Swap Out SMA SB 4000US with SB3.8-1SP-US-41

[RoyGBiv]

I am very much a dyi person but after unpacking the new inverter and pulling the cover off the old one I'm feeling like I got in over my head on this one. Seems like it would be an easy swap but looks like I'm gonna have to cut metal conduit around hot DC wires. Not sure I'm down with that but I suppose I have no choice as I have tried to throw money at local pros and they are clearly not interested. That said, the new inverter has the DC shutoff built into the housing where the old one (Dead btw) had the DC shutoff attached to the bottom, so it would appear the conduit needs to be cut shorter on both sides to accommodate the new inverter. What tools and safety equipment will I need? I have rubber gloves, safety glasses, and a multimeter. I'm worried I'm gonna blow something up, perhaps myself. TIA.

[sdold]

Hi Roy, welcome to the forum. Why do you say you'd have to cut around hot wires? At night the solar panel wires would be dead, and I'd expect a circuit breaker for the AC side of the inverter in the electrical panel. Pictures would really help.

[RoyGBiv]

20230524_194547.jpgYeah, sorry was gonna attach a pic to that op but neglected to do so. Yes, AC side of things is no problem with the breaker, only concern is DC side. I suppose I could work at night but figured with the right equipment switching over during the day wouldnt be an issue.

[bcroe]

Working in the dark is a good plan. I set up a work light and
organize tools first in daylight. Bruce Roe

[RoyGBiv]

Finally done. Only hard part was adjusting the damn conduit. Also, had a hell of a time finding an adapter to go from 1/2" to 1 1/8" to accommodate the cutouts, but these worked well: https://nationalfiresupply.com/product/qsa9303/ . Thanks for your help all!

20230816_111521.jpg

[RoyGBiv]

A question for the experts here. I have 2.94 kW of panels, seems this system peaks out at 2.35 kW. Is this to be expected with 13 year old panels? Azimuth / angle should be close to ideal this time of year, but it is quite hot (90+) here in Colorado these last few days. No idea what output was before but when new I know it actually went a smidge over 3kW from time to time.

[bcroe]

Besides other factors, high temps will reduce panel output power.
The actual panels will be a lot hotter than the air. Bruce Roe

[oregon_phil]

Finally done. Only hard part was adjusting the damn conduit. Also, had a hell of a time finding an adapter to go from 1/2" to 1 1/8" to accommodate the cutouts, but these worked well: https://nationalfiresupply.com/product/qsa9303/ . Thanks for your help all!

20230816_111521.jpg

Nice work in the retrofit!

FWIW, I had an issue with my original 7.7-40. The lower communication cabinet where your emt goes in is made out of cast aluminum that is powdercoated on the outside, but not 100% coated on the inside. If you haven't already done so, use duct seal in your EMT to make sure moisture isn't pumped in and out of the lower communication cabinet from wind/thermal effects. My original installer didn't use duct seal in the EMT for my original 7.7 -40. The inside of the communication cabinet started oxidizing where powdercoating wasn't covering 100%. I had to RMA my first inverter because of the oxidation. Our local energy trust also required all connections inside of the communication cabinet to be squirted with dielectric grease after all connections were made to prevent moisture related failures.

[J.P.M.]

A question for the experts here. I have 2.94 kW of panels, seems this system peaks out at 2.35 kW. Is this to be expected with 13 year old panels? Azimuth / angle should be close to ideal this time of year, but it is quite hot (90+) here in Colorado these last few days. No idea what output was before but when new I know it actually went a smidge over 3kW from time to time.

What's your zip and what's your array's azimuth and tilt ?

Any shade ?

2.35 kW output/2.94 kW STC output = 0.79 of STC output. That's probably not too bad but depends on Plane Of Array ("P.O.A.") irradiance, array orientation, panel fouling and on cell temp. - which is itself largely f(irradiance,ambient air temp, wind vector).

Also, being 13 yrs. old will also lower the panels' max. output by something like maybe 5-7 %.

On the other side of the energy ledger, a lot of CO is quite a ways above sea level which will tend to increase the P.O.A. irradiance.

I'd suspect that, depending on all the above mentioned factors that arrays in places with high(er) elevations like CO can see outputs greater than STC ratings more often than lower elevations, especially in winter.
I've seen such things happen when I lived in NM at 5000 ft.+ elevation.

You can get a rough 1st approx. to maximum power output by doing a PVWatts run with accurate inputs, use the hourly output option and look for a day/date with clear sky output that's close to the date when the 2.35kW output was observed and see what the model's output shows.

