Starting a big diesel with lifepo4

[Phil_Fr17]

Quality LFPs or LTOs are not cheap batteries, otherwise you'd know.
In that case, and given the first post, I'll stick with lead batteries, which will be the cheapest since you're talking about "budget EVE 280AH cells".
In my case, I don't care about the SOC of my 12-volt Winston batteries, it's always 100% because in a car, recharging after starting is very quick. They seem to be doing just fine after more than 7 years.
But they're not cheap and there's yttrium in them, which no-one has tried to copy or test, but that's what should make the difference in terms of durability with a 100% SOC.

[CaptainRivet]

Originally Posted by Phil_Fr17

Quality LFPs or LTOs are not cheap batteries, otherwise you'd know.
In that case, and given the first post, I'll stick with lead batteries, which will be the cheapest since you're talking about "budget EVE 280AH cells".
In my case, I don't care about the SOC of my 12-volt Winston batteries, it's always 100% because in a car, recharging after starting is very quick. They seem to be doing just fine after more than 7 years.
But they're not cheap and there's yttrium in them, which no-one has tried to copy or test, but that's what should make the difference in terms of durability with a 100% SOC.

6xyinglong LTO cells 40AH b-grade ypu can get for 350Euro which have around 35AH and can deliver 350A const and 700A for 10sek. Add a 5 A heltec for 30Euro. They have 30000cycles so >50year lifespan. Means do it right ones and forget about them. A for that 400Euro all in is cheap.

Yes EVE are best AH for the buck and a really good cells IF you operate them at 0.3C and occasionally 0.5C max, perfect storage and housebank cells. The new 304AH with great terminals are much better then 280AH ones. But still budget if you look at the thin alu case wrapped compared to sturdy composite casing of winston or calb.
But the real 4C constant and 8C peak of a the new winston is a different world, thats why they cost 3x of an Eve for the same capacity. Used the older which are 3C/6C 50AH and 90AH cells in cars as only battery with big stereos with no BMS and they eun since 10years without issues....you get what you pay for. Winston are a no brainer for smaller boats where you simple put 4x130AH cells and you can induction cook, start the engine, windlass... all with one battery. Juat strap together put busbars on, connect a cell monitor and a main battery fuse and you are done. There the small euro on top compared to eg EVEs is so worth it.

Different if you have space then 2 or 3p EVE banks can do the same but you have 3xtimes the capacity as Winston for the same price.

[SOLAR SUPPORT]

Originally Posted by SOLAR SUPPORT

The initial inertia of the moving parts of the engine, such as the flywheel, crank, pistons, connecting rods, and additionally the compression pressure that the piston must overcome, forces the starter motor very hard for a few seconds at the start of the first movement. If you cannot provide the high power required to start the starter motor in the first few seconds, the starter motor can not start.

If the required current can be measured during the first start, the CCA value of the battery required for this job becomes clear. These small-capacity Lifepo4 batteries, which are used as starter batteries, must have circuit elements that can accumulate energy and discharge instantly. A power-increasing circuit that allows the 50 Ah Lifepo4 battery to produce 1450 A, albeit for a short time.

LA crank batteries, whose capacity decreases over the years and during cold winter months, may sometimes need support. To avoid troublesome moments at sea with an insufficient LA crank battery, I considered a switched, temporary parallel connection setup.

In order to provide boost energy to the 70Ah LA crank battery that runs the VP D1-30 engine, I wanted to connect my 2P4S, 400Ah Lifepo4 service batteries in parallel to support the LA battery. Before adding a battery switch and wiring for this, I measured the maximum current drawn from the LA crank battery during a cold start of the engine. Measured max. The 306A current dropped to 200A, 40A after about a second. Total cranking time is 5-6 seconds.

Lifepo4 400Ah service battery bms has the capacity to safely deliver a total of 2x150A continuously and 600A for 5 seconds. 2x125A battery mega fuses have the capacity to easily pass high currents for short periods of time. In my first test using the temporary parallel connection switch of the service batteries, the maximum currents observed in the measurement were around 180A for the Lifepo4 bank and 126A for the LA crank battery. It works smoothly.

https://youtu.be/XXUlP5-hBB0?si=iiWH5JN4KIhYPFIv

[CaptainRivet]

Originally Posted by SOLAR SUPPORT

LA crank batteries, whose capacity decreases over the years and during cold winter months, may sometimes need support. To avoid troublesome moments at sea with an insufficient LA crank battery, I considered a switched, temporary parallel connection setup.

