diy solar

diy solar

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

That you did not seen any problems with rigid busbars do not prove that there isn't any.

You also haven't proven that there is ..... to me, it seems logical that flexible bus bars would be the best solution ... but not if they don't meet other criteria.

It seems kind of bold to come on a thread where someone has laid out their build ... and the successes and failures of that build .... and try to lecture them about their bus bars with no evidence they will cause problems.
 
idk honestly i think it’s bad advice to say go solid busbars with 280Ah + cells.

cool if someone built with solid busbars and they haven’t posted in the fire forum

extrapolation station!

no offense. any measurable flexture in the terminals is a big red flag to me. doesn’t have to be a red flag for others.

the most important thing is each persons autonomy in making trade offs..
 
The only caveat I would add is that commercial battery packs are engineer solutions, hopefully with some R&D and testing, and are usually more thoughtfully and precisely secured than a lot of DIY'ers will do. Unless we are talking about the many cheap off-brand drop-in replacement batteries coming out, but most of those are pretty new and untested (in the long term).

I'm not saying there is anything wrong with solid busbars, just that being reassured by commercial use should probably only be reassuring if you have designed the enclosure/compression system, as thoughtfully and robustly as comparable quality commercial offerings. In your case, I know that you have, but most folks still opt for less robust solutions (The Will P. electric tape special.. ?, or putting the cells in a roughly sized plastic battery box with some things wedged in to prevent movement, or in some cases no securing the cells at all beyond the busbars and gravity/inertia). But I suppose that is a bit of a tangent.

Yeah I'm thinking of all the drop-in replacements we've seen teardowns of. Zero compression and solid busbars. And these get placed in mobile environments.

The example being displayed (not by you) is something designed for an EV. Yes, flexible bus bars in that case due to the nature of its environment - far more severe of an environment than mine.
 

My cells aren't anywhere near 90 degrees Celsius.

That you did not seen any problems with rigid busbars do not prove that there isn't any.

And vice-versa. But the overwhelming evidence shows that there are no problems - arguably millions of batteries with solid bus bars with no reported problems as a result of solid bus bars.

"Any commercial"... many serious build use flexible busbars. It's pretty obvious that there will be stress !

In an environment such as an EV, I agree that building with flexible makes more sense - an EV is subject to far worse conditions than most of the batteries we're using/building.

The only way to have no stress on rigid busbars is to let some space between cells AND constrain them, this way there will be enough space between them to inflate THEN the compression would prevent them to inflate further. Busbars would have to be screw at the max acceptable inflation. This way cells can only deflate and create space between cells.

I don't argue this. All I've stated is that I observed 0.1-0.5mm of expansion from 0-100 SoC (and 40F - 85F) and that I believe that's sufficiently small to use solid bus bars.
 
My cells aren't anywhere near 90 degrees Celsius.



And vice-versa. But the overwhelming evidence shows that there are no problems - arguably millions of batteries with solid bus bars with no reported problems as a result of solid bus bars.



In an environment such as an EV, I agree that building with flexible makes more sense - an EV is subject to far worse conditions than most of the batteries we're using/building.



I don't argue this. All I've stated is that I observed 0.1-0.5mm of expansion from 0-100 SoC (and 40F - 85F) and that I believe that's sufficiently small to use solid bus bars.
Still, look graph, temperature is a parameter as I said, contrary to your affirmation. Do not need to go to 90°C.

No there is not vice versa, it's not symmetric, "That you do not see problem with rigid busbars do not prove that there not any".
1- If you can see a problem there sure is one
2 -If you can't see a problem there still can be one hidden, specially on this case where you can see the internals.
Millions of case is not enough to prove something. One counterexample is enough to prove the falsity.

What I mean is that you take a risk, you are confident tha you took enough test to globaly cover your use. And yes it can be sufficient most of the time. But seeing the potential consequences...I feel it's not enough.

" I observed 0.1-0.5mm of expansion from 0-100 SoC (and 40F - 85F) and that I believe that's sufficiently small to use solid bus bars."
From my point of view..0.5mm is already too much.

Let's imagine your charger for a problem and overcharge....now batteries take 2mm on each side of them...and it's not even that much, that's now 4mm distance increase that busbar are refusing to give and transferring to terminals...

I'm not saying what you should do, it's your life. Just that in MY mind those risk are not worth the price differences, reason I discuss the case, you did a good job with the compression springs and all, I'm not saying you did non sense. Just...as in all physics publication, we discuss each other point...to go further ahead and improve.

