Pocster Posted July 28, 2022 Share Posted July 28, 2022 Yes . I suspected issues . Now they are confirmed. Best to share the rewards and the pain ! PW charges off SE ( correct I think ) Then greedy SE steals it back - around the loop we go . Even when PW charges from 00:30 to 04:30 SE gets ready to steal !! Talking to installer as we speak ; also notice SE app doing strange ( incorrect ) things . The lovely effect of cloud apps 🙄😞 1 Link to comment Share on other sites More sharing options...
Pocster Posted July 29, 2022 Author Share Posted July 29, 2022 The SE app has been strange recently. SE is set to only charge from solar and not to discharge between 00:30 and 04:30 otherwise PW grabs it all . This bit works . What SE is doing now I have no idea …. I suspect some firmware balls up on SE side … Link to comment Share on other sites More sharing options...
Thorfun Posted July 29, 2022 Share Posted July 29, 2022 the trials and tribulations of trailblazers. hope you get it all sorted. 1 Link to comment Share on other sites More sharing options...
SteamyTea Posted July 29, 2022 Share Posted July 29, 2022 You need some blocking diodes. Everything s put right with a pocket full of diodes. Link to comment Share on other sites More sharing options...
Pocster Posted July 29, 2022 Author Share Posted July 29, 2022 Apparently installer contacted SE and there are no known issues . So ! I’ll drain the PW into my EV when it’s low . I’ll do this deliberately at the end of the day so SE and PW have no chance of recovering via PV . I’m assuming the same will happen . Link to comment Share on other sites More sharing options...
Pocster Posted July 30, 2022 Author Share Posted July 30, 2022 Right ! I’ve been thinking about this ….. I think the default SE option of “ maximise charge/discharge” is incorrect in this situation. Essentially ( I believe ) it means the SE will charge / discharge when it wants . This I’m 100% certain as I have proof means it will steal from the PW for its own needs . The solution to this is set a discharge only option . But that means it can’t charge from PV . So it might require that it charges from PV for most of the day then I change it to discharge only at a certain time . This should stop all the stealing I think . I need to empty SE and PW to test this . For winter I assume telling SE to charge only from the grid at off peak means ( hopefully ) the PW won’t steal from it … Makes sense in my head . Update when theory proven or disproven ! Link to comment Share on other sites More sharing options...
Pocster Posted July 31, 2022 Author Share Posted July 31, 2022 SWMBO is in with youngest grandson so we are eating kw like there’s no tomorrow. I notice in Tesla app our consumption is 3kwh. This is not possible . SE says solar input is 3kwh . Equally not possible ( SE at 97% ) . It’s stealing from PW !! Quickly go in SE online app set it to discharge only - house load drops to 0.4kwh . There proof !!! Might set SE and PW to both from grid tonight at off peak and see what happens … Link to comment Share on other sites More sharing options...
RobLe Posted July 31, 2022 Share Posted July 31, 2022 A friend of mine had I think a similar, but simpler problem. He had V2H and also a solic 200 diverter. With no sun, the solic should do nothing, and V2H supply the house. What actually happened is that the solic sometimes turned on and ‘stole’ all the V2H energy, everything ramping up to full power. The solic and the V2H had slightly different ideas about what the power flow into the house was, causing the issue. We fixed it by bodging 10 thin turns of wire around the solic sensor in addition to the mains tail, with 100mA ac flowing in it from an (mains)ac-(selv)ac adapter, convincing the solic that there was 1A*240V less available power than reality! The above fix only works because the solic takes power only, and (obviously!) can’t deliver electrical power. Perhaps you could see if either the PW or SE(another home batt i assume) have adjustable dead band regions? That is, if one of the units allowed power flow in either direction of less than 100W or so, then the issue might resolve. The bodge we did will not work for you, if both units can have bidirectional powerflow; you need a software fix, or a more complex bodge. 1 Link to comment Share on other sites More sharing options...
SteamyTea Posted July 31, 2022 Share Posted July 31, 2022 27 minutes ago, pocster said: SE says solar input is 3kwh Link to comment Share on other sites More sharing options...
Ferdinand Posted July 31, 2022 Share Posted July 31, 2022 What's PW, and aren't you in the SW not the SE? Explication required. The only PW acronym I am aware of is the one that Brenda at the Samaritans used to use - some years ago. 1 Link to comment Share on other sites More sharing options...
