|
| Author |
Message |
gcrank1
Member
|
# Posted: 23 Feb 2021 16:49
Reply
My mppt scc is set up for agm's currently on it:
14.6v bulk
13.4v absorb & float due to the limited sun time
No Eq
I know not to discon the bat-bank unless the array is switched off to the scc so I want to leave the two big ups agm's perm hooked up as is.
Wondering if it may be ok to just jumper cable a 50-100ah lfp c/w bms up to the agm bank for whatever charge time I can get per day?
My intension right now is to transfer the lfp home to charge & back and hook it up in cabin with its own inverter and still have the power house 90' away with the array, scc, agm bank and their inverter.
That way I can trial the lfp for usage, still have the agm bank to fall back on, and add some recovery charge if needed to the lfp per good solar day.
|
|
ICC
Member
|
# Posted: 23 Feb 2021 19:58
Reply
LFP cells have a much lower internal resistance than any type of lead-acid cells.
Let's assume a hypothetical situation where each of the LFP and the AGM are at 70% full. For the state of discussion let's also assume the AH capacity of both the AGM and LFP are the same.
So the CC will sense the combined parallel voltage and determine the rate of charge from that. More of the current from the CC will be absorbed by the lower resistance LFP cells. When the LFP reaches full charge the AGM will be far behind. That will allow the CC to to continue with a rate of charge higher than what the LFP needs.
The LFP BMS can only do so much, can only handle so much current. Usually the amperage that the BMS can shift from one cell to another is not all that great.
I think because the AGM would not be as full as the LFP the CC would be "tricked" into continuing to charge at too high a rate for the LFP.
.
FYI, this difference in internal resistance is what can cause mismatched lead-acid batteries to fail. Theoretically, all cells from the same batch of a production run of lead-acid cells should have a very similar resistance, based on the lead mix for the plates coming from the same metallurgic mix.
So, take batteries made in 2 different batches. There is a good chance the internal resistance will be less in one batch. As those batteries are discharged and recharged that difference can cause 2 things. One, the batteries with less resistance can discharge more than the others. Two, as the batteries are charged the batteries with the lower resistance can have more energy go into them. They should get to full charge before the others. The CC can only sense the combined voltage of all batteries in the bank.
The lower resistance cells will be overcharged by some amount. Lead-acid is much more forgiving than LFP when it comes to overcharging. They may boil off more liquid but can sustain that sort of abuse. Eventually, though, the different cells will become more and more mismatched and we end up with some cells that are highly degraded.
I have seen such a thing myself with a bank of golf cart batteries where a string of 4 new ones was added a year after the first 4 were placed into service. It took years for the faults to be so bad it affected the performance of the whole set but there were several really bad cells develop.
With an LFP and AGM mix, I think there would be an early failure of one type or the other, maybe both.
My 2 cents on the topic.
|
|
gcrank1
Member
|
# Posted: 23 Feb 2021 20:17 - Edited by: gcrank1
Reply
All good points, but....
In my 'scenario' the agms would be full, having been on the solar and not depleted nor being actually hooked to the cabin.
I would be using the new lfp for trials in cabin, monitoring the function in use. Lets say we get there of a Fri noonish, I hook up the lfp & inverter and we just start using elec like we are used to. It would likely work fine all aft, evening/night and morning. Then lets say about 9-10am (sometime after b-fast and its a good solar day) to take the lfp out to the pwrhs and jumper it up to the fully charged agm's to replenish some of those used ah's. The agm's would be at a higher soc than the lfp, the solar would be kicking out amps and the scc is set suitably for the bms of the lfp.
I run into shade to the west after 1pm, so the solar becomes puny. Id pull the lfp off, return it to the cabin and repeat.
The whole thing is managed like a manual transfer switch (me) depending upon what I plug into the One cabin input; ie, there is never be more than one storage/gen source powering the cabin at a time.
