diy solar

diy solar

Are there any real thrustworty +100A BMS?

Writing about mosfet...
It does seem that there is the pain point for failing Daly BMS.

Active balancer BMS strangely seems to have it with their voltage sensors.
While their active Balancer doesn't have this problem at all..

My 400A contactor uses 7 watt , about 0.12A at contact. That's a "cheap" knock-off, $50 incl transport to Thailand.
+ $150 units can be as low as 0.03A
For my Powerwall the 5 watt extra is not a problem.
Electromagnets are known technology easy to last ages.

Not a car starter relay build to have contact a few minutes..
They are cheap, but won't last years.

For anyone needing +100A..
Probably a system with contactors is the best way to go.

Where is not really the mosfet fault..
It's their cheap quality that's being used.
@BiduleOhm once made excellent post about it (if I recall correctly)
 
You have a link of Orion Jr.?

To be clear, I'm not considering anything (yet).
I have active Balancer (2, for 2 different battery sets, one 2A for my cosmetic bad 48* 152Ah and one 1A for 32x 280A.

And building the DIYBMS from Stuart Pittaway.
With contactors/controllers and heatsink to boost the already masive 800mA to 1.5A if needed...
Were you building the diyBMS from boards purchased from the online store that has the SMD parts pre-populated? That would help a bit. Also noticed he is working on an updated version of the controller: https://community.openenergymonitor.org/t/diybms-announcements/15515/7
 
Last edited:
Were you building the diyBMS from boards purchased from the online store that has the SMD parts pre-populated? That would help a bit. Also noticed he is working on an updated version of the controller: https://community.openenergymonitor.org/t/diybms-announcements/15515/4
I'm building them via order of JLCPCB.
Sadly seldom the Attiny 841 on stock, so that need to be soldered.

Order via openenergymonitor will give you the PCB, Attiny and contact pins to solder.
After this.. still programming.
At 3* the price of JLCPCB.

At my first order there was no thermistor in-stock so I order both via Digikey.
0403 thermistor is not funny to solder. If I would have known, I would have waited or choose different part that have the same specifications. (Learning curve)
Even with microscope just to damn small to do.

During my order of the controler board I was new to Reddit and the whole Gerber / BOM file world..
Ended up soldering all parts by hand:)
SMD 0805 is not that bad to do.

Once you have all software setup and plugins, it's not too difficult to follow the steps is the videos.
Before you figure out what plug-in is needed.....
(I placed it in the comments section of the tutorial videos)

After soldering 12 of the 20 where functional...
Just too damn small parts that take me weeks to solder.

I do have energy disorder, tasks like those, really costs me.

(Nerve damage after surgery, 15 years ago, been on high dose tramadol/ morphine /oxicodon ever since, after 5 years or so I learned to live with the side effects of tramadol, pain is tolerable, but does eat loads of energy. While I can do "everything" I need to choose.
Do I build one or two BMS board or do groceries.. for that day.
Trying to do 3 will set me back a few days, being able to do about nothing except recharge.

The specialist at the hospital calls me a success story.. as I still am active and do things.

It's strange to teach my 80 year ord father with hart problem how to deal with less energy..

That's a step aside...

Back to topic, most people won't even try to go solder an SMD that they can hardly see without optical help...
Or install several programs on their laptops to program/flash a chip / IC.

I'm trying to figure out what is the reliable BMS today that can give the required feedback of SOC and cell voltage, preferably web based to avoid the probleem of the need to be within 10 meters to check the battery status.

More and more people are going to get Powerwall at their house.
As it's cheaper then ever.
And many electricity companies pay less and less for excess solar power..
They have too much of it.

Besides that, Thailand..... A stable grid is something you won't find here.
Not even in Bangkok.
Brownout/ blackout are regular.
Lower pricing, better stability.
We are living in a golden period for this development.

Perhaps I'm too focused on BMS.
So many post about trouble....
It seems that there aren't much stable options that have all build in.
(Daly, active balancer BMS)
External, Chargery, Batrium, Electrodamus, ....
They seem to be working as they should.

