Cruisers Forum
 


Join CruisersForum Today

Reply
 
Thread Tools Rating: Thread Rating: 33 votes, 4.82 average. Display Modes
Old 11-06-2016, 13:09   #5251
Senior Cruiser
 
senormechanico's Avatar

Cruisers Forum Supporter

Join Date: Aug 2003
Boat: Dragonfly 1000 trimaran
Posts: 5,205
Re: LiFePO4 Batteries: Discussion Thread for Those Using Them as House Banks

Which is why all my electronic measuring equipment is Fluke.
(Well, except for one Beckman DMM in my garage)
__________________

__________________
"I have never understood why it is "greed" to want to keep the money you have earned,
but not greed to want to take somebody else's money"

-Thomas Sowell
senormechanico is offline   Reply With Quote
Old 11-06-2016, 13:14   #5252
Senior Cruiser
 
SV THIRD DAY's Avatar

Cruisers Forum Supporter

Join Date: Aug 2006
Location: Morro Bay, CA
Boat: 1978 Hudson Force 50 Ketch
Posts: 3,181
Re: LiFePO4 Batteries: Discussion Thread for Those Using Them as House Banks

Quote:
Originally Posted by Maine Sail View Post
I use a carefully calibrated Ah counter (Link-Pro/E-Expert-Pro) that is physically and manually reset to full each time the bank hits 13.8V & less than 10A of charge current. Voltage only is very difficult to use with LFP and especially at low current draw....
Exactly what I have been doing and it's been working for over two years now.
__________________

__________________
Rich Boren Living Aboard in Morro Bay, CA and the owner of:
Cruise RO Water High Output Water Makers
Technautics CoolBlue Refrigeration

SV THIRD DAY is offline   Reply With Quote
Old 11-06-2016, 22:58   #5253
Marine Service Provider
 
OceanSeaSpray's Avatar

Join Date: Aug 2013
Location: New Zealand
Boat: Custom 13m aluminium sloop
Posts: 361
Re: LiFePO4 Batteries: Discussion Thread for Those Using Them as House Banks

Quote:
Originally Posted by ebaugh View Post
... In that case, the charging device, a Victron Inverter Charger shows huge (+/- 100 mV) readings of string voltage under charge. The designer/developer claims these readings are real. As explained, there is little to no averaging going on in the readings. But a Fluke multimeter, or anything else I attached does not show these deviations, perhaps explained by automatic averaging in the firmware/software of other devices. Hmmm. Not being able to tell 3.45 from 3.55 is a issue. Further, there is a feeling by the Orion BMS company that the peak voltages are limiting. Not to beat up Victron, shop testing on a Magnum device shows the same.

Not directly related, but a consideration when pushing to 3.6V or more per cell.
Ripple on the output of DC chargers can be very bad news with LFP.

We have an outfit here that has been selling "more-or-less drop-in" LFP cells into the campervan market with silly advice for many years and it has been most interesting to watch them change their tune over time as they keep getting burned. They still haven't learned.

They have had countless disasters. One of the main causes was overcharging combined with the output ripple from some shore power chargers. As long as the battery absorbs all of the current provided, the ripple has limited effect on voltage and is not much noticeable, but as the cells get nearer to full, it causes higher and higher voltage peaks, to the point of starting to break down the electrolyte, even though the average "Fluke" voltage may not look that impressive. What these multimeters display is RMS. If you want to see peak, you can build yourself a little filter with a fast diode and a ceramic capacitor between the source and the instrument and then add back the forward voltage drop of the diode.
Because we are dealing with chemical reactions that require a threshold potential to occur, it doesn't matter how long the peaks last. If the value is exceeded, the reaction is suddenly triggered. This has caused many cells to gas and swell, which requires about 4.2V by the way, but sometimes leads to claims that "3.65V did it". 3.65V RMS doesn't mean anything in terms of peak value.

I treat PWM control with utmost suspicion as well for the same reasons. Limiting charging voltage using a PWM signal has the potential for doing exactly the same with a full battery. The RMS value is ok, but the "on" peaks may not be. While I use switching for charge regulation with solar power, I purely switch based on voltage (not time), the frequency varies and is in fact very low. This ensures that the cells HV limit doesn't get exceeded.

Fixed frequency switch-mode chargers/charge controllers produce square waves and without good filtering and proper charge termination, the peak voltages can be enough to wreck the cells.
The answer would be adding a low-pass filter, like capacitor and inductor, on the output of the charger to smooth the ripple out. That is both expensive for high currents and unnecessary when charging SLAs (if anything the pulses are beneficial to break down plate sulfation), so many units don't have a lot in the way of output filters.

