43 DS - Let's start a fresh discussion on LiFePO4

[andreshs1]

Hi all

After 3 years, my AC Delco 440ah house bank has died.....so we have opted for upgrading to LiFePO4

Setup:
- Tao BMS
- 8 x 3.22V CATL 310ah Grade A cells
- Victron SmartSolar 150/45 MPPT
- Victron Orion Smart TR 12/12 30A non-isolated
- Mastervolt 12/1500 (to be upgraded to 3Kw)
- Fischerpanda 6kw
- 540Watts of solar (and planning to upgrade to 1kw ~ by end of the year)

I am pending the cells in order to do the installation (in theory next week)

I would be interested to hear from those with 'first hand' experience.

What works what doesn't
what should I consider

one of the many benefits that we are expecting to get out is to remove the gas cooker/oven, and replace by Induction and electric oven

Below my design:


Regards
Andres 

[fakinx]

Hello andreshs1,

I'm one of those guys who like hands-on on things regarding my boat so I just installed 700Ah of lithium on board of our SO 42DS.
Big job, not with install itself but with all designing and testing things in my shack for more than a year before transferring everything on board in one single leap. There is still a lot to do with arranging the cabling in order but it's functional and safe for time being.
My 2 cents on your setup are:

1.
MPPT for 540W (2x270W) of solar should really be 100/50 and not 150/45. You will probably (check your specs!) not exceed 100 V with 2 panels in parallel (which is alot!) so you should be fine. More important thing is you leave some spare capacity of MPPT charging not to overheat MPPT nor you want to lose any juice available. In summer there is very good yield from solar because panels are not overheating if they are mounted somewhere free of any obstacles on the bottom side. True, you'll rarely see max capacity charging but it could happen. It's better to operate on 80% instead of 90 or more (of MPPTs) Those gizmos get hot as hell... All said I'm inclined for separate MPPTs for any panel because of redundancy, safety and more flexibility with installation and so on. High voltage (more than 50V) can be dangerous in shaky environment of a boat.

2.
I would think about dedicated LiFePO4 charger and perhaps charge LA starter battery when solar available for charging Li battery and dedicated separate Orion charger and wouldn't really bother with dedicated shore charger for that. If you set up things with at least 2 MPPTs you can design 3-way switch for one of MPPTs to trickle charge starter battery when boat not in use. 

3.
Be aware that even if you kill all loads on board, gizmos for managing your new battery will suck somewhere between 300 and 600 mA so you should tackle that with some remote management or at least monitoring or you could return to the boat one day and your batteries could be dead.  I suggest at least 2 layers of battery management and you should really think about what-if scenarios.

4.
Using just an inverter is fine and you can implement 2-bus configuration but if you choose inverter/charger you need to design more advanced system especially regarding control of inverter/charger. I chose CAN enabled BMS for that.

Well, those were 4 cents but who is counting...   
All the best!

[andreshs1]

Hi there

Thanks for the feedback

I am also doing it all myself, tomorrow I will start the pre-installation, running cables, installing the MPPT and the Orion, etc. 
In parallel to this project, I will be re-wiring the DC system with tinned copper wire, as I have seen some corrosion on some of the original cabling 

I used the Victron calculator to see which I would need. The calculator identified the 150/30, but when shopping, the 150/45 was cheaper, and I am trying to get one more of the same panels I have, thus, increasing to 3 x 270watts and I like the idea to have plenty of spare capacity.

For the starter battery, most likely I will end up replacing that also for LFP in the future

for the 'what-if' scenarios, I am thinking on adding a switch, so I can use the starter battery as house bank and vice-versa in an emergency

for inverter charger I like the Victron, but I will wait for a while, maybe next Xmas 


in your setup, what are the benefits you are looking for? 
what solar array do you have?
water maker?



Cheers

[ianf]

Hi Andres,

I have a similar setup on my 42DS. The only problem I had was in relation to the generator. I don't know the arrangement with your generator installation but with my 4kW Fischer Panda 12 Volt generator it connects directly to the battery bank and the inverter.

I found that when I fired up the generator it would try and pump maximum output into the LiFePO4 batteries which exceeded the maximum charge rate for the batteries and the BMSs would shut off the batteries. My Victron batteries have individual internal BMSs.

