6048- reducing sell amps because of a small battery bank--will it work?

Dusty
Dusty Solar Expert Posts: 271 ✭✭✭
My 6048 is tied to 110 AH of battery capacity. After reading threads regarding A/C ripple when in sell mode, would I protect my battery bank by reducing the sell current to 5 amps instead of the default of 80% of AC1 breaker ratings? My AC1 breakers are 60A (Midnight Solar E-panel).

Is 5 amps too conservative? I'd like to maximize what I can sell without damaging my battery bank or causing the inverter to fault out.

Thank you!
XW6048, 3.4KW PV, Grid-Tied, always tweaking.
«1

Comments

  • Cariboocoot
    Cariboocoot Banned Posts: 17,615 ✭✭✭
    Re: 6048- reducing sell amps because of a small battery bank--will it work?

    Here's a question to your question: what's the point in having a 6kW inverter if you're only going to use it for 1200 Watts?
  • Dusty
    Dusty Solar Expert Posts: 271 ✭✭✭
    Re: 6048- reducing sell amps because of a small battery bank--will it work?

    Answer: I intend to expand the system as funds permit, but right now I only have ~1.1KW PV output, so I haven't invested a lot in batteries. In my area, power outages are really uncommon, and I have a generator to use for hurricanes. Even with hurricane Isabel in 2003, my home only lost power for 12 hours (Chesapeake, VA). My current daily utility usage is 20KW.
    XW6048, 3.4KW PV, Grid-Tied, always tweaking.
  • Cariboocoot
    Cariboocoot Banned Posts: 17,615 ✭✭✭
    Re: 6048- reducing sell amps because of a small battery bank--will it work?

    :confused: Not sure you're going about it the right way, but assuming the 6 kW install is all permitted? Usually they want the whole thing at once, not piecemeal.

    Reducing the SELL current should ease the ripple effect on the batteries, but it's less than ideal.

    For expanding a GT system, microinverters can be a good choice. Or buying several complete 1 kW GT systems over time. Considering you already have a generator for outages, the $3,000 that XW cost probably could have been put to better use. But it's too late now, right?

    Cut the current back and see if it helps. If it doesn't you'll know by the erratic operation of the XW.
  • Dusty
    Dusty Solar Expert Posts: 271 ✭✭✭
    Re: 6048- reducing sell amps because of a small battery bank--will it work?

    I started off with a ground mount system but intend to install a top-of-pole mount to expand the number of panels. I know that I went overboard with the 6KW inverter, but I didn't want to exceed the inverter's capability for future growth.

    I've been pouring over the operator's manual to make sure I know all the ins and outs of properly configuring it for my small system. This forum has been a lifesaver in making sense of all the capabilities of the 6048.

    When I finally get the system commissioned, I'll try cutting back on the sell amps. If it doesn't work, it looks like I'll be ordering larger batteries sooner rather than later! I really thought that I'd upgrade the number of PVs first since I didn't need a lot of autonomy.

    I do like the idea of microinverters to help diagnose individual panels, but you're right--it's a bit late now. Thanks for your help!

    -Dusty
    XW6048, 3.4KW PV, Grid-Tied, always tweaking.
  • BB.
    BB. Super Moderators, Administrators Posts: 33,431 admin
    Re: 6048- reducing sell amps because of a small battery bank--will it work?

    Reducing the XW Inverter output current is probably a good idea--At least to the level of equivalent battery bank of 100 AH per 1kW of inverter output.

    If you have a full >600 AH (@48 volt) battery bank, then it probably does not matter so much.

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • Dusty
    Dusty Solar Expert Posts: 271 ✭✭✭
    Re: 6048- reducing sell amps because of a small battery bank--will it work?

    Since I only have 110AH of total battery capacity, I thought that reducing the sell amps to 5 would limit selling to the grid of about 1KW--including the house loads. Thanks Bill!
    XW6048, 3.4KW PV, Grid-Tied, always tweaking.
  • Joe_B
    Joe_B Solar Expert Posts: 318 ✭✭✭✭✭✭
    Re: 6048- reducing sell amps because of a small battery bank--will it work?

    With that small of a battery bank, I would seriously consider disabling sell mode. Why beat the heck out of your batteries for a few KWH per day? Batteries are expensive and grid KWH are cheap!
  • Dusty
    Dusty Solar Expert Posts: 271 ✭✭✭
    Re: 6048- reducing sell amps because of a small battery bank--will it work?

    Joe, you bring up a very good point. I thought that once the batteries were charged, it was the MPPT charger that was feeding the grid; so why is that beating the heck out of my batteries? They would stay charged, wouldn't they? Without seeing the unit in action, I'm having trouble understanding what you mean. But I appreciate the input!
    XW6048, 3.4KW PV, Grid-Tied, always tweaking.
  • Joe_B
    Joe_B Solar Expert Posts: 318 ✭✭✭✭✭✭
    Re: 6048- reducing sell amps because of a small battery bank--will it work?

    Sell will drain the batteries down to the grid support voltage. You will have to have grid support enabled and sell mode enabled. When in this mode, the batteries will be drained down to the grid support voltage. If you disable sell mode and enable grid support, then put the unit into enhanced interactive mode, the batteries will remain at float voltage and excess PV will be used to support the local loads. This will keep your batteries happy and still use PV energy to support the loads.
  • Dusty
    Dusty Solar Expert Posts: 271 ✭✭✭
    Re: 6048- reducing sell amps because of a small battery bank--will it work?

    I understand. I've read posts on this site that said to set the grid support voltage to just below float--52.5. But isn't the drain happening at the same time as the MPPT-60 and forcing the voltage to stay at or near float? Isn't 52.5v still considered a full battery? I guess I'm having trouble understanding if the battery is never really drained below a full charge, how is selling detrimental to the battery bank? I apologize for being so dense.
    XW6048, 3.4KW PV, Grid-Tied, always tweaking.
  • Joe_B
    Joe_B Solar Expert Posts: 318 ✭✭✭✭✭✭
    Re: 6048- reducing sell amps because of a small battery bank--will it work?

    Dusty, you are not dense, all good questions. You cannot determine battery SOC based on battery voltage alone. If you look at the graph in the manual (fig. 3.8 ), you will see that in fixed battery voltage mode, the inverter will enter sell at GSV + 2 volts and exit sell mode at GSV -.5 volts. Therefore in this mode, the battery voltage will be pumping up and down by 2.5 volts during sell mode. In actuallity, the battery voltage will vary even more than that due to the delays built into the firmware. To address this issue, they provided a mode called enhanced interactive mode. When the unit is configured for this mode, it will not exit and enter sell mode but will support local loads without using any battery energy. Only PV. This assumes that you are using an XW charge controller and the xanbus is connected to the inverter. The inverter will support local loads based on what the charge controller harvest is reporting.