To make it a bit more fancy, I could use a battery isolator, the agm always on like a starting battery would be and the other side would be where I'd hook up the lfp of a day to recharge but never any backflow.
|
|
gcrank1
Member
|
# Posted: 24 Feb 2021 20:42 - Edited by: gcrank1
Reply
Even though I was hoping to configure the scc to be lfp friendly maybe I should just leave it alone as is and run a 120vac to 12v LFP charger off the existing psw inverter that normally powers the cabin?
Or a DC-DC charger straight off the agm's?
The RhinoVoltz chargers arent big, only 20a, but a couple hours a day should put my daily anticipated 30ish ah's back in and I'd still have the whole fla/agm bank in reserve.
Id still be able to run the inv/gen for lfp charging if needed of a stay.
|
|
ICC
Member
|
# Posted: 24 Feb 2021 21:42
Reply
Quoting: gcrank1 run a 120vac to 12v LFP charger off the existing psw inverter that normally powers the cabin? Or a DC-DC charger straight off the agm's?
Conversion losses? Wasteful?
I am just leary of combining different battery chemistries as described. I may be overly cautious but it is my nature.
Just curious... what gauge is your wire that carries 120 VAC 90 feet between PV/Batteries/Inverter and the cabin? How many amps maximum used at one time at cabin? Average?
|
|
gcrank1
Member
|
# Posted: 24 Feb 2021 23:23 - Edited by: gcrank1
Reply
Conversion loss for only a a couple hours of charging of a good solar day using the 120vac chgr shouldnt break the bank. It should be essentially running off the array when paralleled at the scc output to bat-bank? And I could use that same chgr at home on-grid. Within reason isnt using 'free' solar elec not wasteful, some use dump loads. Again, this is not a perm hookup, just a 'couple hours'.
I agree with jumpers direct is the iffy one, thus my inquiry. Even though I think it could work the caviat may be 'when closely monitored', at least initially for proof of concept. Then we all know how that plan can stumble  .... It sure seems that if my scc readout is at 13.4 absorb to the agm's at mid-morn that it would be in the right place for lfp for a couple hours before the sun moves too far west and it all falls off the cliff (kind of self-regulating that). Take the lfp back inside, hook it up and ready for another aft, eve, night and morn.
Using the lfp chargers:
The 120vac chgr off the inv/gen would be safest for sure to the lfp, and unlike the DC-DC chgr no hookup into the solar at all. Though my thought of having the DC-DC for 'car charging' was primarily to have another option of ways to charge. Ie, can one have too many charging options? Lol, that asked by a guy with maybe 7 or 8 fla bat chargers......(so many I cant count them right now).
In fact, the easiest is I could have that 120vac chgr and bat in cabin and have it charge whenever the inv/gen runs. I did that for a while with a marine bat while re-working the solar. It would be even better with an lfp. Were I starting out with cabineering that would be my basic 'starter elec system' that had no need of, but could grow into, some solar generation.
Wiring:
I use two 50' 12ga. contractor's extension cords for that 90' run. At the mid-point junction is where I discon and plug in the inv/gen when we want to run the 700w microwave, toaster, vacuum, etc., only one at a time. Otherwise the loads are light, 1 or 2 led bulbs, some usb, some radio, the ceiling fan or a box fan as needed, all done with a little 300w psw inv.
My multimeter reads the same at inv., 50', 100' and anywhere in cabin.
Fwiw, I do have about 40' of 30a rv cord I had considered using from the powerhouse for the initial run, but when I got the same vac read with the 12ga. cords I just left 'em.
|
|
ICC
Member
|
# Posted: 25 Feb 2021 18:40
Reply
Quoting: gcrank1 My multimeter reads the same at inv., 50', 100' and anywhere in cabin.
Were those measurements taken when loads were applied?
|
|
gcrank1
Member
|
# Posted: 25 Feb 2021 18:50
Reply
No, just stuck the probes into the slots.
|
|
|
ICC
Member
|
# Posted: 25 Feb 2021 19:42
Reply
Quoting: gcrank1 No, just stuck the probes into the slots.