SteamyTea Posted July 31, 2022 Share Posted July 31, 2022 1 minute ago, Ferdinand said: aren't you in the SW not the SE Closer to London than I am to Exeter. Link to comment Share on other sites More sharing options...
Pocster Posted July 31, 2022 Author Share Posted July 31, 2022 42 minutes ago, Ferdinand said: What's PW, and aren't you in the SW not the SE? Explication required. The only PW acronym I am aware of is the one that Brenda at the Samaritans used to use - some years ago. POwer Wall Solar Edge 😁 1 Link to comment Share on other sites More sharing options...
Pocster Posted July 31, 2022 Author Share Posted July 31, 2022 (edited) 1 hour ago, RobLe said: Perhaps you could see if either the PW or SE(another home batt i assume) have adjustable dead band regions? That is, if one of the units allowed power flow in either direction of less than 100W or so, then the issue might resolve. The bodge we did will not work for you, if both units can have bidirectional powerflow; you need a software fix, or a more complex bodge. PowerWall has very limited settings because it’s algorithm does what It wants ! SolarEdge is dumb ( good ! ) so I can set times to charge / discharge etc. It’s default state is “ maximise self consumption and charge / discharge as needed “ . Fine in a normal situation. Here though of course it sees the PW and grabs it . PW then sees SE and grabs it . Installer is oblivious to this issue . I’ve proven it happens so I can stop the yo yo loop . What I’m not sure about is when both need the grid to charge . PW does what it wants ( and I know if it sees the SE in normal mode it will drain it ). So I’m hoping if SE is set to “charge from grid/solar “ it can’t discharge hence PW won’t see it and therefore will use the grid …. Edited July 31, 2022 by pocster Link to comment Share on other sites More sharing options...
Pocster Posted July 31, 2022 Author Share Posted July 31, 2022 (edited) Actually a question PW can charge at max rate of 3.66kwh. So needs 4 hours to fully charge . I can’t find a max rate of charge for SE . Because it’s DC will the affect its rate of charge ??? . Google suggests DC charging is faster but no explanation of how you can compute the rate of charge and hence time required . Edited July 31, 2022 by pocster Link to comment Share on other sites More sharing options...
Pocster Posted July 31, 2022 Author Share Posted July 31, 2022 SE was 97% when I set it to discharge only . It’s now 100% ! 🙄🙄🙄🙄🤔🤔🤔🤔🤔🤔🤔🤔 Link to comment Share on other sites More sharing options...
Pocster Posted July 31, 2022 Author Share Posted July 31, 2022 11 minutes ago, pocster said: SE was 97% when I set it to discharge only . It’s now 100% ! 🙄🙄🙄🙄🤔🤔🤔🤔🤔🤔🤔🤔 Actually re reading docs it can pV charge whilst in discharge . Link to comment Share on other sites More sharing options...
Ferdinand Posted July 31, 2022 Share Posted July 31, 2022 22 minutes ago, pocster said: Actually a question PW can charge at max rate of 3.66kwh. So needs 4 hours to fully charge . I can’t find a max rate of charge for SE . Because it’s DC will the affect its rate of charge ??? . Google suggests DC charging is faster but no explanation of how you can compute the rate of charge and hence time required . "Rate of 3.66kWh" FFS ! 1 Link to comment Share on other sites More sharing options...
RobLe Posted July 31, 2022 Share Posted July 31, 2022 So both PW and SE are home batts, and there's no useful software settings on them to prevent interaction? I think the issue is that the combined system is underconstrained. PW presumably has a current clamp on the main Live tail, and is attempting to make the current there zero, using a software feedback loop with likely a PI(proportional + Integral) controller. SE is presumably the same. While you might think they would both sit there doing a similar thing - slowly feeding on PV say, an equally valid solution is for one to discharge into the other, while still overall feeding on PV. I think if the PW clamp were after PW+House+PV, then SE clamp after PW+House+PV+SE that this stops the PW responding to the SE as it can't "see" that current flow. The SE can still respond to the PW - but the overall feedback loop is broken, so I think this might fix it 🙂 One of the home batts will be exercised much more than the other doing this - the way around I described it, the PW will do all of the work until it cannot, being full or empty or having too high power flow, then the SE will mop up the rest. Disclaimer: I'm just a diy-er with ideas above my station 🙂 2 Link to comment Share on other sites More sharing options...
