 |
|
02-06-2023, 16:07
|
#1
|
Registered User
Join Date: Dec 2012
Location: Chicago, IL
Boat: 40', Farr 38
Posts: 128
|
AGM>LiFePo4 Upgrade design help
I think I'm real close to a proper safe layout design for my upgrade from (2) AGM banks to (1) starter AGM bank & (1) house bank of LiFePo4. I have completely isolated the Start & House banks to make this possible. The Balmar 618 External Alternator regulator as well as the Phoenix Smart charger will all be set for LiFePo4 battery profiles. The DC>DC will be set for AGM.
Any constructive advice, suggestions, or thoughts are most welcome!
|
|
|
02-06-2023, 17:53
|
#2
|
Registered User
Join Date: Apr 2005
Location: Seabroook Texas or Southern Maine
Boat: Pearson 323, Tayana V42CC
Posts: 1,479
|
Re: AGM>LiFePo4 Upgrade design help
Temp sensor for alternator. Personally would not bother with the negative cable switch You do you though. If you are going to switch the alternator B+ wire use a field disconnect switch and also consider installing a device like an alternator protector device (or not) but definitely a field disconnect style battery switch.
Alternator sense wire and fuse to house battery? An Echo charger to supply the start battery may or not be a better idea? No idea what the costs are these days or a V-reg relay maybe. Or the DC-Dc you have.
I'd consider a emergency bypass to use the house bank for the starter. Or just get a jump box which would be easier.
|
|
|
02-06-2023, 18:09
|
#3
|
Registered User
Join Date: Jun 2009
Location: Annapolis, MD
Boat: Sabre 34-1 (sold) and Saga 43
Posts: 1,801
|
Re: AGM>LiFePo4 Upgrade design help
I like it. A lot.
One question. My problem with drop-ins is the limited BMS. On my boat, I can theoretically draw 300A on my inverter (although rarely does it even see 200A). My windlass can draw 100A, not including inrush. I have a power winch which probably(?) draws upward of 100A, not counting inrush (and I'll often run it while my wife is running the windlass, and do this while the engine isn't running, so it's 100% battery draw). And assorted other loads (5A each for fridge, freezer, chartplotter, and wash down pump).
My point is that suspect I can easily see 200A peak loads on my AGM battery. I would find a drop in LFP with a 200A hard coded BMS limit to be a gut wrenching limitation.
|
|
|
02-06-2023, 18:18
|
#4
|
Registered User
Join Date: Jun 2009
Location: Annapolis, MD
Boat: Sabre 34-1 (sold) and Saga 43
Posts: 1,801
|
Re: AGM>LiFePo4 Upgrade design help
Quote:
Originally Posted by Scubaseas
Temp sensor for alternator.
If you are going to switch the alternator B+ wire use a field disconnect switch and also consider installing a device like an alternator protector device (or not) but definitely a field disconnect style battery switch.
|
Lots of good thoughts. But comments on these two:
I fully assumed your Balmar reg is temp sensed. If not, that is critical!
The B+ switching is another serious limitation of drop-ins. Most have no external advanced notification of a disconnect event, so you can't do B+ alternator shut down. However, with a well set up external regulator, the alternator should already be at idle before "something" causes a disconnect event -- so damage "should" be avoided.
|
|
|
02-06-2023, 18:33
|
#5
|
Registered User
Join Date: Apr 2005
Location: Seabroook Texas or Southern Maine
Boat: Pearson 323, Tayana V42CC
Posts: 1,479
|
Re: AGM>LiFePo4 Upgrade design help
Quote:
Originally Posted by sailingharry
Lots of good thoughts. But comments on these two:
I fully assumed your Balmar reg is temp sensed. If not, that is critical!