That would not be an accurate measurement. The load needs to be applied in order to measure the voltage drop at the end of the cable. The drop is not going to be huge until the loads are high. A 5 amp load should drop the voltage by about 2 volts at 100 feet; possibly more depending on the resistance at any plugin connections. A 15 amp load would result in a drop of about 6 volts (5% of 120 VAC) at the same 100 feet with 12 gauge wires.
A maximum drop of 5% is the goal. Less is better.
|
|
gcrank1
Member
|
# Posted: 25 Feb 2021 22:55
Reply
I did not know that!
Now I have another reason to take my new dig-meter 
I had looked at some online chart(s) for 120vac extension cord runs with different gauge wires and the 12ga. had looked to be ok.
Do I guess right that one leaves the plug not seated just enough to get the probes on the prongs?
|
|
darz5150
Member
|
# Posted: 25 Feb 2021 23:04
Reply
You could stick a kill a watt meter at the start, middle, then end.
|
|
ICC
Member
|
# Posted: 26 Feb 2021 09:21
Reply
Quoting: gcrank1 I had looked at some online chart(s) for 120vac extension cord runs with different gauge wires and the 12ga. had looked to be ok.
Depends on the load.
Quoting: gcrank1 Do I guess right that one leaves the plug not seated just enough to get the probes on the prongs?
If you are wanting a reading at halfway, yes. At the cabin though, just turn on the usual loads and measure at any unused outlet that is supplied by the same source. It is what is at the point of use end that matters.
|
|
gcrank1
Member
|
# Posted: 26 Feb 2021 10:12
Reply
This 'wiring' is simple and the loads light to moderate. There is a 15a switched powerstrip inside for the microwave, toaster, vacuum, but only one at a time And Only when on the gen hooked up at the 50' point.
The 100' run is the bat-bank/inverter at the toolshed 'powerhouse' for the light loads, led lights, radio, usb taps, etc. (biggest load would be the small, cheap ceiling fan). The heavy loaded p-strip is never plugged into this 'circuit'.
The only 12vdc is at the bat-bank.
This segregation of loads with essentially 2 'circuits' seemed to be a simple and safe solution for a constantly morphing situation. Each high draw unit runs 'off the cord' just as well as when the gen trial was at 25' c/w a 12ga. cord; no heat detected at plugs or cords at any time even at my max run.
Btw, the inverter has its own fuses and gen its own circuit breakers; they have never blown/tripped.
|
|
gcrank1
Member
|
# Posted: 26 Feb 2021 21:01 - Edited by: gcrank1
Reply
We have a decent sized Amish community around the cabin and they run a very nice lumber and hardware store only 13mi. from my driveway.
I had a nice chat today with a sharp young fellow there. They sell the RhinoVoltz and MillerTech LFP batteries primarily to the Amish (and I think those may actually be Amish companies, not sure). I hadnt really thought about it before but they as a group have been 'living off grid' forever. This 'high school age?' chap had set up the solar system at the store and knew his stuff! They Love the LFP's compared to the old 'marine' deep cycles.
What struck me was that he was from an 'old school culture' and doing new generation high-tech solar power. And his eyes didnt glaze over after only a few minutes talking with me like most do.....
The LFP's are normal stock items for them and they even shave a bit off the msrp. Report is they are holding up well in the community so Im thinkin that will be where I buy.
|
|
ICC
Member
|
# Posted: 26 Feb 2021 22:49
Reply
|
|
paulz
Member
|
# Posted: 27 Feb 2021 15:13 - Edited by: paulz
Reply
Quoting: gcrank1 The 120vac chgr off the inv/gen would be safest for sure to the lfp
Excuse me for butting in. Grank1 I think you went into more detail but I lost track. You must have a pro typist.
I was running my generator yesterday for some power tools, my LFPs were getting low and had one more night to spend. I was about to plug my 110ac LFP charger (the same one I use on grid) to my battery bank but it was still hooked to the solar, so I skipped it, and made it overnight anyway.
