diy solar

diy solar

LiFePo4 addition to a SLA House battery?

It's a 400ah lifepo DIY + a Dynasty UPS12-300MR + my charger.

They're all on 4ga welding wire and 120amp anderson connections.
 
Horsefly said:
Actually, the world of lead acid batteries is a bit more complicated than that. SLA also includes "Maintenance Free" or "Valve Regulated" batteries, that are actually just flooded lead acid batteries (yes, they slosh) that don't have caps for you to open and add water. I believe that most places I've seen refer to SLA as only the sealed version of the flooded, and AGM and GEL are separate.
Click to expand...
"MF" or maintenance free is simply a FLOODED lead-acid "sloshy", that knowing that the typical consumer won't check or add water as needed, is filled with more electrolyte than normal at the factory, and sealed up. The hope is that this will take the battery to the typical 4-year automotive life without the consumer having to check.

"Valve Regulated" is NOT sloshy, and have plates in close contact with each other. In the case of an AGM, the plates are separated by an "absorbed glass matt". Think of a paper-towel saturated with electrolyte. In the case of GEL, the former sloshy electroyte has stabilizers added to make it more like jello. Then they are sealed up forever.

In either case, some of those playing with trash online, pull the caps off of agm's and gel's, and assume that they need water! Not so. It just looks that way since they don't understand the chemistry. They proceed to fill the seemingly empty agm's and gels with water, and unknowingly turned their trash into very poorly performing FLA's, which now require a different method of care. :)
 
Here are the results of my first day of LFP+AGM.

For over a month, I have been charging the chassis 55Ah Optima AGM battery in the van from a 140W panel and through an MPPT SCC . I log stats at 5 min intervals. There was a pattern of morning Volts being around 12.6V just before sun-up. Also there was a pattern of about 0.4A charge current with total daily of around 5Ahs. The AGM runs a dashcam 24/7.

On Oct 11:
Just prior to sun-up, the SCC read the AGM at 12.57V, a perfectly consistent value.

I then attached a 100Ah LFP at around 9:20AM. The LFP Volts had been measured by a clamp-on as 13.18V prior to attachment. Just prior to this attachment, the AGM had reached 13.9V and the SCC was in Float. The now new bank Volts dropped to a low of 13.28V for about 10 min. Then the SCC reverted into CC mode(Bulk) because it could not maintain the bank at the 13.5V float setting. Charging continued, and the SCC remained in bulk/CC until sundown.

The following morning, Oct 12, the SCC detected 13.21V just before sun-up. That is a gain of 0.64V. Also, Instead of the usual 5Ahs charge per day, the first day charging was 35Ahs.

I will be moving the LFP into the trailer later to mate with a 100AH AGM.
 
gelmjw said:
Here are the results of my first day of LFP+AGM.

For over a month, I have been charging the chassis 55Ah Optima AGM battery in the van from a 140W panel and through an MPPT SCC . I log stats at 5 min intervals. There was a pattern of morning Volts being around 12.6V just before sun-up. Also there was a pattern of about 0.4A charge current with total daily of around 5Ahs. The AGM runs a dashcam 24/7.

On Oct 11:
Just prior to sun-up, the SCC read the AGM at 12.57V, a perfectly consistent value.

I then attached a 100Ah LFP at around 9:20AM. The LFP Volts had been measured by a clamp-on as 13.18V prior to attachment. Just prior to this attachment, the AGM had reached 13.9V and the SCC was in Float. The now new bank Volts dropped to a low of 13.28V for about 10 min. Then the SCC reverted into CC mode(Bulk) because it could not maintain the bank at the 13.5V float setting. Charging continued, and the SCC remained in bulk/CC until sundown.

The following morning, Oct 12, the SCC detected 13.21V just before sun-up. That is a gain of 0.64V. Also, Instead of the usual 5Ahs charge per day, the first day charging was 35Ahs.

I will be moving the LFP into the trailer later to mate with a 100AH AGM.
Click to expand...
I'm not sure this test really tells you much. You had a 55Ah AGM, and after you added 100Ah of LFP the amount of harvested energy and the morning resting voltage were higher. Well, yeah. If you would have added 100Ah of AGM, they would have been higher too.

