Xuba Electronics: DEAL - 280AH LiFePo4 cells. Purchase & Review

[Ampster]

Why would one need a machine shop to cut and drill Copper? Do you really think it is that much harder to work with compared to aluminum?

At the time I did not have a drill press and that place was cheap. I did cut the copper with a Sawzall after the chop saw had difficulty. I have used the same chop saw blade and my table saw with aluminum. At any rate that is history. I am committed to try Aluminum this time.

 

[nebster]

The reason PG&E uses a lot of aluminum wire is it is cheaper and less likely to be stolen. They can just push more power thru it to make up for the losses. Cost of copper buss bars is minor and you/we can't make up for losses the same way. I want every bit of power I can squeeze from my battery. While keeping it within the parameters for long life and maximum duty cycles, but still providing the power I need it for.

It's actually more about the weight than cost of the metal, for transmission lines. But your point still stands, IMO.

 

[Sgt Raven]

It's actually more about the weight than cost of the metal, for transmission lines. But your point still stands, IMO.

Copper theft is a real big problem out here. I know there were some remote stretches the phone company ended up put up fiber optic, not because they needed the capacity. They did it because people kept stealing the copper wire. It got so bad, that people were stealing it faster than they could replace it.

People will try and steal wire from live circuits, too. People be crazy the stuff they do. Other than CRV cans and bottles, past a certain amount they have to hold the check for 3 days. I think it's $60. The times I ran up against that, I told them to mail me my check.

 

[LiFer]

Is there a way to determine shipping without entering credit card info? (mcmaster.com)
It's 3:22AM and it could be that my brain isn't functioning. I just can't figure it out...

 

[Sgt Raven]

Is there a way to determine shipping without entering credit card info? (mcmaster.com)
It's 3:22AM and it could be that my brain isn't functioning. I just can't figure it out...

IDK about there, but someone on one of these threads posted a link to copper bars on eBay.
You might want to compare prices there and shipping and time to get them.

If you go to their home page, one of the things is shipping. On the left towards the bottom.
That is for shipping supplies.
Also there was a pop up that their customer service agents are available 24/7 by phone.
They have 5 different locations under contact us. I'd pick the closest to you to call.


https://www.mcmaster.com/covid-19-response/

 

[Delta-V]

Now that we have a good sense of comfort the Xuba cell/ battery capacity is good (I haven’t read anyone really say otherwise). Moving on to determining Solar Charge Controller (SCC) setting options.

After a good amount of tinkering today. I’m gonna see about using the solar charge settings below as a base-line. Gotta start somewhere, ha ha.

The hope/ desire is to see if I can set reasonable solar charge settings for just above the upper knee (upper limit) of the typical LiFePO4 graph (as opposed to going for the usual full 100% charging settings). The upper knee seems to curve up from the flat-line at around 3.35V per cell or so (or 26.8V). Aiming more for the middle 80 – 90% usage range. Don't know til we try, ha ha. Preliminary tests seem promising.

Quick Side-Note
I ended up hooking back up my cheap 15A BMS’s. Apparently, they do work. After my last capacity test with no BMS’s, the cells did get a little out of balance (~180mv) and never recovered after I charged them back up. Once I hooked up the BMS’s, the cells gradually balanced to less than 20mv. I guess I needed to disconnect them to see they were working after all.

Solar Charge Controller (SCC) - EPever Tracer 4215BN 40A with MT50 Remote Meter​

Experimental settings: Subject to change!

Description​	Setting​	Per Cell (V)​	Notes​
Battery Type:​	User​	​	​
Battery AH:​	560 AH​	​	​
Temp Comp Coefficient:​	0 mv/C/2V​	​	Should be zero for LiFePO4​
Rated Voltage:​	Auto​	​	24V (12V or 24V)​
Over Voltage Disconnect:​	27.8V​	3.475V​	Cannot be lower than Charge Limit​
Charge Limit:​	27.3V​	3.4125V​	Cannot be lower than Equalization Charge​
Over Voltage Reconnect:​	27.3V​	3.4125V​	​
Equalization Charge:​	27.3V​	3.4125V​	Cannot be lower than Boost Charge or higher than Charge Limit​
Boost Charge:​	27.3V​	3.4125V​	Cannot be lower than Float Charge or higher than Equalize Charge​
Float Charge:​	26.9V​	3.3625V​	Cannot be lower than Boost Reconnect or higher than Boost Charge​
Boost Reconnect:​	26.7V​	3.3375V​	​
Low Voltage Reconnect:​	23.0V​	2.875V​	​
Under Voltage Reconnect:​	23.1V​	2.8875V​	Cannot be lower than Under Voltage Warning​
Under Voltage Warning:​	23.0V​	2.875V​	Cannot be lower than Low Voltage Disconnect​
Low Voltage Disconnect:​	22.0V​	2.75V​	Cannot be lower than Discharge Limit​
Discharge Limit:​	21.5V​	2.6875V​	​
Equalization Time:​	0 minutes​	​	Disabled​
Boost Time:​	165 minutes​	​	I’ve heard about 30 min per 100 AH​