The B+ switching is another serious limitation of drop-ins. Most have no external advanced notification of a disconnect event, so you can't do B+ alternator shut down. However, with a well set up external regulator, the alternator should already be at idle before "something" causes a disconnect event -- so damage "should" be avoided.
|
Not really related to just drop ins. If you have power on the field wire and loose the B+ or B- connection you can either pop the exciter diodes or spike the voltage. To like 18 or more volts. Which can fry anything that's running on 12V. You can no longer disconnect the plus battery terminal with the car running and not risk frying some really expensive stuff. Boats are pretty much the same. It's one reason why I personally would loose the negative cable switch.
|
|
|
02-06-2023, 18:35
|
#6
|
Registered User
Join Date: Jul 2020
Location: Virginia, USA
Boat: Tayana 37
Posts: 960
|
Re: AGM>LiFePo4 Upgrade design help
Mostly good.
If you use a single fuse 400A then each battery (and the cable connecting it) should be capable of 400A. BMS should not be relied upon as a safety measure. Mosfets can fail closed. I would prefer a seperate fuse for each battery and then wired to a primary bus bar.
The other fuses (like 40A for charger) should be as close to the primary busbar as possible (ideally right on the primary busbar).
Also I would move the battery isolation switches as close to the battery as possible.
Two completely optional thing to consider. The first would be to add an alternator protector (Balmar among others makes them) in case the house bank disconnects without warning. With drop in replacement batteries you have no automated method of handling BMS disconnect. The other optional thing would be an A B (but not combine) switch with alternator disconnect to allow you to emergency start with house bank but if you only have two batts each capable of 100A max it is unlikely they can start the battery but worth considering if you ever upgrade in the future. Depending on your engine you likely need 500 CCA. The other alternative is ignore that complexity and just get a portable battery booster to provide backup starting capability.
|
|
|
02-06-2023, 18:49
|
#7
|
Registered User
Join Date: Apr 2005
Location: Seabroook Texas or Southern Maine
Boat: Pearson 323, Tayana V42CC
Posts: 1,479
|
Re: AGM>LiFePo4 Upgrade design help
"If you use a single fuse 400A then each battery (and the cable connecting it) should be capable of 400A. BMS should not be relied upon as a safety measure. Mosfets can fail closed. I would prefer a seperate switch for each battery and then wired to a primary bus bar.
The other fuses (like 40A for charger) should be as close to the primary busbar as possible (ideally right on the primary busbar).
Also I would move the battery isolation switches as close to the battery as possible."
Good points. I have 150A MRBFs on each of my 4 batteries. Then a T class fuse of 200A for the house main bus to LiFePo4 house bank.
|
|
|
02-06-2023, 19:59
|
#8
|
Registered User
Join Date: Dec 2012
Location: Chicago, IL
Boat: 40', Farr 38
Posts: 128
|
Re: AGM>LiFePo4 Upgrade design help
Thank you. Yep, I've added for the Balmar regulator, Alternator Protection Module, Alt Temp & Battery temp sensors. Just in case the battery BMS system fails.
Quote:
Originally Posted by Scubaseas
Temp sensor for alternator. Personally would not bother with the negative cable switch You do you though. If you are going to switch the alternator B+ wire use a field disconnect switch and also consider installing a device like an alternator protector device (or not) but definitely a field disconnect style battery switch.
Alternator sense wire and fuse to house battery? An Echo charger to supply the start battery may or not be a better idea? No idea what the costs are these days or a V-reg relay maybe. Or the DC-Dc you have.
I'd consider a emergency bypass to use the house bank for the starter. Or just get a jump box which would be easier.
|
|
|
|
02-06-2023, 20:05
|
#9
|
Registered User
Join Date: Dec 2012
Location: Chicago, IL
Boat: 40', Farr 38
Posts: 128
|
Re: AGM>LiFePo4 Upgrade design help
That's a Great point. I think I will return that item!
