Jump to content

Lithium Iron Phosphate batteries: Are they safe?


Recommended Posts

I'm only half serious.  I've heard the stories of them blowing up or catching fire, so I have to ask.  In my boat, a WindRider-16, the battery sits directly behind me, just under the rear deck.  :unsure:

Right now I use a single, 55 Amp/Hr AGM battery to run my Minn Kota Riptide-45 trolling motor, but it weighs a ton and, along with my 225# of meat and bone, pushes the stern down more than I'd like.  It also doesn't give me a lot of run time, given the max Amp draw on the motor is 42A.  

The 100 Amp/Hr LifeP04, lithium iron phosphate battery would just about double my run time, and it only weighs 25lbs, but I'm a little worried about having it sit 6" behind my butt.  

Should I be worried, or are the fire risks only when charging it?    

 

fullsizeoutput_14d1.jpeg

Link to post
Share on other sites
The LiFePO4 batteries are the safest type of Lithium batteries as they will not overheat, and even if punctured they will not catch on fire. The cathode material in LiFePO4 batteries is not hazardous, and so poses no negative health hazards or environmental hazards.
 
  • Like 1
Link to post
Share on other sites

Don't forget a AGM battery shouldn't be discharged to 0% if you want to keep them alive.

So a 100 A.hr Li Ion has more than 2x the USEFUL capacity. Don't know if you flattened the AGM though, so maybe only 2x 

  • Like 1
Link to post
Share on other sites
Quote

 I've heard the stories of them blowing up or catching fire, so I have to ask. 

You heard wrong.   They can overheat and even melt down but do not catch fire.   Blowing up, nope.   There are plenty of dangerous chemistries that can.

  • Like 1
Link to post
Share on other sites
2 minutes ago, solosailor said:

You heard wrong.   They can overheat and even melt down but do not catch fire.   Blowing up, nope.   There are plenty of dangerous chemistries that can.

It’s a common misunderstanding. LiPo’s are dangerous little fuckers, but good at what they do. They also account for 99.99% of people’s concerns when they hear “lithium” mentioned in a battery. 

Link to post
Share on other sites
1 hour ago, Monkey said:

It’s a common misunderstanding. LiMH are dangerous little fuckers, but good at what they do. They also account for 99.99% of people’s concerns when they hear “lithium” mentioned in a battery. 

Fixed it for you. Ferrous phosphate safe, mmkay?

Link to post
Share on other sites

Cool, thanks for all the helpful replies!  I went ahead and ordered one, but I cheaped out and went with the $400 model on Amazon.  

https://www.amazon.com/gp/product/B084DB36KW/ref=ppx_yo_dt_b_asin_title_o00_s00?ie=UTF8&psc=1

I know, I know, Chinese batteries are junk, but I just couldn't justify putting the $900+ Relion in a boat worth less than that.  83% of the 957 Amazon reviews were 5-Stars, and many of them were from people who used it for my intended purpose.    

I'll be putting it in a box to make sure it doesn't get dunked in salt water, and I have a NOCO Genius-5 that's said to be sufficient to keep it charged.  

Hopefully I'm good to go.  If not, I'll buy the Relion RB100 and chock it up to another one of my expensive lessons learned the hard way..  

fullsizeoutput_1620.jpeg

Link to post
Share on other sites
14 hours ago, Winston29 said:

I'm only half serious.  I've heard the stories of them blowing up or catching fire, so I have to ask.  In my boat, a WindRider-16, the battery sits directly behind me, just under the rear deck.  :unsure:

Right now I use a single, 55 Amp/Hr AGM battery to run my Minn Kota Riptide-45 trolling motor, but it weighs a ton and, along with my 225# of meat and bone, pushes the stern down more than I'd like.  It also doesn't give me a lot of run time, given the max Amp draw on the motor is 42A.  

The 100 Amp/Hr LifeP04, lithium iron phosphate battery would just about double my run time, and it only weighs 25lbs, but I'm a little worried about having it sit 6" behind my butt.  

Should I be worried, or are the fire risks only when charging it?    

 

fullsizeoutput_14d1.jpeg

Safe ? just ask the guy who owned this boat 

6EAB248C-C45A-4EDD-AD28-3E7A69CDCE24.jpeg

B280F00F-117E-4742-B447-93A05C8FF8BD.jpeg

Link to post
Share on other sites
12 hours ago, Winston29 said:

Cool, thanks for all the helpful replies!  I went ahead and ordered one, but I cheaped out and went with the $400 model on Amazon.  

https://www.amazon.com/gp/product/B084DB36KW/ref=ppx_yo_dt_b_asin_title_o00_s00?ie=UTF8&psc=1

I know, I know, Chinese batteries are junk, but I just couldn't justify putting the $900+ Relion in a boat worth less than that.  83% of the 957 Amazon reviews were 5-Stars, and many of them were from people who used it for my intended purpose.    

I'll be putting it in a box to make sure it doesn't get dunked in salt water, and I have a NOCO Genius-5 that's said to be sufficient to keep it charged.  

Hopefully I'm good to go.  If not, I'll buy the Relion RB100 and chock it up to another one of my expensive lessons learned the hard way..  

fullsizeoutput_1620.jpeg

I fully understand not spending that kind of money. (Employee discount for the win!) I wasn’t recommending any particular brand, I just went with that one because the BMS is built in and because the price was right. 

  • Like 1
Link to post
Share on other sites
13 hours ago, Max Rockatansky said:

Fixed it for you. Ferrous phosphate safe, mmkay?