Unless I'm missing the point. What do you think this test told you?

I think what would make your test much more interesting is to measure the current going in, out, and between the two during charging and discharging.
 
Horsefly said:
...

Unless I'm missing the point. What do you think this test told you?

I think what would make your test much more interesting is to measure the current going in, out, and between the two during charging and discharging.
Click to expand...
That I can get the AGM floated overnight for free?

I did measure the charge current once about 2 hours after they were joined. Each battery has a separate post on the bus bars. I did not parallel at the battery. My clamp on read 0.66A to the AGM and 4.10A to the LFP. That agreed with the total out from the SCC into the bus bars. I was pleased to see this imbalance because I want the LFP to get the majority of the charge. LFP has better charge efficiency and that avoids wasting daylight waiting for AGM to absorb. The LFP can take all night to float the AGM.

By keeping the AGM floated overnight, there is less need for absorption during solar hours. Hand-in-glove!

I would love to have automatic data logging of all of the current flows. I will have to settle for ad-hoc manual measurements.

I do suspect that one time only, right after connection, that the AGM transferred current to the LFP.

This LFP was acquired to be installed in the Travel Trailer. I am waiting on parts.
 
OK, I just measured -1.2A, with the combined bank at 13.28V at 20:40 CDST. I would say the AGM is wanting to decrease. I will measure Amps again right after sunrise.
 
gelmjw said:
That I can get the AGM floated overnight for free?

I did measure the charge current once about 2 hours after they were joined. Each battery has a separate post on the bus bars. I did not parallel at the battery. My clamp on read 0.66A to the AGM and 4.10A to the LFP. That agreed with the total out from the SCC into the bus bars. I was pleased to see this imbalance because I want the LFP to get the majority of the charge. LFP has better charge efficiency and that avoids wasting daylight waiting for AGM to absorb. The LFP can take all night to float the AGM.

By keeping the AGM floated overnight, there is less need for absorption during solar hours. Hand-in-glove!

I would love to have automatic data logging of all of the current flows. I will have to settle for ad-hoc manual measurements.

I do suspect that one time only, right after connection, that the AGM transferred current to the LFP.

This LFP was acquired to be installed in the Travel Trailer. I am waiting on parts.
Click to expand...
I guess there are different goals for different people in looking at this shotgun marriage of AGM with LFP. I think that some people see the toughness of the AGM as protecting the LFP: The LFP can be cut off by its BMS, but it won't cause a problem for the alternator because the AGM will step up and take it. I can't tell, but I think that some people are more familiar with AGM, and think they can dip their toe in the water by just slapping on some LFP.

I have no experience in this, and don't pretend to know. I just look that the discharge curves that are so dissimilar, I have to believe that as the combined batteries cycle up and down, most of the energy is moving between the two, rather than out to loads and in from charging sources. Maybe I'm wrong. I think to figure this out requires some constant current monitoring with shunts on both banks. Not sure anyone is going to bother with that.

AGMs (which is the only thing I really knew much about until about a year ago), are really mysterious. You don't really know how they are doing, or how much longer they will last. They really like to be charged to full, and stay in absorption until the current goes to 0.01C.

LFP doesn't like any of that. LFP is OK with being charged to full, but prefers to stay "between the knees" of 95% SoC and 10% SoC (or so). The preferred AGM bulk / absorption voltage for most brands is too high for an LFP bank.

So what do you do? Sacrifice what's best for one in order to keep the other happy? Optimize one knowing that you are prematurely sacrificing the other? I dunno.
 
Horsefly said:
LFP doesn't like any of that. LFP is OK with being charged to full, but prefers to stay "between the knees" of 95% SoC and 10% SoC (or so). The preferred AGM bulk / absorption voltage for most brands is too high for an LFP bank.
Click to expand...
My 48V bank bulk charges at 56.4V and floats at 54.0V. That's right on spec for my bank.

For a 16s LFP bank that would mean bulk charging at 3.525V/cell and floating at 3.375V/cell. That just seems a nice match.
 
gelmjw said:
That I can get the AGM floated overnight for free?

