cpbotha

I just wanted to add my datapoint: Discovered that my GoodWe GW5048D-ES firmware version 1242416 (aka 2424G) was not submitting data to SEMS, then checked but SolarWIFI*** access point completely absent, and WiFi light on the front completely dead. Tried wifi reset (short press button behind meter panel), reload (longer than 3s press) and then even removing the WiFi module completely and keeping disconnected for a minute or more: WiFi light remained dead. Only when I did the full shutdown and cold start (change-over to grid, PV off, grid supply off, ES rotary power switch off, battery disconnect, wait until system dead, reverse process, thanks @Bobster. in thread below) did the wifi fortunately come fully back up, with solidly burning light and data going to SEMS.

Update: Thanks to the GeyserWise manual switches I have just now established that indeed one of the geysers was accidentally moved to the backup circuit during the 60A -> 80A upgrade work. One of the uber patient electricians will hopefully remedy this tomorrow. By the way, GoodWe with BMS and correct SOC works substantially better than out.

I would like to report back on my post-March 2019 model GoodWe GW5048-ES + PylonTech US3000*2 adventures: It turned out that firmware 151507 had a bug w.r.t. the BMS communication with these batteries. After a quite painful process of getting the thing upgraded, we the BMS communication working perfectly. Here also Segen was super helpful. Notes to others: Although the manual says explicitly that one should NOT use EzFlash on post-March 2019 versions of this inverter, and only to use DataSend, the advice we got from Segen was that sometimes, if a flashing procedure went bad, one HAS to do the whole procedure: EzFlash master.out, EzFlash slave.out, datasend .bin, wifi firmware upgrade, reset to factory. In my case, this took the inverter from red fault led always on, to an almost 100% working state. I say almost, because at the moment, the WiFi went a bit wonky (it's mostly up, but sometimes mysteriously down), and the PV Master app is reporting my non-backup-circuit loads as being on the backup circuit.

Congratulations TTT, well done!!

How sure are you about this w.r.t. the newer models? I can find mention of the EzConverter in some of the older manuals, but it's not mentioned in the latest GoodWe ES literature at all. I would like to be corrected. My 9-series S/N GoodWe 5084-ES is not talking BMS to my US3000s yet.

That's indeed what I'm doing at the moment, but of course I would prefer the BMS to work for better charging and better SOC reporting. I've sent GoodWe support an email with details. I will of course report back here if I make any progress.

You can add the US3000, but it has to be the new master battery, i.e. its "link 0" socket is empty, and its link 1 is connected to the link 0 of your second battery, and so on.

That would be a post by @Youda let's see if I can link it now: https://powerforum.co.za/topic/2734-axpert-settings-vir-pylontech-35-kw/?tab=comments#comment-52016

This GoodWe officially supports the US3000, at least in the PV master app there's a whole section for US3000 batteries. Have tried three different BMS cables: the one that comes attached to the goodwe, the one that comes with the pylontech cable pack, and a separate patch cable I usually use to connect one of my laptops to ethernet hub. No dice. Will consider trying crossover tomorrow... It does look like the dip switch settings were wrong. According to the below post by Yoda, and further readings of the manual, dip switches on both pylontechs should be all 0. However, doing this and powercycling the batteries does not seem to have helped. I'll open op the GoodWe again tomorrow morning to make sure the BMS cable is securely in the internal RJ45 socket. (ARGH ARGH ARH why is sitelock blocking my replies?! I was quoting a post by Yoda on this site, and then sitelock blocked my replies numerous times. Really?)

We're having similar problems with a just-installed newer GoodWe with explicit support for US3000. GoodWe 4.6 ES (beige) with 2 x US3000: tried the cable delivered with the goodwe, tried another ethernet cable confirmed working with laptop, tried the cable delivered with the PylonTech connection kit: Battery Communication Fail. @Niel let's keep each other up to date.

@Rautenk can you shed some more light on #2? My application for the 80A was approved, I've paid the people, and work is about to be scheduled. I didn't mention anything about SSEG on my application, although it is in fact the plan to integrate the GoodWe 4.6kVA ES. However, there are 6 people living on my premises, so maybe that helped the request?

Not exactly. CoCT also defines grid-tied as "connected to the utility grid either directly or through a consumer’s internal wiring". For the sake of these definitions, even if your inverter is for some or other reason not capable to feed back into the grid, but it's still connected to the grid in order to be able to use incoming AC, it is still classified as grid-tied.

I'm just trying to keep things in perspective. If this is the same prepaid meter than the one CoCT would have put in any way (at my cost), then indeed the upgrade cost is "only" the R5k SNC. In the end we pay, that we don't have much control over. We do have some degree of control over how we feel about that. (that's true about many things in life.