Not sure what you’re correcting. I said LiPo. Lithium polymer. Not safe, mmkay?

Link to post
Share on other sites
5 hours ago, slug zitski said:

Safe ? just ask the guy who owned this boat 

 

 

What are the details behind the pictures?

Was that actually the battery going into thermal runaway?  Was it a LiFePO4 chemistry battery?

There are so many variables in the cause of a fire.  But inherently the chemistry in LiFePo4 does not experience thermal runaway.

However according the ABYC the DC system on any boat is the leading cause of fires.

 

  • Like 1
Link to post
Share on other sites
1 minute ago, SimonGH said:

What are the details behind the pictures?

Was that actually the battery going into thermal runaway?  Was it a LiFePO4 chemistry battery?

There are so many variables in the cause of a fire.  But inherently the chemistry in LiFePo4 does not experience thermal runaway.

However according the ABYC the DC system on any boat is the leading cause of fires.

 

Don’t know 

they “say “ lithium batteries 

fire department couldn’t extinguish the fire 

 It was in the water at the time 

 

Link to post
Share on other sites

Another classic low-information/N=1 post.

LiFePo with a BMS (internal or external) that can cut the battery off from loads and charging sources is pretty damned reliable and safe. Your propane system and the remainder of your electrical system are a greater threat.

Link to post
Share on other sites

the devil's in the details...

A Li-ion battery in thermal runaway is a scary thing.  I've literally seen them melt their way out of aircraft.

But that's a completely different chemistry than LiFePo4

 

  • Like 1
Link to post
Share on other sites
2 hours ago, Leeroy Jenkins said:

If you went from FLA to LiFePo4 on a cruising boat would you need to rethink solar controllers and shore powered chargers?  Alternators?   

 

if it’s a modern boat you may be able to reuse some components

Typically every component must be upgraded 

also the battery cables , lithium is capable of absorbing very high amps  

Link to post
Share on other sites
21 minutes ago, slug zitski said:

 

if it’s a modern boat you may be able to reuse some components

Typically every component must be upgraded 

also the battery cables , lithium is capable of absorbing very high amps  

This is BS. You need to change your charging profiles and not float the batteries at 100% SOC but charger upgrades are optional. If your battery cables are sized appropriately for your current FLA batteries and chargers, they'll work fine for those same chargers with LFP. 

LFP batteries typically charge at up to C/2, so if you've got a 400Ah house bank, you can charge it at 200A. However, if you've only got a 100A charger, it'll work just fine. You won't be taking full advantage of the batteries' fast charging capabilities but there's no harm done. Likewise, because LFP can accept that C/2 charge for most of its recharge cycle, your alternator may taper its output due to heat but as long as its got good thermal protection, there's no harm done. 

Link to post
Share on other sites
1 hour ago, IStream said:

This is BS. You need to change your charging profiles and not float the batteries at 100% SOC but charger upgrades are optional. If your battery cables are sized appropriately for your current FLA batteries and chargers, they'll work fine for those same chargers with LFP. 

LFP batteries typically charge at up to C/2, so if you've got a 400Ah house bank, you can charge it at 200A. However, if you've only got a 100A charger, it'll work just fine. You won't be taking full advantage of the batteries' fast charging capabilities but there's no harm done. Likewise, because LFP can accept that C/2 charge for most of its recharge cycle, your alternator may taper its output due to heat but as long as its got good thermal protection, there's no harm done. 

Generally true for the wiring and battery charger, but not always for an alternator. C/2 is still faster than led acid technologies can charge, and an LFP bank will take that C/2 charge rate for longer than lead acid batteries. The result is often a cooked alternator, which can burn itself up delivering that charge current. To prevent this, you can use an external smart regulator that can derate the alternator based on its temperature. Balmar makes the most well known alternators and smart external regulators.  If you have an alternator with an internal regulator but an external voltage sense, you can use a less expensive charge reference controller and your existing alternator. All these devices allow you to select your charging profile, so you can use them with pretty much any battery chemistry.   

  • Like 1
Link to post
Share on other sites

What George said.

The only other concern is if you have pure LiFePo system, then the BMS can abruptly turn off charging in certain situations.  If that happens while charging via an alternator, that can be an issue...  I believe the Balmar system can handle that so as not to damage your alternator, but if you have a separate FLA starter battery it's actually simpler to just leave something FLA in the system to absorb the alternator current if/when the LiFePo BMS shuts off...

Link to post
Share on other sites
6 hours ago, George Dewey said:

Generally true for the wiring and battery charger, but not always for an alternator. C/2 is still faster than led acid technologies can charge, and an LFP bank will take that C/2 charge rate for longer than lead acid batteries. The result is often a cooked alternator, which can burn itself up delivering that charge current. To prevent this, you can use an external smart regulator that can derate the alternator based on its temperature. Balmar makes the most well known alternators and smart external regulators.  If you have an alternator with an internal regulator but an external voltage sense, you can use a less expensive charge reference controller and your existing alternator. All these devices allow you to select your charging profile, so you can use them with pretty much any battery chemistry.   

I don't disagree with your statement but I thought "your alternator may taper its output due to heat but as long as its got good thermal protection, there's no harm done" was a sufficiently qualified statement.

Link to post
Share on other sites
1 hour ago, SimonGH said:

What George said.