Pocster Posted July 31, 2022 Author Share Posted July 31, 2022 28 minutes ago, RobLe said: So both PW and SE are home batts, and there's no useful software settings on them to prevent interaction? I think the issue is that the combined system is underconstrained. PW presumably has a current clamp on the main Live tail, and is attempting to make the current there zero, using a software feedback loop with likely a PI(proportional + Integral) controller. SE is presumably the same. While you might think they would both sit there doing a similar thing - slowly feeding on PV say, an equally valid solution is for one to discharge into the other, while still overall feeding on PV. I think if the PW clamp were after PW+House+PV, then SE clamp after PW+House+PV+SE that this stops the PW responding to the SE as it can't "see" that current flow. The SE can still respond to the PW - but the overall feedback loop is broken, so I think this might fix it 🙂 One of the home batts will be exercised much more than the other doing this - the way around I described it, the PW will do all of the work until it cannot, being full or empty or having too high power flow, then the SE will mop up the rest. Disclaimer: I'm just a diy-er with ideas above my station 🙂 Did wonder about the location of the CT clamps . Will try my experiment first 😁 Link to comment Share on other sites More sharing options...
Pocster Posted July 31, 2022 Author Share Posted July 31, 2022 29 minutes ago, RobLe said: One of the home batts will be exercised much more than the other doing this - the way around I described it, the PW will do all of the work until it cannot, being full or empty or having too high power flow, then the SE will mop up the rest. This is exactly how I expected it to work . SE slave to PW Link to comment Share on other sites More sharing options...
kommando Posted July 31, 2022 Share Posted July 31, 2022 Rob is on to something, here is a similar issue with competing surplus diversion boxes getting their knickers in a twist. https://community.openenergymonitor.org/t/arduino-based-mk2-pv-diverter-being-resurrected/11713/2 To quote from the thread The user whom I helped had just two systems - I think battery and water heating, and his problem was the two systems oscillated, stealing power from each other alternately. I suggested that he ran the wire feeding his immersion heater backwards through the grid c.t. of the other system, so its current in the main cable was cancelled out. So it knew nothing about the immersion heater, while the immersion PV system saw the battery charger as part of the normal house load, which automatically had priority. That solved the problem. 1 1 Link to comment Share on other sites More sharing options...
MikeSharp01 Posted July 31, 2022 Share Posted July 31, 2022 1 hour ago, RobLe said: Disclaimer: I'm just a diy-er with ideas above my station 🙂 Not sure it sounds like that! Link to comment Share on other sites More sharing options...
MikeSharp01 Posted July 31, 2022 Share Posted July 31, 2022 Firstly can I admire and respect your wish to give this setup a try. It feels relatively obvious that unless they were designed to, or can by use of their settings be made to, work symbiotically that they will work parasitically - everything tends to chaos. Both the PW & SE assumes that they are "the only battery in the village" and behaves accordingly. Which does what to which probably comes down to the sensitivity of the respective front end - IE what they think they see happening on the wires and a fractional difference in interpretation will be enough to upset the applecart. As I see it you will need some pretty tricky control electronics, perhaps a sort of rules based MQTT broker like device (function of not form of) to get these two talking to one another and their respective PV arrays nicely - have a word with @Radian cos he seems to be the master of this sort of stuff! 1 Link to comment Share on other sites More sharing options...
PeterW Posted July 31, 2022 Share Posted July 31, 2022 Surely just buying a second PW would have been easier ..?? Link to comment Share on other sites More sharing options...
Pocster Posted July 31, 2022 Author Share Posted July 31, 2022 6 minutes ago, PeterW said: Surely just buying a second PW would have been easier ..?? No availability. Also require 3 phase ( 10k for me !! ) Link to comment Share on other sites More sharing options...
Recommended Posts
Create an account or sign in to comment
You need to be a member in order to leave a comment
Create an account
Sign up for a new account in our community. It's easy!
Register a new accountSign in
Already have an account? Sign in here.
Sign In Now