    I do not believe that you can have sell enabled when in enhanced interactive mode but I could be wrong about this. In any case, look at the economics, you are using 20KW per day on your local loads. When the sun is cranking, you can only pull down about 1KW. It would make sense to use that 1KW to directly offset some of your AC load usage rather than pump it into the battery and then sell it back to the grid through the inverter. The battery is horrendously inefficient and the inverter has losses as well. given your low power levels, these inefficiencies are significant.

    Use the solar harvest to offset your loads, what do you get by selling back to the grid? about 15 cents per KWH? Say you have full sun for 6 hours a day, that is only about 90 cents a day of sell back discounting inefficiedncies. If you pull energy out of your batteries to sell to the grid, you will be adding cycles to the battery bank and minimizing the life of the bank. Given your system size, it makes more sense to keep the batteries at 100% in case the power does fail and using the excess PV energy locally to offset your grid draw.
  • RCinFLA
    RCinFLA Solar Expert Posts: 1,484 ✭✭✭✭
    Re: 6048- reducing sell amps because of a small battery bank--will it work?
    Dusty wrote: »
    Joe, you bring up a very good point. I thought that once the batteries were charged, it was the MPPT charger that was feeding the grid; so why is that beating the heck out of my batteries? They would stay charged, wouldn't they? Without seeing the unit in action, I'm having trouble understanding what you mean. But I appreciate the input!

    It's the ripple current that beats up the battery. It is the storage capacitor that gets fed the constant DC current (for given illumination level) from PV controller and delivers the 120 Hz half sinewave shaped pulses of current to the inverter necessary for the 60 Hz single phase AC grid.
  • Dusty
    Dusty Solar Expert Posts: 271 ✭✭✭
    Re: 6048- reducing sell amps because of a small battery bank--will it work?

    Thank you!!! I have a Xantrex LinkLite battery monitor that I used with my little Morningstar 300W inverter. I know that the 6048 doesn't have a shunt, but the Midnight Solar E-panel does, so I'll definitely be hooking that up. Even so, I could use the shunt that came with the LinkLite.

    I had the incorrect picture that the battery stayed topped off while the Xantrex MPPT-60 fed the "sell" mode to the grid. My loads during the day don't harvest the max out of my current panels, and the a/c won't be tied into the 6048 anyway, so I'll have at least half of the harvest lost each day. That seems like such a waste. Why upgrade if the current system outmatches my consumption during the day, and I can't sell back to use the harvest at night? Looks like a larger battery bank is the only answer.

    Once I get the system commissioned, I'll enjoy testing the different features. I haven't read about the Enhanced Interactive Mode, so I'll have to dig deeper.

    Thanks again for taking the time to explain all this!!!

    Regards,

    Dusty

    Joe_B wrote: »
    Dusty, you are not dense, all good questions. You cannot determine battery SOC based on battery voltage alone. If you look at the graph in the manual (fig. 3.8 ), you will see that in fixed battery voltage mode, the inverter will enter sell at GSV + 2 volts and exit sell mode at GSV -.5 volts. Therefore in this mode, the battery voltage will be pumping up and down by 2.5 volts during sell mode. In actuallity, the battery voltage will vary even more than that due to the delays built into the firmware. To address this issue, they provided a mode called enhanced interactive mode. When the unit is configured for this mode, it will not exit and enter sell mode but will support local loads without using any battery energy. Only PV. This assumes that you are using an XW charge controller and the xanbus is connected to the inverter. The inverter will support local loads based on what the charge controller harvest is reporting.

    I do not believe that you can have sell enabled when in enhanced interactive mode but I could be wrong about this. In any case, look at the economics, you are using 20KW per day on your local loads. When the sun is cranking, you can only pull down about 1KW. It would make sense to use that 1KW to directly offset some of your AC load usage rather than pump it into the battery and then sell it back to the grid through the inverter. The battery is horrendously inefficient and the inverter has losses as well. given your low power levels, these inefficiencies are significant.

    Use the solar harvest to offset your loads, what do you get by selling back to the grid? about 15 cents per KWH? Say you have full sun for 6 hours a day, that is only about 90 cents a day of sell back discounting inefficiedncies. If you pull energy out of your batteries to sell to the grid, you will be adding cycles to the battery bank and minimizing the life of the bank. Given your system size, it makes more sense to keep the batteries at 100% in case the power does fail and using the excess PV energy locally to offset your grid draw.
    XW6048, 3.4KW PV, Grid-Tied, always tweaking.
  • Joe_B
    Joe_B Solar Expert Posts: 318 ✭✭✭✭✭✭
    Re: 6048- reducing sell amps because of a small battery bank--will it work?

    Dusty,

    A little more clarification on how the XW does grid support. When you have a fully charged battery, the XW will pull as much current as required to support the loads based on the battery voltage. As the battery SOC depletes, the voltage will drop for a given load. When the XW senses that the voltage is dropping, it will throttle back the current it is pulling from the bank to maintain the battery voltage at the GSV setting. This is an inherent flaw in the way the XW manages battery energy. True SOC is determined by amp hours removed from the bank, not the voltage of the bank. As a result, if you just set GSV to some value and the XW draws current from the bank, the battery voltage drops. At this point, the XW will decrease the load on the bank to maintain the voltage AND CONTINUE TO PULL POWER OUT OF THE BANK at lower and lower current levels. By doing this, the SOC will go much lower than you expect it to. I have confirmed this with my battery monitor as well as the specific gravity of the electrolyte. You cannot set the GSV to a certain level and expect the battery to be depleted to a given SOC. Most people overcome this issue by either running their systems in the enhanced interactive mode or by ensuring that they harvest more PV energy than they require for the loads.

    My system uses the latter method, I have a total array capacity of 2.5 KW and I usually get about 12KWH per day on a good day. I use about 9 KWH per day for my loads and I have a bank capacity of 740 AH (370 AH @50% dod) which works out to about 8.8 KWH. So on a good day, I can run my loads and recharge my bank to 100%. The issue is on a cloudy day, I am pulling power out of my bank at a higher rate than I can harvest energy so the SOC of my bank decreases. I use my Victron battery monitor to control a contactor that will limit the discharge of my bank to about 80% of full charge before I switch back on to the grid. I do not rely on battery voltage but true amp hours drawn from the bank to determine when to transition to the grid. However, on extended periods of cloudy WX, I do have to fudge with the thresholds on the battery monitor to keep the battery healthy. Occasionally I have to enable the grid charger to avoid leaving the SOC of the pack below 80% and also to do an occasional EQ.

    One other issue is the battery bank efficiency, to recharge a bank depleted to say 50% dod, you will be putting in energy equivalent to about 80%. In other words, you put in about 20% more energy than you get out. So cycling the battery bank not only kills the bank faster in terms of cycle life but also costs you about 20% of your energy.

