EG4 peak shaving mode CT location

[broot5]

I just finished up installing solar on my home and am using the EG4 18kpv inverter with a wall-mount battery. I got the battery for both partial home backup for a critical loads panel I installed and to do peak shaving since I'm grid tied. I'm in Phoenix with on-peak hours 4-7.

I currently have the battery set to force discharge at 5kw from 5-7 (there's enough solar coming in 4-5 to cover the first hour). This has been working mostly fine as I've been able to have 0 usage from my utility for going on 3 weeks now. But this is cause I watch it like a hawk. I really would like to use the peak shaving feature, but the information on this feature seems limited and conflicting online.

When I called into EG4, it was explained that if I have the CT sensors on the 2 hots coming from my main meter into my main panel with Peak shaving enabled, then the inverter will fill the energy demands of my main panel from the battery if the solar is not producing enough. That's only if I have the preset of "grid peak-shaving" set to 0, meaning I'm telling the system to not use any energy from the grid during the times I set it. I've had 2 EG4 support staff tell me this is the case, and 2 signature solar staff say this is not the case. The EG4 staff made it sound like there's a lot of confusion that people are not putting the CT sensors in the right spot to allow the peak shaving working mode to work properly

I currently just have my CT sensors in my inverter on my 2 hots to grid ports so I can measure what goes to the grid. I don't really want to run the CT sensors to my main box as my inverter is in my garage about 60 ft from the main service entrance UNLESS I know for sure it will allow peak shaving to work.

Has anyone figured out the peak shaving? I only really want advice from someone that has actually used it properly with the CT sensors in the main panel.

 

[BentleyJ]

In order to have accurate peak load shave for both the critical loads subpanel AND the main panel, the CT's need to be placed in the main panel on the incoming wiring feeding into the main breaker. This allows the 18kpv to regulate the amount of power backfeeding to the main panel to support loads but without wasting battery power selling to the grid.

 

[broot5]

Thanks for the answer @BentleyJ.

A couple more points of clarification - the eg4 manual specifies to place CT clamps on their respective L1 and L2 AC lines. However, those lines travel from my inverter through 2 disconnects, a bi directional meter, then into a meter socket adapter before meeting up with the grid. So if I'm installing them in the main meter, how would I know which line is L1 vs L2 or does that really matter? Also do I still have the arrows engraved on the CT clamps pointing back to the inverter so I have them in the right direction?

Something that's complicating this a little for me is I have bare-metal hot bars coming into my main panel instead of 2 sleeved 120v lines - see attached picture. So it's a little more dangerous for me to be trying to troubleshoot this with those live bare bars as opposed to just clamping onto sleeved wires and easily removing them if it doesn't work out. That said, I'm trying to get this right the first time cause getting those clamps back off those metal bars could get pretty sketchy - I've been putting on full arc-flash ppe anytime I'm messing with those main bars coming in cause of some of the horror stories I've heard.

Have you or anyone else been able to utilize the peak shaving mode to see that the mode works well with the CT's in the main panel? That's one of my main hang-ups; I have yet to hear about a single account of someone with real-life experience in utilizing eg4's peak shaving mode. I'd be a lot more comfortable with hearing first hand experience from someone to know this will work the way I want before going through the trouble. something a little more anecdotal than "eg4 recommends..."

 

[rogerheflin]

I just finished up installing solar on my home and am using the EG4 18kpv inverter with a wall-mount battery. I got the battery for both partial home backup for a critical loads panel I installed and to do peak shaving since I'm grid tied. I'm in Phoenix with on-peak hours 4-7.

I currently have the battery set to force discharge at 5kw from 5-7 (there's enough solar coming in 4-5 to cover the first hour). This has been working mostly fine as I've been able to have 0 usage from my utility for going on 3 weeks now. But this is cause I watch it like a hawk. I really would like to use the peak shaving feature, but the information on this feature seems limited and conflicting online.

When I called into EG4, it was explained that if I have the CT sensors on the 2 hots coming from my main meter into my main panel with Peak shaving enabled, then the inverter will fill the energy demands of my main panel from the battery if the solar is not producing enough. That's only if I have the preset of "grid peak-shaving" set to 0, meaning I'm telling the system to not use any energy from the grid during the times I set it. I've had 2 EG4 support staff tell me this is the case, and 2 signature solar staff say this is not the case. The EG4 staff made it sound like there's a lot of confusion that people are not putting the CT sensors in the right spot to allow the peak shaving working mode to work properly

I currently just have my CT sensors in my inverter on my 2 hots to grid ports so I can measure what goes to the grid. I don't really want to run the CT sensors to my main box as my inverter is in my garage about 60 ft from the main service entrance UNLESS I know for sure it will allow peak shaving to work.

Has anyone figured out the peak shaving? I only really want advice from someone that has actually used it properly with the CT sensors in the main panel.

"peak shaving" does not do what you think it does. Disable it.

"peak shaving" just seems to prevent/limit AC charging during the window.

It is piss-poorly named. Check the help that comes up when you put your mouse on it.