So, can I just plug my LFP charger to the generator and the battery bank without disconnecting the CC? Seems to me the CC will see the increase in battery voltage and slow down it's charging?
|
|
gcrank1
Member
|
# Posted: 27 Feb 2021 17:24 - Edited by: gcrank1
Reply
I was wondering when you'd come around Paul  , good question.
With the lead acid batteries you can have all source charge inputs; gen, solar, wind, hydro, grid all hooked up at once and multiples thereof. The battery doesnt know or care where the regulated input is coming from as long as each is appropriate to the battery.
Each charging device has its own control 'box', no 'dumb' chargers used by any of us, right? The bat-bank will only take what it can take, the smart chargers will react to that and pull back as needed, if needed.
The caveat I think is that all charging sources combined should not exceed the max C rate of the battery bank spec.
I would hope that the same is true of LFP! I have been in the habit of having my smart-charger on (with the agm's on the scc too) whenever the gen is running and would expect to do the same with the LFP. We best not assume on this one; its gonna take someone at a higher pay grade than me to tell us.
|
|
paulz
Member
|
# Posted: 27 Feb 2021 19:26
Reply
I've charged these LFP (Valence brand) batteries over and over with this charger, but with them disconnected from anything else. My concern is what the SCC is going to think when battery voltage start rising, to potentially 14.6. The SCC could be disconnected first, which means switching off the panels and then the battery, but I'd rather not if not necessary. I know others use their gennys to augment the solar, and I don't believe they switch off, just want to be sure it's OK.
|
|
gcrank1
Member
|
# Posted: 27 Feb 2021 19:48
Reply
The diysolarforum.com site is the 'go-to' for stuff like this; they are not only big on solar but LFP too.
|
|
paulz
Member
|
# Posted: 28 Feb 2021 10:01
Reply
Yes I should get over there but the guys here have been most helpful. It's been a very slow learning process but by hook or crook I have two working systems (cabin and shop).
Read a few other forum posts, looks like it's OK to connect the LFP charger, going to try it next time batteries get low
|
|
ICC
Member
|
# Posted: 28 Feb 2021 10:42
Reply
A system can have multiple input sources each with its own charge controller. I have 3 solar CC from panel arrays pointing different directions, plus a genset if I need. A friend has 2 solar and two wind, each independent, plus a diesel welder used as a genset.
There is a danger that if all inputs were running that the total current to the battery bank could get too high for battery health. Depends on how much the input total up to and what type of batteries and the total capacity. One must do the math and be certain.
|
|
Steve_S
Member
|
# Posted: 28 Feb 2021 11:36 - Edited by: Steve_S
Reply
Interesting.
Right now, at this moment I am charging my 910AH LFP Bank. Now I am getting about 80A off my Solar Controller + I'm also pushing another 80A from my Genset via the Samlex Inverter/Charger. So 160A being pushed to four battery packs so they'll be topped off to 100% by mid-afternoon.
The nice thing about LFP is that it will charge pretty fast and they will take .5C Charge Rate which for example is 50A for a 100AH Battery.
There can be multiple Solar Controllers and even a Charger going simultaneous as long as no battery pack exceeds it's 0.5C Maximum Rate IF all other battery packs in a bank are shut down due to being full.
SCC & Disconnecting from Batteries.
IE When a BMS cuts-off a battery being charged.
- PWM Controllers DO NOT Handle this well and will quite often suffer damage. Plain & Simple.
- MPPT Controllers can generally handle this without issues, very few will fail if the battery is disconnected.
When an SCC powers on, it 1st senses the battery connected to determine its Voltage Settings, then it senses the panel input and completes it's startup process. Once up & running, it will retain the settings. Most do NOT retain the battery voltage profile, this is a safety mechanism, a PITA but for good reason.
As the SCC charges it "should" be programmed to cutoff at a set voltage and/or a specific "End-Amps" value*. For a 12V Pack Lo Volt Cutoff should be no lower than 10.4V and Hi Volt cutoff at 14.4V The main power curve is 12V-13.6V, set Float to 13.9V.