The only other concern is if you have pure LiFePo system, then the BMS can abruptly turn off charging in certain situations.  If that happens while charging via an alternator, that can be an issue...  I believe the Balmar system can handle that so as not to damage your alternator, but if you have a separate FLA starter battery it's actually simpler to just leave something FLA in the system to absorb the alternator current if/when the LiFePo BMS shuts off...

Your alternator(s) should have clamping diodes across the outputs for self protection whether or not you've got LFP's installed. 

Link to post
Share on other sites
2 hours ago, SimonGH said:

What George said.

The only other concern is if you have pure LiFePo system, then the BMS can abruptly turn off charging in certain situations.  If that happens while charging via an alternator, that can be an issue...  I believe the Balmar system can handle that so as not to damage your alternator, but if you have a separate FLA starter battery it's actually simpler to just leave something FLA in the system to absorb the alternator current if/when the LiFePo BMS shuts off...

Xantrex handles it by dumping the charge to a start battery. At least, thats what it seems to do. I’ve done the research but not made the plunge as I’d have to replace the alternator, add a BMS and the batteries. Cost of a kite…

Link to post
Share on other sites
3 hours ago, SimonGH said:

The only other concern is if you have pure LiFePo system, then the BMS can abruptly turn off charging in certain situations.  If that happens while charging via an alternator, that can be an issue...  I believe the Balmar system can handle that so as not to damage your alternator, but if you have a separate FLA starter battery it's actually simpler to just leave something FLA in the system to absorb the alternator current if/when the LiFePo BMS shuts off...

 

1 hour ago, IStream said:

Your alternator(s) should have clamping diodes across the outputs for self protection whether or not you've got LFP's installed. 

Actually, Balmar specifically states that their alternators will be damaged by this and they will not cover that under warranty. They recommend that a Sterling alternator protection device be installed. This is clamping diodes IStream mentioned. Or as SimonGH said, you can have a small FLA battery to absorb that spike. 

I recently installed a Balmar 6 series with a new 618 regulator. If anyone is looking for a a great deal on a brand new, never used, Balmar 614 regulator, PM me ;-)

 

Link to post
Share on other sites
1 hour ago, IStream said:

I don't disagree with your statement but I thought "your alternator may taper its output due to heat but as long as its got good thermal protection, there's no harm done" was a sufficiently qualified statement.

Fair enough, and don't mean to split hairs, but it's really important for people to understand that most stock alternators lack that thermal protection, and won't taper their output. Instead they will just run until they melt, and the time it takes to get to the melting stage is not very long. 

Link to post
Share on other sites
7 minutes ago, Raz'r said:

Xantrex handles it by dumping the charge to a start battery. At least, thats what it seems to do. I’ve done the research but not made the plunge as I’d have to replace the alternator, add a BMS and the batteries. Cost of a kite…

I guess it depends upon what kind of boat you have and how big a kite, but this can be done for probably under $3,000. Real money to be sure. 

Link to post
Share on other sites
1 minute ago, George Dewey said:

I guess it depends upon what kind of boat you have and how big a kite, but this can be done for probably under $3,000. Real money to be sure. 

I'd need a new alternator, external regulator, BMS and 2 100AH batts(I've decided if I do it, I'd get brand name ones).  More than $3k. Even if I do the install, which I would.

 

Other than weight, and a bit shorter time charging, there's not much more in it, since I have 2 year old Firefly Carbon/Foam AGMs that are pretty bulletproof (and can discharge to 70%)

 

Link to post
Share on other sites
1 minute ago, Raz'r said:

I'd need a new alternator, external regulator, BMS and 2 100AH batts(I've decided if I do it, I'd get brand name ones).  More than $3k. Even if I do the install, which I would.

 

Other than weight, and a bit shorter time charging, there's not much more in it, since I have 2 year old Firefly Carbon/Foam AGMs that are pretty bulletproof (and can discharge to 70%)

 

Brand name batteries will boost the cost to be sure. I did the conversion to get more capacity for offshore racing, so I can charge less frequently, and more quickly when I do charge. If people are not offshore racing then I agree there is limited benefit, mostly weight reduction. The Firefly Carbon Foam batteries are very nice. 

Link to post
Share on other sites
9 minutes ago, George Dewey said:

Brand name batteries will boost the cost to be sure. I did the conversion to get more capacity for offshore racing, so I can charge less frequently, and more quickly when I do charge. If people are not offshore racing then I agree there is limited benefit, mostly weight reduction. The Firefly Carbon Foam batteries are very nice. 

The ability to rapidly charge is the main reason you buy lithium 

failure to upgrade the rest of the system ,to take advantage of this , is a fools game 

even the carbon foam batteries benefit from a charging system upgrade 

Link to post
Share on other sites
8 hours ago, George Dewey said:

C/2 is still faster than led acid technologies can charge, and an LFP bank will take that C/2 charge rate for longer than lead acid batteries. The result is often a cooked alternator, which can burn itself up delivering that charge current. 

First sentence only half true. In fact a replacement LFP battery bank may charge at a lower rate. I've got a 440 ah Lifeline AGM bank which will readily take 280 amps for 30 minutes or more when down to 50% (280 is all I've got). Lifelines are not current limited on charge. In contrast, most LFP manufacturers of these size batteries suggest C/2 as a max and many call out C/4 if you want maximum life. If replacing a 440 ah AGM bank I'd go with a 200 or 300 ah LFP to get the same useable capacity. The max charge rate would be 100 or 150 amps. Even with a 400 ah bank, 200 amps max. Now, if you are replacing some Walmart flooded cells, maybe. I was surprised at the lower rate of charge when looking at LFP. 

