LFP FLA hybrid

[john61ct]

Overcharging is less black and white.

To me 14.6V is way overcharging, but just reducing cycles lifespan to mfg rating

so "damage" remains invisible for many years.

Going higher even than that, dunno, at some point heat becomes an issue.

High amps is rarely a problem, for the cells anyway.

My point is, you really have to work to push past safety at the high end.

While on the low side, just ignore them long enough and they turn to scrap.

[Delfin]

Originally Posted by john61ct

For the long tail functionality, a 60-100AH LFP bank would be plenty to service even a very large FLA one.

The key variable is what SoC do you want to stop running your ICE charging. On say an 800AH bank you may need to go to 90%, where a 400AH bank you could stop sooner at 80%.

Either way you are saving many ICE runtime hours per year, the bigger the LFP bank the more you save.

Really? A LA battery is about 60% efficient in the taper phase of charging. So, if you have a "very large FLA" bank of oh, say, 1000 Ah, and the taper of the FLA is the last 150 amps, you need 250 amps or so to fill it. Add house loads during this process and your estimate that a 60 - 100 Ah LFP bank doing the job is grossly inadequate. Try doubling it, and you might get close.

Now, since that "very large FLA bank" of 100 Ah MAY deliver 400 amps of useful energy if you expect that FLA bank to last, a 600 Ah LFP bank can completely replace it and last for years and years. So why would you want to install a 300+ Ah LFP bank to top off a FLA bank, when a 600 Ah LFP bank will do the entire job AND deliver stable voltage throughout the discharge process?

[john61ct]

I was thinking much lower AH, the last couole hours may only be from 97%-100%.

But OK, maybe start by testing wuth

large Main bank is LFP, smaller Reserve is lead.

[CatNewBee]

Originally Posted by john61ct

CNB, you make a lot of exaggeratedly negative comments about lead / FLA.

Not going to argue them here, if you want to, start a new thread.

Just wanted to post a heads up for relative noobs to take them with a grain of salt.

With other members prejudiced in the other direction, too much against LFP, I guess things b balance out 8-)

I don't think it is too negative, it is realistic in my opinion.

No reason to glorify the FLA, they are consumables with a life span depending on use (amps), charging and storage between 1 year and 10 years, they stand a limited number of cycles and need to be replaced all together if one fails. FLA is not equal FLA so, there are different dessigns for different purposes at different price points.

I am just saying, get a differentiated view on the technologies and decide whats fits you best.

BTW, same for LiIon, LiFePO4, LiFeYPO4,, LiMnO2, Lead Crystal, GEL, AGM, Spiral cell AGM, Lead-Calcium, SMF or open to refil water batteries and all other chemistries and geometries out there.

The point is to make an educated decission and sort out prejudices and false claims. Neither technology is perfect, but one is often a better fit than the other to what you intent to do, for how long you want to use it, and what you are prepared to invest upfront.

[Dockhead]

Originally Posted by Delfin

Really? A LA battery is about 60% efficient in the taper phase of charging. So, if you have a "very large FLA" bank of oh, say, 1000 Ah, and the taper of the FLA is the last 150 amps, you need 250 amps or so to fill it. Add house loads during this process and your estimate that a 60 - 100 Ah LFP bank doing the job is grossly inadequate. Try doubling it, and you might get close.

Now, since that "very large FLA bank" of 100 Ah MAY deliver 400 amps of useful energy if you expect that FLA bank to last, a 600 Ah LFP bank can completely replace it and last for years and years. So why would you want to install a 300+ Ah LFP bank to top off a FLA bank, when a 600 Ah LFP bank will do the entire job AND deliver stable voltage throughout the discharge process?

Yeah, that was exactly my analysis. Except that the tail charge is actually much less efficient than 60% if you do a long low amp absorption charge like lead batteries like, so it's even worse than you state. The charge efficiency approaches 0% at the end -- you're using energy mostly just to heat up the lead batteries.



I think this pretty much invalidates the hybrid bank idea for me, if the lead is meant to be cycled regularly.



It might still be an OK solution for me, though, as the lead bank can still perform a very useful function providing uninterruptible backup to the lithium, and providing occasional use reserve capacity. For remote cruising or for expedition boats, these can be very valuable functions.



