Questions on 25kW Residential Cap in NY

[aleenoor]

My suggestions to the OP:

1. Can you get a 2nd separate service for your castle ? say your in-laws / servant / peasant accommodations ? That way you can you can get 2x 25kW systems.
2. If above doesn't work, I would take a serious look at reducing your consumption. Its always cheaper to use less. Upgrade HVAC, Pool Pump, W/H to be more efficient. Upgrade the heck out of insulation / windows etc
3. Another option is to add an off-grid system on top of your 25kW grid tied system. Move predictable day-time loads like pool-pumps, water-well etc. to the off-grid system with a smallish size battery backup.

I would find out if 25kW limit is on the AC size of DC size. If it is AC, then by all means, get a 30 to 35kW DC system with an AC output limit set at 25kW. Your system will exceed 25kW 1% of the times so not much will be lost.
Good-luck.

[pclausen]

Sorry for resurrecting this old thread. I'm in a similar pickle. I currently have a 20kW system which matched pretty closely what my consumption was. I in-fact was starting to build up a "bank" with the POCO (I have a net metering agreement with them). But then I got into crypto coin mining and currently have about a 8 kW draw 24/7 from my mining rigs, plus everything else in my house.

My POCO (State of Virginia actually from what I understand) has a 20kW limit for residential systems. But, given that I'm always shaving 8 kW off whatever solar production makes it to the meter, I figured I'd be safe going to a 28kW system, at least, without them ever figuring it out.

Before mining, I was making and consuming about 2,800 kWh a month. Now I'm still making about 2,800 kWh, but my consumption has jumped to almost 10,000 kWh. So is the POCO really going to care if my peak delivery back to them goes above 20 kW at high noon each day? I suppose I can discuss converting to a commercial account with them, but that will only drop my gross kWh rate from $0.07 to $0.04 and the base fee each month is higher and there is this thing they call peak load that will start taking into account.

[cebury]

Before mining, I was making and consuming about 2,800 kWh a month. Now I'm still making about 2,800 kWh, but my consumption has jumped to almost 10,000 kWh. So is the POCO really going to care if my peak delivery back to them goes above 20 kW at high noon each day? I suppose I can discuss converting to a commercial account with them, but that will only drop my gross kWh rate from $0.07 to $0.04 and the base fee each month is higher and there is this thing they call peak load that will start taking in.

Yes. When will they notice? Dont know. There are technical reasons the limits are in place. They arent going to trust your promise that you will always consume the extra 8kw of juice from your larger 28kw system. They have to ensure their infrastructure supports your system and wont negatively affect quality or reliability down the line. Just as I am sure you realize you cant just add extra panels to make 28kw. Your inverter will need upgrading, your wires will need upsizing, service panel changes, or you will need two independent systems. If your intention is to go larger, like 35kw, that is not trivial even though its only a 10kw upgrade.

I would dig into the billing for commercial Peak Load because that may affect you more than you realize, not sure for your state.

So which crypto coin minjng you doing? Bitcoin?

[peakbagger]

I expect that you would be talking to the utility interconnect group. They may be amenable to installing protective current relaying to trip the system off line if it exceeds the rated backfeed current. It usually means a couple of calibrated utility grade meters connected to the utility breaker. It may cost you tens of thousands of dollars as you will need an interconnect study which is oriented towards large power generators. In general once you go commercial the utility gets considerably more difficult and expensive to deal with.

What you are getting into is that many utility substations were not designed to have generated power entering the station, they were designed to distribute power from the grid and may not be equipped with protective relaying to ensure that power being independently produced. Your inverter is supposed to be designed to prevent exporting when the grid id down but they cant trust your equipment. They may need to upgrade their substation ad may require you to install something called Direct Transfer Trip which is something you really dont ever want to deal with (lots of bucks). Regulators "twisted the arms" of the regulators to make small solar simple to connect but the trade off was there would be an upper limit on output and unfortunately it looks like you want to trip the limit so you get to deal with the commercial side.

[bcroe]

I currently have a 20kW system which matched pretty closely what my consumption was. I in-fact was starting to build up a "bank" with the POCO (I have a net metering agreement with them). But then I got into crypto coin mining and currently have about a 8 kW draw 24/7 from my mining rigs, plus everything else in my house.

My POCO (State of Virginia actually from what I understand) has a 20kW limit for residential systems. But, given that I'm always shaving 8 kW off whatever solar production makes it to the meter, I figured I'd be safe going to a 28kW system, at least, without them ever figuring it out.

Before mining, I was making and consuming about 2,800 kWh a month. Now I'm still making about 2,800 kWh, but my consumption has jumped to almost 10,000 kWh. So is the POCO really going to care if my peak delivery back to them goes above 20 kW at high noon each day?