Thank you
Quote:
Originally Posted by Scubaseas
Not really related to just drop ins. If you have power on the field wire and loose the B+ or B- connection you can either pop the exciter diodes or spike the voltage. To like 18 or more volts. Which can fry anything that's running on 12V. You can no longer disconnect the plus battery terminal with the car running and not risk frying some really expensive stuff. Boats are pretty much the same. It's one reason why I personally would loose the negative cable switch.
|
|
|
|
02-06-2023, 20:30
|
#10
|
Registered User
Join Date: Dec 2012
Location: Chicago, IL
Boat: 40', Farr 38
Posts: 128
|
Re: AGM>LiFePo4 Upgrade design help
Thank you. All these High amp switches, batteries, shunts, fuses, etc., are within 1 meter, quite close. Interestingly, I had this drawn w/separate fuses to each 200A battery and another source suggested the single fuse. The 2/0 cable to the single fuse would each be identical in length and no more than .5 meter. As I've separated the starter draw from the house bank downsizing the protective fuses to a more reasonable 200A makes sense. I'll see if I can do a buss bar, it's a very tight install.
Quote:
Originally Posted by Statistical
Mostly good.
If you use a single fuse 400A then each battery (and the cable connecting it) should be capable of 400A. BMS should not be relied upon as a safety measure. Mosfets can fail closed. I would prefer a seperate fuse for each battery and then wired to a primary bus bar.
The other fuses (like 40A for charger) should be as close to the primary busbar as possible (ideally right on the primary busbar).
Also I would move the battery isolation switches as close to the battery as possible.
Two completely optional thing to consider. The first would be to add an alternator protector (Balmar among others makes them) in case the house bank disconnects without warning. With drop in replacement batteries you have no automated method of handling BMS disconnect. The other optional thing would be an A B (but not combine) switch with alternator disconnect to allow you to emergency start with house bank but if you only have two batts each capable of 100A max it is unlikely they can start the battery but worth considering if you ever upgrade in the future. Depending on your engine you likely need 500 CCA. The other alternative is ignore that complexity and just get a portable battery booster to provide backup starting capability.
|
|
|
|
02-06-2023, 21:37
|
#11
|
Registered User
Join Date: Dec 2012
Location: Chicago, IL
Boat: 40', Farr 38
Posts: 128
|
Re: AGM>LiFePo4 Upgrade design help
Given the great comments, I've made a few changes. Specifically changing the fuse arrangement for the batteries and the elimination of the negative switch.
Other thoughts?
|
|
|
03-06-2023, 06:39
|
#12
|
Registered User
Join Date: Jun 2009
Location: Annapolis, MD
Boat: Sabre 34-1 (sold) and Saga 43
Posts: 1,801
|
Re: AGM>LiFePo4 Upgrade design help
Quote:
Originally Posted by Capt Ben
Given the great comments, I've made a few changes. Specifically changing the fuse arrangement for the batteries and the elimination of the negative switch.
Other thoughts?
|
That surface mount MRBF fuse block looks like a nifty piece of gear. But I'm not entirely sure how much it gains you.
* In the it unlikely event that you have some sort of a short in the 18-in cable between the battery and the fuse block, it provides no protection at all. But this is an extremely unlikely scenario.
* In the more likely scenario of an overload condition downstream of the block, the first and only fuse to blow will be the 300A fuse.
* In the unique failure mode where you have a load in excess of 200A, and less than 300A, and one battery disconnected for some reason, a battery fuse would blow. Using 2/0 cable between the battery and the fuse would eliminate this concern.
* As drawn, none of these fuses would theoretically ever blow because your BMS would trip first. An argument could be made that the fuses should be low enough to protect the BMS.
I would personally look at the system as drawn as adding complexity, cost, and space with very little return on that investment. I would size the battery cables to individually support the 300a fuse (2/0), and only have the one fuse.
EDIT: I just realized there is one failure mode that might benefit from this system. Some sort of internal short in one battery (that somehow doesn't cause that one battery to detonate!) would cause the second battery to unload through the failed battery. This would be very high current with no protection. The battery fuses would protect this failure mode. If it is even a legitimate failure mode.
|
|
|
03-06-2023, 07:25
|
#13
|
Registered User
Join Date: Apr 2005
Location: Seabroook Texas or Southern Maine
Boat: Pearson 323, Tayana V42CC
Posts: 1,479
|
Re: AGM>LiFePo4 Upgrade design help
Also ABYC 11.10.1.1.1 Overcurrent Protection Device Location - Ungrounded conductors shall be provided with overcurrent protection within a distance of seven inches (178 mm) of the point at which the conductor is connected to the source of power measured along the conductor (see FIGURE 14). ...There are of course exceptions to this.