Second sentence, true. The faster recovery rate of LFP over good AGM isn't the higher rate of charge - which can actually start lower - it is the lack of a long charge tail. 

  • Like 1
Link to post
Share on other sites
1 hour ago, George Dewey said:

 

Actually, Balmar specifically states that their alternators will be damaged by this and they will not cover that under warranty. They recommend that a Sterling alternator protection device be installed. This is clamping diodes IStream mentioned. Or as SimonGH said, you can have a small FLA battery to absorb that spike. 

I recently installed a Balmar 6 series with a new 618 regulator. If anyone is looking for a a great deal on a brand new, never used, Balmar 614 regulator, PM me ;-)

 

I thought there was a way to tell the balmar external regulator (the 614) that the BMS shut off charging and it would turn off the alternator output?

For my application where I have a dedicated engine start battery, it just seemed to make sense to leave that as FLA.

Link to post
Share on other sites
56 minutes ago, slug zitski said:

The ability to rapidly charge is the main reason you buy lithium 

failure to upgrade the rest of the system ,to take advantage of this , is a fools game 

even the carbon foam batteries benefit from a charging system upgrade 

It may be the reason you buy lithium but it's not necessarily the reason other people buy lithium. In my case, it was size and weight. I was able to replace four 8D AGMs with three group 27 LFPs and recover a lot of storage and give up about 300lbs.

As it turns out, even without charging system upgrades you may be able to charge faster with LFP because they don't have the long, low charge acceptance behavior of lead (as DDW points out). In my case, my 125A shore charger can refill my house bank from 10% SOC to 90% SOC in about 3 hours, which is about half what it took before. My alternators are much more effective whenever my bank is over 70% SOC. And my 870W of solar is unthrottled all day long.

  • Like 1
Link to post
Share on other sites
17 minutes ago, SimonGH said:

I thought there was a way to tell the balmar external regulator (the 614) that the BMS shut off charging and it would turn off the alternator output?

For my application where I have a dedicated engine start battery, it just seemed to make sense to leave that as FLA.

There is, as long as the BMS has such an output signal. But, the BMS should not be acting as a charge controller anyhow. The regulator charge profile should be set for the max voltage, and the BMS cutoff should be higher than that, so as long as things are working right the BMS never shuts off the charge current. 

I agree that FLA is a fine solution for a dedicated start battery.  

Link to post
Share on other sites
1 hour ago, George Dewey said:

There is, as long as the BMS has such an output signal. But, the BMS should not be acting as a charge controller anyhow. The regulator charge profile should be set for the max voltage, and the BMS cutoff should be higher than that, so as long as things are working right the BMS never shuts off the charge current. 

I agree that FLA is a fine solution for a dedicated start battery.  

I believe that technically to be ABYC compliant you're supposed to have a way to warn the operator that the BMS is going to shut down before it does...  so you can be warned that your instruments / autopilot is going to shutdown in a situation where the BMS kills the battery output...

Link to post
Share on other sites
9 minutes ago, SimonGH said:

I believe that technically to be ABYC compliant you're supposed to have a way to warn the operator that the BMS is going to shut down before it does...  so you can be warned that your instruments / autopilot is going to shutdown in a situation where the BMS kills the battery output...

Well, that's an excellent point. I know Xantrex has a wireless monitor that runs on an app on your phone, but that's nothing like a flashing red light that kicks off.

Link to post
Share on other sites
33 minutes ago, SimonGH said:

I believe that technically to be ABYC compliant you're supposed to have a way to warn the operator that the BMS is going to shut down before it does...  so you can be warned that your instruments / autopilot is going to shutdown in a situation where the BMS kills the battery output...

Yeah that's the only thing my BMS is missing that I would like it to do. 

Link to post
Share on other sites

This is one really nice feature of my Valence BMS. It'll put out a warning signal for over-temp or other issues well in advance of cutoff that can be used to kill the alt and otherwise gracefully prepare for the batteries going offline.

Link to post
Share on other sites
On 10/5/2021 at 3:21 PM, gbkersey said:
The LiFePO4 batteries are the safest type of Lithium batteries as they will not overheat, and even if punctured they will not catch on fire. The cathode material in LiFePO4 batteries is not hazardous, and so poses no negative health hazards or environmental hazards.
 

Uhm......  Skip ahead to 2:18.  :unsure:

 

 

Link to post
Share on other sites

Since there’s a bunch of folks more experienced with LifePo’s here, what recommendations do y’all have for my boat. The 100 aH battery has a built in BMS. Single battery, shore power will be a simple 10 or 20 amp Lithium capable charger. Engine is just a 15 hp outboard with a 15 amp stator. I might add a small amount of solar that would only be connected when distance racing. Rapid charging isn’t a concern. It’s a race boat with minimal house loads and a pull start engine, so nothing too fancy. Do I really need a fancy charge controller?  I’m redoing every inch of electrical on the boat before I launch in the spring, so might as well do it mostly right. 

Link to post
Share on other sites
11 minutes ago, Winston29 said:

Uhm......  Skip ahead to 2:18.  :unsure:

 

 

 

Good thing this kid with the batteries is not a sailor. "I'm going to stay down wind of it"... and then walks upwind to get out of the smoke!

  • Like 1
Link to post
Share on other sites

There are many videos of LFP batteries being drilled, cut, punctured and so on with no flame. If you dead short it, maybe, but that's true of pretty much any battery. 