Somehow people around here think I know a lot about solar :)
Even drive +7 hours to look at my not yet finished setup...
(And was happy he took the trip!)

Where giving some electricity is easy, a few panels, PWM/MPPT, inverter and LA or small LiFePO4, it's a total different story when you build to power a home or small resort.
10-20-40 kw...
That emerging market does need stable products.
 
Last edited:
When all is said and done I'll easily have $9-12k+ into my system and I'm using a Chargery for over/under/safety control and disconects. It isnt pretty or slick but it just works. Battery SOC monitoring is done by a Bogart Trimetric w/ the Chargery as a backup. Active balancing is done by 2 Heltec 5A balancers and a string of QNBBM.

In my VERY strong opinion separating out the functions of balancing, monitoring, and safety control is a 'Best Practice' because sooner or later (especially in a system that may be in place for a decade or more) electronics WILL fail and if everything is in one gadet then you're dead in the water for all functions until repair or replacement.

My requirements were to have a wired remote display - especially one that's not on a bloody phone, and to drive external contactors. The only thing out there that wasnt stupid expensive that met those requirements was the Chargery.
BTW, IMO anybody who is trusting more than 20A to chinese mosfets of unknown provenance is asking for trouble.

At my max amperage of 280A (surge) the only thing I trust is a gigavac contactor to handle that kind of load, I work with high power industrial electronics and electrical gear and its ALL contactors, and for pretty good reasons.
 
@Maast ,
I totally agree with separate main functionalities.

Bogart Trimetric is new for me.
You have a link?

Same goes for the Heltec 5A balancers. You have a link?
No clue what "a string of QNBBM." Is.

For me the 1A active balancer already does a great job for S16, 560Ah

I guess more doesn't hurt :)
 
I'm sure you don't really understand how daly BMS works, or otherwise you would not have posted this.


Daly used this way can only trigger one.

Unless you open up the BMS, and tap into the mosfets..
But that is getting close of building your own :)

Well at the default thresholds ( 2.2v and 3.75v) of a dumb Daly, something is Really wrong with the charging source or load so a complete shutdown might be in order.

But as you point out the Daly could only control a single contactor, so... Buy 2 ( $18.27 ea) it is the contactors that run up the cost. These dumb Daly appear to be simple analog devices with a chip per channel. There shouldn't be any problem with hooking them up in parallel, one for the charge contactor, one for discharge.

By using contactors with a coil voltage consistent with the pack voltage one could use simple, low current, SPST switches to either turn off the contactor to work on the system or rig an emergency bypass to the bms. All analog, all simple.

No modification of bms is required.

Of course this is far less fun than soldering your own (which I would be likely to do as well). But does seem consistent with the goal of the original post, "Trustworthy, 100A+"
 
Well at the default thresholds ( 2.2v and 3.75v) of a dumb Daly, something is Really wrong with the charging source or load so a complete shutdown might be in order.

But as you point out the Daly could only control a single contactor, so... Buy 2 ( $18.27 ea) it is the contactors that run up the cost. These dumb Daly appear to be simple analog devices with a chip per channel. There shouldn't be any problem with hooking them up in parallel, one for the charge contactor, one for discharge.

By using contactors with a coil voltage consistent with the pack voltage one could use simple, low current, SPST switches to either turn off the contactor to work on the system or rig an emergency bypass to the bms. All analog, all simple.

No modification of bms is required.

Of course this is far less fun than soldering your own (which I would be likely to do as well). But does seem consistent with the goal of the original post, "Trustworthy, 100A+"
:)

You are correct.
If you buy a split port BMS and not common port, it might even work.

Just use the mosfets to switch low current needed to switch contactor :)
 
:)

You are correct.
If you buy a split port BMS and not common port, it might even work.

Just use the mosfets to switch low current needed to switch contactor :)
And yes, separate ( split) port bms with two contactors seems like it would work as well. Everything is low current except contactor switched charging and loads. And Daly claims these analog bms's are waterproof as well. I would guess that might be important for you (or a marine installation).
 
