rectified yes = 2X ac frequency as that is the 'new' ac frequency content.

Tripletts all in phase, which is why 180 Hz and 540 Hz mentioned earlier, as triplets do not cancel in a 3phase bundle, and high mag forces can result. (note: a common cause of common mode inductors due to core heating due to saturation or delta in delta-wye if high harmonics)
 
ATSman said:
Yeah, my age is showing :weeping: What I meant to say was the NEUTRAL and GROUND conductors (120v hot, neutral, ground 3-wire) were crossed at the plug/socket connection causing the neutral current to return on the ground paths back to the supply subpanel. These net currents generated the EMI like a transmitter antenna that cased the screen jitter. Not sure if this is possible with the newer type LCD/ LED monitors. Can anyone comment on that issue?
Click to expand...

Oh well that makes a lot more sense, thanks!

Wouldn't that cause a similar result to if you just bonded the neutral to the ground in a subpanel?

For what it's worth, I deployed a cheap Gauss meter phone app yesterday and definitely picked up fields bouncing back and forth between the conduits. I don't know how much stock to put in the numerical values, peaking around 8 Gauss, but qualitatively they were 6 to 8x between conduits compared to what I measured anywhere else around their circumference and about 10 to 20x background levels. Nowhere did it affect the LCD on my phone, and nowhere in the building have occupants reported problems with computers or screens, only the annoying audible impacts.
 
junkhound said:
Not to confuse OP.

Never found anything you have ever said to disagree with, but perhaps should rethink that sentence.
Apply 100Hz ac to loudspeaker, you hear 100 Hz, not 200 Hz:
Or put another way - apply a dc bias and you hear the same, not shifted. etc. Note: full wave Rectified yes = 2X ac frequency as that is the 'new' ac frequency content.

Tripletts all in phase, which is why 180 Hz and 540 Hz mentioned earlier, as triplets do not cancel in a 3phase bundle, and high mag forces can result. (note: a common cause of common mode inductors due to core heating due to saturation or delta in delta-wye if high harmonics)
Click to expand...

I did try what I think is a slightly better audio spectrum analyzer phone app yesterday - this one seemed to filter out background noise a bit better but might need some slight calibration. I got pretty well defined peaks at about 124 and 248 Hz, which I would imagine if I had good calibration would actually be exactly 120 and 240. These were very comparable values to what I was able to measure near the utility pad transformer outside the building.
 
Reporting out results from 3rd site visit

Reporting out results from 3rd site visit

I spent another few hours at the site yesterday investigating the mysterious conduit hum. This time we had the installing electrician there - and he does in fact appear very highly motivated to get the bottom of it and wants it fixed as much as anybody. We also consulted over the phone with the utility rep who oversaw the power quality monitoring effort at the inverter combiner main, ACSD, and buildng MSP several months ago.

I am still compiling my notes for reporting out to the client but briefly just to address the main questions that have been put to me by this crowd:


  • The parallel 4” EMT has a total of 3 EMT 90° bends plus one pull box for a total of 4 corners.
  • The EMT runs at slightly varying distances apart from one bend to the next, from a couple inches to about a foot apart.
  • Conduit is clamped tightly to (hung from) horizontal uni-strut that is bolted directly to horizontal building frame members along about 80% of its length. For some reason on the remaining 20% the used much less rigid 2-hole hangers bolted directly to the building frame. This less rigid section is at the end toward the “load side” or farthest from the inverters. I’m not certain yet whether this corresponds to where it gets the loudest or not – still haven’t had good full sun weather so my own personal observations of the phenomena have been limited.
  • Because the building frame member spans are 6 feet in most places, the average span between conduit clamps is 12 feet.
  • Downloaded Advanced Spectrum Analyzer Pro to my phone and found the hum appears to have resonant peaks very near 120 and 240 Hz. They are probably exactly those values but I imagine this tool is not very finely calibrated.
  • Not sure how to approach the math but is it likely there's a relationship between the resonant frequency and the distance between clamps?
  • It appears to be resonating most loudly along the longest run that goes right through the middle of the building, which is also where they are farthest apart. Need to spend a bit more time doing a methodical audio survey along its length when conditions are better.
  • EMF readings from a phone app Gauss meter were positive. I did not have time to plot out the values linearly along the conduit but will return to do so if it’s sunny this afternoon.
  • Can’t recall if I mentioned it before, but if you physically grab the conductors in the ACSD, there is a very obvious vibration that can be felt in all three phases, with a notably greater magnitude in the B phase. The vibration is there even when current is low enough there’s not much of an audible hum. Without vib analysis it’s not feasible to quantify. Hmmm…phones have accelerometers in them now, is there an app for that? YEP. Will install and try that out next vist too.
  • We measured current along all 3 phases simultaneously with a single CT and found there is definitely a non-zero current there, ranging from 1.5 to 3 amps as solar output varied between about 60 and 90 kW (about 75% capacity), or up to 240A per phase. At the same time we saw up to 10A total on the parallel grounds, so – if I understand correctly – when you factor in RMS that sort of balances out even if it’s an undesirable way to get those electrons home. This positive result would seem to be pretty significant although it’s still unclear what’s causing it. We don’t have monitoring equipment available to trend this over a few days but I’m trying to track something down to accomplish that.
  • The utility rep did not think this result was unusual, that 1 to 1.5% current unbalance is common in large solar arrays. Still seems significant to me though.
  • We have not, nor as far as I know has the utility, evaluated much on the utility line side. I’m still very interested in the integrity of the ground and neutral bonds in the pad transformer. I did look in the MSP with FLIR and saw nothing unusual…perhaps other than the fact that the main ground pin is not where we thought it would be, and in fact we couldn’t positively find it. There is a small section of exposed copper we think might be it but somebody enclosed it in drywall. Turns out the one I thought was it was just connecting the outdoor backup genset to the building frame, interrupted along its path by a connection to the sprinkler system – which itself had the electrician scratching his head because he came up through trades learning to never do that since the sprinkler system is likely floating and you don’t necessarily want to ground it like you would a cold water inlet or gas pipe. It also turned out that particular conductor, while firmly bonded to the sprinkler pipe, was very poorly bonded to the building frame. Really don’t see how that affects the solar system but we’ll remediate and see if it makes any difference.
  • The electrician will return as soon as possible to address the compromised components we identified with FLIR, starting with hypress lugs etc at the inverter combiner box main switch.
  • OK, maybe this wasn’t so brief but there are a ton of variables here and I’m trying to get my arms around all of them and eliminate them one by one until we’re down to something we can point to as the source of our issue.
 
Last edited:
180613-1130 EDT

junkhound:

A speaker is a DC magnetic filed with an interacting AC magnetic field. Thus, the fundamental force frequency is the fundamental of the AC field.

Two AC magnetic fields that are synchronous to each other double the fundamental frequency of the force between them compared to the fundamental of an individual field. +/- or -/+ fields attract, while +/+ or -/- push apart.

.
 
180613-1154 EDT

In a balanced three phase system when one conductor is at its peak current, the other two conductors have a combined current that is equal in magnitude and opposite in phase. Thus, balance and net zero measured current with a clamp-on around all three conductors. But there are internal forces between the conductors.

.
 
You sure that roof mounted AC units aren't causing the pipe to vibrate, since it sounds like it's hung from the underside of the roof. You say it's worse in the afternoons when the AC's would be running more.


I'm kind of a look at everything guy.
 
cowboyjwc said:
You sure that roof mounted AC units aren't causing the pipe to vibrate, since it sounds like it's hung from the underside of the roof. You say it's worse in the afternoons when the AC's would be running more.


I'm kind of a look at everything guy.
Click to expand...

That's a great question but we're in a unique temperate coastal climate. 70°F is considered a hot day, average year round temps in the 50's. There are heat pumps up there but AC loading is very minimal here. Aside from that, the audible effects are highly transient and there is an extremely strong correlation to when the PV system is rapidly ramping up as clouds clear away, then it will rapidly fall off as clouds move back in, all of which tracks 100% with observable inverter data, current readings, etc.
 
Redwood Infrared said:
That's a great question but we're in a unique temperate coastal climate. 70°F is considered a hot day, average year round temps in the 50's. There are heat pumps up there but AC loading is very minimal here. Aside from that, the audible effects are highly transient and there is an extremely strong correlation to when the PV system is rapidly ramping up as clouds clear away, then it will rapidly fall off as clouds move back in, all of which tracks 100% with observable inverter data, current readings, etc.
Click to expand...

Have you tried opening the PV AC disco or interconnection breaker when the noise is occurring to see if it stops immediately?
 
ggunn said:
Have you tried opening the PV AC disco or interconnection breaker when the noise is occurring to see if it stops immediately?
Click to expand...

Yes, and before they brought me in they actually did a test where they turned off the inverters 1 by 1, then brought them back online in reverse order. Regardless of whether it's a function of insolation or # of inverters, there is a strong correlation between the sound level and net current.

