diy solar

diy solar

WWWS (what would Will Say) EG4 vs SOK server rack battery

Yep I hit into that same wall. I had already bought a Sunny Boy and was in the process of getting the rest of the equipment and bailed out due to the vague and complex nature of the install.
I went with Sol-Ark instead and I am so happy I did. It's just so much simpler to setup and has a lot more features.
SMA stuff is probably the most robust systems on the Market but they are so complex to setup and the customer support is not geared at all towards DIY.
I am also toying with the idea of keeping the Sunnyboys and going with brand new forklift batteries. KISS. I have REC BMS parts in order but they aren’t here yet at the vendor and I could cancel.

Probably 20-25 year life with a more simple setup albeit with a tad bit more maintenance. And they are somewhat rebuildable.
 
FLA doesn’t really bother me…parents have been running off of cheap Sam’’s Club deep cycle batteries for 25 years on a old Trace and Outback 12 volt system.
 
I am also toying with the idea of keeping the Sunnyboys and going with brand new forklift batteries. KISS. I have REC BMS parts in order but they aren’t here yet at the vendor and I could cancel.

Probably 20-25 year life with a more simple setup albeit with a tad bit more maintenance. And they are somewhat rebuildable.
Well if you do go with SMA just make sure that forum member "Hedges" is your best friend :)
That guy knows more about SMA equipment than all the SMA USA Techs I talked to.
I got so tired of hearing "We will have to call Germany and get back to you"
In all honesty it was during early 2021 so maybe the senior staff had stayed home due to Covid.
 
after reading this whole thread I can only come t the conclusion....yep magnum 4448 pae's are ugly... old tech and the company that bought them out (sensata) are a$$holes, but the gear itself is pretty solid and it just works when it comes to making AC from DC.
 
Previous versions of SOK tools (the pc programming software) only had 5 options for protocols. The new version that came out Yesterday now has far more, so I'm back to the drawing board testing SMA, considering the software now has a button that says SMA. This likely solves the problem, but will test it ASAP. If it works we will be restoring our SMA support once again.

The good news - Under RS485 we have a LOT more options. Previously we didn't have Voltron_485, we only had pylon 485. So i'm going to be testing this also today to see if it fixes some of the issues with EG4 inverters not communicating (because no one could update the firmware)


1659616924357.png


1659616978716.png
 
I got ancy and called Current Connected reference Sunny Boy and SOK communication. They are still testing it. At least it wasn’t a “no.” So I guess maintain holding pattern for now.

I’m just getting nervous as to whether I should keep the REC BMS order and go with CALB 230 volt cells (I value quality and no magic smoke over capacity and price) or if the SOK will work and be a hell of a lot easier to install.
 
I got ancy and called Current Connected reference Sunny Boy and SOK communication. They are still testing it. At least it wasn’t a “no.” So I guess maintain holding pattern for now.

I’m just getting nervous as to whether I should keep the REC BMS order and go with CALB 230 volt cells (I value quality and no magic smoke over capacity and price) or if the SOK will work and be a hell of a lot easier to install.
The rec is a top quality piece of gear, and the CALB230's are the best of the best cells.

I did more testing today...not just with one battery but with a rack of batteries. So far, I can say the SMA has a much easier time with 3 batteries (in terms of not triggering OVP) as compared to just 1 battery. Before we say officially "Yes" we want to make sure we've covered everything fully.

Some customers are running SMA already with the known bug. I haven't gotten much feedback from those customers.
 
@HighTechLab so if you have an older “non smart” or non IOT inverter how do these server rack units work? Just like a DIY pack w/xxx BMS? Or is there something one would need to do to use it?
Just set the voltages for Absorb/float/LVD and it would work fine.
 
The rec is a top quality piece of gear, and the CALB230's are the best of the best cells.

I did more testing today...not just with one battery but with a rack of batteries. So far, I can say the SMA has a much easier time with 3 batteries (in terms of not triggering OVP) as compared to just 1 battery. Before we say officially "Yes" we want to make sure we've covered everything fully.

