is noise an issue with SunnyBoy inverters ?

[RShackleford]

Considering where to install the inverter (SunnyBoy 3.8-US-41) in my system. I'm currently planning to put it on side of house, just outside the guest bedroom.

So I'm concerned about noise/hum from the inverter coupling into that room; how bad is it ? And does it only occur when there's sun on the panels and it's outputting AC power, or any time it's connected to the grid ? If it's only when sunny, probably not a concern (since nobody is likely to be in there sleeping).

Otherwise, could flip off main panel breaker for inverter connection, at night when there's a guest. Bit of a PITA though. I suppose one could imagine a relay driven by a timer, that disconnects the inverter every night. (I did something similar for my water heater, worked out great). Not sure what the long-term effect would be of disconnecting the inverter every night though.

[J.P.M.]

Google: "Analysis of noise Emissions of Solar Inverters" , a M.S. thesis by Joni Malen, and what turns into a pretty decent primer on the subject.

Note table 1, section 5, P. 27 of the thesis "Devices to be investigated and fig. 34 on P.51. An SMA 6000 is among the i nverters tested.

I have an old $50 Radio Shack sound pressure level meter that seems to have useful accuracy compared to a $15K professional SPL meter from Bruel & Kjaer I had occasion to use 1X/awhile when before I retired.

I've got a Power One inverter in my garage. Using my Radio Shack SPL, that inverter it seems to have a noise spectrum response and accuracy similar to the one the thesis shows for the Power One which would lead me to think the SMA values for the thesis might be representative of common values found in the field.

FWIW, the Power One in my garage increases the ambient noise environment in my garage by maybe about 10 DBA or so at 1 m distance, from maybe 20-30 Dba to maybe 30-40 DBA. It's hardly noticeable, if at all.

If you already know this, forget I mentioned it, but if not, keep in mind through all this that apparent sound levels are not linear. A sound needs to be 10 times as intense to be perceived as twice as loud.

Take what you want of the above, Scrap the rest.

[RShackleford]

Wow, thanks, pretty comprehensive paper.

Encouraged that "The cooling fan is responsible for the majority of noise emission below 10 kHz, whereas the chokes are the major sources of noise above 10 kHz. For the majority of devices, noise above 10 kHz represents a significant portion of total noise ...". Fan noise seems fairly benign, somewhat like a white-noise sleep-aid machine. Choke noise is probably awful like an old CRT television; but inaudible to a male in his 60s, much less so to a younger and/or female individual. And of course, such high frequency noises tend not to carry as far, especially through structural materials (although there is double-glazed a window a few feet away).

Haven't read the paper in detail yet but I don't think they explicitly discuss it; however I take away that noise is probably negligible when no power is being generated. "High-frequency noise can be explained by increased currents in the choke assembly causing increasing magnetostriction." People will be spending very little time in that room when the sun is out.

I sure hate to complete the installation and find the noise level is unacceptable. Switching off the breaker at night, on the infrequent occasions when someone sleeps in there, is a fall-back, but I sure hope it's not necessary.

[J.P.M.]

Wow, thanks, pretty comprehensive paper.

Encouraged that "The cooling fan is responsible for the majority of noise emission below 10 kHz, whereas the chokes are the major sources of noise above 10 kHz. For the majority of devices, noise above 10 kHz represents a significant portion of total noise ...". Fan noise seems fairly benign, somewhat like a white-noise sleep-aid machine. Choke noise is probably awful like an old CRT television; but inaudible to a male in his 60s, much less so to a younger and/or female individual. And of course, such high frequency noises tend not to carry as far, especially through structural materials (although there is double-glazed a window a few feet away).

Haven't read the paper in detail yet but I don't think they explicitly discuss it; however I take away that noise is probably negligible when no power is being generated. "High-frequency noise can be explained by increased currents in the choke assembly causing increasing magnetostriction." People will be spending very little time in that room when the sun is out.

I sure hate to complete the installation and find the noise level is unacceptable. Switching off the breaker at night, on the infrequent occasions when someone sleeps in there, is a fall-back, but I sure hope it's not necessary.

FWIW, you're welcome.
I'd snoop around the neighborhood or PVoutput.org for someone nearby with an SMA inverter and ask if you could stop by and hear for yourself.

The ones in my HOA seem to run about as noisy as my Power One, that is, not very to not noticeable.

NOMB, but why is this an issue now, that is, so far into the design ?

[Mike90250]

i think the "sounding board" effect will be the predominant noise path. The outer wall (inverter mounted) shares the same building studs as the inner wall sheetrock and any 120Hz hum/buzz will be coupled right into the room. Daytime only. Although my old Xantrex 3kw inverter in the garage was pretty quiet as I recall. I had my bedroom over the garage and never noticed anything

[RShackleford]

NOMB, but why is this an issue now, that is, so far into the design ?

I've been lurking and asking around here for many months, and probably started construction in earnest about two months ago. But I haven't exactly been driven - more like a hobby that I work on a couple hours from time to time.

And I never felt that I had a fully fleshed-out design. Home projects, I tend to do design by iteration, if you will; and it has almost always worked out very well for me, in fact, significantly better than if I'd tried to do a complete design from the get-go. This project may be the exception, but honestly, I'm not sure the noise issue would've come up anyhow; virtually all my learning has been from reading and posting on this forum, and I don't recall ever hearing the noise issue mentioned.