Alternators are variable frequency, 3-phase rectified, unfiltered power sources by the way, but they start with nice sine waves and, with three phases combined, the difference between RMS and peak is quite modest (8.9% higher, if I didn't screw up), which is why they never cause us much grief.
At 3.50V/cell RMS, the theoretical peak (with zero charging current) wouldn't exceed 3.66V and that is a long way from breaking down the electrolyte.
If we do the same calculation for an equivalent single-phase charger, peak is 29% higher than RMS: 4.51V, goodbye LFP cell! That stands for the common basic AC mains transformer + rectifier battery charger.

I filter the cell voltages on my BMS boards and we don't want millisecond reaction times for battery protection anyway, it would be far too cranky and achieve nothing.
Keeping an eye on peak voltage is not silly however considering what can sometimes get hooked up to lithium batteries. It is a very interesting observation that you have made here, thanks.
__________________
"The case for elimination: the only equipment that never needs maintenance and never breaks down is the one you don't have on board."
OceanSeaSpray is offline   Reply With Quote
Old 12-06-2016, 01:46   #5254
mrm
Registered User

Join Date: Feb 2011
Location: Poland, EU
Boat: crew on Bavaria 38 Cruiser
Posts: 552
Re: LiFePO4 Batteries: Discussion Thread for Those Using Them as House Banks

Quote:
Originally Posted by OceanSeaSpray View Post
Ripple on the output of DC chargers can be very bad news with LFP.

[...]even though the average "Fluke" voltage may not look that impressive. What these multimeters display is RMS.[...]
I was under the impression, that RMS display was restricted to AC ranges of multimeters and did not apply to DC ranges. Live and learn, I guess. Thanks for the info.
__________________
mrm is offline   Reply With Quote
Old 12-06-2016, 04:31   #5255
Marine Service Provider
 
OceanSeaSpray's Avatar

Join Date: Aug 2013
Location: New Zealand
Boat: Custom 13m aluminium sloop
Posts: 361
Re: LiFePO4 Batteries: Discussion Thread for Those Using Them as House Banks

Quote:
Originally Posted by mrm View Post
I was under the impression, that RMS display was restricted to AC ranges of multimeters and did not apply to DC ranges. Live and learn, I guess. Thanks for the info.
Actually, you will be right in many cases about the RMS range being AC-coupled. I sometimes switch to AC when measuring DC voltages to see if they are noisy and then I get the RMS value of the noise with the DC component blocked...
It will usually display a filtered average on the DC range and there will be small differences between instruments on noisy DC signals.
__________________
"The case for elimination: the only equipment that never needs maintenance and never breaks down is the one you don't have on board."
OceanSeaSpray is offline   Reply With Quote
Old 12-06-2016, 05:36   #5256
Registered User
 
ebaugh's Avatar

Join Date: Dec 2011
Location: On the boat
Boat: DeFever 44
Posts: 525
Re: LiFePO4 Batteries: Discussion Thread for Those Using Them as House Banks

I was negligent in leaving out a couple additional observations on the Orion. The voltage fluctuations were also observed on discharge. But the proof that the Orion was actually reading something real, was on discharge it depended on the load. An inverter load showed this, but when a similar current load, 3 DC sailing winch motors,was observed, the voltage fluctuation mostly disappeared.

This observed fluctuation was very dependent on current flow, at low values it was small and increased as additional current was applied. Offhand it appeared to be linear, i.e. 100A showed 10x what 10A did. I attached a couple pictures, one when the charger came on and another where we had a 5 ton A/C compressor short cycling on the inverter.

Maybe some of the engineers here can comment, but could you surmise that the peaks and valleys are really always there, but the bank acts as a big capacitor making even the more sensitive Orion unable to sense?

If this behavior is really potentially detrimental, perhaps it needs to be measured accurately.

I have parameters with the ADCs in my home grown BMS to shorten the sample time and decrease averaging, but never tinkered with the values. Have to try this one day.

Click image for larger version

Name:	ImageUploadedByCruisers Sailing Forum1465730677.639486.jpg
Views:	90
Size:	102.2 KB
ID:	125988Click image for larger version

Name:	ImageUploadedByCruisers Sailing Forum1465730696.392268.jpg
Views:	137
Size:	103.3 KB
ID:	125989


Sent from my iPad using Cruisers Sailing Forum
__________________
ebaugh is offline   Reply With Quote
Old 12-06-2016, 06:36   #5257
Registered User
 
ebaugh's Avatar

Join Date: Dec 2011
Location: On the boat
Boat: DeFever 44
Posts: 525
Re: LiFePO4 Batteries: Discussion Thread for Those Using Them as House Banks

I can add a couple more unrelated observations from this install that may be interesting. Physically it required 2 banks of 1000Ah each due to space limitations. There were 2 BMS systems, one on each bank. Each bank had about 25 ft of cable where it connected to the ships main DC bus. This cable was 2 4/0 for positive and another 2 4/0 for negative. Normal operations both banks were kept in parallel.