But the lithium bank in this configuration cannot be reduced in size on account of having the lead in the system.

[Dockhead]

Originally Posted by john61ct

I was thinking much lower AH, the last couole hours may only be from 97%-100%.

But OK, maybe start by testing wuth

large Main bank is LFP, smaller Reserve is lead.

Reserve bank doesn't even need to be small --


just don't cycle it regularly.



In my case this would be determined by the existing system -- I would leave the house bank as is -- 210 amp/hours (@24v) of lead. If my lithium bank blew up in some remote place, I could run the boat on that. Not luxuriously, but safely.


And roughly 105 amp/hours of power could be added to the capacity of the lithium occasionally when I don't feel like running the generator, say on a long passage under sail. That's roughly 8-10 hours of sailing on my boat in average conditions.


Means the lithium would need to be about 300 amp/hours * 24v.

[john61ct]

Originally Posted by Dockhead

Except that the tail charge is actually much less efficient than 60% if you do a long low amp absorption charge like lead batteries like, so it's even worse than you state. The charge efficiency approaches 0% at the end -- you're using energy mostly just to heat up the lead batteries.

If you mean the ratio between the total potential output of an ICE source, and that going into increasing SoC of the lead bank then that is true.

Like an 80A alt output where trailing amps is below 10A for hours.

Exactly why I want to avoid using ICE for that long tail.

I had thought you guys were talking about the ratio between what is accepted by the bank, vs the useful AH going into increasing SoC e.g. 25AH over three hours.

If it is the latter, any links to a test / demo of the ratio being that low?

You're saying a batt-batt connection will remove something like 100+AH from source, in order to get 25 usable AH into the target?

[john61ct]

Originally Posted by Dockhead

Reserve bank doesn't even need to be small -- just don't cycle it regularly.

Well that is true, by definition.

Say it is an expensive Lifeline bank, might only need a top-up every couple weeks.

Or if Firefly, just quarterly.

[john61ct]

This was my original design scenario Main / Reserve rather than Starter / House, when I posted the idea over a year ago.

I do think it is worth prototype testing IRL.

[rgleason]

Amen!

Like an 80A alt output where trailing amps is below 10A for hours.

Exactly why I want to avoid using ICE for that long tail.

[Qayaq]

Hi Forum,
Thank you for all your glorious posts.

Regarding my question as to what I could use to control my HVC and LVC?

It would seem that Arduino would be able to accomplish that.

HVC. I would want it to open circuit my charge bus to LFP line at X volts and not cut back in until LFP has dropped to X volts. Also to go Open Circuit if there is no charge source available.

LVC. I would want it to open circuit LFP switchboard line at X volts and not cut back in until LFP has risen to X Volts and also switch my FLA into the switchboard when the LFP is out.

So basically a fully custom programmable multiple voltage sensor that could trigger LVC and HVC.

I’m hoping that someone will be able to point me down the right Arduino road as to which Arduino shields might be available to accomplish this. Would be good to short circuit the Arduino learning and mistakes process.

As a pre-emptive to a lot of posts as to why I’m doing this, is it worth it, what about the cost, will it fail, why don’t you just keep FLA……LA LA

The reason why is that in my opinion it will be an improvement on the FLA system that I have.

Also because I WANT TO DO IT. Nothing like a bit of a hack to put a smile on my face.

I haven’t come up with a unit for LVC max 25 amp and HVC an unlikely max 100amp, so if anybody has any suggestions for this it would be helpful.

Cheers and Thanks
Simon

[CatNewBee]

Nice project. how do you want to measure cell voltages? A simple arduino shield will not do the trick...

A nice solenoid is the ML-RBS from BlueSea. It is good up to 500A, bi-stable, and is triggered by a positive impulse on the SET respective RESET inputs with very little current (built-in amplifier and logic). it does not use any power when not switching and can be switched manualy - if necessary.

https://www.bluesea.com/products/770..._-_12V_DC_500A

[Dockhead]

Originally Posted by Qayaq

Hi Forum,
Thank you for all your glorious posts.