[RShackleford]

i think the "sounding board" effect will be the predominant noise path. The outer wall (inverter mounted) shares the same building studs as the inner wall sheetrock and any 120Hz hum/buzz will be coupled right into the room. Daytime only. Although my old Xantrex 3kw inverter in the garage was pretty quiet as I recall. I had my bedroom over the garage and never noticed anything

I wonder about 60Hz or harmonics thereof; the big paper mentioned above seem to say that fan noise and choke "whine" were the main culprits. I don't think frequencies above 10KHz would couple through stud walls very well. And to the extent it's daytime-only, I'm pretty re-assured.

There is a local installer I can ask. In fact he just did a 22kW job for a neighbor (55 panels !) using 3 big SunnyBoy inverters, but on the side of a big barn. I guess I can go over there on a sunny day and have a listen.

[inetdog]

I wonder about 60Hz or harmonics thereof; the big paper mentioned above seem to say that fan noise and choke "whine" were the main culprits. I don't think frequencies above 10KHz would couple through stud walls very well. And to the extent it's daytime-only, I'm pretty re-assured.

There is a local installer I can ask. In fact he just did a 22kW job for a neighbor (55 panels !) using 3 big SunnyBoy inverters, but on the side of a big barn. I guess I can go over there on a sunny day and have a listen.

Has anybody tried using some sort of vibration blocking mounting to avoid the sounding board effect?
I would also imagine that a double studded wall where the outside wall connects to one set of studs and the inside wall surface connects to the other (used for better thermal insulation) would also help with sound transmission.

[RShackleford]

Has anybody tried using some sort of vibration blocking mounting to avoid the sounding board effect?
I would also imagine that a double studded wall where the outside wall connects to one set of studs and the inside wall surface connects to the other (used for better thermal insulation) would also help with sound transmission.

If that high-frequency whine from the choke is the main issue, I don't think it'll go through even non-soundproofed walls; the windows are a little more worrisome.

If hum were an issue, yeah a staggered-stud wall would probably help a lot; ain't gonna re-frame the wall though . Soundproofing between inverter and wall could probably help a lot, but there'd still be connectors making a "link" for the sound, absent some more elaborate attachment measures. But from that paper, hum doesn't seem like a big issue; perhaps because of no conventional transformer (in most modern string inverters) ? The word "hum" doesn't even appear in the paper, but presumably it's translated from Finnish; and actually, their graphs don't go below 100Hz, need to look into that more.

Here's the paper: https://core.ac.uk/download/pdf/80710277.pdf It's well done, and lots of background info on acoustic theory.

[solarix]

The hum from SunnyBoy inverters is quite low but does vary from unit to unit. Never been a problem for any of our customers. The smaller, (<6kw?) sizes do not have cooling fans.

[sabersix]

I have a SMA SB5000TL-US-22 mounted inside my garage. There is no appreciable noise at any time except when it powers up in the morning at which point there are a series of metallic clicks. I'd say three rapid clicks over a 5 minute interval. Those clicks may be audible, but I don't think you'll notice any hum or harmonics from the unit when operating.

[J.P.M.]

Has anybody tried using some sort of vibration blocking mounting to avoid the sounding board effect?
I would also imagine that a double studded wall where the outside wall connects to one set of studs and the inside wall surface connects to the other (used for better thermal insulation) would also help with sound transmission.

Usually to secure equipment using fixations such as bolting an inverter to a wall, any spacer material gets compressed and winds up losing a lot of or most of any dampening advantage gained compared to the spacer material's unstressed/uncompressed state.

Aside from the important considerations about joint strength and design, it's possible, at least in theory, to "tune" an elastic spacer by varying compressive stress between parts to be connected so the vibration transmission is 180 deg. out of phase with the 60 hz. component of the inverter vibration, but as a practical matter, it won't work.

Stuff with (numerically) low moduli of elasticity usually have low vibration transmission when unstressed because the vibrational energy gets dissapated (damped) more easily "inside" the material. When compressed, it's harder to damp the spacer material and so, more of the vibrational energy gets transmitted. A bolted joint that has a spacer that in effect also acts in a similar fashion to a gasket may not do as much as thought to reduce vibration transmission, particularly for connections that require compression between the parts connected or a secure joint.

I suppose it's possible to mount the inverter on the ground on (vertical) springs. But then seismic considerations may be of some concern if you pull a porky inspector.

I once suspended a phonographic turntable using rubber straps to reduce vibration transmission from the room structure to the platter. Seemed to do the trick to kill feedback loops to the stylus at high SPL's.

A double studded wall as you describe would also help reduce transmission of vibrational energy for reasons that are pretty analogous to why such walls also reduce transmission of thermal energy - a more tortuous path for the energy to get through.

All in all, and as Solarix points out and as my limited experience tells me, while this may be an interesting subject from an academic standpoint, the magnitude of inverter noise is usually very low. And, if still of concern, a trip to a neighbor's or any nearby home with an SMA inverter will provide easily accessible and practical information.

All this is much ado about nothing. Interesting maybe, and fun if you're into that sort of thing, but really not worth the electrons we've burned up discussing it.

[RShackleford]

Listened to neighbor's 22kW rig, on this fine sunny day - three big versions of my inverter, probably 7kW each. A slight buzzing, 60 or 120Hz I guess; no hint of high-frequency whine (also not discernible by wife's female ears).

[solarix]

I had a guy once complain about the start-up relay clicks on an SMA Sunnyboy. We had noticed a tent in the back yard when doing the install, and it turns out he would sleep out in the back yard in the summer and the clicks would wake him up at the crack of dawn..... Sorry, not a warranty issue.