I feel a bit dumb for realizing each day there's yet another app for the smartphone I hadn't thought of but on my next visit I'll deploy my new Android vibration analyzer and try the sequential inverter shutdown test again to verify whether there's a correlation between current, vibration intensity, and hum volume. I imagine I will also see a relationship between mechanical vibration and audio frequencies as well.
 
Redwood Infrared said:
That's a great question but we're in a unique temperate coastal climate. 70°F is considered a hot day, average year round temps in the 50's. There are heat pumps up there but AC loading is very minimal here. Aside from that, the audible effects are highly transient and there is an extremely strong correlation to when the PV system is rapidly ramping up as clouds clear away, then it will rapidly fall off as clouds move back in, all of which tracks 100% with observable inverter data, current readings, etc.
Click to expand...
Gotta throw it out there sometime, I once walked up on two electricians an and engineer trying to solve a problem on why a motor wasn't working, I asked if they checked to see if a mouse had gotten in and chewed on the windings and shorted it out, they all laughed and kept checking stuff, I asked again and they told me that if I wasn't going to be serious I could leave. I told them I was serious and they asked why? I said because I can see the mouse and the arc mark it left on the side of the motor. Sometimes the simplest answer is the one. :happyyes:
 
cowboyjwc said:
Gotta throw it out there sometime, I once walked up on two electricians an and engineer trying to solve a problem on why a motor wasn't working, I asked if they checked to see if a mouse had gotten in and chewed on the windings and shorted it out, they all laughed and kept checking stuff, I asked again and they told me that if I wasn't going to be serious I could leave. I told them I was serious and they asked why? I said because I can see the mouse and the arc mark it left on the side of the motor. Sometimes the simplest answer is the one. :happyyes:
Click to expand...

One thing that's true of infrared thermography that I've found is really a pretty good basis for a personal philosophy is that many subjects emit frequencies we can't see with the naked eye. It is helpful to shift one's perspective by whatever means available so that the broader spectrum can be made observable.
 
180613-2112 EDT

Redwood Infrared:

When working with electronic transducers you need to make sure that unwanted signals are not what you are observing.

You mentioned seeing 120 Hz in your cellphone acoustic measurement. This I can expect from 60 Hz single phase. I don't know how likely it would be to see this as dominate from three phase wiring.

High electric and magnetic fields might couple directly into your cellphone transducer and obscure any acoustic signal. This is why I suggested you might want a non-electronic transducer right at the cable and sound pressure coupled via a plastic tube to a more remote electronic transducer. I have not played with such a method.

Also note that acoustic tubes can create resonances. A tight wire might be a useful sound transmitter. Mechanics use screwdrivers to couple engine or other noises to their ear.

If you have a sound level change as you move along a conduit, then you might have a mechanical resonance of a length of conduit. I don't know what to expect, but changing mass on the conduit can change the resonance point, or added damping material can change the amplitude. Both mass and damping material will change resonance. Changing the resonance point also changes the amplitude .

.
 
My guess, some sort of resonance created with the setup between the wire AC current and the metal (magnetic) EMT.
 
Redwood Infrared said:
...did a test ....QUOTE]

Think we all are breathlessly awaiting anyof that test data to be shown. Sure would be helpful to share an audio waveform, voltage and current waveforms.

Need to have something factual on which to base conjectures vs. just that it drives folks out of the office.

All we have so far factual is that something makes noise and it is related to the level of sunshine and the inverter model number.

Would be great if you could get a release to share the THD, voltage, and current measurements taken by the poco.
Click to expand...
 
Last edited:
Great discussion. I would like to know how this all plays out in the end.
One thing I have learned thru the years (48) in T-shooting problems is that my mind tends to think of the most complicated reasons for the cause of the failure. When all is said and done it usually turns out to be something simple. I would like to know if this is a normal reaction and have others felt the same way.
But I will not rule out the times when I have pulled my hair out on some weird problems....ha! :D
 
ATSman said:
Great discussion. I would like to know how this all plays out in the end.
One thing I have learned thru the years (48) in T-shooting problems is that my mind tends to think of the most complicated reasons for the cause of the failure. When all is said and done it usually turns out to be something simple. I would like to know if this is a normal reaction and have others felt the same way.
But I will not rule out the times when I have pulled my hair out on some weird problems....ha! :D
Click to expand...
Yes, minor things like the shipping hardware not loosened or removed on AC Units at installation.
 
Status
Not open for further replies.
Top