    Hope this helps you,

    Joe
  • Dusty
    Dusty Solar Expert Posts: 271 ✭✭✭
    Re: 6048- reducing sell amps because of a small battery bank--will it work?

    I remember reading the thread about how you set up your Victron monitor to enable a relay to control your state of charge. I know that my LinkLite monitor has an aux system that will drive a relay as well, but I just haven't absorbed that whole process. I was impressed with your setup but haven't thought about how to do that with my system.

    You really opened my eyes about SOC loss even though the voltage seems like still a full charge. Thank you! I have one question about your explanation about when the SOC of the battery depletes--isn't the MPPT-60 CC charging at the same time? Why should the SOC ever really drop if there are electrons are flowing in to the battery at the same time electrons are flowing out? Doesn't the Xanbus regulate this??? By cutting down the Sell Amps to 5, shouldn't the MPPT-60 keep up without depleting the batteries at all?

    I can already harvest more PV energy during the day than my loads can use, but I certainly can't supplement my a/c use this summer--since that circuit won't even be wired into the 6048.

    So in summary: if I get a larger (i.e. 600ah or greater) battery bank, I'll have sufficient days of autonomy to run the inverter and support my loads for several days of inclement weather, but it's probably cheaper to use my generator on those rare occasions I lose power. However, I can't support the 6KW output of the inverter without a large battery bank to act as a ballast, so I've overspent on the inverter (yup, I got that), and I shouldn't sell any power back to the utility (at .11 a KWH) without spending literally thousands more in batteries and thousands more in PV panels.....

    So, I've spent a little bit in small capacity AGM batteries, and I'll get to watch the meter spin backwards for only a couple years. Or, I can spend thousands more and watch the meter spin backwards for a longer time before having to spend thousands again years later....

    I guess I'm just getting a little frustrated. No reflection on your explanation, which I really appreciate! After all, I didn't start this science project to save money. I was just intrigued about using solar power.

    Regards,

    Mark
    Joe_B wrote: »
    Dusty,

    A little more clarification on how the XW does grid support. When you have a fully charged battery, the XW will pull as much current as required to support the loads based on the battery voltage. As the battery SOC depletes, the voltage will drop for a given load. When the XW senses that the voltage is dropping, it will throttle back the current it is pulling from the bank to maintain the battery voltage at the GSV setting. This is an inherent flaw in the way the XW manages battery energy. True SOC is determined by amp hours removed from the bank, not the voltage of the bank. As a result, if you just set GSV to some value and the XW draws current from the bank, the battery voltage drops. At this point, the XW will decrease the load on the bank to maintain the voltage AND CONTINUE TO PULL POWER OUT OF THE BANK at lower and lower current levels. By doing this, the SOC will go much lower than you expect it to. I have confirmed this with my battery monitor as well as the specific gravity of the electrolyte. You cannot set the GSV to a certain level and expect the battery to be depleted to a given SOC. Most people overcome this issue by either running their systems in the enhanced interactive mode or by ensuring that they harvest more PV energy than they require for the loads.

    My system uses the latter method, I have a total array capacity of 2.5 KW and I usually get about 12KWH per day on a good day. I use about 9 KWH per day for my loads and I have a bank capacity of 740 AH (370 AH @50% dod) which works out to about 8.8 KWH. So on a good day, I can run my loads and recharge my bank to 100%. The issue is on a cloudy day, I am pulling power out of my bank at a higher rate than I can harvest energy so the SOC of my bank decreases. I use my Victron battery monitor to control a contactor that will limit the discharge of my bank to about 80% of full charge before I switch back on to the grid. I do not rely on battery voltage but true amp hours drawn from the bank to determine when to transition to the grid. However, on extended periods of cloudy WX, I do have to fudge with the thresholds on the battery monitor to keep the battery healthy. Occasionally I have to enable the grid charger to avoid leaving the SOC of the pack below 80% and also to do an occasional EQ.

    One other issue is the battery bank efficiency, to recharge a bank depleted to say 50% dod, you will be putting in energy equivalent to about 80%. In other words, you put in about 20% more energy than you get out. So cycling the battery bank not only kills the bank faster in terms of cycle life but also costs you about 20% of your energy.

    Hope this helps you,

    Joe
    XW6048, 3.4KW PV, Grid-Tied, always tweaking.
  • Joe_B
    Joe_B Solar Expert Posts: 318 ✭✭✭✭✭✭
    Re: 6048- reducing sell amps because of a small battery bank--will it work?

    Dusty, what happens in my system with regard to the XW SCC is once the battery is up to 100%, the SCC will only load the panels enough to support the loads on the inverter. In my case this results in losing out on maximum harvest capability. It's true that when the sun is shining, electrons will be flowing directly from the SCC to the inverter but other things are going on here. Say for example, a cloud passes by, your solar output will drop and the electrons will be drawn from the battery. Also, the SCC is an MPPT charger and regularly does a "scan" to find the max power point. during the scan time, the battery will support the load. Also, if the load exceeds the amount of energy that the PV can supply, the energy comes from the battery.

    In my area, it is rare to have a completely sunny day, more often than not, some part of the day we have clouds passing by. Unless you "ride the controls" you will be pulling mini cycles out of the battery. I have found the latency of my xanbus is pretty bad. I run a lot of stuff 2 XW SCC, SCP, AGS, gateway, and inverter. the response times can be pretty long. An example of this is sometimes one of my SCC's will be pulling max power off the array and the other wont because my loads dont need it. If something like my well pump kicks on, it will take about 10 seconds (or longer) for the second SCC to throttle up. During this time, my battery is honking big amps to support the system. This is why I need such a big battery to stabilize things.

    With regard to your specific question, since the XW does not have a current shunt it can only estimate the SOC of the battery based on voltage and the data on the xanbus. It is not very accurate. Quite frequently my XW will think the battery is at 100% when it is not. If left alone, my banks SOC would slowly stairstep down. I have found that to properly manage the SOC and keep it cranking up to 100% I must trick the system from time to time by using the grid charger or setting my float voltages artificially high and using my dump load to shunt he voltage down. The other way I found to trick it was to lie to it about the bank size but that whole can of worms is best left for another discussion.
  • Dusty
    Dusty Solar Expert Posts: 271 ✭✭✭
    Re: 6048- reducing sell amps because of a small battery bank--will it work?

    Thanks Joe,

    My misunderstanding was that at times when clouds passed in front of the sun and the load exceeded the output of the MPPT-60, the grid would make up the difference and not the batteries. My misconception of that is why I thought that a small battery bank would be okay since it was only used for brief backup use if the utility should drop. I did not know that the latency (response) of the MPPT was so slow. That's not something I remember in the Xantrex literature, but I'll double check the specifications. Perhaps that's due to the MPPT scanning for the right power point due to the change in sunlight as you mentioned (ie clouds).

