LiFePo4 update - emergency engine start question

Minerva

Well-Known Member
Joined
16 Oct 2019
Messages
2,004
Visit site
I’m planning on a lithium batter upgrade this winter. Already got it clear in my hear regarding alternator + shore power charger -> engine start battery. Solar to -> lifepo4 housebank. Both batteries with separate isolator switch. Between the engine start battery and the lifepo4 house bank a DcDc charger activated from a switch.

Question I’m mulling over, is say my engine start battery goes flat. How best do I rig an “in emergency only” engine start feed from the lifepo4 domestic bank?

Thanks.
 
installations i have seen allow you to charge the starting battery from the house bank, even if the house bank is lithium and the starting battery is, say, AGM.

so, you charge the starting battery from the house bank and then use the starting battery to start the engine.
 
I have a fused length of wire with a switch that I can fit on my starter motor from my battery. Nice to have to be able to bypass key etc. You can them use any battery to start the engine. Would that be ok with what you have in mind?
 
There are a few variations of isolator switches, VSRs with over-ride function, 1B2 switch, jump leads if you like.

The thing to be aware of is that 'modest' sized Lifepo4 batteries are not rated to start a diesel engine.
So you need to understand your BMS.
Will it shut down if you try starting the engine off it?


I think this can be a big deal, because if you've got a solar panel or anything topping up your engine start battery, you don't get much warning of the battery dying. It's on life support then it flat lines.
Mate of mine found that this year. One day the engine started with no hint of trouble, a week later the battery is deader than a dead thing having a rest.

It's not an issue if you are talking about monsta LIfepo4 systems of >500Ah.
I think it is an issue for smaller boats with 200Ah or less of house bank.

Work arounds might be some other flavour of Lithium such as a 'jump pack' or a spare start battery, or just buy a new start battery well before the old one dies and look after it.

Or you could get a starting handle?
There are legends of people starting engines by wrapping the mainsheet around the flywheel and crash gybing or zummat, but personally, I'm in favour of having a second battery which is actually intended to start an engine.

Rationally, a start battery is not big money, just buy a new one every three years ?
I'm not comfortable with no back-up, but that's maybe all about misdeeds of my, and my mates', younger days.
 
I'm not comfortable with no back-up, but that's maybe all about misdeeds of my, and my mates', younger days.
+1 I think we must have shared a misspent youth :)

What is the current draw on the starter? IIRC, it's likely to be rated at less than 1000w, so will draw less than 80 amps. If you have four 12v batteries in parallel, that's <20a each, which shouldn't be a stretch for them. However, I'm no expert, so someone who knows what they're talking about will be along in a minute to explain why I'm wrong.
 
+1 I think we must have shared a misspent youth :)

What is the current draw on the starter? IIRC, it's likely to be rated at less than 1000w, so will draw less than 80 amps. If you have four 12v batteries in parallel, that's <20a each, which shouldn't be a stretch for them. However, I'm no expert, so someone who knows what they're talking about will be along in a minute to explain why I'm wrong.
The inrush current on a starter motor will be a lot more than 80A.
It will depend on the battery, the wiring and how you measure it.

Starter motors are kind of optimised to be used with a typical lead-acid battery.
It's often going to be the V/I characteristic of the battery, 'internal resistance' which determines the size and shape of the current spike.
Other battery chemistries have different characteristics.

Basically, the motor initially looks pretty much like a short circuit, if the BMS works fast enough to see this, it will cut the battery off.

As you say, share it around several batteries or any bank of hundreds of Ah, much less likely to be a problem.
I think it's something one should look into before buying a couple of drop-in LifePo's for a smaller boat.
There's a lot Lifepo 'leisure battery replacements' which appear not to support engine starting.
 
It's often going to be the V/I characteristic of the battery, 'internal resistance' which determines the size and shape of the current spike.
Are "parallel" batteries literally that, or separately connected to a control unit to perform a moderation process between them? If literally, the internal resistance of each will govern how the total current is split between them, which may not be equitable...
 
if the BMS works fast enough to see this, it will cut the battery off.
This is looking at it backwards. They almost always are fast enough to "see" it, but in almost every case have a built in delay on purpose to handle momentary spikes. A BMS is a software component that makes a logical decision to do things with hardware, it's nothing like a fuse.

For what it's worth, my Fogstar Drift battery starts my MD2040 quite happily with the standard emergency parallel switch setup found on modern yachts.
 
I’m planning on a lithium batter upgrade this winter. Already got it clear in my hear regarding alternator + shore power charger -> engine start battery. Solar to -> lifepo4 housebank. Both batteries with separate isolator switch. Between the engine start battery and the lifepo4 house bank a DcDc charger activated from a switch.

Question I’m mulling over, is say my engine start battery goes flat. How best do I rig an “in emergency only” engine start feed from the lifepo4 domestic bank?

Thanks.
The issue might be the BMS. They will isolate the battery in a fraction of a second if they think the load is over the threshold limit. Are you planning to buy a battery or build one?
If you build one, it would be very easy to take a parallel connection from the battery terminals to bypass the BMS for emergency engine starting. What size lithium battery are you considering?
 