  • Like 2
Link to post
Share on other sites
On 10/6/2021 at 3:22 AM, Max Rockatansky said:

Make sure you have a BMS for your batts that disconnects, not just gives warning…

I'm not following...   Disconnects what, when?  

The only thing I could find in the specs, in regards to the BMS and "disconnecting", was a low temp, charging disconnect, but I won't be charging below 32*.  I'll be charging it indoors with (hopefully) a NOCO Genius-5, 5A smart-charger.  

Or did you mean something else?  

This new battery will be getting very little use.  Worst case scenario, I'll have to motor 2.5 - 3.0 miles, at 1/2 - 3/4 throttle (less than 42A), back to the dock, if the wind dies.  (The entrance to the channel is about 100 yards past those distant high-tension towers, in this photo).  

My main reason for getting the LiFeP04 is because it will reduce the battery weight in the boat by 50%.  Doubling the Amp/Hrs is a nice bonus, but one I won't likely be using much.  

fullsizeoutput_b9a.jpeg

Link to post
Share on other sites
15 minutes ago, Winston29 said:

I'm not following...   Disconnects what, when?  

The only thing I could find in the specs, in regards to the BMS and "disconnecting", was a low temp, charging disconnect, but I won't be charging below 32*.  I'll be charging it indoors with (hopefully) a NOCO Genius-5, 5A smart-charger.  

Or did you mean something else?  

This new battery will be getting very little use.  Worst case scenario, I'll have to motor 2.5 - 3.0 miles, at 1/2 - 3/4 throttle (less than 42A), back to the dock, if the wind dies.  (The entrance to the channel is about 100 yards past those distant high-tension towers, in this photo).  

My main reason for getting the LiFeP04 is because it will reduce the battery weight in the boat by 50%.  Doubling the Amp/Hrs is a nice bonus, but one I won't likely be using much.  

 

AAUI, one of the jobs of the BMS is to protect the battery if the charging/discharging exceeds safe parameters. In those cases, it can take the battery "offline" - effectively dead at that time. I've also read that some of the integral BMSs that you can get, can fail on their own. Again, no battery. That's why some folks recommend external BMSs, so you can bypass if needed. Some will recommend "build your own" so that if one of the 4-cell batteries goes out, you can reconfigure and get back some battery capacity. Not sure how all this works in practice, except it tells me it's not yet "install and forget"

Link to post
Share on other sites
2 hours ago, Winston29 said:

The only thing I could find in the specs, in regards to the BMS and "disconnecting", was a low temp, charging disconnect, but I won't be charging below 32*.  I'll be charging it indoors with (hopefully) a NOCO Genius-5, 5A smart-charger.  

High voltage cut off, low voltage cut off, over-current/short circuit cut off,  over temp cut off and low temp cut off are typical, I think. Some folks are worried about blowing alternator diodes, frying windings or having fancy electrical systems crash. I don't know what the options list is for a WindRider-16 but it's just about possible that some of these issues may not be tiny trimaran type terrors.

 

 

  • Like 1
Link to post
Share on other sites
40 minutes ago, weightless said:

High voltage cut off, low voltage cut off, over-current/short circuit cut off,  over temp cut off and low temp cut off are typical, I think. Some folks are worried about blowing alternator diodes, frying windings or having fancy electrical systems crash. I don't know what the options list is for a WindRider-16 but it's just about possible that some of these issues may not be tiny trimaran type terrors.

 

 

Thanks for that link!  I've seen that guy's channel, but not that video.  (I'm a little afraid to watch it). :P

I'll just be running a small, 45# thrust, trolling motor.  Nothing else.  

Link to post
Share on other sites
1 hour ago, Winston29 said:

(I'm a little afraid to watch it). :P

 

I take it back.  He gives it very high marks for build quality in all areas except low temp charge protection, which it seems it doesn't have.  No big deal for me.  My part of California rarely drops to below freezing.  If it gets that cold, I won't be sailing anyway, so charging won't be an issue.  

  • Like 1
Link to post
Share on other sites

I’ve been rocking the (then) cheapest 10Ah LiFePO4 12v battery (w/ built in BMS) from Amazon for a little over 3 years now and it’s been great for getting the weight down in my boat. 

  • Like 1
Link to post
Share on other sites
Quote

 In fact a replacement LFP battery bank may charge at a lower rate.  If replacing a 440 ah AGM bank I'd go with a 200 or 300 ah LFP to get the same useable capacity. The max charge rate would be 100 or 150 amps. 

You are incorrect about the LFP charge rates.   It's not DIVIDED by the capacity, its multiplied.   Most LFP pack are rated at 2-3C meaning 2-3 times the banks size. An LFP 100Ah pack will easily charge at 200A or higher.   I've never seen a quality pack recommend a fraction charge rate of the pack.   Now maybe all these CHEAP, CRAPPY built inexpensive packs like a 100Ah for $400-700 from Amazon.   All of these packs DO NOT AND CANNOT support the charge/discharge rate of the actual batteries because they are using an inexpensive BMS with tiny wiring.   Watch some of the youtube videos of inexpensive packs being taken apart....   they certainly don't have proper sized relays or wiring (a 16g wire isn't capable of 100amps, needs a fat cable).   A proper relay alone is $150.   A decent BMS that still needs assembly is $500+, more like $1k for a complete BMS build.   