The test was we decoupled the banks, charging one to 100% (3.6V), discharging one to about 30%. Then with the charge imbalance, paralleled the banks allowed them to settle, then recharged. Found 2 interesting observations, both really characterized by the voltage drop in the cable.

1) Peak current on combining the banks was 451A, which decayed to under 100A in 2 minutes. There are those who might say combining would lead to a disaster in this scenario. Simply not true, provided there is sufficient voltage drop in the cabling. And sufficient voltage drop is likely present, unless you were bent on specifically disproving this observation.

2) On charge, both banks re-synced in a reasonable time. In the beginning the low bank trailed the high bank by 1/2V, measured at the bank. But at 3.6V, given about 30 minutes or so in the CV portion of the charge, both banks evened out with the current dropping from 250A to < 5A.

While "normal" charge was to more like 3.45V, it would have never synced the banks very fast, but probably would have over time and cycles once the high bank was discharged. This part is a theory, and needs more testing. However, at 3.6V the "high" bank probably had fractionally more charge, but not enough to worry about.

My conclusion is there isn't any reason to worry about combining multiple banks if:

1) Each bank has a BMS should an individual string have a fault, removing the bank.

2) The voltage drop in the cabling for each bank must be engineered to be similar for the common loads and charging sources.

The only advantage to this scenario might be if you carried it out on a system using 400Ah cells, in a 1600Ah system, and did 4 banks and BMS systems. Then a cell failure could be isolated, leaving 3/4 of the bank available until there was an opportunity to resolve the problem. Still...I would normally think the complexity would would outweigh the benefit for a simple house bank. But if the physical space available dictates multiple banks, it can be done.



Sent from my iPad using Cruisers Sailing Forum
__________________
ebaugh is offline   Reply With Quote
Old 12-06-2016, 10:33   #5258
Senior Cruiser
 
senormechanico's Avatar

Cruisers Forum Supporter

Join Date: Aug 2003
Boat: Dragonfly 1000 trimaran
Posts: 5,205
Re: LiFePO4 Batteries: Discussion Thread for Those Using Them as House Banks

I don't use my shore charger as I have plenty of solar, but out of curiosity I'm going to hook up my Fluke Scopemeter to the battery bank and see what I get from the BlueSky 2512 IX mppt controller when charge is nearly done.
At the dock, that's mid AM.
__________________
"I have never understood why it is "greed" to want to keep the money you have earned,
but not greed to want to take somebody else's money"

-Thomas Sowell
senormechanico is offline   Reply With Quote
Old 12-06-2016, 15:58   #5259
Marine Service Provider
 
OceanSeaSpray's Avatar

Join Date: Aug 2013
Location: New Zealand
Boat: Custom 13m aluminium sloop
Posts: 361
Re: LiFePO4 Batteries: Discussion Thread for Those Using Them as House Banks

Quote:
Originally Posted by ebaugh View Post
I was negligent in leaving out a couple additional observations on the Orion. The voltage fluctuations were also observed on discharge. But the proof that the Orion was actually reading something real, was on discharge it depended on the load. An inverter load showed this, but when a similar current load, 3 DC sailing winch motors,was observed, the voltage fluctuation mostly disappeared.

This observed fluctuation was very dependent on current flow, at low values it was small and increased as additional current was applied. Offhand it appeared to be linear, i.e. 100A showed 10x what 10A did. I attached a couple pictures, one when the charger came on and another where we had a 5 ton A/C compressor short cycling on the inverter.

Maybe some of the engineers here can comment, but could you surmise that the peaks and valleys are really always there, but the bank acts as a big capacitor making even the more sensitive Orion unable to sense?

If this behavior is really potentially detrimental, perhaps it needs to be measured accurately.

I have parameters with the ADCs in my home grown BMS to shorten the sample time and decrease averaging, but never tinkered with the values. Have to try this one day.
You are showing a +/-10mV swing charging and +/-50mV with the inverter load. That is what you can expect each time. Most AC chargers "pulse" current into the battery to some extent and inverters are not nice DC loads at all, they grab power in pulses as required for creating the AC output waveform.