Regarding my question as to what I could use to control my HVC and LVC?

It would seem that Arduino would be able to accomplish that.

HVC. I would want it to open circuit my charge bus to LFP line at X volts and not cut back in until LFP has dropped to X volts. Also to go Open Circuit if there is no charge source available.

LVC. I would want it to open circuit LFP switchboard line at X volts and not cut back in until LFP has risen to X Volts and also switch my FLA into the switchboard when the LFP is out.

So basically a fully custom programmable multiple voltage sensor that could trigger LVC and HVC.

I’m hoping that someone will be able to point me down the right Arduino road as to which Arduino shields might be available to accomplish this. Would be good to short circuit the Arduino learning and mistakes process.

As a pre-emptive to a lot of posts as to why I’m doing this, is it worth it, what about the cost, will it fail, why don’t you just keep FLA……LA LA

The reason why is that in my opinion it will be an improvement on the FLA system that I have.

Also because I WANT TO DO IT. Nothing like a bit of a hack to put a smile on my face.

I haven’t come up with a unit for LVC max 25 amp and HVC an unlikely max 100amp, so if anybody has any suggestions for this it would be helpful.

Cheers and Thanks
Simon

If I may echo CatNewBee -- this is a great project, and I hope if you do it, you will start a new thread and post about it day by day. You will get a lot of interest and a lot of good ideas from the folks here.


The logic of such a device would be simple enough -- besides Arduino I would think you could even use a programmable logic controller.

[Dockhead]

Originally Posted by john61ct

If you mean the ratio between the total potential output of an ICE source, and that going into increasing SoC of the lead bank then that is true.

Like an 80A alt output where trailing amps is below 10A for hours.

Exactly why I want to avoid using ICE for that long tail.

I had thought you guys were talking about the ratio between what is accepted by the bank, vs the useful AH going into increasing SoC e.g. 25AH over three hours.

If it is the latter, any links to a test / demo of the ratio being that low?

You're saying a batt-batt connection will remove something like 100+AH from source, in order to get 25 usable AH into the target?

No, I was talking about amp/hours in vs. amp/hours stored. I don't have the literature right to hand, but I understand that this ratio approaches infinity at the end of the tail charge. What we would want to know is how many amp/hours of power would it take to go from 85% or whenever we shut off the generator, to a nice full 100%. If it's a 200 amp/hour bank, then 15% of that is 30 amp/hours. I guess that will be something well above double, or 60 amp/hours, if not triple, then you have to add the efficiency losses in the charger, which are not trivial. I can easily imagine 100 amp/hours being needed which is, for God's sake, equal to the whole usable power of a 200 amp/hour lead bank


So as a vivid illustration of the futility of my original idea -- you would be better off just not using the 100 amp/hours out of the lead bank and just spending the 100 amp/hours straight from the lithium. And even if it's only 60 amps to do the tail charge, vs. 100 amps you will be able to use from the bank, then you can see how little there is to gain from this, by doing the tail charge off the lithium batteries. It was a bad idea.




And as you say, if you are running a generator doing nothing else but carry this light load, you have all that inefficiency (and possible damage to the generator). My generator uses 0.7 l/h just to run; then whatever power you take out of it requires another about 0.43 l/h per kW/h. So that's a huge inefficiency to run for the long tail, as you pointed out.


So this - charge inefficiency during the tail charge -- is just a giant disadvantage of lead/acid technology altogether, and this giant disadvantage is only masked, not solved, by having solar power. Solar is great because it has no overhead so doesn't mind working all day at low output, but the efficiency losses of the long tail are still there. Basically, unless you have massive overkill of solar capacity, you will have to do a lot of the bulk charging by generator or main engine in order to leave the solar time free for the long tail charge. With lithium you can capture and use all this wasted power, so you could have a smaller solar installation (less windage!) or capture and use more power.

[Qayaq]

Originally Posted by Dockhead

If I may echo CatNewBee -- this is a great project, and I hope if you do it, you will start a new thread and post about it day by day. You will get a lot of interest and a lot of good ideas from the folks here.


The logic of such a device would be simple enough -- besides Arduino I would think you could even use a programmable logic controller.

This is the thread that I started about this project.