I did measure the charge current once about 2 hours after they were joined. Each battery has a separate post on the bus bars. I did not parallel at the battery. My clamp on read 0.66A to the AGM and 4.10A to the LFP. That agreed with the total out from the SCC into the bus bars. I was pleased to see this imbalance because I want the LFP to get the majority of the charge. LFP has better charge efficiency and that avoids wasting daylight waiting for AGM to absorb. The LFP can take all night to float the AGM.

By keeping the AGM floated overnight, there is less need for absorption during solar hours. Hand-in-glove!

I would love to have automatic data logging of all of the current flows. I will have to settle for ad-hoc manual measurements.

I do suspect that one time only, right after connection, that the AGM transferred current to the LFP.

This LFP was acquired to be installed in the Travel Trailer. I am waiting on parts.
Click to expand...
Ding Ding Ding we have a winner!
 
wattmatters said:
My 48V bank bulk charges at 56.4V and floats at 54.0V. That's right on spec for my bank.

For a 16s LFP bank that would mean bulk charging at 3.525V/cell and floating at 3.375V/cell. That just seems a nice match.
Click to expand...
Yeah, I think we've been here before (like on the previous page of this thread). I am mostly unaware of any AGMs other than yours that this works for. I've never heard of or seen anything about your AGMs, other than the company's main business seems to be backup power for telecoms and such (with the exception of their acquisition of Outback). So I give you that your AGMs seem to fit better with LFP, but the discharge curve is still a slope vis the flat discharge curve of the LFP. So like I said - most of the energy will be between the AGMs and the LFPs, rather than to your loads and from your charging sources. I'm not certain of this, and would love to see some real world evidence to the contrary.
 
Horsefly, Yes, the curves are different. That's WHY it works. It cycles the lead batteries very little overall, but the extra top and bottom end juice is there if you need it. I played with it for a few days with a shunt on the lead battery (and bluetooth BMS on the lithium battery) - You aren't losing much, maybe 2-3 amp hours into the lead battery until everything settles down to 13.8.

Also, keep in mind who the audience is. If someone was building a brand new system, it would be silly to set things up this way. But in many peoples cases, we have lead batteries already there. Why not use them? In my case, I have two group 24 lead batteries in my travel trailer. I like having lead there in case it gets cold, or the trailer brake needs to be activated and something is wrong with the BMS for some reason. I like having an absolute bullet proof battery in there just in case.
 
Horsefly said:
Well, not so fast. Sounds like @gelmjw will report back with more data on the current transients.

Besides, spending the money on an LFP battery just so it can get your AGM to float doesn't seem like "free".
Click to expand...
I did check again at 6:00AM. I read -1.2A at the LFP neg cable. Given the 10 hours since the reading last night results in a total of -12Ahs.

Then I realized I had left a dome light on last night! When I turned it off, the reading was -0.40. Let's say that was the true discharge from the LFP to the AGM. Only -4Ahs were consumed to float the AGM.

The bank Voltage dropped from 13.33/3.33 to 13.29/3.32 Volts over the 10 hours.

Free?
 
To funny. I have been running a hybrid system for over two years. My CG2 FLA's are over 6 years old. I see 15 to 30 tenths an amp per hour going into the FLA's at night. (the amps are dependant on the SOC of the LFP bank.) My amps would be lower due to the higher resistance of FLA over AGM.
 
Last edited:
This is an interesting graph of how the 2 chemistries behave during discharge. LFP at the bottom. Does not seem to be a lot of contribution from lead until the LFP is depleted.

 
That's why it's interesting to have it set up like this. Lithium goes away, and you have a reserve bank there if you need it.
 
You must log in or register to reply here.

Similar threads

D
Automatic toggling of power source (Vehicle vs. House battery)
2
Replies
21
Views
885
0truck0
0truck0
A
How to make LifePo4 "safer" ?
2 3
Replies
41
Views
1K
hwy17
hwy17
O
I'm looking to power a mobile workshop.
2 3
Replies
42
Views
873
justinm001
J
J
Should I add a fuse to my battery pack?
Replies
1
Views
529
stromy
S
R
Which battery for RV?
Replies
10
Views
623
corn18
corn18
Back
Top