To get around this I had to route the generator output to the batteries through a bank of Victron DC-DC chargers and, to allow full battery capacity to drive the inverter when the generator is not running and block the generator output by-passing the DC-DC chargers, I put some high capacity blocking diodes in the line from the batteries to the inverter rigged in parallel.  I did this as a temporary measure and when I get back to the boat this year I'm going to swap out the diodes for a single pole double throw (SPDT) high power relay as I don't like diodes.

I notice in your diagram you don't show your generator.

[fakinx]

Regarding MPPT that's odd. 270W panel should max out at +20A @12v nominal. Sure, it's charging LPO up to 14,4V with (up to) 18,7A but that's roughly 93% so it will be hot. For 3 such panels you lack roughly 20A with your current lone MPPT! Like I said before if you choose your location carefully you should see max yield in the summer now and then. But then again I do not know any technical details about your PV panels. 

My what-if reference didn't aim at total power loss of LPO battery since you plan 8 cells in 2P4S configuration. You can always reconfigure the bank and loose only half of the capacity if ONE cell goes haywire. But that is only in case you catch that. If you let one bad cell drag the rest of the pack down you're screwed. The same case is when one cell is not performing and get full way earlier than the others. In that case you need some sort of balancing so cells stay rather close to each other. I speak generally about HV LV HT LT and BMS failure scenarios and what your system will do automatically on such occasion. And all that should be considered on a cell basis and not on the pack as whole. That's why I pointed to two layers of management at least. One should be BMS and if that goes in electrical haven what then? Needless to say your BMS should perform as planned in 99,99% of the time and should be tested if it's able to do so.

My setup (for now) consists of 4x700Ah Winston cells (I know, they are heavy and I should use smaller ones but there are reasons for my decision about that), REC ActiveBMS (CAN enabled), bunch of electronics, 3x500A bi-stable switch/contactor with LV switch off capability, Victron Multi inverter/charger, 2xMPPT 100/50 with 780W of solar, Orion DC/DC 30A charger and Victron GX. Along with "smart" functions embedded in those gizmos I think (with moderate certainty) our system should perform as designed. When I'll get a Wakespeed regulator for my alt (but that means I'll need to alter my alternator and change belts and pulleys) I'll use Orion for charging starter battery instead of LPO charge that is configured today. Once in a future I'll change my remaining LA batteries (starter and BT) to LPO also and will implement another Orion for charging BT battery which is far away in the bow. For testing I used smaller 60Ah Winston cells which will serve perfectly for starter or BT battery but I didn't get to that yet. And from what I learned heat can be a serious problem when Orion gets to full output. I also tested 390W PV panel with 75/15 MPPT and it got very very hot so I decided to go with oversized MPPT for final config (and there is still some reserve in it if I ever decide to go with more powerful or additional PV panels - originally I had 2x200W and in less than 10 years they managed to squeeze almost 2x amount of energy from the same PV panel dimensions). Who knows what will near future bring in that field...

My goal is to equip totally self-sufficient (and quiet! so no generator) boat and since we are power hungry when on board LPO was obvious choice when our LA batteries (almost completely) died last year after lightning strike in our marina and our surge protection switches went off. We couldn't get to the boat in time due to all covid restrictions for travel. Watermaker is next big thing and I'll build that sucker myself too... I already have some spare membranes (although maybe too big) from my other project.

Good luck with your project.

[andreshs1]

Feb 25, 2022 8:43:40 GMT ianf said:

Hi Andres,

I have a similar setup on my 42DS. The only problem I had was in relation to the generator. I don't know the arrangement with your generator installation but with my 4kW Fischer Panda 12 Volt generator it connects directly to the battery bank and the inverter.

I found that when I fired up the generator it would try and pump maximum output into the LiFePO4 batteries which exceeded the maximum charge rate for the batteries and the BMSs would shut off the batteries. My Victron batteries have individual internal BMSs.

To get around this I had to route the generator output to the batteries through a bank of Victron DC-DC chargers and, to allow full battery capacity to drive the inverter when the generator is not running and block the generator output by-passing the DC-DC chargers, I put some high capacity blocking diodes in the line from the batteries to the inverter rigged in parallel.  I did this as a temporary measure and when I get back to the boat this year I'm going to swap out the diodes for a single pole double throw (SPDT) high power relay as I don't like diodes.

I notice in your diagram you don't show your generator.