You might back up to my 28 June post on this thread. My system can only deliver 15KW, but it
may exceed 3000 KWH summer months. Adding panels to catch the rising and setting sun
won't increase your 20KW peak, won't require more inverters or plant, but will extend the
length of your productive day. Bruce Roe

[pclausen]

My current system consists of 80 panels with micro inverters. 6 strings totals, each going to a 20A breaker in a sub-panel in my shop located some 200' from the service entrance.

Somewhat dated schematic of my setup (doesn't include the 24 panel ground mount that was added about 1 month after I first went online):



Here's the 400A service entrance to the house:



I'm feeding the shop 200A sub-panel through the old 400A meter base at the shop (I used to have separate 400A services, one to the shop and one to the house). Here's the pass through at the shop:



Nothing is currently connected to the PV Subpanel tabs. As stated above, all PV connections are via 20A breaker in the shop sub-panel on the other side of the wall shown in the above pic.

So I have the infrastructure in place to add another 20kW worth of solar to the taps show above. I'd mount another dedicated panel outside next to that 400A meter base at the shop.

So this is the plan I'm currently on with my POCO:



And my other 2 choices are:



And:



I'm not sure what they mean by Demand Charge per kW on the larger power service plan, but the usage and energy charges per kWh are certainly better than the other 2 plans.

Depending on what I can and cannot do within my current installation, it might make sense to re-install a separate 4000A service to the ship, move some of my miners down there, and do separate solar arrays for that meter. My original plan was to keep both 400A services and tie the solar into the shop service, but since my primary service was at the house some 200' away, and given my generally very low usage at the shop, the solar had to be tied to the house service. So I had to dig up the 500MCM that was running from the shop to the house and run it up to the house instead.

I suppose another option would be to just leave everything the way it is, and setup a bunch of East and West facing panels, and connect those to a regular inverter. This way my mid day peak would stay the same, but I would add move production before and after high noon, and keep doing that until my mid-day peak reach, say 28 kW.

I'm mining various alt coins (ZCash, ZenCash, Eth, etc) and holding some and converting some to BTC. I'm running GPU rigs exclusively right now, but plan to add some ASIC miners as well.

[pclausen]

You might back up to my 28 June post on this thread. My system can only deliver 15KW, but it
may exceed 3000 KWH summer months. Adding panels to catch the rising and setting sun
won't increase your 20KW peak, won't require more inverters or plant, but will extend the
length of your productive day. Bruce Roe

Exactly. I was just posting the above reply where I put that down as an alternative option. So how did you determine the optimal angle for the E and W facing panels?

[bcroe]

Exactly. I was just posting the above reply where I put that down as an
alternative option. So how did you determine the optimal angle for the E and W facing panels?

Your app may be more difficult with micro inverters. To use the same inverter plant, you would need to
tie 2 or 3 panels in parallel tied to a micro inverter. Need the paralleled panels all be the same size,
they aren't here? Using strings here, I don't have a clearcut micro design. Maybe 2 panels tilted east
and west could be optimized for a micro. Here is an experiment with varying tilt of panels facing E, S,
and W. The sum of the E and W panels has about the same peak as the S panel, but much more over
the rest of the day. Bruce Roe

PVm17Jn16.jpg

[AzRoute66]

I so love that graph.

[pclausen]

Very cool. I'm at 38 degrees lat. Out of my 80 Micro Inverters, I had 20 fail on me so far. I raised a stink with the vendor, and they send me 80 more. So I have 60 'free' 250W inverters to play with. I'll pick up a pallet of panels and a bunch of treated 2x4's and have some fun!

So if I wire each pair of E-W panels to a single 250W inverter, I'm wondering what Watt rating panels I should get? 290W or so? Maybe something like this?

http://www.renvu.com/Solar-Products/Mono-Poly-Crystalline-Solar-Modules/UPSOLAR-290W-All-Black-Mono-PV-Module-60cell-35mm

Also, my inverters uses the standard MC4 connectors, so I'm hoping someone makes a MC4 "harness" that allows me to connect 2 panels in parallel (that is what I want as opposed to series right?) to each inverter, and then connect multiple inverters in parallel to form a string.

[bcroe]

Very cool. I'm at 38 degrees lat. Out of my 80 Micro Inverters, I had 20 fail on me so far. I raised a stink with the vendor, and they send me 80 more. So I have 60 'free' 250W inverters to play with. I'll pick up a pallet of panels and a bunch of treated 2x4's and have some fun!