Perhaps check "cells fire" on this forum...it can be nasty.

Ohthetrees build, this is, to my point of view...a good solution, it's a parallel build..but the flexible connections are good.
IMG_0041B .jpg
 
Last edited:
A very significant part of engineering and product design is risk management and assessment. Just because a failure can happen, doesn't mean it will with any frequency, and while lack of failures doesn't necessarily prove a problem doesn't exist, it does build a strong case for reliability and risk. I have seen a fair number of entry level design engineers get trapped in the "what if" stage, and overdesign for every conceivable failure. Or simply insist that design practices are a absolute law, and neglect to assess the actual risk

The idea that there is a hidden pandemic of failures in the millions of solid busbar batteries out there is a bit silly.

I also find it somewhat hilarious that folks are arguing that a fraction of a mm of expansion justifies going with flex bars. Everything flexes, and the terminals on these cells flex quite readily at the scales we are talking about. The terminal is set in plastic, which is sealed with an O ring, which itself mates into a thin metal case. The whole stack can easily accommodate the flex cinergi has measured. This is supported by all the batteries in service without leaks or terminal failures.

I have seen bus bars cause damage in big DC applications. It essentially requires a much longer length, and several times the temperature swings seen in batteries like this.
 
This reminds me of an interview with Elon Musk I watched the other day, he said something that really struck me, as someone who's done a lot of manufacturing. He said "the problem with good engineers is they will spend an enormous amount of time and effort optimizing a component that's not really necessary."
 
This reminds me of an interview with Elon Musk I watched the other day, he said something that really struck me, as someone who's done a lot of manufacturing. He said "the problem with good engineers is they will spend an enormous amount of time and effort optimizing a component that's not really necessary."

That shouldn't be a very common occurrence. Lots of stuff is needed to put a design together. Why waste any time on unimportant parts? Just throw in something simple that will work. Put the effort into achieving desired performance, reliability, margin. Even in the event of extreme overloads, if appropriate.

What is common (with not "good" or inexperienced engineers) is to use a electronic component for the volts/amps/whatever rating it has at 25 degrees C. At the very least, need to suitably derate it for expected environment, and a maximum junction temperature about 25 degrees below data sheet absolute max. For anything dissipating power, some thermal design is needed (often knowing Ohms law is all that is required to also perform thermal engineering.)

The other thing to do is think through failure modes. Avoid anything catastrophic or dangerous, and ensure design will survive and recover from any reasonable fault with no more than power cycling.
 
I want 0 lateral pushing on the terminals of my cells over the operational lifetime of the cell. Believe that there is a simple geometric reason to want the layers inside the cell to *not* be differentially compressed on top and bottom. Like an SoC seesaw. No disrespect to anyone who uses solid busbar with 280/320Ah class cell. No disrespect at all. Want to learn general truth, as opposed to guidance meant for a narrow case.

Also have a 100Ah Frey/Fortune cell pack, and it exhibits essentially no observable flexing over charge cycle. Solid busbar with that and no worry in my mind.

Calling flex busbar "overoptimization" seems... misguided. Increasing pack cost by 300% to get flex busbar? Yeah maybe not. Increase cost by 5-15% to reduce aging factor? Sign me up? 0.5 mm over each cycle? it's essentially a lever, and the layers inside are on the order of microns, I'm not interested. And honestly, I don't care if other people wag a finger at me saying the effort is wasted, because the cultivation of skills is also an intrinsically valuable endeavor :) the skills will inevitably be transferable

All due respect to people who have built up their skills to get to the point of an operating DIY LiFePO4 battery pack. It's amazing to me that anyone is able to DIY a pack of >5kWh to be honest. This whole forum and the projects on it impress me and inspire me constantly.

Thanks everyone for seeking to discover The LiFePO4 Path ?
 
A very significant part of engineering and product design is risk management and assessment. Just because a failure can happen, doesn't mean it will with any frequency, and while lack of failures doesn't necessarily prove a problem doesn't exist, it does build a strong case for reliability and risk. I have seen a fair number of entry level design engineers get trapped in the "what if" stage, and overdesign for every conceivable failure. Or simply insist that design practices are a absolute law, and neglect to assess the actual risk

The idea that there is a hidden pandemic of failures in the millions of solid busbar batteries out there is a bit silly.