Link to post
Share on other sites
8 minutes ago, solosailor said:

You are incorrect about the LFP charge rates.   It's not DIVIDED by the capacity, its multiplied.   Most LFP pack are rated at 2-3C meaning 2-3 times the banks size. An LFP 100Ah pack will easily charge at 200A or higher.   I've never seen a quality pack recommend a fraction charge rate of the pack.   Now maybe all these CHEAP, CRAPPY built inexpensive packs like a 100Ah for $400-700 from Amazon.   All of these packs DO NOT AND CANNOT support the charge/discharge rate of the actual batteries because they are using an inexpensive BMS with tiny wiring.   Watch some of the youtube videos of inexpensive packs being taken apart....   they certainly don't have proper sized relays or wiring (a 16g wire isn't capable of 100amps, needs a fat cable).   A proper relay alone is $150.   A decent BMS that still needs assembly is $500+, more like $1k for a complete BMS build.   

Umm, no.

Victron, page 25: https://www.victronenergy.com/upload/documents/Lithium_Battery_Smart/Lithium_Smart_Battery_Manual-en.pdf

Mastervolt, page 31: https://www.mastervolt.com/products/li-ion/mli-ultra-12-1250/

Valence and others, all 0.5C. It is possible to charge faster if you can keep the batteries cool but for routine use, 0.5C is the recommended rate pretty much across the board.

Link to post
Share on other sites
Quote

I take it back.  He gives it very high marks for build quality in all areas except low temp charge protection, which it seems it doesn't have. 

Maybe high marks for good solders and zip ties but no way it can handle hundreds of amps, simply impossible.   That is why most of these packs have VERY LOW amp ratings for discharge and charge compared to what the cells can actually do.

Link to post
Share on other sites
Quote

Valence and others, all 0.5C. It is possible to charge faster if you can keep the batteries cool but for routine use, 0.5C is the recommended rate pretty much across the board.

Lithionics is 1C.    It's the BMS that is the limitation.   All LFP cells can handle multiple C rate of charge.  I've worked on so many packs that I can't remember but many are setup to charge at 2C+ with no issued to the cells.   Think 240V AC charger and you get the idea.  Of course the BMS has wiring and contractors to handle that load as well.   

Link to post
Share on other sites
3 minutes ago, solosailor said:

Maybe high marks for good solders and zip ties but no way it can handle hundreds of amps, simply impossible.   That is why most of these packs have VERY LOW amp ratings for discharge and charge compared to what the cells can actually do.

Yes. Very expensive to build reliable high current electronics. Lower current (fractional C) BMS devices might be addressing standby power for a computer system. However, even a low current BMS requires a clever design to prevent damage from unexpectedly high discharge faults. The BMS industry appears to be about 90% amateur engineering.

Link to post
Share on other sites

Yes, the BMS in that tear down looks to be a $20-40 eBay/Amazon BMS.   I just ordered an Orion Jr2 BMS system.   My dealer cost for the components were over $1k for the BMS, wiring looms, wifi adaptor, SOC monitor and still needs a $150 contractor.    Of course this BMS can do Dual Bus (separate charge/discharge) like the Lithionics and is fully programmable.

Link to post
Share on other sites

Also, another thing to know about these inexpensive packs with integrated BMSs is that if any fault is triggered the battery is shut off and you NO LONGER have access to your amps in the packs.   That is why I usually build with an external BMS and put in a physical bypass switch to get around the BMS.   Tell me you like the idea of a shitty solder on a 22g wire to make it so you can't use  your VHF in an emergency.   Even more critical in propulsion systems that use a BMS.

  • Like 1
Link to post
Share on other sites
14 minutes ago, solosailor said:

Also, another thing to know about these inexpensive packs with integrated BMSs is that if any fault is triggered the battery is shut off and you NO LONGER have access to your amps in the packs.   That is why I usually build with an external BMS and put in a physical bypass switch to get around the BMS.   Tell me you like the idea of a shitty solder on a 22g wire to make it so you can't use  your VHF in an emergency.   Even more critical in propulsion systems that use a BMS.

For many reasons, I agree with favoring an external BMS over a built in one. There are many choices, as people have said, ranging from $20 units to very expensive ones. If you are moving a lot of current around you need an external "contactor" or relay, as opposed to FETs. I use the Overkill Solar BMS which is not ideal, but it is well supported and meets my needs since I am not moving significant current. It's made with quality FETs (I reviewed the datasheet) and I'm passing less than 100A through it for charging, and discharging at a max of maybe 20A tops. I did size the wires correctly. 

One thing to consider with the contactors is the amount of current they consume just to stay closed, typically 1A or more. These devices tend to be physically heavy and power hogs. Now if you have a ginormous battery bank and routinely run 30+ amps of appliances and inverters and such, not such a big deal. If you have an offshore racer, different story. I did try a self latching relay from 123 BMS (the 123 Smart Relay I think) and it worked very nicely, and can pass about 100A (from memory - double check). This is a nice solution if you don't want to risk FETs, but it needs to be paired with a BMS that supports it. Still, it was one more thing to deal with so it's sitting on a shelf somewhere. 

As for power for emergencies, big consideration. With my luck, the BMS will cut power during a MOB situation half way to Bermuda. Many people have a small FLA type battery in conjunction with their LFP bank for this reason. This is an important consideration; lives could depend upon it. 

Link to post
Share on other sites
Quote

One thing to consider with the contactors is the amount of current they consume just to stay closed, typically 1A or more.

Many do....   the Givavac I've been using draws 0.68A.

Link to post
Share on other sites
18 minutes ago, weightless said:

I dunno, so, I googled bi-stable dc contactor. 1st hit:  https://www.littelfuse.com/products/dc-solenoids-and-relays/bi-stable-latching-relays.aspx

Why not?