The battery on the other hand has some internal resistance, so surges do alter the cell voltage and the higher the current, the greater the deviations; the Orion clearly makes instantaneous measurements and then filters the noise in the software, which is why it can show the noise. It needs to do that, because it internally calculates the correlation (i.e. battery internal resistance) between voltage and current.
Batteries are not "big capacitors", at a minimum, they need to be viewed as an ideal battery in series with a resistor. The ideal battery just offers a relation between voltage and SOC; the resistor adds the dependency between cell voltage and current.

What would be interesting to measure for you is how big the charger noise gets when the battery is near-full and accepting less and less current. This is what differentiates a charger that is acceptable for LFP cells from one that will wreck them.

To be honest, there is nothing that makes me more nervous with LFP than AC chargers. So far, I have managed to stay clear of any installations where people wanted to plug the DC system into shore power. We used shore power to run refrigeration etc directly on AC mains instead, and small on-board DC loads were still completely taken care of by solar without any needs for an AC charger.
If I was to charge using AC, I would source a high-quality DC power supply unit or a DC charger of equivalent quality with stated maximum ripple and voltage regulation specifications.
__________________
"The case for elimination: the only equipment that never needs maintenance and never breaks down is the one you don't have on board."
OceanSeaSpray is offline   Reply With Quote
Old 12-06-2016, 20:45   #5260
Senior Cruiser
 
senormechanico's Avatar

Cruisers Forum Supporter

Join Date: Aug 2003
Boat: Dragonfly 1000 trimaran
Posts: 5,205
Re: LiFePO4 Batteries: Discussion Thread for Those Using Them as House Banks

I put my Fluke Scopemeter on the 200 ah LiFepo4 bank (only bank on the boat) when mostly charged, but still in bulk phase.
the battery voltage was 13.7 according to the BlueSky. I read the voltage directly at the battery with my Fluke 77 at the same time 13.6.

With the Fluke ScopeMeter, the pulse = about + 0.045 volts pretty much in a very low duty cycle. (3%) I tried to take a photo but the reflections on the scope screen were awful. You will just have to trust.
I suspect one of my cells isn't quite up to snuff as it's always the highest before charging is done and lowest under a big load, although I need to check my cell connections before I draw any conclusions. Delta about 0.04 volt @ 100 amps.
__________________
"I have never understood why it is "greed" to want to keep the money you have earned,
but not greed to want to take somebody else's money"

-Thomas Sowell
senormechanico is offline   Reply With Quote
Old 12-06-2016, 21:06   #5261
Registered User
 
ebaugh's Avatar

Join Date: Dec 2011
Location: On the boat
Boat: DeFever 44
Posts: 525
Re: LiFePO4 Batteries: Discussion Thread for Those Using Them as House Banks

I don't have any data saved to refer to. But I don't recollect any substantial difference at 3.6V versus lower voltages. For sure, as the current decreased in the CV phase, the line straightened out in both a normal charge to 3.4-3.45V, or a maintenance/test charge to 3.6V

For smaller yachts, many loads tend to be DC, and at anchor, charging more solar based. But for bigger or power yachts (mine small at 44', but power), the loads are essentially all AC and the primary charging is the genset. So LFP is functionally an AC battery, something that allows us to only run the genset 3-4 hours a day instead of 24.

Obviously every cruising style different, but we found even cruising, we spent 50% at anchor, 50% at dock. Sometimes security related (Colombia), sometimes convience or lack of good anchorages (Puerto Rico mainland and Aruba).

Maybe LFP isn't the best tool in the arsenal for this application?
__________________
ebaugh is offline   Reply With Quote
Old 12-06-2016, 21:25   #5262
Marine Service Provider
 
OceanSeaSpray's Avatar

Join Date: Aug 2013
Location: New Zealand
Boat: Custom 13m aluminium sloop
Posts: 361
Re: LiFePO4 Batteries: Discussion Thread for Those Using Them as House Banks

Yes, but that should remain fine if the chargers are reasonably well filtered and there is still no reason why you wouldn't run all your AC loads straight off the shore mains when you have access to that.
Why cycle the battery and incur all the power conversion?