I have asked the CoCT to explain what the shared network charge (SNC) is for. Also, RED Engineering (my amazing solar installer) warned that CoCT would replace my meter with a prepaid one in any case, even if I didn't request the 80A upgrade, simply due to the SSEG system.

After more than a month of waiting, and repeated contact with CoCT (during these occasions they were always friendly and helpful, the whole process just takes so long), I finally received my quote for the 60A to 80A upgrade. The quote is identical to that of @Niel: 7231.95 in total, consisting of the R5k shared network charge, and the single phase split prepayment meter. (the wording is also identical) I did not mention *anything* about SSEG in my application, I only requested the 60A to 80A upgrade. My only reason for doing this is indeed the sign-off of a soon-to-be-installed GoodWe 4.6kVA inverter. The planned system is zero feed-in. I am going to pay, because I want to have my system signed-off, but this does leave a bitter taste in one's month.

I would appreciate some of the more experienced members chiming in here. While we wait for them, allow me to contribute my 2 cents: I also thought that monos were still better at higher temps, but after having looked at the specifications of modern polys, such as the Canadian Solar CS3U PolyGEN4 modules, I'm also starting to think that polys have closed the gap and perhaps even overtaken many of the monos. I think the general message is to interpret each product's specifications and make your decision based on those performance characteristics, not only on whether they're polys or monos. If you can get a poly panel with the desirable energy density and an energy coefficient better than the monos you have on your list, go for the polys, and vice versa.

That's just a sign that you should rather spend your time here and on the Victron forums where you are unreservedly appreciated.

Very nice, thank you! Does your design include batteries? Because in theory, your 5kW inverter could then use the estimated 3.5kW from your 4.4kW PV, and some watts from the storage system to generate more than the 3.5kVA limit. I am asking this question to try and understand how well (or not) CoCT interpreted your response. If you have no batteries in your design, then it seems OK, but if you do, it looks like it could be an oversight on their part. P.S. or it's semantics. PV is technically generation, whilst batteries are not generation. However, together they can still mean your setup inserts more than 3.5kVA into your AC wiring system, as per @Rautenk's definition.

This is fantastic news @Fuenkli congratulations! How exactly did you argue the point, did you just reply to their email, or did you go and see someone in person? With regard to your intelocked change-over switch: This was coincidentally discussed on the first page of this long and valuable thread. See for example https://powerforum.co.za/topic/3029-anyone-like-to-look-over-my-proposed-sseg-application-and-diagram/page/2/?tab=comments#comment-47281 and the posts just after it. Quoting @Bulldog: "The changeover switch selects your loads to be connected either to grid or to the inverter. It must be mechanically interlocking which means it is not possible to select both grid and inverter even in the case of a fused contact. A switch with a "all off" position is also acceptable. For purposes of official interpretation you will be operating your system either on grid or on inverter using a manual changeover switch."

I think we have moved far beyond the point of what the correct interpretation is (because CoCT also does not know) and more into the regions of what will get signed off. Also in the definition that you quote, there is still the utter mystery of what is meant by "the storage can be used to REDUCE the effective size of the generator", only underlining my observation above.

@Elbow thank you for posting this! (I did read through everything, but more from the standpoint of learning from what you wrote. I have that form partially completed in my solar project folder also.

I continued the conversation with GoodWe, and now have more clarity. The 27.8% oversizing they claim, is on the nominal PV input of 3600W, bringing the maximum to 4600W. In other words, the maximum PV input they state in the specs, already has the oversizing calculated in. The 3600W nominal PV input is not mentioned in the manual anywhere, but can be derived from the 36 in the model number 3648.

Hi there @GreenFields - have we recently been on the phone by any chance? (green fields solar?) I would like to underline @plonkster's point -- even if you have perfect reverse flow blocking (which does not exist), if your generator (inverter in this case) trips, all of a sudden the loads you had on your own inverter are now transferred to the grid, so your house is suddenly pulling e.g. 3.5kW of extra power FROM the grid. This is why the NRS definitions mention specifically "instantaneous change in power flow" and "generator trips": This goes in both directions, and they want to ensure that the delta in both directions can never be more than 25% of the supply All of that being said: Things are not looking bright even for my GoodWe 3648 plans. Judging by the answer @Fuenkli got, they might even throw out the 3.6kW Goodwe, because it's over 3.5...

Hi there @Rautenk - could you add batteries to this example to try understand also the effect of the storage on the generation limits that is mentioned in the NRS definitions?

Hi there @The Terrible Triplett - when you completed this form in Oct / Nov, did you fill in your own details under D. INSTALLER DETAILS AND DECLARATION, or were you working together with someone at that point? I'm considering doing this myself as well, as I can't seem to find an installer who has time to help me at the moment. (I am exploring the idea of finding a talented electrician and roof climber (for example) so we can install the whole thing together.)