    That's why this forum is so important--I don't have to reinvent the wheel--just know enough to ask (hopefully) the right (coherent) questions. Thank you for the answers!

    I have another question in response to one thing you said; "I must trick the system from time to time by using the grid charger..." Does this mean that in a normal grid tie/sell mode, you would not have the 6048's charger enabled and just rely on the MPPT-60 all the time? In that case, wouldn't the grid still make up the difference to support the loads and let the MPPT-60 keep the batteries floating?

    Thanks again!!! :confused:
    Joe_B wrote: »
    Dusty, what happens in my system with regard to the XW SCC is once the battery is up to 100%, the SCC will only load the panels enough to support the loads on the inverter. In my case this results in losing out on maximum harvest capability. It's true that when the sun is shining, electrons will be flowing directly from the SCC to the inverter but other things are going on here. Say for example, a cloud passes by, your solar output will drop and the electrons will be drawn from the battery. Also, the SCC is an MPPT charger and regularly does a "scan" to find the max power point. during the scan time, the battery will support the load. Also, if the load exceeds the amount of energy that the PV can supply, the energy comes from the battery.

    In my area, it is rare to have a completely sunny day, more often than not, some part of the day we have clouds passing by. Unless you "ride the controls" you will be pulling mini cycles out of the battery. I have found the latency of my xanbus is pretty bad. I run a lot of stuff 2 XW SCC, SCP, AGS, gateway, and inverter. the response times can be pretty long. An example of this is sometimes one of my SCC's will be pulling max power off the array and the other wont because my loads dont need it. If something like my well pump kicks on, it will take about 10 seconds (or longer) for the second SCC to throttle up. During this time, my battery is honking big amps to support the system. This is why I need such a big battery to stabilize things.

    With regard to your specific question, since the XW does not have a current shunt it can only estimate the SOC of the battery based on voltage and the data on the xanbus. It is not very accurate. Quite frequently my XW will think the battery is at 100% when it is not. If left alone, my banks SOC would slowly stairstep down. I have found that to properly manage the SOC and keep it cranking up to 100% I must trick the system from time to time by using the grid charger or setting my float voltages artificially high and using my dump load to shunt he voltage down. The other way I found to trick it was to lie to it about the bank size but that whole can of worms is best left for another discussion.
    XW6048, 3.4KW PV, Grid-Tied, always tweaking.
  • Joe_B
    Joe_B Solar Expert Posts: 318 ✭✭✭✭✭✭
    Re: 6048- reducing sell amps because of a small battery bank--will it work?
    Dusty wrote: »
    Thanks Joe,

    My misunderstanding was that at times when clouds passed in front of the sun and the load exceeded the output of the MPPT-60, the grid would make up the difference and not the batteries. My misconception of that is why I thought that a small battery bank would be okay since it was only used for brief backup use if the utility should drop. I did not know that the latency (response) of the MPPT was so slow. That's not something I remember in the Xantrex literature, but I'll double check the specifications. Perhaps that's due to the MPPT scanning for the right power point due to the change in sunlight as you mentioned (ie clouds).

    Well you have touched on yet another issue with the XW Dusty, namely the sharing of grid power and battery power. There is a feature called load shaving, normally when running in grid support mode the load is split at around 50% between the grid and the battery. This is very annoying when you have more than enough PV to support the load and the XW is still pulling power from the grid. So load shaving will cause this 50-50 balance to be offset towards the battery but you can never get it to go 100% off grid without physically disconnecting AC1 and running in invert mode. Also, the firmware that controls the balance between battery and grid is VERY slow. It ramps over the course of minutes. There is no mechanism to shift the balance from the battery to the grid.

    The speed of which these things happen is somewhat dependent on the speed of the xanbus. If you have a lot of nodes in the network, the latency goes up. The xanbus only runs at 250 Khz and there is a lot of data flying around (primarily to support the SCP). So in a big system the response times can be quite slow. This is not something xantrex brags about and I dont even recall ever even seeing a spec on the response times for a given system configuration.

    The MPPT is slow because it has to ramp the load on the panels around to find the max power point for the panels, in my experience, this scan takes about 4-5 seconds and happens a lot. This causes the inverter to suck more battery power when it happens.
    I have another question in response to one thing you said; "I must trick the system from time to time by using the grid charger..." Does this mean that in a normal grid tie/sell mode, you would not have the 6048's charger enabled and just rely on the MPPT-60 all the time? In that case, wouldn't the grid still make up the difference to support the loads and let the MPPT-60 keep the batteries floating?

    Yes, normally I disable the grid charger and just use the SCC to run the system. The reason I say trick the system is that because the XW does not have a true current shunt, it tries to keep track of the battery SOC based on voltage and time. They dont use any kind of Peukert compensation or efficiency correction on the calculation so as time goes on, the system thinks the battery is full and it really is not. To reset these counters, one way is to let the grid charger do a full charge to float. Another way I have found is to lie to the XW and tell it that the battery is much smaller than is actually is. My value should be set to740 AH but if I set it to that value, the charger exits absorb too quickly and the battery is left low. By setting the value to 320 AH, it will stay in absorb longer and top off the batteries. This is because the XW charge algorithm has a fixed value of tail current set at 2% of the pack capacity. I want the charger to stay in absorb mode until the current tails off to about 1% so I lie to the algorithm.

    These values are very dependent on what you have in your system and must be arrived at by a lot of tweaking and experimentation. I have been messing around with this system for years and am still tweaking certain parameters. I use a contactor based on SOC to run the XW in invert mode when I have sufficient PV and the battery is charged. I also have a 1KW dump load that I use the aux output settings to control to keep the panels loaded when my PV exceeds my loads. Might as well heat water rather than unload the panels and lose the energy right?

    I have basically set up priorities in my system, first comes the battery, I always make sure the battery is fully charged before I switch in other "features". Second comes my grid draw, once the batteries are happy, I ramp into grid support mode then into invert mode to kill the grid. Once my loads are satisfied, I enable the dump contactor. Of course all of this is automatic and highly variable and dynamic based on how much energy I am getting off the panels at any given time.
  • Dusty
    Dusty Solar Expert Posts: 271 ✭✭✭
    Re: 6048- reducing sell amps because of a small battery bank--will it work?

    Well, at least I'm in the ballpark, and I'm understanding configuration a lot more (and the capabilities and limitations of the XW) thanks to you.

    I thought that load shaving was only a feature to prevent me from exceeding a certain load limit where exceeding it would incur an additional charge from the power company. With that in mind, I intended to leave that function disabled. So this is why I read in an older post to enable load shaving and set the amps to zero? By having the charger on the 6048 disabled, is there less likelyhood that the grid will be used when there is excess PV? I guess I relied on Xanbus to track all the capabilities and optimize the harvest. I have a lot to learn....