Hi Ian

the generator spends more time broken that operational, thus, the plan is to get rid of it, the only dependency we have on it right now is the aircon, which, will in time be replaced by a DC capable one....


and also, the generator is not connected to the battery bank, so no risk from that side

Cheers

[andreshs1]

Feb 25, 2022 9:18:17 GMT fakinx said:

Regarding MPPT that's odd. 270W panel should max out at +20A @12v nominal. Sure, it's charging LPO up to 14,4V with (up to) 18,7A but that's roughly 93% so it will be hot. For 3 such panels you lack roughly 20A with your current lone MPPT! Like I said before if you choose your location carefully you should see max yield in the summer now and then. But then again I do not know any technical details about your PV panels. 

My what-if reference didn't aim at total power loss of LPO battery since you plan 8 cells in 2P4S configuration. You can always reconfigure the bank and loose only half of the capacity if ONE cell goes haywire. But that is only in case you catch that. If you let one bad cell drag the rest of the pack down you're screwed. The same case is when one cell is not performing and get full way earlier than the others. In that case you need some sort of balancing so cells stay rather close to each other. I speak generally about HV LV HT LT and BMS failure scenarios and what your system will do automatically on such occasion. And all that should be considered on a cell basis and not on the pack as whole. That's why I pointed to two layers of management at least. One should be BMS and if that goes in electrical haven what then? Needless to say your BMS should perform as planned in 99,99% of the time and should be tested if it's able to do so.

My setup (for now) consists of 4x700Ah Winston cells (I know, they are heavy and I should use smaller ones but there are reasons for my decision about that), REC ActiveBMS (CAN enabled), bunch of electronics, 3x500A bi-stable switch/contactor with LV switch off capability, Victron Multi inverter/charger, 2xMPPT 100/50 with 780W of solar, Orion DC/DC 30A charger and Victron GX. Along with "smart" functions embedded in those gizmos I think (with moderate certainty) our system should perform as designed. When I'll get a Wakespeed regulator for my alt (but that means I'll need to alter my alternator and change belts and pulleys) I'll use Orion for charging starter battery instead of LPO charge that is configured today. Once in a future I'll change my remaining LA batteries (starter and BT) to LPO also and will implement another Orion for charging BT battery which is far away in the bow. For testing I used smaller 60Ah Winston cells which will serve perfectly for starter or BT battery but I didn't get to that yet. And from what I learned heat can be a serious problem when Orion gets to full output. I also tested 390W PV panel with 75/15 MPPT and it got very very hot so I decided to go with oversized MPPT for final config (and there is still some reserve in it if I ever decide to go with more powerful or additional PV panels - originally I had 2x200W and in less than 10 years they managed to squeeze almost 2x amount of energy from the same PV panel dimensions). Who knows what will near future bring in that field...

My goal is to equip totally self-sufficient (and quiet! so no generator) boat and since we are power hungry when on board LPO was obvious choice when our LA batteries (almost completely) died last year after lightning strike in our marina and our surge protection switches went off. We couldn't get to the boat in time due to all covid restrictions for travel. Watermaker is next big thing and I'll build that sucker myself too... I already have some spare membranes (although maybe too big) from my other project.

Good luck with your project.

Hi

I am a bit confused, according to the panel specifications, out of 270w I should expect 9A in an ideal world. (see attached the highlighted one)


Following your comments, I have also checked the Renogy panels, and the expected performance is very similar

do you mind elaborating a bit more on your calculations? I want to make sure I am not missing anything

as mentioned, I based my selection on the Victron calculator, where I input each parameter from my panel

cheers

[fakinx]

Oh, I see.
You are calculating current at nominal voltage of your panel which is 29,81V. That's PV side, the input one. For MPPT it's the first number the second is on charge side of your battery which is 12V if I'm correct. Currents are different at the same power at different voltage. Power = Voltage * Amperage
That means you need to choose MPPT which has sufficient voltage capability on the input side and current capability on the charge side. On the charge side all Victron MPPTs auto sense voltage (12/24V) so no problem with that.

On the input side you are looking for numbers that can handle your voltage of all panels connected to one unit added together and on the other side about 70-85% of max current PV panels are able to produce. Sure, you can go higher with percentage but you'll need to deal with excessive heat in peak times of solar radiation and even perhaps loose some juice due to lower efficiency at higher temps.