And yes, separate ( split) port bms with two contactors seems like it would work as well. Everything is low current except contactor switched charging and loads. And Daly claims these analog bms's are waterproof as well. I would guess that might be important for you (or a marine installation).
Waterproof..
Not so important at all.

My personal system is as good as ready, except not all parts together yet.

People looking at me for advice, who I'm happy not provide.
But I'm not responsible for bad quality from Daly.
If you give advice based on past experience... And it does go wrong, the people who where so happy with the advice will blame me for it...

Slippery path...
What's the most tested and proven functional for long, long time...

Indeed the low amp BMS.
Where +100 is "new"
Limitations like "ow yeah, you can do 100A... (Peak) continuous 75A.
And charging? 50A peak..
Lol.
So actually a 75A choacked for charging BMS that can handle overload to 100A....

@Maast , separate the parts.
It's clearly not grown enough to be able to handle the new developments.

Where I even like to separate the cell modules from the BMS.

Perhaps make a "what to buy" guide for the newcomers.

It doesn't have to be top dollar.
Chargery does a fine job.

Probably 20A smart Split port Daly would be little cheaper.

Thing is, adding up voltages is just not as accurate as measure each individual cell (or parallel cell set) understand it's easy, and "accurate enough".
As soon as you are dumping 30mA from cell 5, number 6 reading will be inaccurate, and so will the rest up to 16..

That's why I don't like that type of measurement.

There must be some multy contact system like Bluetooth.
Where BT just connect one in one.
16 cells,
16 "clip on" modules.
No wires (why we still use?)
Just poll every X time to see if all is OK.
Would be micro amps of power consumption.
Easy to replace a cell module if needed, or to add a few if it's a "running cell" (different Internal resistance) that needs to be hold back.

Lol...
Time for the drawing board ;-)
 
Bogart Trimetric is new for me.

Same goes for the Heltec 5A balancers. You have a link?

No clue what "a string of QNBBM." Is.
Bogart Engineering has been around for a very long time and they make products that are deadnuts reliable. I think it was something like $170 but some things are worth spending money on. IMO they make the very best battery monitor on the market with the Victron one being a close second. Again not pretty or slick but they just work - forever.
http://www.bogartengineering.com/products/battery-monitor.html

The Heltec active balancers are something I found just a month or so ago. If I'd have known about them I'd have never have spent the money I did on the QNBBMs. The basic Heltec active balancers are the best no frills high amperage active balancers on the market. Whats really nice is they move 5A even when the cell voltage differential is only in the millivolt range. They make more expensive ones that what I've linked with bluetooth, etc, etc, etc but these have the same functionality for only $50 a pop for a 16S. I have to have a high amperage balancer because my BYD module cell groups are horribly mismatched in capacity even with all my tinkering with them.
https://heltec-bms.com/project/5a-capacitive-active-equalization-active-balancer/

The QNBBMs come in 1S modules that you string together for as many cells/cell groups as you want, they also make pre-ganged 8S modules, they're expensive and I spent $320 for 16 1S modules. The highest amperage I've seen them move while testing BYD modules is only 1.4A with a .25V differential. Now that I've found the Heltecs I may just keep them in reserve because they make an annoying mosquito whine while operating.
https://www.aliexpress.com/item/32788338337.html

The reason that I insist on wired display connections is reliability. I break phones, a lot, and I don't want to be stuck not being able to see whats going on the next time I'm in-between phones. I also don't want a radio-sensitive device that can be interfered with or cooked by a nearby lightning strike or whatever. However the biggest reason is hackability, I dont want a virus or a worm having the capability to screw with my power systems with the potential to break it or even potentially start a fire that burns down my house. Among the other things I do at work is I'm the lead ISSO and do you know how many 'smart' internet connected gadgets (other than phone and computer) I have around my house? None. zero, zip, nada. WAY too easy to hack and provide an increased attack surface to my critical systems.
 