NOTE: Some settings are dependent on other settings. A Parameter Error means one or more other settings need to be adjusted first.

Observations so far:

• I still had the Battery AH setting set to 280 AH instead of 560 AH. Not sure if the SCC uses this setting or not for its charging algorithm. The SCC didn’t seem to be charging right lately. Could be because I had this set wrong when I recently doubled the battery from 8 to 16 cells (8S2P).
• From past experience, in User settings, I had only seen the highest voltage go to the Float Charge setting. Today, I just saw for myself, the Boost Charge (Absorb) is for a limited time only (Boost Time). When testing, with solar charging, and using a Toaster oven, I’d get the voltage to drop below the Boost Reconnect setting, turn off the toaster oven, then watch the voltage charge back up until it got to the Boost Charge setting (ie. 27.3V). After 90 minutes of testing, all of a sudden the voltage wouldn’t go above the Float Charge setting (ie. 26.9V) during peak Sun hours. The Boost Time was set to 90 minutes at the time. I ended up physically disconnecting/ reconnecting (power off/ on) the SCC to reset the Boost Time. Then it would charge back to the Boost Charge setting (ie. 27.3V) until it would time out again. Probably considered a Rookie mistake in my case. It makes sense now. Figured I’d mention this as this may be considered a given for some, but not for others (like myself). Guess I never really paid attention to this until now.
  • Apparently, the Boost Time begins once the voltage reaches the Boost Charge voltage. After the time runs out then the voltage won’t go above the Float Charge setting. Click!
• I was using a coffee maker, microwave, and toaster oven in various tests today (Intentionally forcing a big enough drop in voltage to trigger the various upper knee/ upper limit SCC charging settings). The voltage would drop below Boost Reconnect setting and the SCC would bulk charge until Boost Charge was reached.
• After the Boost Time timed-out, voltage would only go to the lower Float Charge setting. Depending on how low the voltage went during discharge tests, the charging Amps (current) would adjust accordingly. The bigger the gap between the discharged voltage and Float Charge setting, the higher the charging Amps. The charging Amps would gradually dwindle the closer the voltage got back to the Float Charge voltage setting.
  
• Reading the Drok meter, I could see I used 600 Wh, but as the discharged voltage bounced back some and was still close to the Float Charge setting, the solar charging current would be low Amp-wise. Meaning, as quick as the power was used by my various tests, it took longer to replenish. So, the closer the discharged voltage is to the Float Charge setting, the longer it takes to level-off back to the Float Charge voltage.
  • I don’t think this is a big deal as the battery had hardly been discharged at these settings. When the voltage gap was bigger, though, it would bulk charge (higher current) until it got close to the Float Charge voltage again.

Test Scenarios
Trying to determine some test scenarios and get a handle on the some real-world variables to consider (ie. load vs no-load, continuous use vs resting periods, voltage offsets, other unknowns, if these things even matter, etc…).

Thinking out-loud here:

Test #1

• Use battery, low C-rate discharging while solar charging (under load, not resting)
• Charge to Boost Charge setting (Absorb)
• Wait for charging Amps to be low (level-off), less than 1A – May have to wait for the Boost Time to finish.
• Stop discharging – Shut everything off
• Stop charging
• Let battery rest, measure resting voltage
• See how much the voltage dropped

Test #2

• Use battery, low C-rate discharging while solar charging (under load, not resting)
• Charge to Boost Charge of 27.3V (Absorb)
• Confirm Boost Time finishes
• Let voltage level-off at Float Charge voltage
• Stop discharging – Shut everything off
• Stop charging
• Let battery rest, measure resting voltage
• See how much the voltage dropped
• Compare results with Test #1

Test #3

• Based on Test #1 and Test #2 results, adjust settings (if necessary), test again.