I also find it somewhat hilarious that folks are arguing that a fraction of a mm of expansion justifies going with flex bars. Everything flexes, and the terminals on these cells flex quite readily at the scales we are talking about. The terminal is set in plastic, which is sealed with an O ring, which itself mates into a thin metal case. The whole stack can easily accommodate the flex cinergi has measured. This is supported by all the batteries in service without leaks or terminal failures.

I have seen bus bars cause damage in big DC applications. It essentially requires a much longer length, and several times the temperature swings seen in batteries like this.
Well...you talk about risk management, I do not see disadvantages for flexible bread or cable and they do not cost really that much.
So.. an ingenior will chose them over rigid busbar. The only downside I see is that it cost more time to build, it's harder to industrialize the process.

There is no problem with bloated cells or stress on terminals..?
And now...still think it's useless....?
 

Attachments

  • 1319B482-6EE1-4942-8E68-D713715B6039.jpg
    1319B482-6EE1-4942-8E68-D713715B6039.jpg
    114.9 KB · Views: 46
  • tumblr_mvf5lji2hM1qa2swjo1_1280.jpg
    tumblr_mvf5lji2hM1qa2swjo1_1280.jpg
    97.3 KB · Views: 46
This reminds me of an interview with Elon Musk I watched the other day, he said something that really struck me, as someone who's done a lot of manufacturing. He said "the problem with good engineers is they will spend an enormous amount of time and effort optimizing a component that's not really necessary."
Cause it's their job... They are searching for the graal. If they got a certain amount of time, let's say 4 days and found an acceptable solution in the first 2 days...they will most of the time use those 2 remaining days trying to improve this first solution.
Musk just want optimisation for incomes, he is in an other "dimension".
 
I want 0 lateral pushing on the terminals of my cells over the operational lifetime of the cell. Believe that there is a simple geometric reason to want the layers inside the cell to *not* be differentially compressed on top and bottom. Like an SoC seesaw. No disrespect to anyone who uses solid busbar with 280/320Ah class cell. No disrespect at all. Want to learn general truth, as opposed to guidance meant for a narrow case.

Also have a 100Ah Frey/Fortune cell pack, and it exhibits essentially no observable flexing over charge cycle. Solid busbar with that and no worry in my mind.

Calling flex busbar "overoptimization" seems... misguided. Increasing pack cost by 300% to get flex busbar? Yeah maybe not. Increase cost by 5-15% to reduce aging factor? Sign me up? 0.5 mm over each cycle? it's essentially a lever, and the layers inside are on the order of microns, I'm not interested. And honestly, I don't care if other people wag a finger at me saying the effort is wasted, because the cultivation of skills is also an intrinsically valuable endeavor :) the skills will inevitably be transferable

All due respect to people who have built up their skills to get to the point of an operating DIY LiFePO4 battery pack. It's amazing to me that anyone is able to DIY a pack of >5kWh to be honest. This whole forum and the projects on it impress me and inspire me constantly.

Thanks everyone for seeking to discover The LiFePO4 Path ?
Well..you explained my point of view in a far better way then I tried too...?

And I would add an 2 other advantage, if the rigid busbars got oval holes it will tend to unscrew nuts, not on flexible breaded or cables connections.
It will support vibrations without transferring stress on terminals.

I can't see real disadvantages of flexible connectors.
I'm building some for a friend, it's not hard nor expensive : 50mm2 cable, crimper, 50mm2 lugs....I need roughly 1m of cable and 30 lugs for 16s pack.
15€ lugs, 10€ for the cable...50-100€ for the crimper...yeah, reason I do it for friends.
 
Last edited:
I tried flexible and had performance problems so I went back to rigid. Might be fine for most but it wasn't for me.
Also, I'm not gonna have the u-shaped issue or anything similar - my pack is in a compression frame. Nor will I have over-SoC. I have two protection mechanisms to prevent that and a third that protects against over-expansion.
Nothing is perfect. I'm not saying that. I'm saying I think it's good enough and there's evidence to support "good enough" - as others said, it's risk tolerance ... informed risk tolerance.
 