Last edited:
  • Like
Reactions: Dzl
Thank you for the links.

I learn a lot this thread!
Yeah, the QNBBM seems to be highly overpriced.

I don't know about your cells.
If you have parallel or single rows.
I just got an YR1035 LiFePO4 internal resistance meter.
Funny to see that my car key, tip to base, have higher resistance then my 152Ah cells :)
I measured 24 of the 48, they are between 0.28 and 0.32 mOhm.

Having imbalance in your system might have to do with less accurate matched cells.

I don't know yet what's "normal" difference.

I'm just going to try to get all as much in the middle as possible.
Parallel sets of 3, so that should be possible to stay within 0.01 mOhm.


I do have a few "inductance active balance" (China, clone? Same price range)IMG_1834_副本.jpg
@Will Prowse stopped his test of these after he fried it, and concluded it doesn't do active balance...
Because his battery tester with active balancing function did not balance.
Boy... Is he wrong there.
They do work. Excellent.
And yes, talking out of the battery connection while turned on will fry it. It'snot a BMS :)

My biggest issue with these is that they only start @0.1v difference between cells.

That's too high for me.
I like to a little bit "all the time" instead to try to catch up at a level I don't even want the imbalance to go.

Their capacitance balancers
IMG_5196_副本2.jpg
Seem good, work "always" up to 0.005v difference.
Same as my active Balancer :)
(Who have BT)
images (61).jpeg
Active balancing is a "religion"...
If you want to regular top charge your battery, you do NOT usually need it.
If you want to stay under de 93% SOC (apparently that's the magic top number) and if ever reached above 5%, the normal top balancing where the excess charge is dumped doesn't happen.
To keep your cells happy, this will help.
Depending on the capacity of your array and internal imbalance / mismatch it might even be needed for those who do charge above 93%

The religion part is if you believe that charge to max 93% will give many extra cycles.
I haven't seen real life field installation proof of this.
Still... I'm a believer :)

Most difficult part is 93%..
As voltages on average are just a rough indication that easy can be 20% off...
I guess I'm between 85 and 95% when I stop my charge.

I have some battery SOC meters and shunts, not yet installed.
Once they are, set a baseline and it's more accurate to calculate.
 
Last edited:
There was another recent thread here where someone tested those capacitance balancers (or a clone of that one) and couldn't get them to work even when the cells were very unbalanced.
Thanks for the feedback.

Yeah...
I do remember reading about it.

The inductance do work.
I've tested at home with 4 cells, one on 2.5 the rest on 3.35...
After a few days...
All equal.
And crazy....
I connected only the positive leads.
(What is automatic also connected to the negative lead of the other cell)
So 4 wires to 4 positive leads and the thing started blinking.

I thought, what the heck, just leave it like this.
3 days later....
Balanced. (152Ah.)
Nothing else connect, just 4 cells and S4 active balancer.

They do work.
Friend of mine tried also, one worked, one he wiggled a bit... Dark..
Sensitive stuff.
Low priced.
Once connected, Do not touch.

But really actually does work.

To bad of the 0.1v
I wish they let it start @ 0.05v
 
@HRTKD

I'm in the process of building DYIBMS that using contactor/relays and can do 800mA balancing per cell.
What's involved in the DIYBMS? One main controller board and an extra voltage measurement board for each cell? That doesn't sound too compact. How far off are you from completion? I may be interested.
 
I'm amazed people have a tough time with daly bms. I own around 20 (a whole box of LTO/LiFePO4/NMC bms) and I have only killed one (I mixed up two of the balance wires). And it was my fault.

I set and forget daly bms on my systems and friends, never an issue. Need to short out p- and b- to turn it on, but that's all. They have worked great for me.

How do you guys kill yours?

But I absolutely despise daly customer service.
Hi Will, I have looked at these and I am concerned about the low voltage battery cutoff (only 2.2V). Would you be concerned with this or is it ok? thank you.
 
Back
Top