Test #4

• At new adjusted upper limit (upper knee) settings
• Disconnect charging
• Do discharge capacity test
• Note various lower limit (lower knee) ranges
• Compare results from previous capacity tests, adjust settings (if necessary), test again.

Constructive criticism, suggestions, recommendations welcome. If you see any obvious no-no’s, please chime-in. Even if these settings are way off, perhaps they could still be used as a comparison to improved settings. Worst case, know what not to do, ha ha.

 

[Ampster]

This is great. I am going to read it a few times and relate some of the setting terms to those used by Outback. I can disable equalize but it still has to be set and it controls other settings even though it is never used.

 

[Sgt Raven]

Now that we have a good sense of comfort the Xuba cell/ battery capacity is good (I haven’t read anyone really say otherwise). Moving on to determining Solar Charge Controller (SCC) setting options.

After a good amount of tinkering today. I’m gonna see about using the solar charge settings below as a base-line. Gotta start somewhere, ha ha.

The hope/ desire is to see if I can set reasonable solar charge settings for just above the upper knee (upper limit) of the typical LiFePO4 graph (as opposed to going for the usual full 100% charging settings). The upper knee seems to curve up from the flat-line at around 3.35V per cell or so (or 26.8V). Aiming more for the middle 80 – 90% usage range. Don't know til we try, ha ha. Preliminary tests seem promising.

Quick Side-Note
I ended up hooking back up my cheap 15A BMS’s. Apparently, they do work. After my last capacity test with no BMS’s, the cells did get a little out of balance (~180mv) and never recovered after I charged them back up. Once I hooked up the BMS’s, the cells gradually balanced to less than 20mv. I guess I needed to disconnect them to see they were working after all.

Solar Charge Controller (SCC) - EPever Tracer 4215BN 40A with MT50 Remote Meter​

Experimental settings: Subject to change!

Description​	Setting​	Per Cell (V)​	Notes​
Battery Type:​	User​	​	​
Battery AH:​	560 AH​	​	​
Temp Comp Coefficient:​	0 mv/C/2V​	​	Should be zero for LiFePO4​
Rated Voltage:​	Auto​	​	24V (12V or 24V)​
Over Voltage Disconnect:​	27.8V​	3.475V​	Cannot be lower than Charge Limit​
Charge Limit:​	27.3V​	3.4125V​	Cannot be lower than Equalization Charge​
Over Voltage Reconnect:​	27.3V​	3.4125V​	​
Equalization Charge:​	27.3V​	3.4125V​	Cannot be lower than Boost Charge or higher than Charge Limit​
Boost Charge:​	27.3V​	3.4125V​	Cannot be lower than Float Charge or higher than Equalize Charge​
Float Charge:​	26.9V​	3.3625V​	Cannot be lower than Boost Reconnect or higher than Boost Charge​
Boost Reconnect:​	26.7V​	3.3375V​	​
Low Voltage Reconnect:​	23.0V​	2.875V​	​
Under Voltage Reconnect:​	23.1V​	2.8875V​	Cannot be lower than Under Voltage Warning​
Under Voltage Warning:​	23.0V​	2.875V​	Cannot be lower than Low Voltage Disconnect​
Low Voltage Disconnect:​	22.0V​	2.75V​	Cannot be lower than Discharge Limit​
Discharge Limit:​	21.5V​	2.6875V​	​
Equalization Time:​	0 minutes​	​	Disabled​
Boost Time:​	165 minutes​	​	I’ve heard about 30 min per 100 AH​

NOTE: Some settings are dependent on other settings. A Parameter Error means one or more other settings need to be adjusted first.

Observations so far:

• I still had the Battery AH setting set to 280 AH instead of 560 AH. Not sure if the SCC uses this setting or not for its charging algorithm. The SCC didn’t seem to be charging right lately. Could be because I had this set wrong when I recently doubled the battery from 8 to 16 cells (8S2P).
• From past experience, in User settings, I had only seen the highest voltage go to the Float Charge setting. Today, I just saw for myself, the Boost Charge (Absorb) is for a limited time only (Boost Time). When testing, with solar charging, and using a Toaster oven, I’d get the voltage to drop below the Boost Reconnect setting, turn off the toaster oven, then watch the voltage charge back up until it got to the Boost Charge setting (ie. 27.3V). After 90 minutes of testing, all of a sudden the voltage wouldn’t go above the Float Charge setting (ie. 26.9V) during peak Sun hours. The Boost Time was set to 90 minutes at the time. I ended up physically disconnecting/ reconnecting (power off/ on) the SCC to reset the Boost Time. Then it would charge back to the Boost Charge setting (ie. 27.3V) until it would time out again. Probably considered a Rookie mistake in my case. It makes sense now. Figured I’d mention this as this may be considered a given for some, but not for others (like myself). Guess I never really paid attention to this until now.
  • Apparently, the Boost Time begins once the voltage reaches the Boost Charge voltage. After the time runs out then the voltage won’t go above the Float Charge setting. Click!
• I was using a coffee maker, microwave, and toaster oven in various tests today (Intentionally forcing a big enough drop in voltage to trigger the various upper knee/ upper limit SCC charging settings). The voltage would drop below Boost Reconnect setting and the SCC would bulk charge until Boost Charge was reached.
• After the Boost Time timed-out, voltage would only go to the lower Float Charge setting. Depending on how low the voltage went during discharge tests, the charging Amps (current) would adjust accordingly. The bigger the gap between the discharged voltage and Float Charge setting, the higher the charging Amps. The charging Amps would gradually dwindle the closer the voltage got back to the Float Charge voltage setting.
  
• Reading the Drok meter, I could see I used 600 Wh, but as the discharged voltage bounced back some and was still close to the Float Charge setting, the solar charging current would be low Amp-wise. Meaning, as quick as the power was used by my various tests, it took longer to replenish. So, the closer the discharged voltage is to the Float Charge setting, the longer it takes to level-off back to the Float Charge voltage.
  • I don’t think this is a big deal as the battery had hardly been discharged at these settings. When the voltage gap was bigger, though, it would bulk charge (higher current) until it got close to the Float Charge voltage again.

.......snip......
Constructive criticism, suggestions, recommendations welcome. If you see any obvious no-no’s, please chime-in. Even if these settings are way off, perhaps they could still be used as a comparison to improved settings. Worst case, know what not to do, ha ha.

In your usage pattern, what would it look like cutting them off at 3.5V on charging and where would they settle after resting a short time? After the sun goes down, how quickly would you use the part from 3.5V to 3.35V?

In my long slow test, my 12V battery went from 3.475V to 3.35V in 2 hours and stayed there for days.
My test ended at 3.2V because the fridge settings and the voltage drop trying to start the compressor.

I did see a 0.5V difference on the fridge volt meter when it was running vs off. I.E. 12.9V running and 13.4V off...
From the battery it was 2' of 12AWG, a fused distribution block, Cig plug adapter, cig plug and 8' of what looks to be 16AWG wire to the fridge plug. The cig plug adapter and cig plug have been replaced with Anderson Powerpoles, but I haven't brought the battery back in to see if it makes a difference, yet.

 

[Sgt Raven]

This is great. I am going to read it a few times and relate some of the setting terms to those used by Outback. I can disable equalize but it still has to be set and it controls other settings even though it is never used.

On my Kisea equalize is only used with FLA batteries, not SLA, AGM, Gel, Lithium, or Custom. You have to purposely turn it on, or it's off.

 

[Delta-V]

In your usage pattern, what would it look like cutting them off at 3.5V on charging and where would they settle after resting a short time? After the sun goes down, how quickly would you use the part from 3.5V to 3.35V?

In my long slow test, my 12V battery went from 3.475V to 3.35V in 2 hours and stayed there for days.
My test ended at 3.2V because the fridge settings and the voltage drop trying to start the compressor.

I did see a 0.5V difference on the fridge volt meter when it was running vs off. I.E. 12.9V running and 13.4V off...
From the battery it was 2' of 12AWG, a fused distribution block, Cig plug adapter, cig plug and 8' of what looks to be 16AWG wire to the fridge plug. The cig plug adapter and cig plug have been replaced with Anderson Powerpoles, but I haven't brought the battery back in to see if it makes a difference, yet.

You could set them to 3.5V. It won't hurt, but not much more benefit either. You would use the capacity between 3.5V to 3.35V pretty quickly (negligible watts). The real workhorse voltage range is around 3.35V to 3.1V (that flat-line in the graph). I'm just trying to get as close to the upper knee as reasonable without too much baby-sitting of the system, still getting good potential usage capacity, and good longevity.

 

[Ampster]

In your usage pattern, what would it look like cutting them off at 3.5V on charging and where would they settle after resting a short time?

My guess from working with other LFP cells and looking at the discharge curve is that they will settle around 3.32 to 3.35 no matter whether you terminate the charge at 3.4 or 3.6. I think @Delta-V said the same thing in a different way.