    I'll just have the 6048, MPPT-60 and SCB connected via Xanbus, so I hope that the latency on the bus won't an issue for me.

    And I didn't know the MPPT scan takes 4-5 seconds. That's good to know too!

    Wow, your system is really complex. I can see why you're still tweaking it after several years. Thank you again for lending your experience!
    Joe_B wrote: »
    Well you have touched on yet another issue with the XW Dusty, namely the sharing of grid power and battery power. There is a feature called load shaving, normally when running in grid support mode the load is split at around 50% between the grid and the battery. This is very annoying when you have more than enough PV to support the load and the XW is still pulling power from the grid. So load shaving will cause this 50-50 balance to be offset towards the battery but you can never get it to go 100% off grid without physically disconnecting AC1 and running in invert mode. Also, the firmware that controls the balance between battery and grid is VERY slow. It ramps over the course of minutes. There is no mechanism to shift the balance from the battery to the grid.

    The speed of which these things happen is somewhat dependent on the speed of the xanbus. If you have a lot of nodes in the network, the latency goes up. The xanbus only runs at 250 Khz and there is a lot of data flying around (primarily to support the SCP). So in a big system the response times can be quite slow. This is not something xantrex brags about and I dont even recall ever even seeing a spec on the response times for a given system configuration.

    The MPPT is slow because it has to ramp the load on the panels around to find the max power point for the panels, in my experience, this scan takes about 4-5 seconds and happens a lot. This causes the inverter to suck more battery power when it happens.



    Yes, normally I disable the grid charger and just use the SCC to run the system. The reason I say trick the system is that because the XW does not have a true current shunt, it tries to keep track of the battery SOC based on voltage and time. They dont use any kind of Peukert compensation or efficiency correction on the calculation so as time goes on, the system thinks the battery is full and it really is not. To reset these counters, one way is to let the grid charger do a full charge to float. Another way I have found is to lie to the XW and tell it that the battery is much smaller than is actually is. My value should be set to740 AH but if I set it to that value, the charger exits absorb too quickly and the battery is left low. By setting the value to 320 AH, it will stay in absorb longer and top off the batteries. This is because the XW charge algorithm has a fixed value of tail current set at 2% of the pack capacity. I want the charger to stay in absorb mode until the current tails off to about 1% so I lie to the algorithm.

    These values are very dependent on what you have in your system and must be arrived at by a lot of tweaking and experimentation. I have been messing around with this system for years and am still tweaking certain parameters. I use a contactor based on SOC to run the XW in invert mode when I have sufficient PV and the battery is charged. I also have a 1KW dump load that I use the aux output settings to control to keep the panels loaded when my PV exceeds my loads. Might as well heat water rather than unload the panels and lose the energy right?

    I have basically set up priorities in my system, first comes the battery, I always make sure the battery is fully charged before I switch in other "features". Second comes my grid draw, once the batteries are happy, I ramp into grid support mode then into invert mode to kill the grid. Once my loads are satisfied, I enable the dump contactor. Of course all of this is automatic and highly variable and dynamic based on how much energy I am getting off the panels at any given time.
    XW6048, 3.4KW PV, Grid-Tied, always tweaking.
  • Joe_B
    Joe_B Solar Expert Posts: 318 ✭✭✭✭✭✭
    Re: 6048- reducing sell amps because of a small battery bank--will it work?

    Dusty,

    The grid charger and load shaving have nothing todo with each other. What load shaving is used for is to shift the balance of the grid vs. battery usage to support loads. with load shaving disabled, the XW will pick a ratio based on the battery SOC. (obvious drawbacks) and what I have found is that if the XW "thinks the battery is charged, this ratio is about 20% grid, 80% battery. As the XW determines that the battery SOC is dropping, the ratio will shift more to favor the grid. The algorithm is not too good and it typically settles in at about 50-50. When you enable load shaving, you affect the decision point in favor of more battery BUT, you cant have complete control over it. In other words, if load shaving is enabled and load shave amps are set to zero. You will favor the battery but the grid will still be supplying a lot of power. Basically it doesnt work right. If the XW thinks the battery is at 100%, and load shave amps are set to zero, then you will still be pulling about 100 watts from the grid regardless of your PV yield. Hence my incorporation of a contactor to just override everything and run 100% off the battery. Also you willnotice that when you are grid interactive, the grid draw is very unstable. The values will bounce around quite a bit at about 1 Hz or so. Hence, when I specify the percentages above, it is an average of the bouncing.

    Yeah my system is complex and sometimes it takes me quite a while to figure out what is going on but the way I want my system to work is to #1 keep the batteries happy and #2 maximize solar harvest over grid useage. Consequently I am adjusting certain parameters based on seasonal conditions. There is definately not one set of parameters that will work all the time. Primarily what I adjust the most is the discharge floor on the BVM and the GSV setting but I sometimes have to mess around with the chargers custom battery settings as well and as I said previously I have to do an occasional EQ off the grid. One other think I find I have to occasionally change is enabling / disabling grid support. We have had about 4 days of continuous rain here this week and when this happens, rather than squeaking every watt out of the battery, I just bite the bullet and go on grid. I get a few hundred watts off the PV when it is this bad and I use that to keep the batteries at float but today I am pretty much 100% on grid. This weekend we are supposed to have good sun again and I will reenable grid support. When my BVM reaches 100% SOC, the contactor will open and I will again be 100% off grid.

    Joe
  • Dusty
    Dusty Solar Expert Posts: 271 ✭✭✭
    Re: 6048- reducing sell amps because of a small battery bank--will it work?

    Thanks again Joe,

    I suspected that the charger and load shaving were not connected, but thank you for confirming that. It seems like the main function of Load Shaving (from what I've read in the operation manual) is to use battery/PV to maintain grid usage below a set level that will cause a financial penalty from the power company if exceeded. That said, using load shave with grid amps set to zero is a kludge to force the balance of usage to be weighed as heavily as possible on the battery/PV output. Very clever, and maybe not a perfect solution--but definitely not documented in the operator's guide as a function. THANK YOU!!!

    I feel your pain about constant days of rain where you are. When I was running my little off-grid system, it was with heavy heart when I looked at the abismal harvest from the panels that day--but even that was a learning experience for me. Right now, my panels start to harvest around 0900 EST and finish producing around 1900-ish. My panels were only harvesting around 5KWH on those days, but I didn't expect much more from my small system.