I had put your numbers in Victron calc for ONE panel and got SmartSolar MPPT 100/20. If I put in 3 in series calc says 150/60 Tr or 150/60 MC4 BUT

If you expand calculations tab you can find out that suggested MPPT max out at 57,8A max temp of the unit so that means in great sunshine in the middle of the day, wind is blowing underneath of the panels to cool them down and batteries are hungry you can expect 19,3A of charging current from ONE panel where there should be some reserve not to running hot. Your panel can produce max 270/12 which is 22,5A when batteries are empty and 270/14,4 which is 18,7A when batteries are full. That means that in ideal world you'd be running your MPPT at 100% which also means max heat. Of course, you'll very rarely see the max capacity of your panels and probably things will work just fine but I'm one of those guys who always look both ends of the stick... 

Now that I know your numbers your MPPT (150/45) is good for 2 panels (you will be running at max 88%) but that is it. For third one you'll need another MPPT which is fine because they work very well together when connected to network through BT.

[vasko]

Feb 24, 2022 21:41:24 GMT fakinx said:

Hello andreshs1,

I'm one of those guys who like hands-on on things regarding my boat so I just installed 700Ah of lithium on board of our SO 42DS.
Big job, not with install itself but with all designing and testing things in my shack for more than a year before transferring everything on board in one single leap. There is still a lot to do with arranging the cabling in order but it's functional and safe for time being.
My 2 cents on your setup are:

1.
MPPT for 540W (2x270W) of solar should really be 100/50 and not 150/45. You will probably (check your specs!) not exceed 100 V with 2 panels in parallel (which is alot!) so you should be fine. More important thing is you leave some spare capacity of MPPT charging not to overheat MPPT nor you want to lose any juice available. In summer there is very good yield from solar because panels are not overheating if they are mounted somewhere free of any obstacles on the bottom side. True, you'll rarely see max capacity charging but it could happen. It's better to operate on 80% instead of 90 or more (of MPPTs) Those gizmos get hot as hell... All said I'm inclined for separate MPPTs for any panel because of redundancy, safety and more flexibility with installation and so on. High voltage (more than 50V) can be dangerous in shaky environment of a boat.

2.
I would think about dedicated LiFePO4 charger and perhaps charge LA starter battery when solar available for charging Li battery and dedicated separate Orion charger and wouldn't really bother with dedicated shore charger for that. If you set up things with at least 2 MPPTs you can design 3-way switch for one of MPPTs to trickle charge starter battery when boat not in use. 

3.
Be aware that even if you kill all loads on board, gizmos for managing your new battery will suck somewhere between 300 and 600 mA so you should tackle that with some remote management or at least monitoring or you could return to the boat one day and your batteries could be dead.  I suggest at least 2 layers of battery management and you should really think about what-if scenarios.

4.
Using just an inverter is fine and you can implement 2-bus configuration but if you choose inverter/charger you need to design more advanced system especially regarding control of inverter/charger. I chose CAN enabled BMS for that.

Well, those were 4 cents but who is counting...   
All the best!

Sounds good

Personally I would not buy anything Victron anymore after recent experience with their support and warranty - for that price tag it shiukd be way wa better - in practice worse then aliexoress

[mehmet]

Feb 24, 2022 14:59:13 GMT andreshs1 said:

Hi all

After 3 years, my AC Delco 440ah house bank has died.....so we have opted for upgrading to LiFePO4

Setup:
- Tao BMS
- 8 x 3.22V CATL 310ah Grade A cells
- Victron SmartSolar 150/45 MPPT
- Victron Orion Smart TR 12/12 30A non-isolated
- Mastervolt 12/1500 (to be upgraded to 3Kw)
- Fischerpanda 6kw
- 540Watts of solar (and planning to upgrade to 1kw ~ by end of the year)

I am pending the cells in order to do the installation (in theory next week)

I would be interested to hear from those with 'first hand' experience.

What works what doesn't
what should I consider

one of the many benefits that we are expecting to get out is to remove the gas cooker/oven, and replace by Induction and electric oven

Below my design:


Regards
Andres 


<button disabled="" class="c-attachment-insert--linked o-btn--sm">Attachment Deleted</button>

I have VICTRON LiFePO4 system since a year and am totally satisfied with it. Victron installer connected the anchor winch to the batteries dierect, that means they have bypassed the voltage control systems. You shoul concider this aswell. 
As an induction stove and electric oven my suggestion os GN Espace from UK. I am very happy with it.

[andreshs1]

Feb 26, 2022 8:52:06 GMT fakinx said:

Oh, I see.
You are calculating current at nominal voltage of your panel which is 29,81V. That's PV side, the input one. For MPPT it's the first number the second is on charge side of your battery which is 12V if I'm correct. Currents are different at the same power at different voltage. Power = Voltage * Amperage
That means you need to choose MPPT which has sufficient voltage capability on the input side and current capability on the charge side. On the charge side all Victron MPPTs auto sense voltage (12/24V) so no problem with that.