When you are cruising, LFP is undoubtedly the best answer to energy storage.
__________________
"The case for elimination: the only equipment that never needs maintenance and never breaks down is the one you don't have on board."
OceanSeaSpray is offline   Reply With Quote
Old 13-06-2016, 06:37   #5263
Registered User
 
toddedger's Avatar

Join Date: Mar 2012
Location: Michigan
Boat: Morgan 27
Posts: 191
Re: LiFePO4 Batteries: Discussion Thread for Those Using Them as House Banks

I kept an AGM in my system as a starter battery. My house loads has an A/B switch. When at the dock I switch loads to the AGM battery to avoid having to worry about automated cycling my LiFePO4. My newer Pronautic charger has the ability to change from 13.8 LI profile to an AGM profile. In fact if I just leave the LI contactor off, my charge also has a power supply mode that doesn't even need a battery in the loop. Why wast your LiFePO4 batteries at the dock when you don't need them?
__________________
toddedger is offline   Reply With Quote
Old 13-06-2016, 16:41   #5264
Marine Service Provider
 
OceanSeaSpray's Avatar

Join Date: Aug 2013
Location: New Zealand
Boat: Custom 13m aluminium sloop
Posts: 361
Re: LiFePO4 Batteries: Discussion Thread for Those Using Them as House Banks

Quote:
Originally Posted by senormechanico View Post
I put my Fluke Scopemeter on the 200 ah LiFepo4 bank (only bank on the boat) when mostly charged, but still in bulk phase.
the battery voltage was 13.7 according to the BlueSky. I read the voltage directly at the battery with my Fluke 77 at the same time 13.6.

With the Fluke ScopeMeter, the pulse = about + 0.045 volts pretty much in a very low duty cycle. (3%) I tried to take a photo but the reflections on the scope screen were awful. You will just have to trust.
I suspect one of my cells isn't quite up to snuff as it's always the highest before charging is done and lowest under a big load, although I need to check my cell connections before I draw any conclusions. Delta about 0.04 volt @ 100 amps.
As long as the battery still loads the charger properly, voltage ripple is usually not much of a concern, the battery itself keeps it down. Problems can start when the current tapers down.
This is how cells can get completely destroyed even by very small chargers.
__________________
"The case for elimination: the only equipment that never needs maintenance and never breaks down is the one you don't have on board."
OceanSeaSpray is offline   Reply With Quote
Old 14-06-2016, 00:58   #5265
Registered User

Join Date: Oct 2008
Posts: 63
Re: LiFePO4 Batteries: Discussion Thread for Those Using Them as House Banks

Hi floks,

Can this noise form the Chargers not be eliminated by "just" adding few big Capacitors in Parallel? How big would they need to be for 1000ah bank?
Anybody has experience whit the noise of the Victron 24V/5000/100A combo?

Other topic:
For the Alternator and other peack induced Voltages, has anybody used big current Z Diodes to limit peace Voltages like 18V or 36V for 12 or 24V systems, they can pull up to 100A for short periods whit in nanoseconds, so few of those at critical points should do the job in order to prevent the V to run away in case your
Bank disconnects, and before the Alt. Regulator kicks in.
I do have a normal lead acid Bank in Parallel via a bipolar diode that gives me 0.7V drop, maybe i even put 2 of those in so 1.4V drop, so there should not be to much drain current from that, it's the bow truster batt anyhow so i need it (has sep. Charger).

Any comments on that?
Many thx.
Lagoon
__________________

__________________
lagoon100 is offline   Reply With Quote
Reply

Tags
battery, lifepo4, LiFePO4 Batteries, sailing

Thread Tools
Display Modes Rate This Thread
Rate This Thread:

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off


Similar Threads
Thread Thread Starter Forum Replies Last Post
LiFePO4 Batteries - Okay Tear Me Apart ;-) jallum Electrical: Batteries, Generators & Solar 395 03-12-2015 13:19
Voltage drop under load, amps read 99% ?? VVD Electrical: Batteries, Generators & Solar 25 28-06-2011 16:25


Our Communities

Our communities encompass many different hobbies and interests, but each one is built on friendly, intelligent membership.

» More about our Communities

Automotive Communities

Our Automotive communities encompass many different makes and models. From U.S. domestics to European Saloons.

» More about our Automotive Communities

Marine Communities

Our Marine websites focus on Cruising and Sailing Vessels, including forums and the largest cruising Wiki project on the web today.

» More about our Marine Communities


Copyright 2002- Social Knowledge, LLC All Rights Reserved.

All times are GMT -7. The time now is 10:42.


Google+
Powered by vBulletin® Version 3.8.8 Beta 1
Copyright ©2000 - 2016, vBulletin Solutions, Inc.
Social Knowledge Networks
Powered by vBulletin® Version 3.8.8 Beta 1
Copyright ©2000 - 2016, vBulletin Solutions, Inc.

ShowCase vBulletin Plugins by Drive Thru Online, Inc.