    It seems like I might as well shut the inverter off after the sun goes down. The grid will run the limited load (which they'd do anyway at night plus the overhead of the inverter itself), and the batteries won't discharge much overnight until the MPPT-60 can output power and float them. Sorry for going off topic a little bit.
    Joe_B wrote: »
    Dusty,

    The grid charger and load shaving have nothing todo with each other. What load shaving is used for is to shift the balance of the grid vs. battery usage to support loads. with load shaving disabled, the XW will pick a ratio based on the battery SOC. (obvious drawbacks) and what I have found is that if the XW "thinks the battery is charged, this ratio is about 20% grid, 80% battery. As the XW determines that the battery SOC is dropping, the ratio will shift more to favor the grid. The algorithm is not too good and it typically settles in at about 50-50. When you enable load shaving, you affect the decision point in favor of more battery BUT, you cant have complete control over it. In other words, if load shaving is enabled and load shave amps are set to zero. You will favor the battery but the grid will still be supplying a lot of power. Basically it doesnt work right. If the XW thinks the battery is at 100%, and load shave amps are set to zero, then you will still be pulling about 100 watts from the grid regardless of your PV yield. Hence my incorporation of a contactor to just override everything and run 100% off the battery. Also you willnotice that when you are grid interactive, the grid draw is very unstable. The values will bounce around quite a bit at about 1 Hz or so. Hence, when I specify the percentages above, it is an average of the bouncing.

    Yeah my system is complex and sometimes it takes me quite a while to figure out what is going on but the way I want my system to work is to #1 keep the batteries happy and #2 maximize solar harvest over grid useage. Consequently I am adjusting certain parameters based on seasonal conditions. There is definately not one set of parameters that will work all the time. Primarily what I adjust the most is the discharge floor on the BVM and the GSV setting but I sometimes have to mess around with the chargers custom battery settings as well and as I said previously I have to do an occasional EQ off the grid. One other think I find I have to occasionally change is enabling / disabling grid support. We have had about 4 days of continuous rain here this week and when this happens, rather than squeaking every watt out of the battery, I just bite the bullet and go on grid. I get a few hundred watts off the PV when it is this bad and I use that to keep the batteries at float but today I am pretty much 100% on grid. This weekend we are supposed to have good sun again and I will reenable grid support. When my BVM reaches 100% SOC, the contactor will open and I will again be 100% off grid.

    Joe
    XW6048, 3.4KW PV, Grid-Tied, always tweaking.
  • Dusty
    Dusty Solar Expert Posts: 271 ✭✭✭
    Re: 6048- reducing sell amps because of a small battery bank--will it work?

    I think I get it. That's why I couldn't expect to output 6KW (even if I had the panels to support it) with such a small battery bank. I'm hoping that by limiting the sell mode to 5 amps, that will keep the ripple to an acceptable level for a 1KW output until I upgrade to a larger battery bank.

    Right now I'm not using a/c at all, so it's theoretically possible that I wouldn't use all my available PV harvest, even with the small setup I have now. I have a small window unit in my attached garage that WILL be one of the output circuits of the 6048, so once I turn on that air conditionaer, I could turn off sell and not be concerned with wasting any harvest. And by the time I get this system set up (I'm trying to be a sub-contractor under a master electrician--brother of my neighbor), that air conditioner will be in full use.

    And once I upgrade my system's firmware so that it supports Enhanced Interactive Mode, hopefully the system will leave the batteries alone and just take whatever excess power is needed off the grid.

    Sorry I didn't reply to your post sooner. I didn't notice it at first. :blush:

    Regards,

    Dusty
    RCinFLA wrote: »
    It's the ripple current that beats up the battery. It is the storage capacitor that gets fed the constant DC current (for given illumination level) from PV controller and delivers the 120 Hz half sinewave shaped pulses of current to the inverter necessary for the 60 Hz single phase AC grid.
    XW6048, 3.4KW PV, Grid-Tied, always tweaking.
  • Dusty
    Dusty Solar Expert Posts: 271 ✭✭✭
    Re: 6048- reducing sell amps because of a small battery bank--will it work?

    Just an update....

    So far, I have had no faults/warnings (other than an occasional F25 fault for grid frequency being too high) with regard to selling to the grid with only a 110 AH battery bank. I have 10 190-watt panels feeding the XW MPPT-60, the XW 6048 GSV is set to 64V to enable Enhanced Interactive Mode, and sell amps is set to 8. When I install an additional 4 190 panels next week, I'll increase sell amps to 10. When my current AGM batteries' life has expired, I'll upgrade to 250AR 6-volt AGMs. So far, so good!
    XW6048, 3.4KW PV, Grid-Tied, always tweaking.
  • Dusty
    Dusty Solar Expert Posts: 271 ✭✭✭
    Re: 6048- reducing sell amps because of a small battery bank--will it work?

    Hi Joe,

    I hope you don't mind my asking another question or two that popped in my head as I re-read your comments.

    Regarding how the XW shares between grid and battery when supporting loads....let's assume for now that Load Sharing is disabled and grid support and sell are enabled. I've received input on a number of threads that grid power is cheaper in the long run when compared to overusing and thus replacing my battery bank earlier than necessary.

    In an ideal world, I'd want to use all available PV power during the day (~8-10Kw on a good day), selling any excess back to the grid for re-use during the night. My normal electrical usage during the day (unless HVAC is running) is about 250 watts, so there is some excess produced during the day--about 3-4Kw on a sunny day. Any surges or electrical use in excess of PV output should come immediately from the grid--leaving the batteries untouched and fully charged in case the grid drops. (I realize that this is not how the XW works--but this is how I would have liked it to work.)

    My question is, why do you want to force the system to use your batteries (via load shave enabled, amps set to zero, and sell disabled) to maximize the use of PV, if the XW's current functionality is to use the batteries so heavily to do so? I understand that you have set up a special system with a battery monitor to prevent dropping battery SOC below your specified level, but even so, won't your batteries wear out sooner and outweigh any cost savings of maximizing PV harvest when the batteries have to be replaced that much sooner?

    Does Enhanced Interactive Mode actually work in sell mode? I have my GSV set to 64v so that the Inverter/charger tracks the MPPT-60 SCC (connected via Xanbus), but since there is so little documentation on this function I have no way of really knowing that it's working.

    I just wish the XW would leave the batteries alone (in float), maximize PV harvest for loads and sell any excess back to the grid for later use at night (allowing me to keep the battery bank small), and only use the grid during the day to augment PV output when needed. Any settings I can adjust on the current XW system to match my ideals as closely as possible is what I'm looking for.