So if I wire each pair of E-W panels to a single 250W inverter, I'm wondering what Watt rating panels I should get? 290W or so? Maybe something like this?

http://www.renvu.com/Solar-Products/Mono-Poly-Crystalline-Solar-Modules/UPSOLAR-290W-All-Black-Mono-PV-Module-60cell-35mm

Also, my inverters uses the standard MC4 connectors, so I'm hoping someone makes a MC4 "harness" that allows me to connect 2 panels in parallel (that is what I want as opposed to series right?) to each inverter, and then connect multiple inverters in parallel to form a string.

I would start with 250W panels; tilted for most daily hours, there won't be much (if any) clipping. Evil Bay will sell
you 2 into 1 adapters for parallel; I use them here as taps for my voltmeter. In the end it will be your design, keep
us informed with pictures and results. Put a spacer between treated wood and aluminum panels, which don't play
well. I use a rectangle of linoleum with a hole for a SS bolt. At 42 degrees lat. here. Bruce Roe

[DrLumen]

This is a bit off topic but... Bruce, it is good to see charting is not a lost art but, if you want, I will create a spreadsheet doc that will generate those charts for you. You could open it and make the data changes from excel or google sheets. Nothing wrong with your charts but they are so much easier to make when the data is in a spreadsheet or database.

[bcroe]

This is a bit off topic but... Bruce, it is good to see charting is not a lost art but, if you want, I will create a spreadsheet doc that will generate those charts for you. You could open it and make the data changes from excel or google sheets. Nothing wrong with your charts but they are so much easier to make when the data is in a spreadsheet or database.

Yes a formal presentation would need such a big cleanup effort. I'm not there yet, and more
worried about some concrete just now. But maybe that will happen later. thanks, Bruce Roe

[pclausen]

Thanks Bruce. I see plenty of those parallel adapters in "the bay". Cool. So I was walking around trying to decide where to put these E-W panels and if I should do one more array pointing South as well.

So here's a shot from abut 300' in front of the shop building:



So I was thinking the E - W arrays could start where the front of the trailer is, and then go back towards to camera. I do have some pretty tall trees, especially towards the East. Here's a shot showing them:



And a shot from the house looking down, showing the "valley" on the other side of the shop where I was planning to put the E - W panels.



I'm also thinking I can get away with putting up another South facing 24 panel array behind the existing one.

Here's a graph showing my production and consumption this afternoon:



So I'm only putting 9kW onto the grid, well below my 20kW limit. Being that it is quite hot today, the panels aren't all that efficient. Here's another shot from a few weeks ago, on a cool day, when the sun just came out from behind some clouds:



Here I was putting about 11kW into the grid, again well below my 20kW limit.

So I have a pretty constant 7 - 8 kW load from the miners before HVAC and other household appliances. They all run some flavor of Linux with watchdogs that would restart them if they sensed a given rig wasn't mining. So it is very unlikely that some/all of the miners would die during high noon one day. But even if they did, the conditions would have to be perfect to me to spike over 20kWh.

So I would my spare microinverters for the 2nd ground mounted South facing array, which will be close to the shop and easy to wire to the existing 200A sub-panel at the shop. It has 225A bus bars, and the main breaker inside it has already been downgraded to 150A. My voltage runs high from the POCO, typically 250 volt, even at night when the panels are pulling it up. So 104 panels @ 260W = 27,040 DC Watts. Divide that by 250V and you get 108A AC before losses. So I think having a total of 104 micro inverter based panels will work just now.

For the E - W facing panels, I'm thinking that it would be better to go with a single large inverter. If I was to go with 20 panels East and 20 panels West, I should only need to size the inverter to be able to handle 24 panels, correct? So let's say I went with 275W panels, that would require about a 7 kW inverter.

I have not looked into a DC array before, but given the 400 ft run up the house, I'd want to run as close to 600V as possible, right. So that would be about 20 panels, correct?

I still have all the literature from the county inspector, that includes a thorough section on DC. Since my current system is all micro inveter based, I just skimmed that section before, but I seem to recall that everything needs to be in steel conduit and I need HIGH DC VOLTAGE warning stickers every 10 feet or something like that on all exposed conduit.

[bcroe]

Now you are opening up other options. My idea was to increase hours with panels, without investing more
in the inverter plant. That keeps your peak power at the same limit. Changing things at the AC level will
require a complete rework.

Strings lend themselves to the kind of experiment I'm doing, but don't deal well with shadows. My curve
above no doubt will change with latitude, and it will change with seasons. Probably there will be experiments
with the tilt over months. The E-W panel arrangement produces enough power here, I was planning to
remove some S panels. Just what is the best winter arrangement isn't clear, considering the clouds
and short sun days. Bruce Roe