diy solar

diy solar

Cinergi's 28 kWh / 4 kW Solar / 10 kW inverter RV build

Nope. The shunt is used for SoC and overall amperage calculations. In order to figure out the DC-only loads, the inverters "measure" (poorly) their DC inputs and subtract it from the shunt's value... thus the inaccurate value.
And the kWh graphs are AC-only.
It's still amazing stuff. I do wish I could see/graph/measure my DC side ... I was off-inverter for a while and my DC load consumption was shocking... I need to figure out where it's all going.
What if you added a second shunt in your system? Install it on your 12V battery that supplies the 12V RV house loads. I'm not sure how a second shunt would tie into the Cerbo. Or, even if two shunts could be fed into a Cerbo. But, maybe a BMV-712 shunt that only sees the 12V house system could give you enough data to evaluate your 12V loads. Just a guess.

Also, just a wild, wild guess based on absolutely NO first-hand knowledge: Could the propane furnaces be using a lot of 12V power? Furnace fans?
 
What if you added a second shunt in your system? Install it on your 12V battery that supplies the 12V RV house loads. I'm not sure how a second shunt would tie into the Cerbo. Or, even if two shunts could be fed into a Cerbo. But, maybe a BMV-712 shunt that only sees the 12V house system could give you enough data to evaluate your 12V loads. Just a guess.

Also, just a wild, wild guess based on absolutely NO first-hand knowledge: Could the propane furnaces be using a lot of 12V power? Furnace fans?

That's specifically the feature that I asked Victron to implement -- measure the DC loads via a separate shunt. Right now, you can't get that page to use a separate shunt. You can, of course, separately monitor the second shunt on your own.

The furnace fans do use a lot of power, yes. I don't use that, though - I use the mini-split heat pump. I just need to take the time to measure individual circuits. I already figured out that my jacks, even in standby mode, use a low of power, so those have their own switch to completely cut them off. There are probably other similar ghost loads I can turn off.
 
I already figured out that my jacks, even in standby mode, use a low of power, so those have their own switch to completely cut them off.
I wouldn't have guessed that either. I wonder if because they are always monitoring level.
 
Nope. The shunt is used for SoC and overall amperage calculations. In order to figure out the DC-only loads, the inverters "measure" (poorly) their DC inputs and subtract it from the shunt's value... thus the inaccurate value.
And the kWh graphs are AC-only.
It's still amazing stuff. I do wish I could see/graph/measure my DC side ... I was off-inverter for a while and my DC load consumption was shocking... I need to figure out where it's all going.
Im in the same boat. With everything off, and I mean everything. My 12v loads are a minimum 85w. I never thought to look though. Thanks for the tip on the leveling system, I will put a clamp meter on it and find out!

I too had the problem with my SolarAssistant keeping track of DC loads, and after I switched to the IP22 battery charger, which ran off of 120v, instead of DC->DC, then I started to read all my 12v loads as well!

Its a small bonus to having a 120v->12v charger instead of 48v->12v. The big downside is the lower efficiency factor.
 
Its a small bonus to having a 120v->12v charger instead of 48v->12v. The big downside is the lower efficiency factor.
I found it interesting that Victron lists the efficiency of its Orion 48/12-30 as 87%.
While the IP22 12/30 is 93% with 0.5W idle.

There's still inverter efficiency to add in the mix. But it may not be wasting as much as I assumed.
 
Cell group 4 (2 cells) has developed bad IR and as I approach 100% (only at 92% now), it's running away - likely because one of the 2 cells isn't properly participating as a parallel group. I've been sitting here for months (no movement) so it's interesting to see that I've developed bad IR (likely busbar connection and not a cell problem). I'm gonna go clean the connections again and see what happens...

HighCell4.jpg
 
Curious to hear whether cleaning the terminals of those cells and re-torquing fixes the problem. My money says it will.
 
Curious to hear whether cleaning the terminals of those cells and re-torquing fixes the problem. My money says it will.

I bet it is, too. I've had to do it twice already (the first time was probably because I used linty q-tips). It's on my todo list this weekend.

EDIT: I did check the torque already - it's good.
 
Did you use noalox or similar on original installation?

I used an anti-ox cleaner (I forget which) but no "paste" of sorts, and this may become a learning experience about that! Because my bus bars are tin plated - I shouldn't need it from what we've all been told. And I'm in AZ - extremely dry.
 
i created flexible "bus bars" with tinned lugs and 1/0 cable. I still used noalox, though. So far i've been good in florida *knocks on wood*
Batteries purring along, solar purring along, and loads purring along.

I didnt glop the noalox on, just a tiny dab spread around with my finger (I had nitrile gloves on) both on the lug and on the terminal.
 
I pulled the bus bars and did the needful. However, I had to PRY one of the busbars up and drill out the two end holes a little bigger so I could put the busbar back on. Even under compression, there's still enough expansion that something like this happened - and I bet is the reason I had higher resistance for this cell group. Despite my measurements and experimentation, this may be the first real-world proof I've seen here that, even with a compression fixture, flexible busbars are needed with EVE 280 cells. I specify EVE here because the Lishen 272's I have didn't expand/contract during my testing.
I'm letting the system stay at 100% so it can fully balance and I'll keep an eye on it after that.
 
The graphs look normal again, so definitely some sort of issue with the busbar <-> terminal contacts. And given that they're tinned and it's dry here and I haven't been moving, the only explanation I can think of is the small movements from cell expansion/contraction with rigid bus bars. I guess I need to convert to flexible. That's not going to be fun. Or cheap. And while I know that I never push more than 100 amps through, having these large busbars have kept my resistances very low and voltage fluctuations low which has kept the Orion BMS happy; high voltage fluctuations previously (when I used wires for busbars) caused the BMS to prematurely trigger various voltage-based actions (SoC drift, balancing, reducing max charge amperage, etc).
 
Not sure if it was mentioned here, but these look promising. Basically CALB style laminated tinned copper. I believe they are available in several spacing options.


I have had good like with the MG chemicals carbon paste, that might be worth a try. Though with that much work involved, its probably not worth experimentation.
 
Back
Top