    I've tried load shaving after installing a VICTRON battery monitor. When I got home from work, the PVs were still outputting close to max output, but the battery SOC was 0% according to the monitor readout (I believe that means the batteries were really at 50% SOC). I don't want to do that on a daily basis, so I turned load shaving back off.
    Joe_B wrote: »
    Well you have touched on yet another issue with the XW Dusty, namely the sharing of grid power and battery power. There is a feature called load shaving, normally when running in grid support mode the load is split at around 50% between the grid and the battery. This is very annoying when you have more than enough PV to support the load and the XW is still pulling power from the grid. So load shaving will cause this 50-50 balance to be offset towards the battery but you can never get it to go 100% off grid without physically disconnecting AC1 and running in invert mode. Also, the firmware that controls the balance between battery and grid is VERY slow. It ramps over the course of minutes. There is no mechanism to shift the balance from the battery to the grid.
    XW6048, 3.4KW PV, Grid-Tied, always tweaking.
  • Joe_B
    Joe_B Solar Expert Posts: 318 ✭✭✭✭✭✭
    Re: 6048- reducing sell amps because of a small battery bank--will it work?
    Dusty wrote: »
    Hi Joe,

    I hope you don't mind my asking another question or two that popped in my head as I re-read your comments.

    Regarding how the XW shares between grid and battery when supporting loads....let's assume for now that Load Sharing is disabled and grid support and sell are enabled. I've received input on a number of threads that grid power is cheaper in the long run when compared to overusing and thus replacing my battery bank earlier than necessary.

    No argument there, batteries are expensive and grid power is relatively cheap.
    In an ideal world, I'd want to use all available PV power during the day (~8-10Kw on a good day), selling any excess back to the grid for re-use during the night. My normal electrical usage during the day (unless HVAC is running) is about 250 watts, so there is some excess produced during the day--about 3-4Kw on a sunny day. Any surges or electrical use in excess of PV output should come immediately from the grid--leaving the batteries untouched and fully charged in case the grid drops. (I realize that this is not how the XW works--but this is how I would have liked it to work.)

    Yeah, me too, but it doesnt as we both realize.
    My question is, why do you want to force the system to use your batteries (via load shave enabled, amps set to zero, and sell disabled) to maximize the use of PV, if the XW's current functionality is to use the batteries so heavily to do so? I understand that you have set up a special system with a battery monitor to prevent dropping battery SOC below your specified level, but even so, won't your batteries wear out sooner and outweigh any cost savings of maximizing PV harvest when the batteries have to be replaced that much sooner?

    In order to fully load my panel array with an XW system, it is necessary to be in invert mode. As you know, if you are grid tied, then the system will always pull about 20-30% of its load from the grid. This sucks when you have enough PV to fully run the loads. My solution was to make the system run in invert mode. When doing this, it pulls the maximum from the PV panels, some goes to the loads and some goes to recharging the battery. In my system, any excess over that goes into a dump load to heat DHW. So I never really pull any grid power, I cycle my battery bank down to about 85% over night and the next day, they hit 100%.

    These shallow discharge cycles dont impact the life of the battery bank to any great extent because the SG is never really below 1.250 for any length of time (1.260 is 100% for me). If the SOC of my bank drops below 85% because of cloudy days, then the system will reconnect to the grid to supplement loads. If the grid is down, I still have my bank sitting at 85% to run my loads until the grid comes back, If the grid is not on when my SOC hits about 60%, then the generator will start.
    Does Enhanced Interactive Mode actually work in sell mode? I have my GSV set to 64v so that the Inverter/charger tracks the MPPT-60 SCC (connected via Xanbus), but since there is so little documentation on this function I have no way of really knowing that it's working.

    I believe that it does but I dont run my system is sell mode so I cant be sure.
    I just wish the XW would leave the batteries alone (in float), maximize PV harvest for loads and sell any excess back to the grid for later use at night (allowing me to keep the battery bank small), and only use the grid during the day to augment PV output when needed. Any settings I can adjust on the current XW system to match my ideals as closely as possible is what I'm looking for.

    I beleive that you have your system set up the best it can be given the limitations of the XW.
    I've tried load shaving after installing a VICTRON battery monitor. When I got home from work, the PVs were still outputting close to max output, but the battery SOC was 0% according to the monitor readout (I believe that means the batteries were really at 50% SOC). I don't want to do that on a daily basis, so I turned load shaving back off.

    Something is wrong there Dusty, are you sure that the Victron is set up and installed correctly? Did you manually sync it? Are the parameters set up correctly? I have never seen my Victron say the battery SOC was at 0%. Can you properly display the current and voltage of the bank? By the way, there are a lot of bugs in the load shaving algorithm, I dont ever enable that mode.

    Joe
  • Dusty
    Dusty Solar Expert Posts: 271 ✭✭✭
    Re: 6048- reducing sell amps because of a small battery bank--will it work?
    Joe_B wrote: »
    In order to fully load my panel array with an XW system, it is necessary to be in invert mode. As you know, if you are grid tied, then the system will always pull about 20-30% of its load from the grid. This sucks when you have enough PV to fully run the loads. My solution was to make the system run in invert mode. When doing this, it pulls the maximum from the PV panels, some goes to the loads and some goes to recharging the battery. In my system, any excess over that goes into a dump load to heat DHW. So I never really pull any grid power, I cycle my battery bank down to about 85% over night and the next day, they hit 100%.

    These shallow discharge cycles dont impact the life of the battery bank to any great extent because the SG is never really below 1.250 for any length of time (1.260 is 100% for me). If the SOC of my bank drops below 85% because of cloudy days, then the system will reconnect to the grid to supplement loads. If the grid is down, I still have my bank sitting at 85% to run my loads until the grid comes back, If the grid is not on when my SOC hits about 60%, then the generator will start.

    Ok, that answers my question about battery life expectancy. If you don't lower SOC below 85%, there's little change in battery life.

    I believe that it does but I dont run my system is sell mode so I cant be sure.

    When you mentioned always getting 20-30% from the grid, I wondered about Sell Mode. Since I'm selling back to the grid, am I actually selling 20-30% less than I could be because of the balancing issue? It sounds like I am. Since you don't use Sell Mode, this is a rhetorical question. I was just thinking that with Sell enabled, I'm maxing out the panels at all times to sell any excess back to the grid.

    I believe that you have your system set up the best it can be given the limitations of the XW.

    Let's hope that future firmware updates improve this!

    Something is wrong there Dusty, are you sure that the Victron is set up and installed correctly? Did you manually sync it? Are the parameters set up correctly? I have never seen my Victron say the battery SOC was at 0%. Can you properly display the current and voltage of the bank? By the way, there are a lot of bugs in the load shaving algorithm, I dont ever enable that mode.

    I did manually synch it. I seem to recall that the instructions for SOC said that the meter bases SOC on the DOC desired. I'll have to read that again. I'm pretty sure I set it for 40% DOD when I first configured it, so the SOC of 0 would really mean it was down to 60%. Since I only have a 110AH battery bank compared to yours, I'm not surprised you've never seen yours go to zero! :roll:

    Regarding Load Shave, my mistake in thinking that it was you that used that feature. Must have been someone else that recommended that by setting it to zero, you'd reduced the grid usage to about 10% instead of 20-30%. However, it drives my SOC too low for my liking, so I no longer use it.
    XW6048, 3.4KW PV, Grid-Tied, always tweaking.
  • Joe_B
    Joe_B Solar Expert Posts: 318 ✭✭✭✭✭✭
    Re: 6048- reducing sell amps because of a small battery bank--will it work?
    Dusty wrote: »
    Ok, that answers my question about battery life expectancy. If you don't lower SOC below 85%, there's little change in battery life.