[RoyGBiv]

Nice work in the retrofit!

FWIW, I had an issue with my original 7.7-40. The lower communication cabinet where your emt goes in is made out of cast aluminum that is powdercoated on the outside, but not 100% coated on the inside. If you haven't already done so, use duct seal in your EMT to make sure moisture isn't pumped in and out of the lower communication cabinet from wind/thermal effects. My original installer didn't use duct seal in the EMT for my original 7.7 -40. The inside of the communication cabinet started oxidizing where powdercoating wasn't covering 100%. I had to RMA my first inverter because of the oxidation. Our local energy trust also required all connections inside of the communication cabinet to be squirted with dielectric grease after all connections were made to prevent moisture related failures.

I have compression fittings where the conduit enters the lower cabinet, that is water tight. The rest of the emt has been there for 13 years so far. I see what youre saying about the uncoated inside. I will keep an eye on it, thanks!

[RoyGBiv]

What's your zip and what's your array's azimuth and tilt ?

Any shade ?

2.35 kW output/2.94 kW STC output = 0.79 of STC output. That's probably not too bad but depends on Plane Of Array ("P.O.A.") irradiance, array orientation, panel fouling and on cell temp. - which is itself largely f(irradiance,ambient air temp, wind vector).

Also, being 13 yrs. old will also lower the panels' max. output by something like maybe 5-7 %.

On the other side of the energy ledger, a lot of CO is quite a ways above sea level which will tend to increase the P.O.A. irradiance.

I'd suspect that, depending on all the above mentioned factors that arrays in places with high(er) elevations like CO can see outputs greater than STC ratings more often than lower elevations, especially in winter.
I've seen such things happen when I lived in NM at 5000 ft.+ elevation.

You can get a rough 1st approx. to maximum power output by doing a PVWatts run with accurate inputs, use the hourly output option and look for a day/date with clear sky output that's close to the date when the 2.35kW output was observed and see what the model's output shows.

No shade. Actually the numbers I was seeing were on the SMA dashboard chart, those must be averages. Yesterday I was taking a closer look and did see the real-time output around 2.8kW for short periods of time, right around that 5-7% number.

[RoyGBiv]

Now that this project is behind me I must say, it amazes me how little help there is from local installers to do such a simple job. I contacted five, and they clearly werent interested. Perhaps they were too busy, but I would have happily paid a pro $300 for what amounted to 3 hours of work for me (a complete noob). I spent most of the time fumbling around with the conduit. Id say a pro could have done the job in 30 - 45 minutes. An easy $300- $400 . The way this new enclosure is designed I could swap to a new inverter (the top part) in 15 round minutes. This is the way it should have been designed in the first place.

That said, I cant be the only one around here with a dead string inverter being they last 10-13 years. My impression from talking to installers is that people just replace the entire system. Why the hell would people do that? I calculate I crossed the breakeven point of my initial investment at around 11 years, why would I start over? Insane. Seems you could easily make a business out of this sort of thing, just replacing old inverters... heck I bet most people dont even pay attention to whether or not an old string inverter is functioning anymore outside of a utility bill increasing, which they probably figure is just rates going up these days. lol.

[J.P.M.]

No shade. Actually the numbers I was seeing were on the SMA dashboard chart, those must be averages. Yesterday I was taking a closer look and did see the real-time output around 2.8kW for short periods of time, right around that 5-7% number.

Feel like sharing your zip and array orientation ?
Unless you're at a high elevation, that 2.8 kW seems kind of high for a quasi steady state output. Maybe for a few seconds or a minute or so under pt. cloudy skies due to cloud albedo.

[RoyGBiv]

Feel like sharing your zip and array orientation ?
Unless you're at a high elevation, that 2.8 kW seems kind of high for a quasi steady state output. Maybe for a few seconds or a minute or so under pt. cloudy skies due to cloud albedo.

Zip is 80128 array orientation is east. Not sure of the roof pitch, probably typical for this area 6/12 (?). Watch says altitude is 5640 ft.

[J.P.M.]

Zip is 80128 array orientation is east. Not sure of the roof pitch, probably typical for this area 6/12 (?). Watch says altitude is 5640 ft.

Thank you.

FWIW:
A quick PVWatts model run suggests the max. hourly clear sky output for your supplied inputs over 8,760 contiguous hours to be ~ 2.33 kW. Your results will vary some but the model seems to usually get pretty close to actual outputs for operating arrays when given good input.

FYI, the model also shows an annual output for a 2.95 STC kW array to be ~ 3,775 kWh/yr. using your supplied inputs.