They will isolate the battery in a fraction of a second
Most are a lot longer than you think. The BMS protects the battery, but it’s not a breaker and not designed to be fast acting. Many will happily supply full amps for half a second as no damage would result from that. Most damage is heat related in this instance, to the battery at least. It may destroy everything around it in that time, but that’s what the big fuse is for 😂
 
Most are a lot longer than you think. The BMS protects the battery, but it’s not a breaker and not designed to be fast acting. Many will happily supply full amps for half a second as no damage would result from that. Most damage is heat related in this instance, to the battery at least. It may destroy everything around it in that time, but that’s what the big fuse is for 😂
My JK BMS logs a fault every time I turn on the inverter. It tells me it isolated the bank. I don't see the shut down as my other, far larger battery doesn't shut down. I think it will depend very much on the size of battery and the bms. May work for some but not others
 
This is looking at it backwards. They almost always are fast enough to "see" it, but in almost every case have a built in delay on purpose to handle momentary spikes. A BMS is a software component that makes a logical decision to do things with hardware, it's nothing like a fuse.

For what it's worth, my Fogstar Drift battery starts my MD2040 quite happily with the standard emergency parallel switch setup found on modern yachts.
Thanks Lustyd. Your emergency parallel switch set up; is that just in effect a cable from +domestic bank to + engine start bank with a switch in the middle?
 
Thanks all for your input. I’ve 3x100ah lifepo4 batteries to install with a max continuous discharge of 100a. I’m pretty confident that the bms will handle starting the engine in extremis.
 
Thanks Lustyd. Your emergency parallel switch set up; is that just in effect a cable from +domestic bank to + engine start bank with a switch in the middle?
It's the standard (modern) setup. House bank has a main switch between it and house loads on the positive cable. Engine bank has a switch between it and the engine on the positive cable. The load side of each switch is also connected to a third switch (and originally a VSR in many instances) which connects the load side of the switches together. This allows you to use either battery or both to power either engine, house or both.
With LiFePo4, remove the VSR from the system if you have one as it will connect the lithium directly when the engine is on, which you don't want.


EDIT - sorry just checked and it seems the switch is usually battery side of the switches, not load side. This according to the BEP wiring diagram
 
It's the standard (modern) setup. House bank has a main switch between it and house loads on the positive cable. Engine bank has a switch between it and the engine on the positive cable. The load side of each switch is also connected to a third switch (and originally a VSR in many instances) which connects the load side of the switches together. This allows you to use either battery or both to power either engine, house or both.
With LiFePo4, remove the VSR from the system if you have one as it will connect the lithium directly when the engine is on, which you don't want.
Perfect. Thanks for clarifying.
 
Perfect. Thanks for clarifying.
I just found a blue sea systems diagram showing the switch on the load side so it seems there is no standard since BEP do it the other way. I can see advantages to both based on various scenarios.
 
I bought a 120amp cheap eBay LifePo4 battery to use with my electric outboard. Plan was to charge it via inverter and its own 240 volt charger when on shore power or when the engine was running. I never got as far as that. My old, occasional use Toyota Amazon got a flat battery and I stupidly in ignorance decided to use it to start the car. I fried the BMS and that was the end of that. I am not sure if it was the starter current draw or the alternator charge or both that killed it. The Toyota being a 4.7 litre V8 probably did not help
 
Interesting video from Rod at marine how to. He measures inrush current at around 500A. He also states that this can cause some Mosfet BMSs issues in the medium term. Newer BMSs are often better as are larger BMSs (amp rating).


Tldr; ensure “drop ins” are engine start rated or check the BMS spec. More BMS / larger mosfets the better (usually).
 
I bought a 120amp cheap eBay LifePo4 battery to use with my electric outboard. Plan was to charge it via inverter and its own 240 volt charger when on shore power or when the engine was running. I never got as far as that. My old, occasional use Toyota Amazon got a flat battery and I stupidly in ignorance decided to use it to start the car. I fried the BMS and that was the end of that. I am not sure if it was the starter current draw or the alternator charge or both that killed it. The Toyota being a 4.7 litre V8 probably did not help
Have you considered opening the pack and replacing the BMS?
 
Interesting video from Rod at marine how to. He measures inrush current at around 500A. He also states that this can cause some Mosfet BMSs issues in the medium term. Newer BMSs are often better as are larger BMSs (amp rating).


Tldr; ensure “drop ins” are engine start rated or check the BMS spec. More BMS / larger mosfets the better (usually).
Very interesting. I measured the inrush current on my 1kW 1GM10 starter and got 80A. This was with a cheapish clamp meter and using an old 70Ah lead acid battery. I will have to try measuring again using my Fluke meter as I now have a shunt and a newish 110Ah lead acid battery.
I would assume that the internal resistance of the battery along with any cable resistance has a significant effect on peak current. I notice that the video used very short thick cables.
My 1GM started very easily with an old 24Ah gel battery and 3m long 10mm2 cables when I had taken the main batteries home for charging. Volt drop for this is only 1V from the cables and probably the same again for the battery internal resistance but a 500A peak would be severely limited by this arrangement with a 10V drop.
 
Top