T class fuses are FAST. MRBFs not so much but I had them and installed them on my system incase there was some kind of short while working in the area. I'd rather replace a MRBF than a BMS board. If he can get two T class fuses close to the battery (one for each within 7" ideally) sure the MRBFs are redundant and not needed.
MRBFs are quick and easy to install and have good covers which solves another ABYC requirement. Yes, technically you do not need them at all with the T fuse as drawn but then you are not individually trying to protect a what $900 or more battery?
|
|
|
03-06-2023, 07:44
|
#14
|
Registered User
Join Date: Jul 2020
Location: Virginia, USA
Boat: Tayana 37
Posts: 960
|
Re: AGM>LiFePo4 Upgrade design help
Quote:
Originally Posted by Capt Ben
Thank you. All these High amp switches, batteries, shunts, fuses, etc., are within 1 meter, quite close. Interestingly, I had this drawn w/separate fuses to each 200A battery and another source suggested the single fuse. The 2/0 cable to the single fuse would each be identical in length and no more than .5 meter. As I've separated the starter draw from the house bank downsizing the protective fuses to a more reasonable 200A makes sense. I'll see if I can do a buss bar, it's a very tight install.
|
By close I mean on your diagram. Right now your battery isolation switches don't isolate the batteries making them kinda useless. The switch should be positioned so that when it is off there is no connectivity between the battery and the rest of the system. Alternatively you can get rid of them completely. Having a switch which cuts off the battery from some but not all of the system doesn't really serve any purpose and doesn't meet any requirement or guideline.
As for the fuses yeah each battery should having its own fuse is ideal. LFP is serious power and with low resistance amperages in fault conditions can be significant. Multiple batteries in parallel increase that amperage. You can start to run into AIC limits.
In your updated diagram not sure if you need the second fuse after battery fuses. Can you explain your reasoning? Also if the BMS is limiting current to 100A why 200A battery fuses and wiring and then 300A combined fuse?
If the BMS is limited to 100A per battery then 100A fuse on each battery and you are done. Ensure each battery wire can handle 100A and once you parallel them ensure the wiring can handle 200A.
I see you switched to MRBF fuses from Class T. Be sure to lookup or compute short circuit current for the batteries and make sure MBRF has sufficient AIC.
|
|
|
03-06-2023, 08:06
|
#15
|
Registered User
Join Date: Apr 2005
Location: Seabroook Texas or Southern Maine
Boat: Pearson 323, Tayana V42CC
Posts: 1,479
|
Re: AGM>LiFePo4 Upgrade design help
Have a look at Waytekwire.com , search MRBF fuse holders. Both Easton and LittleFuse make MRBF holders and covers that are substantially cheaper than BlueSeas. BlueSeas does make great stuff however. The covers on the LittleFuse MRBF holders are very well made and work a lot better than the Blue Seas ones IMHO.
I apologize for inserting the 150A MRBF on the batteries into the conversation. My LFPs cut out at 200A but I have no load that would be any where near that so decided the 150A which matches my cable size at lengths involved for my house bank. I have a separate AGM batteries for windlass (in the bow area) and engine starter/genset starter. I also run a separate dedicated LFP for my autopilot and instruments with select switch to change power source for the AP to house or it's own battery.
|
|
|
 |
|
Thread Tools |
Search this Thread |
|
|
Display Modes |
Rate This Thread |
Linear Mode
|
|
Posting Rules
|
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts
HTML code is Off
|
|
|
|
Advertise Here
Recent Discussions |
|
|
|
|
|
|
|
|
|
|
|
|
Vendor Spotlight |
|
|
|
|
|