PancakeMan Posted June 30, 2022 Share Posted June 30, 2022 Hi Power forum, First time poster here. Recently installed an RCT AXPERT MAX 8KVA/8KW with 3x pylontech US300C batteries and 18x395w Jinko panes (2 strings 10/8). Everything is running well, except for some strange issues with regard to "Blended" power from Solar/Grid Mode. I am seeing inverter temperatures above 70C. The inverter is located undercover with ample air at cool winter temperatures <20C. Has anyone else seen issues like this with their inverters? Iv added a picture of todays test although this happens every time the inverter goes into Solar/Grid Mode. As soon as I manually change it to Solar/Battery mode temps get much better. Any insights would be great. The inverter is ~1 month old. The_Duk3_ 1 Quote Link to comment Share on other sites More sharing options...
Steve87 Posted June 30, 2022 Share Posted June 30, 2022 Do you have picture of this installation to see exactly how this inverter is placed on the site. Jacques Ester 1 Quote Link to comment Share on other sites More sharing options...
WannabeSolarSparky Posted June 30, 2022 Share Posted June 30, 2022 (edited) The spikes of temp coincides with the load spikes, which means the inverter is doing heavy duty inverting Usually your fans should ramp up at the same time too, which by the looks of your graphs is exactly what is happening. The graphs are very useful, inverting solar to ac does generate a lot of heat. On the graph you can see when the grid kicks to help the load spikes then the temp drops as it is not having to work the mppts as hard. That all looks pretty normal AFAIK. Edited June 30, 2022 by WannabeSolarSparky Quote Link to comment Share on other sites More sharing options...
PancakeMan Posted June 30, 2022 Author Share Posted June 30, 2022 Thanks @Steve87 and @WannabeSolarSparky . Please find the pic of the setup attached. Awesome thanks @WannabeSolarSparky good to know. Any reason why we don't see the same temperatures when it is running in SBU. Is the AC-AC with solar DC-AC really that inefficient. It seems like the DC-AC only is a lot more efficient. Could the losses have something to do with synchronizing with the grid? The_Duk3_ 1 Quote Link to comment Share on other sites More sharing options...
PancakeMan Posted June 30, 2022 Author Share Posted June 30, 2022 (edited) Here is another snapshot. Running SBU from 10:26 with much greater loads(upto 8kW), hotter ambient temperatures and PV power close to 5kW, and the inverter didn't move once over 60C. It just seems like the SUB mode is flawed. I will run a test tonight in SUB more and see if there is still the same issue without solar production. Edited June 30, 2022 by PancakeMan The_Duk3_ 1 Quote Link to comment Share on other sites More sharing options...
Coulomb Posted July 1, 2022 Share Posted July 1, 2022 21 hours ago, PancakeMan said: Could the losses have something to do with synchronizing with the grid? No, I don't believe so. Synchronising is "merely" a matter of adjusting the inverter output frequency (and hence phase when the frequency is close) by very small amounts. All that happens in the end is that the period value in a few DSP registers gets changed by a small value. This would not affect the power losses materially; the transistors are still switching at the same frequency. I suppose that if the frequency of the grid is very jittery, then the control algorithm might not be able to catch up for a significant part of the time, which might cause power to spike in one direction or the other more frequently. But it seems unlikely to me that this would have a noticeable effect on losses and hence on temperature. I've worked on code to perform synchronisation in the past, but I've not worked with South Africa utility power (Eskom). Quote Link to comment Share on other sites More sharing options...
PancakeMan Posted July 1, 2022 Author Share Posted July 1, 2022 Thank you @Coulomb this is very insightful. Im really lost for reasons why SUB creates so much heat. When running SUB in the evenings I get low temperatures. Quote Link to comment Share on other sites More sharing options...
PancakeMan Posted July 1, 2022 Author Share Posted July 1, 2022 Looking at an hour segment the frequency does seem quite jittery. Although I don't think too different from everyone else connected to ESKOM. Has anyone seem similar issues with their Axpert inverters running SUB during solar production times? Quote Link to comment Share on other sites More sharing options...
jumper Posted July 1, 2022 Share Posted July 1, 2022 Could this maybe have something to do with you battery SOC being lower in the morning? The erratic temps seem to also happen when there is a dip in PV with a load and maybe a "flatter" battery causes more AC pull to compensate and higher temps to recharge until later on when the battery is charged (not sure how long they take). Would be interesting to see this plotted with battery voltage or SOC. Quote Link to comment Share on other sites More sharing options...
Coulomb Posted July 2, 2022 Share Posted July 2, 2022 6 hours ago, jumper said: Could this maybe have something to do with you battery SOC being lower in the morning? Brilliant! I never thought about the implications of the low transformer ratio when in SUB mode. To cater for the 0.1% of users with Trojan flooded batteries which require charging to some 60 V and occasionally to some 64 V, they changed the turns ratio of the transformer from 1:8 to 1:7. As a result, the maximum battery voltage increased from 58.4 V to 64.0 V clips when the battery voltage is below about 48.1 V. That's 3.2 VPC on a 15S LFP battery. Under load, this could be a perfectly normal SOC, especially in the morning. I posted about this on the AEVA forum. So that means that the tops of the generated AC waveform are clipped. That's bad enough, causing heating in motors and other problems. But in SUB mode when blending with utility power, it means that at the peaks of the utility waveform, suddenly a lot of power is flowing back towards the battery. It's only for a small part of the cycle, though the lower the battery voltage, the longer this happens, and the higher the current back to the battery. These high peak currents will cause much higher I²R losses, and hence more heating. I think it was a huge mistake for Voltronic Power to change all the later models to 64 V maximum voltage, without offering an alternative. Add to that the unfortunate fact that battery voltages are on average a fair bit lower than they were a few years ago, because of the proliferation of 15S LFP and even worse the 13S NMC batteries, and we have this situation. At least the Axpert MKS is still available (or it was, I hope it still is) in a 58.4 V model. Voltronic have made this even worse by not stating the maximum battery voltage on the barcode stickers, and the resellers often don't include it in their model naming, so you can't be sure what you're ordering. Unfortunately, apart from changing to a 16S LFP battery or going back to the dark ages of lead acid, I can't see any way around this issue. Of course, all this only applies if I'm right about this theory. @PancakeMan, what is your battery voltage around 9am when the temperature is peaking? jumper 1 Quote Link to comment Share on other sites More sharing options...
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.