On the input side you are looking for numbers that can handle your voltage of all panels connected to one unit added together and on the other side about 70-85% of max current PV panels are able to produce. Sure, you can go higher with percentage but you'll need to deal with excessive heat in peak times of solar radiation and even perhaps loose some juice due to lower efficiency at higher temps.

I had put your numbers in Victron calc for ONE panel and got SmartSolar MPPT 100/20. If I put in 3 in series calc says 150/60 Tr or 150/60 MC4 BUT

If you expand calculations tab you can find out that suggested MPPT max out at 57,8A max temp of the unit so that means in great sunshine in the middle of the day, wind is blowing underneath of the panels to cool them down and batteries are hungry you can expect 19,3A of charging current from ONE panel where there should be some reserve not to running hot. Your panel can produce max 270/12 which is 22,5A when batteries are empty and 270/14,4 which is 18,7A when batteries are full. That means that in ideal world you'd be running your MPPT at 100% which also means max heat. Of course, you'll very rarely see the max capacity of your panels and probably things will work just fine but I'm one of those guys who always look both ends of the stick... 

Now that I know your numbers your MPPT (150/45) is good for 2 panels (you will be running at max 88%) but that is it. For third one you'll need another MPPT which is fine because they work very well together when connected to network through BT.

uhmmm, very good point
I guess that I will need to run more tests once I receive/install the new cells to better understand the performance of my panels. as for now, my tests have been using only the lead acid batteries

Cheers

[andreshs1]

Feb 27, 2022 14:49:05 GMT andreshs1 said:

Feb 26, 2022 8:52:06 GMT fakinx said:

Oh, I see.
You are calculating current at nominal voltage of your panel which is 29,81V. That's PV side, the input one. For MPPT it's the first number the second is on charge side of your battery which is 12V if I'm correct. Currents are different at the same power at different voltage. Power = Voltage * Amperage
That means you need to choose MPPT which has sufficient voltage capability on the input side and current capability on the charge side. On the charge side all Victron MPPTs auto sense voltage (12/24V) so no problem with that.

On the input side you are looking for numbers that can handle your voltage of all panels connected to one unit added together and on the other side about 70-85% of max current PV panels are able to produce. Sure, you can go higher with percentage but you'll need to deal with excessive heat in peak times of solar radiation and even perhaps loose some juice due to lower efficiency at higher temps.

I had put your numbers in Victron calc for ONE panel and got SmartSolar MPPT 100/20. If I put in 3 in series calc says 150/60 Tr or 150/60 MC4 BUT

If you expand calculations tab you can find out that suggested MPPT max out at 57,8A max temp of the unit so that means in great sunshine in the middle of the day, wind is blowing underneath of the panels to cool them down and batteries are hungry you can expect 19,3A of charging current from ONE panel where there should be some reserve not to running hot. Your panel can produce max 270/12 which is 22,5A when batteries are empty and 270/14,4 which is 18,7A when batteries are full. That means that in ideal world you'd be running your MPPT at 100% which also means max heat. Of course, you'll very rarely see the max capacity of your panels and probably things will work just fine but I'm one of those guys who always look both ends of the stick... 

Now that I know your numbers your MPPT (150/45) is good for 2 panels (you will be running at max 88%) but that is it. For third one you'll need another MPPT which is fine because they work very well together when connected to network through BT.

uhmmm, very good point
I guess that I will need to run more tests once I receive/install the new cells to better understand the performance of my panels. as for now, my tests have been using only the lead acid batteries

Cheers

Thanks, I will consider connecting the windlass directly, I have to be honest...I am not sure whether this is connected to the house or the starter battery...

The GN Espace have very cool staff, thanks!
I will have to check that out with the admiral, and see what she says

Cheers

[slo0079]

Feb 26, 2022 21:39:31 GMT vasko said:

Feb 24, 2022 21:41:24 GMT fakinx said:

Personally I would not buy anything Victron anymore after recent experience with their support and warranty - for that price tag it shiukd be way wa better - in practice worse then aliexoress

What are reasonable alternatives?

[fakinx]

Victron is the leading player in the field. And it comes with a hefty pricetag…
Alternatives are surely around but I’m not aware of any other system that capable.