With the CT setting at your meter and zero-export being set it should just about zero your usage. I typically have been using .1-.2kwh per day with mine setup that way and exporting a minimal amount.

You probably also do not need to use force discharge.

 

[BentleyJ]

I'm not an EG4 user so will have to defer to @rogerheflin. Its possible the PLS function is not the same as what we believe it is. First determine if the 18kpv even has the functionality you need.

That said, the meter collar adds a completely different dimension to this issue. Sell back power is bypassing the main panel completely so the CT's have to be placed on the incoming utility cable at the meter socket.
The meter collar is essentially a "Y" connector where the power needs to flow to the main panel via one branch and the CT's need to sample the other side of the Y going back to the utility wiring.

 

[broot5]

@rogerheflin - I DO want to export to the grid in my situation. My utility company pays me the same rate per kwh as they sell to me at the off peak rate. So if I'm normally producing 85-90kwh on a given day, and only using around 20-40kwh for my energy needs, then there'd be no reason to not sell the excess back to the grid in my situation as those monetary credits are banked for me to use in the summer months when my energy needs are much higher.

That said, my primary goal with the batteries is to get me to an absolute 0 reliance on the grid between 4-7 as the rates are akin to rape during those hours. I have been using forced discharge as my work-around to peak shave. I can still run heavy loads during peak hours even with hardly any solar coming through as long as my kw forced discharge rate exceeds my the load needs of my main panel. The problem is that's just a set rate that doesn't respond to real-time energy needs that fluctuate.

I've spent hours researching the peak shaving feature. Most say what you say about it - it's poorly named and doesn't actually peak shave as it only limits ac charging during those hours. But yet 2 eg4 technical support staff members claim peak shaving will utilize the following priority to cover main panel loads: PV, then battery, then grid. And that as long as there's enough pv and/or battery during the preset time, peak shaving can bring grid usage to zero.

The consensus: more confusion on the topic. I may just hook up the ct's at the main and test this myself. I just wish I had the normal wires instead of the stupid metal bars feeding the main.

 

[BentleyJ]

That said, my primary goal with the batteries is to get me to an absolute 0 reliance on the grid between 4-7 as the rates are akin to rape during those hours. I have been using forced discharge as my work-around to peak shave. I can still run heavy loads during peak hours even with hardly any solar coming through as long as my kw forced discharge rate exceeds my the load needs of my main panel. The problem is that's just a set rate that doesn't respond to real-time energy needs that fluctuate.

I've spent hours researching the peak shaving feature. Most say what you say about it - it's poorly named and doesn't actually peak shave as it only limits ac charging during those hours. But yet 2 eg4 technical support staff members claim peak shaving will utilize the following priority to cover main panel loads: PV, then battery, then grid. And that as long as there's enough pv and/or battery during the preset time, peak shaving can bring grid usage to zero.

The consensus: more confusion on the topic. I may just hook up the ct's at the main and test this myself. I just wish I had the normal wires instead of the stupid metal bars feeding the main.

You have the concept of PLS correct. However, placing the CT's on the bus bars connecting to the main breaker will completely defeat the purpose of PLS and prevent power from traveling past the CT's. Assuming the 18kpv can do what the EG4 tech said.

Think of the CT as a check valve of sorts in that the readings from the CT are used to regulate the output of the inverter. IF the firmware in the inverter is done properly it will allow 100% of excess PV to flow upstream to the main panel and grid sell which the CT's should ignore since they can sense direction. As soon as PV falls off, battery should be used to make up the difference. The CT's should attempt to keep net power flow in either direction close to 0.

Therefore the CT's have to be on the utility side of the "Y" connection in the meter collar not in the main panel.

 

[Norwasian]

I had an issue where the inverter was single phase but the meter was 3-phase and supplied current to multiple breaker boxes within the building, each on a separate phase. We installed the inverter on the phase we thought was used most. Due to issues with the utility being unwilling to transfer the 3-phase meter account from a personal to an institutional name, we ended up discontinuing the 3-phase and had an electrician rewire everything, joining all of the building into a single-phase meter. But our CT was still on just one leg of what had formerly been the 3-phase arrangement. This meant that our solar production was limited to only a fraction of what our total usage was, and we had to hire someone to climb the power pole, remove the CT from its original location, and reinstall it next to the meter before any of the wires branched out toward the building.

In other words, placing the CT as near the source as possible means it will be most accurate in sensing the amount and direction of current passing through the meter. It will be inaccurate if the CT is placed somewhere beyond a fork in the road, as it can only sense what passes through its own branch, and not anything else--despite the fact that the wires are connected.

Imagine it like a water meter installed into a pipe. If one places the water meter after a T in the pipe (on one side of the T), the only water metered will be the water that passes through the meter's side of the T. The other leg of the T will not get metered. But if the meter is installed before the T, it will not matter which leg of the T carries the water--it will all have been metered before reaching the T.

If there are no branches in the wiring, and no possible loads or points of resistance between the CT and the where the utility power enters for the entire building, it may not matter how far up the line the CT is placed. But it must be placed before any branching or usage/load in order to get an accurate reading of the total electrical current.