    I got this from Trojan FLA battery techs, the battery life depends on two (main) things, Depth of discharge / number of cycles and how long they are left in a discharged state. The latter is what sulfates the plates, when the SG drops, the sulfur in the electrolyte can deposite on the plates and essentially insulate them. Of course this is a simplistic view but according to the people at Trojan, these are the worst two mechanisms that can decrease FLA battery life.

    When you mentioned always getting 20-30% from the grid, I wondered about Sell Mode. Since I'm selling back to the grid, am I actually selling 20-30% less than I could be because of the balancing issue? It sounds like I am. Since you don't use Sell Mode, this is a rhetorical question. I was just thinking that with Sell enabled, I'm maxing out the panels at all times to sell any excess back to the grid.

    Like I said, I dont do sell but in non sell mode, it sucks from the grid.

    Let's hope that future firmware updates improve this!

    Dont hold your breath, I have been waiting for a firmware update for over a year. I pointed out a severe bug to them, they agreed it was a problem, and I have not heard a thing since. I call them about once a month and get the same answer; "Keep checking the website for the fix". I wont be holding my breath...

    I did manually synch it. I seem to recall that the instructions for SOC said that the meter bases SOC on the DOC desired. I'll have to read that again. I'm pretty sure I set it for 40% DOD when I first configured it, so the SOC of 0 would really mean it was down to 60%. Since I only have a 110AH battery bank compared to yours, I'm not surprised you've never seen yours go to zero! :roll:

    No, SOC is based on the size of the bank you program into it and the ampunt of amp hours you put in or pull out. The TTG calculation is dependent on the discharge floor. The SOC will go to zero if you set it 110 AH and pull that amount out but if your bank was really at 100% and the meter was reading zero SOC, there is an issue. Did you set up the charge efficiency and peukert compensation for AGM?
    Regarding Load Shave, my mistake in thinking that it was you that used that feature. Must have been someone else that recommended that by setting it to zero, you'd reduced the grid usage to about 10% instead of 20-30%. However, it drives my SOC too low for my liking, so I no longer use it.

    I have experimented with load shave and elected not to use it due to bugs. If you can get the unit to go into load shave and stay there, then you will see the grid draw approach 10% at times. On my system, the unit will not stay in load shave mode for more than a few minutes at a time and therefore my grid draw in grid tie mode is bouncing between 20 and 30%. The only way I can get my system to use zero grid power is to physically disconnect AC1 from the grid with a contactor. Its just me, I hate to be drawing 200-300 watts from the grid when my PV is capable of over 2.5KW. According to the techs at Schneider, if I could get it into load shave, it would still draw 80-100 watts from the grid at all times, that IMO is just stupid. They have also acknowleged to me that the load shave mode has issues and there are no immediate plans to fix it. :roll:
  • Dusty
    Dusty Solar Expert Posts: 271 ✭✭✭
    Re: 6048- reducing sell amps because of a small battery bank--will it work?
    Joe_B wrote: »

    Dont hold your breath, I have been waiting for a firmware update for over a year. I pointed out a severe bug to them, they agreed it was a problem, and I have not heard a thing since. I call them about once a month and get the same answer; "Keep checking the website for the fix". I wont be holding my breath...

    I held onto the Dongle just in case....


    No, SOC is based on the size of the bank you program into it and the ampunt of amp hours you put in or pull out. The TTG calculation is dependent on the discharge floor. The SOC will go to zero if you set it 110 AH and pull that amount out but if your bank was really at 100% and the meter was reading zero SOC, there is an issue. Did you set up the charge efficiency and peukert compensation for AGM?

    Checking the manual, What I thought was setting DOD was something called "discharge floor," but it appears to only be used to set a relay. The default was 50, but I'm pretty sure I set it to 60.

    I'm using 2 banks of 55AH Optima D34/78 "yellow top" AGM batteries. I do not have the efficiency or Peukert compensation for these batteries. I only saw a SOC of zero once when I was using Load Shave. Perhaps I didn't do the manual synch correctly. I'm relocating the meter this weekend, so I'll have to do another manual synch anyway.

    I think I was remembering the Xantrex LinkLite settings.....Unfortunately, that meter is only good for 12/24V systems. What a shame. I liked that meter. It had a "fuel gauge" that showed when I was approaching the DOD I had programmed into it and would alarm if I reached the DOD I selected.

    I have experimented with load shave and elected not to use it due to bugs. If you can get the unit to go into load shave and stay there, then you will see the grid draw approach 10% at times. On my system, the unit will not stay in load shave mode for more than a few minutes at a time and therefore my grid draw in grid tie mode is bouncing between 20 and 30%. The only way I can get my system to use zero grid power is to physically disconnect AC1 from the grid with a contactor. Its just me, I hate to be drawing 200-300 watts from the grid when my PV is capable of over 2.5KW. According to the techs at Schneider, if I could get it into load shave, it would still draw 80-100 watts from the grid at all times, that IMO is just stupid. They have also acknowleged to me that the load shave mode has issues and there are no immediate plans to fix it. :roll:

    Well, one can only hope that they monitor this forum for ideas to upgrade Firmware to what functionality the customers are looking for. :roll:
    XW6048, 3.4KW PV, Grid-Tied, always tweaking.
  • Dusty
    Dusty Solar Expert Posts: 271 ✭✭✭
    Re: 6048- reducing sell amps because of a small battery bank--will it work?

    Joe,

    Reading through the manual for the Victron BMV600, I cannot find a procedure for manually synching the unit. It seems like it automatically synchs when the battery charger is in Float. I'll make sure the SCC float setting is in accordance with the mfr (13.2-13.8 is an appropriate float voltage for these AGM batteries, according to the MFR spec sheet, so I'll set the SCC for 54.0V. I've sent Optima an email to ask about efficiency and Peukert compensation values).

    My LinkLite had a manual synch button sequence, but I do not see anything like that for the Victron. Is there a way to do it that's not in the documentation?

    -Dusty
    XW6048, 3.4KW PV, Grid-Tied, always tweaking.
  • Blackcherry04
    Blackcherry04 Solar Expert Posts: 2,490 ✭✭✭
    Re: 6048- reducing sell amps because of a small battery bank--will it work?

    My experience with Optima Batteries is that they will always draw on a charger. I can put a charger on a regular AGM battery and it will sit there with zero amp draw and put the same charger and clip in onto a Optima and it will be constantly pushing back into the Optimas 24/7. There is something in their chemistry that causes them to it.