I tried flexible and had performance problems so I went back to rigid. Might be fine for most but it wasn't for me.
Also, I'm not gonna have the u-shaped issue or anything similar - my pack is in a compression frame. Nor will I have over-SoC. I have two protection mechanisms to prevent that and a third that protects against over-expansion.
Nothing is perfect. I'm not saying that. I'm saying I think it's good enough and there's evidence to support "good enough" - as others said, it's risk tolerance ... informed risk tolerance.
What strange is that you did not find the reason of bad performance with the flexible...where they too weak by design or was it a bad crimp.
About risk, it all depend on counter measure that are deployed. Joe that jump from an airplane with a parachute and a spare ...should be ok, his friend Eduardo with a single parachute may have problems...and their maniac friend Viny that jump totally nude expecting god to save him from death ..is just delusional...?

I would say... If your pack is in an fire protective environment then you are a Joe. If it's not... You are between Joe and Eduardo.
 
So, in light of that, I finally installed the Progressive Industries EMS-LCHW50 electrical management system to protect my Victron's from Bad Stuff at the Campground... I tested that I can plug into 50a or 30a no problems and it passes through properly to the Victron and they do the right thing whether on 50 (split phase) or 30 (single phase). So I'm ready. I'm gonna try not to plug in. I'm at almost 3 months now without plugging in.
Have you tried running your generator since installing the PI EMS?
Many have that EMS shut down when cord connected to a generator like your Champion. It's looking for a N_G bond upstream and won't see it on a generator like it does on a 30amp pedestal.
Easy fix with N-G bond plug in the generators 20amp receptacle.

With the amount of power you can generate you may never need a generator. Still, the one time you may need it. Might want to give it a test.

I was happy to see the Quattros were able to function with that very Champion generator that I also have.
I'm now expecting the same with a soon to begin split phase dual Multiplus II 48/3000 build.

I've been re-reading your thread and ran across the EMS install.
Awesome! I'm not so bold as to build my own batteries. But man, do you have a powerhouse system!
 
Have you tried running your generator since installing the PI EMS?
Many have that EMS shut down when cord connected to a generator like your Champion. It's looking for a N_G bond upstream and won't see it on a generator like it does on a 30amp pedestal.
Easy fix with N-G bond plug in the generators 20amp receptacle.

Yes I have to bond NG but it works perfectly with that.

With the amount of power you can generate you may never need a generator. Still, the one time you may need it. Might want to give it a test.

When I was in the Badlands with 100+ temps I needed it a lot, actually. I produced 20-23 kWh per day but I needed like 40.
I also am having a problem here in AZ but I'm getting a lot of shading so I'm only able to produce 8 kWh per day on a good day. Not quite enough.

I was happy to see the Quattros were able to function with that very Champion generator that I also have.
I'm now expecting the same with a soon to begin split phase dual Multiplus II 48/3000 build.

Me, too! They work beautiful and slowly ramp up the load on the generator. Amazing.

I've been re-reading your thread and ran across the EMS install.
Awesome! I'm not so bold as to build my own batteries. But man, do you have a powerhouse system!

Thanks! It's definitely a huge RV install but I could use more panels. Even with the 3300 roof and 400 ground deploy, I've come up short sometimes. It's rare, and I'm power-hungry so ... no surprise really. It's performing better than I expected!
 
Question about your LG split system. Do you have any odd noises during defrost? My unit make a brief, but loud whine/whistle/drone during the start of defrost. Its audible at the indoor units, and its obviously refrigerant flowing. Otherwise the units work great. It was 10F last night, and the overnight power usage was ~10kw-hr.
 
Is that kW per hour or 10 kWhs for overnight? My LG split system for two rooms take about 1.5 kWs per hour but of course it depends on the outside temperature which rarely gets below 40 F.
10 kwh over 12 hours, averaging 830 watts power. Of course if its really windy (I haven't closed all my air leaks yet) it can be double that. That was with the temp setpoint at 15C, which is the lowest the IDUs will go. The lowest power level the unit will run at is around 900W, so there was obviously some cycling.

When the unit is at full power in the teens, it will pull nearly 4kw. Max current appears to be 16A, and is limited by the compressor inverter.

If its near freezing, and really humid (wet snow or freezing rain especially), the unit will defrost every couple hours, which does increase power consumption.
 
Last edited:
Question about your LG split system. Do you have any odd noises during defrost? My unit make a brief, but loud whine/whistle/drone during the start of defrost. Its audible at the indoor units, and its obviously refrigerant flowing. Otherwise the units work great. It was 10F last night, and the overnight power usage was ~10kw-hr.
The noise is most likely th reversing valve shifting to cooling mode to run the defrost cycle.
 
Back
Top