Well I did mention self-latching relays above. Some of these are controlled by pulses of power, as opposed to continuous power. Your BMS needs to be able to properly control it.

Additionally, some people are concerned that if the BMS fails, the relay will remain latched, allowing power to continue to be drawn until the battery is destroyed. With a traditional contactor, the failure mode is usually to open the circuit (that is, cut off the load) when the BMS fails, so the battery can not be drawn too low and destroyed. This is a valid concern, but again, it all depends upon the BMS and how it's all connected. 

  • Like 1
Link to post
Share on other sites

FWIW, the Gigavac GV200 500A contactors are pretty efficient dual-coil units that switch using a high current coil but hold with a low current coil. Mine draws 0.23A at 12V (5.5Ah/day). The most efficient 500A electromechanical I've found is the TE Kilovac EV200 series, which draws about half what the Gigavac GV200 does. You'll pay for it, though, as they're over US$300. 

The best solid state relays I've found are from Nominal Controls. Their bidirectional unit maxes out at 300A and draws about 40mA. You can find them rated for 500A but only in one direction (charge or discharge). I'd say they're not quite ready for this application. 

  • Like 1
Link to post
Share on other sites
1 minute ago, IStream said:

The best solid state relays I've found are from Nominal Controls. Their bidirectional unit maxes out at 300A and draws about 40mA. You can find them rated for 500A but only in one direction (charge or discharge). I'd say they're not quite ready for this application. 

I experimented with a few different solid state relays (SSRs) and they tended to fail, and when the did, they failed closed. Plus there was a slight voltage drop. I also decided not to use them. 

Link to post
Share on other sites
1 minute ago, George Dewey said:

I experimented with a few different solid state relays (SSRs) and they tended to fail, and when the did, they failed closed. Plus there was a slight voltage drop. I also decided not to use them. 

Yeah, kinda defeats the purpose. I use and love SSRs for less critical stuff and actually have one in place between my solar panels and my charge controller so the BMS can cut them off when the batteries reach 95% SOC. Same with the ignition voltage to my Balmar regulator, but both SSRs are operating well below their ratings and if they were to fail closed, it wouldn't be as big a deal as the main charge/discharge contactor, which is the last line of defense.

  • Like 1
Link to post
Share on other sites
2 minutes ago, IStream said:

... but both SSRs are operating well below their ratings and if they were to fail closed, it wouldn't be as big a deal as the main charge/discharge contactor, which is the last line of defense.

Yup, good design. I'm using FETs for the cutoff, because I don't want the weight or current draw of a mechanical contactor. But if someone can spare that power and weight then it's the best way to go. 

Link to post
Share on other sites

I'm not clear on the functional differences between an SSR and a FET in a dc circuit. A pattern that I've seen with FETs is a p-channel with a pullup driven by an open collector. I suppose that an SSR would keep the body diode from conducting and maybe add some kind of flyback protection for the FET for inductive loads? Maybe? Are there conventions about which is appropriate where?

It seems like most SSRs are made for AC where their ability to switch near the zero crossing may be desirable. That doesn't seem meaningful in a dc circuit.

Link to post
Share on other sites

Many electronic products protect their power circuits without power consumption or voltage drop by using a relay or solid state device to create a short which causes a fuse to blow. Good design if the event is rare. 

Link to post
Share on other sites
1 minute ago, El Borracho said:

Many electronic products protect their power circuits without power consumption or voltage drop by using a relay or solid state device to create a short which causes a fuse to blow. Good design if the event is rare. 

"Crowbar"?

  • Like 1
Link to post
Share on other sites
16 minutes ago, weightless said:

I'm not clear on the functional differences between an SSR and a FET in a dc circuit. A pattern that I've seen with FETs is a p-channel with a pullup driven by an open collector. I suppose that an SSR would keep the body diode from conducting and maybe add some kind of flyback protection for the FET for inductive loads? Maybe? Are there conventions about which is appropriate where?

It seems like most SSRs are made for AC where their ability to switch near the zero crossing may be desirable. That doesn't seem meaningful in a dc circuit.

It's basically the same thing except an SSR is packaged as an integrated component with a basic heatsink, mounting holes, and an indicator LED. You wire it up via two load terminals, two control terminals, and you're done.

You can get them for both DC and AC use. I'm using the Crydom DC200D20's for my DC circuits because that's what I had on hand. They do impose a ~0.75V diode drop.

http://www.crydom.com/en/products/panel-mount/perfect-fit/dc-output/powerplus-series/

  • Like 1
Link to post
Share on other sites
1 hour ago, IStream said:

They do impose a ~0.75V diode drop.

I'm struggling to read today so please forgive me if it's just sitting right there in front of me but I don't see why there'd be a diode drop. I see rds-on is 0.11 ohms so I understand that drop.

image.png.bcf66f1a2f034ef971159779db6060d2.png

Link to post
Share on other sites

The spec sheet for the DC200D20 claims 0.78V drop at rated current. I just assumed it was a diode drop because it's about the right voltage for it but I hadn't looked at the circuit diagram and I agree that it doesn't make sense. Sorry for being stupid. It looks like the current should get through that MOSFET without any significant nonlinear drop. I haven't actually measured the drop across my SSRs. 

  • Like 1
Link to post
Share on other sites

@weightless The DC200D20 (200V/20A) has RDS-ON of 0.039 V. At 20 A that will be 0.78 V. You may have seen the 400 V part spec.