 

[Sgt Raven]

My guess from working with other LFP cells and looking at the discharge curve is that they will settle around 3.32 to 3.35 no matter whether you terminate the charge at 3.4 or 3.6. I think @Delta-V said the same thing in a different way.

It just depends on your SCC and when it switches from Bulk to Absorb and Float and if you can do a custom profile. IIRC, I think I saw an Adam Welsh YT on a similar EPever SCC, but he had to hook a computer to the SCC to get to the custom settings. I'm not sure they were able to be custom set from the MT50. It was Delta-V with the Epever. so my answer may have confused the 2 of you.

 

[Delta-V]

It just depends on your SCC and when it switches from Bulk to Absorb and Float and if you can do a custom profile. IIRC, I think I saw an Adam Welsh YT on a similar EPever SCC, but he had to hook a computer to the SCC to get to the custom settings. I'm not sure they were able to be custom set from the MT50. It was Delta-V with the Epever. so my answer may have confused the 2 of you.

The User setting is the custom profile. The difference between the MT50 and the PC Solar Station software looks like you can go one more decimal place. Example, 3.45V instead of 3.4V. Not sure I wanna get that meticulous.

 

[Ampster]

On my Kisea equalize is only used with FLA batteries, not SLA, AGM, Gel, Lithium, or Custom. You have to purposely turn it on, or it's off.

Yes that is the same as my Skybox where I have to set the equalization charge below the Charge limit even though it is turned off. I suppose that is some fail safe decision if someone switched batteries and turned on equalization.

 

[Ampster]

Example, 3.45V instead of 3.4V. Not sure I wanna get that meticulous.

For my 16S pack that .05 volt at the cell level is almost a volt on the charge limit.

 

[LiFer]

I was using a coffee maker, microwave, and toaster oven in various tests today

Did you notice any immediate changes in the battery voltage when all of these turned on?
I'm just curious if any loads or surges triggered your charger in unexpected ways.
Thanks.

 

[Delta-V]

For my 16S pack that .05 volt at the cell level is almost a volt on the charge limit.

Hmmm... Not sure I'd look at it the same way.
16S * 3.2V = 51.2V (48V Nominal)
0.05V * 16 = 0.8V
0.8V ÷ 51.2V = 1.6%

The + or - accuracy of the equipment most folks are using probably wouldn't be able to make much difference from a couple percent one way or the other.

Not saying it can't be done, just don't think it's worth the manual effort/ benefit in the long-run. At this level of accuracy, though, perhaps having a computer of sorts (ie. Arduino) would be more suited to automatically determine best settings on-the-fly.

Even better, get Outback, Midnite, Morningstar, EPever, etc... build this logic into the SCC so we can just chose between high, medium, low efficiency modes then we wouldn't have to worry about it, ha ha.

 

[Ampster]

Even better, get Outback,

That is the perspective I was coming from, but I didn't realize that the other equipment had less accuracy.

 

[Sgt Raven]

That is the perspective I was coming from, but I didn't realize that the other equipment had less accuracy.

Well one question is at what level of accuracy is it capable of. I have 3 DMMs. two are 3 1/2 hand held meters. One is DCV 0.5% and the other is DCV 0.05%. My 3rd one is a 5 1/2 bench Fluke and it's DCV is 0.005% If all 3 were calibrated today, the Fluke Bench top will give the most accurate readings. So your setup is only as good as the weakest link in the chain.

 

[Delta-V]

Did you notice any immediate changes in the battery voltage when all of these turned on?
I'm just curious if any loads or surges triggered your charger in unexpected ways.
Thanks.

Oh yes, the battery voltage would drop a good 1.5V to 2V with any of these high wattage appliances. Actually tripped the power-strip on one test. I had the coffee maker, microwave, and toaster oven plugged into it. Apparently it trips at around 15A@120V (~1,800W). I think it was between the microwave (1,500W) and the toaster oven (1,200W) when it tripped. For the most part, I was testing one of the 3 at any given time plus an oscillating fan. Enough to pull the voltage down enough to get the SCC to react accordingly based on the settings I had set.

I was using the EPever solar charger at the time. It would adapt and go wide-open current-wise. Example, I may be pulling 8A from Solar charging at 27V at the battery. Then start the microwave, the battery voltage drops to 25V, the solar charge controller would soon adapt and start charging at 30A to compensate. Then when the microwave stops. The amperage would soon revert back to previous charging levels.