If the BMS cuts off due to being Full the SCC should just drop output to nil. Remember there is No Way for any external device to know what the individual cells states are.
IF you have an Inverter that is also drawing power while the SCC is charging, if the batteries hit full, the power coming from the SCC will be absorbed by the inverter draw, the SCC does not know what is taking the power, it will just provide whatever is demanded within its capability, if more is demanded than the SCC can provide, the balance is drawn from the battery system.
*End-Amps: This is workable with ONE LFP battery but when multiple packs are on a bank, it is not useful as each pack will vary.
IMPORTANT NOTE:
Calibrate your equipment !
SO many Mess up on this !
You must take a good quality DMM/DVOM and check your Battery Voltage @ the Battery Pack Terminals, then at the Battery Bank BusBars (if running parallel batteries) then at the SCC Batt Terminals and at the Inverter BATT terminals. Correct / Compensate the Voltage Settings on the SCC & Inverter so that they are accurate compares to the batteries. Even a Half Volt can muck up the works with Lithium Based Batteries. There IS a Voltage Drop from the Batt Terminals to the Other devices, all depends on the wires, the number of connections, lugs, etc, it all adds up and it can be significant. ALSO if there is a Big Voltage Drop noted, that can be an indicator of a faulty connection or lug etc.
I wrote up a diddy ion this located here:
https://diysolarforum.com/threads/calibrating-voltage-of-your-system-to-ensure-optima l-operations-scc-inverter-charger-voltage-matching.5030/
Hope it Helps.
Steve
EDIT:
Charging started with solar at 08:30 & genset on at 10:30 adding 80A. 910AH/23.296kWh battery bank at 46% (424AH remaining) SOC reached 100% SOC by 14:00. Amps varied from 80-145 depending on solar input (partly cloudy day).
|
|
paulz
Member
|
# Posted: 28 Feb 2021 13:52
Reply
Great article Steve!
|
|
ICC
Member
|
# Posted: 28 Feb 2021 14:16
Reply
|
|
gcrank1
Member
|
# Posted: 1 Mar 2021 09:56
Reply
Thanx for that, Steve
|
|
Steve_S
Member
|
# Posted: 1 Mar 2021 10:39
Reply
You welcome. I did EDIT the last post I made after charging was done. I hope that adds a bit more "Real World" perspective to fill in the picture a bit more.
The charging time & such will change once I complete the final leg of the Battery Bank Build with final tweaks. That including adding the final 280AH Battery Pack and converting the two 175AH Packs into one 24V/350AH pack as 70AH difference is much easier to contend with than 105AH which creates it's own issues.
The END Build will be 3x 24V/280AH & 1x 24V/350AH. Each pack will have a Chargery BMS8T-300 & DCC-300, QNBBM-8S Active Balancer and MRBF Fuse for each pack.
* The BMS8T's may get swapped to BMS8P models with Active Balancing built-in and better comms which would also allow me to remove the QNBBM's as well... This is being pondered on at this time.
REF Links for the Inquisitive.
*nb: The Chargerystore errors on the certificate with a couple of browsers it's being fixed.
BMS8T-300A
300A DCC Contactor
QNBBM 8S equalizer
BlueSea MRBF Terminal Fuse Block - 30 to 300A (Made Bussman/Eaton)
BMS "P" Series still in development & testing, no links available yet.
|
|
gcrank1
Member
|
# Posted: 10 Mar 2021 20:09
Reply
Hey ICC, I metered the plugs on my 100+' of run today.
My inexpensive dig-multimeter that reads to tenths on the 200vac setting showed .1 diff from the inverter right at the bat-bank to the multi-plug in cabin with 31w of 4x led light bulbs and the ceiling fan on low (a bit more than I usually run but easy to do and repeat).
That is with the one plug in at inverter, one butt to plug in the middle of the run and the other butt inside to the power strip.
Looks like no appreciable loss with that load.
I will check again when I take the gen to run the mw, etc. just to know.
|
|
|