Some customers are running SMA already with the known bug. I haven't gotten much feedback from those customers.

EDITed out this whole post because I was misunderstanding the issue being discussed and mixing it up with another issue... Shutting down because you're pushing the volts too high is a good thing! And awesome that you're testing out the fine points of getting it working on customer gear when I really think about it.
 
Last edited:
I don't think Dexter deserves the above treatment at all.
He's been and open and honest as you can ask for, and I don't think he's intentionally mis-lead anyone.
All manufacturers have issues, it's how they deal with them that sets them apart.
I would have bought my batteries from him, but he doesn't currently offer 12V Rack style units.
 
This seems like a very substantial post detailing how SOK batteries have all the same issues as an EG4, can't start larger inverters, don't have working comms... Yet everywhere else on the internet people will be claiming the SOK has never had a complaint. Looks like after today we'll all just be pointing out they didn't do their due diligence and find that you were already disclosing these issues here, as long as it's just buried in a thread and not in the latest video I'm sure it won't affect your sales. EG4 kept selling strong under the same conditions right?

Nice bait and switch with claiming you were working on SMA comms months ago and then pulling it quietly! It was even more disingenuous considering you don't seem to have a partnership relationship with them that would cause them to want to support you in this effort.
Um, wrong. SOK has never had the issue of not starting inverters. I've owned both the lifepower and the SOK and the SOK started and ran my inverter with one battery without issue. The lifepower would not start the inverter under any scenario.

The issues they are discussing are communication based. Totally different and not BMS/pre-charge related.

The SOK is measurably a better battery in many different ways.

And Dexter and Current Connected are one of the most open, honest vendors i've ever dealt with.
 
Um, wrong. SOK has never had the issue of not starting inverters. I've owned both the lifepower and the SOK and the SOK started and ran my inverter with one battery without issue. The lifepower would not start the inverter under any scenario.

The issues they are discussing are communication based. Totally different and not BMS/pre-charge related.

The SOK is measurably a better battery in many different ways.

And Dexter and Current Connected are one of the most open, honest vendors i've ever dealt with.
Yeah I do agree about it being better and all of that. I went back and read the statement, it is about OVP (I need to read the thread again!), not the inverter startup. Clearly a different issue than inverter startup issues. Thank you for correcting me.
 
I did more testing today...not just with one battery but with a rack of batteries. So far, I can say the SMA has a much easier time with 3 batteries (in terms of not triggering OVP) as compared to just 1 battery. Before we say officially "Yes" we want to make sure we've covered everything fully.

With what number/wattage of components?

Is this about "load dump" where Sunny Boys are cranking out power, loads are consuming power, but when loads suddenly switch off Sunny Island has to shove power down the throats of battery for a couple seconds? (Does SI drive "surge" charge current higher than peak rated? Higher than configured maximum?)

Can you detail the wattages involved, and maybe peak watts and seconds the batteries can absorb it vs. their voltage?
It would seem like how well cells are balanced and how far below their top-balance point the voltage is, affects that.

SMA says GT PV wattage can be 2x Sunny Island wattage, but I would think they couldn't handle "load dump" in excess of their peak charging capability. In which case I don't see where GT PV wattage limit comes from, so long as loads come and go in a staircase fashion. But surge capability of 2x would let SI start non-motor loads and carry until GT PV ramps up.
 
The rec is a top quality piece of gear, and the CALB230's are the best of the best cells.

I did more testing today...not just with one battery but with a rack of batteries. So far, I can say the SMA has a much easier time with 3 batteries (in terms of not triggering OVP) as compared to just 1 battery. Before we say officially "Yes" we want to make sure we've covered everything fully.

Some customers are running SMA already with the known bug. I haven't gotten much feedback from those customers.
Hello Dexter, Any update on getting the SMA Sunny Island to communicate correctly?
 
Last edited:
Back
Top