More importantly, that schematic implies the current is intended to flow from +DC terminal to -DC terminal. My guess is it will not function as intended if the          -DC terminal is at a higher voltage than the +DC terminal. Both the trigger circuit power supply and the circulation diode are problematic when reverse biased. So not a good choice for charge/discharge protection of a battery.

  • Like 2
Link to post
Share on other sites
34 minutes ago, El Borracho said:

You may have seen the 400 V part spec.

Yep. I did.

The equivalent circuit isn't quite what I was expecting either. I thought there would be back to back FETs or a diode. I don't know anything about SSRs. Maybe this is typical of them? Worth keeping in mind if replacing a mechanical relay with a solid state one.

Link to post
Share on other sites
8 minutes ago, weightless said:

Yep. I did.

The equivalent circuit isn't quite what I was expecting either. I thought there would be back to back FETs or a diode. I don't know anything about SSRs. Maybe this is typical of them? Worth keeping in mind if replacing a mechanical relay with a solid state one.

Yes, some unexpected complexity there. Helps explain why the BMS suppliers get the big bucks when they do it right. 

Link to post
Share on other sites
21 minutes ago, weightless said:

News to me: they make SSRs that are for AC only, for DC only (as above) or that are agnostic but don't do the the zero crossing thing. eg:

(tiny relay just to show the concept)

https://media.digikey.com/pdf/Data Sheets/Fairchild PDFs/HSR312(L),412(L)_DS.pdf

image.thumb.png.6570e7a44372dd9d8b40bfa4e2e452d5.png

Yes, everything is easy with miniscule currents. For large currents those bypass (circulating? flyback?) diodes are going to be problematic. Voltage drop and the resulting heat must be addressed. Also the drive capability of the opto isolator would need to be looked at. Power MOSFETS have significant gate charge currents (Qg? Cgs+Cgd? It's all a long time ago.) which is why the CRYDOM part has a driver circuit. Slow switching causes smoke. Isolation may not be needed for your application, perhaps. Complicated. That Overkill Solar engineer seems to have a handle on it, among others.

Link to post
Share on other sites
1 hour ago, El Borracho said:

For large currents those bypass (circulating? flyback?) diodes are going to be problematic.

I was thinking those were just depictions of the body diodes of the FETs. No?

For sure a BMS for a big battery is a job for the pros. I'm just trying to get a high level understanding. I am stoked to have learned some things about SSRs that might be generally useful.

In theory, I suppose one could make up a high current bidirection SSR with similar logic as above from discreet SSRs in series. I think there are details that will matter. While it's a bit short on specifics I found this interesting technical note from crydom:

http://www.crydom.com/en/tech/newsletters/solid statements - using ssrs in parallel and-or series.pdf

image.png.9e8d777410c65169c240783f4f456f1a.png

image.thumb.png.6f2e2cb65bb9b710b5e2ed80b7573fcd.png

Link to post
Share on other sites

Doesn't matter where the diode is. It is still a diode with voltage drop. That circuit does address the bidirectional issue. But the booby prize is both a forward junction diode voltage and a MOSFET RDS(on) resistance. They hint at this with the "higher power dissipation" and "larger heat sink" phrasing. Sounds lossy. Also the RDS(on) resistance specs are idealized. At a low Tj temperature and VGS - which of course is always worse than the brochure conditions. Perhaps by much.

Try it and report back. I'm liking the crowbar idea. MOSFET or TRIAC. Carry a some spare fuses.

(Yeah, I just now rediscovered the cool subscript button...)

Link to post
Share on other sites
4 minutes ago, El Borracho said:

Doesn't matter where the diode is. It is still a diode with voltage drop.

I think the diode is a byproduct of how mosfets are constructed. It is bypassed when the mosfet is on to get rds-on so no diode drop. Someone else drew an 8x10 color glossy picture with circles and arrows and a paragraph on the back.

https://electronics.stackexchange.com/questions/389406/how-should-i-understand-the-intrinsic-body-diode-inside-a-mosfet

image.png.6203d2c92535b4f9fbb4806d5a87a4b3.png

Link to post
Share on other sites
13 minutes ago, weightless said:

I think the diode is a byproduct of how mosfets are constructed. It is bypassed when the mosfet is on to get rds-on so no diode drop. Someone else drew an 8x10 color glossy picture with circles and arrows and a paragraph on the back.

https://electronics.stackexchange.com/questions/389406/how-should-i-understand-the-intrinsic-body-diode-inside-a-mosfet

image.png.6203d2c92535b4f9fbb4806d5a87a4b3.png

The OFF relay (SSR) gives the diode drop in series with the ON relay Rds(on).

 

Link to post
Share on other sites
3 minutes ago, El Borracho said:

The OFF relay (SSR) gives the diode drop in series with the ON relay Rds(on).

 

Both should be off or both should be on. Then you get two times RDS-ON with no diode drop. Both the reverse and forward biased diodes get bypassed when both are on.

Link to post
Share on other sites
21 minutes ago, weightless said:

Both should be off or both should be on. Then you get two times RDS-ON with no diode drop. Both the reverse and forward biased diodes get bypassed when both are on.

I don't think that is how the CRYDOM devices work. I don't think they can be ON when reverse biased. So the diode conducts.

Re your body diode. Yes, it appears when the device is reverse biased. A freebie. They draw it as a separate device for clarity. Same function. Same losses.

Link to post
Share on other sites