In order to provide boost energy to the 70Ah LA crank battery that runs the VP D1-30 engine, I wanted to connect my 2P4S, 400Ah Lifepo4 service batteries in parallel to support the LA battery. Before adding a battery switch and wiring for this, I measured the maximum current drawn from the LA crank battery during a cold start of the engine. Measured max. The 306A current dropped to 200A, 40A after about a second. Total cranking time is 5-6 seconds.

Lifepo4 400Ah service battery bms has the capacity to safely deliver a total of 2x150A continuously and 600A for 5 seconds. 2x125A battery mega fuses have the capacity to easily pass high currents for short periods of time. In my first test using the temporary parallel connection switch of the service batteries, the maximum currents observed in the measurement were around 180A for the Lifepo4 bank and 126A for the LA crank battery. It works smoothly.

https://youtu.be/XXUlP5-hBB0?si=iiWH5JN4KIhYPFIv

To measure the surge and spike you need a oscilloscope, forget what you measured with a current clamp.

The D1-30 as well as D2 have a 2000W starter, so average is 180A but surge is around 450-500A for some ms and then gets to about 300A for another 1/2 sec, thats what your clamp meter measured.

Forget paralleling, just throw the lead out and connect it to your service, BMS and cells can handle that. Put LCV at 3.0V and you have always enough AH left to start that little engine. Wanna do better get another 4 cell and BMS and raise bank to 600AH, so you have 3 batteries and eg can start the engine, for all 3 parallel its piece of cake.

By paralleling you stressing the LFP additionally as current will flow from LFP to lead and starter b because when lead drops in voltage the LFP is supporting and at the same time deliver the juice for the starter.
2nd throw the megafuses out and replace with class T or NH fuses, in a LFP bank short they will be arced=useless.
For LFP main battery fuse you MUST use clas T or NH, nothing else

[CapsizedVeteran]

For what it's worth the BMS configuration on my off brand Chinese lithium batteries doesn't just have one over current setting, it has three different levels. It's a 200a JDB BMS, but it actually allows up to 640a for 640ms and 2000a for 5us. So the initial starter surge likely wouldn't trip anything for the BMS. Although personally I'm more worried about the transient voltage for a huge inductive load like the starter frying the mosfets. I'm considering adding a large capacitor to my system to handle both current surges and transient voltages.

[CaptainRivet]

Originally Posted by CapsizedVeteran

For what it's worth the BMS configuration on my off brand Chinese lithium batteries doesn't just have one over current setting, it has three different levels. It's a 200a JDB BMS, but it actually allows up to 640a for 640ms and 2000a for 5us. So the initial starter surge likely wouldn't trip anything for the BMS. Although personally I'm more worried about the transient voltage for a huge inductive load like the starter frying the mosfets. I'm considering adding a large capacitor to my system to handle both current surges and transient voltages.

Thats quite tight...i would be hesitant on putting 200A through as their mosfets are not matched for sure (not possible for the price they sell them) and definitly not a 2000W starter...that will kill the FETs, if not immediate then quite quick. 2 in parallel yes like solar support. A capacitor won't fix this...
If you want to use it as a hybrid bank i advise to use contactor based BMS or at least for a FET based rating minimum 1.5x better twice the current rating of the starter so 2x150A=300A minimum.
The reason i use ElectrodacusBMS, not current carrying at all, as its planed from start as hybrid bank.

[CapsizedVeteran]

Originally Posted by CaptainRivet

Thats quite tight...i would be hesitant on putting 200A through and definitly not a 2000W starter...that will kill the FETs, if not immediate then quite quick. 2 in parallel yes like solar support. A capacitor won't fix this...
If you want to use it as a hybrid bank i advise to use contactor based BMS or at least for a FET based twice the current rating of the starter.

That's not my settings, that's how they're programmed from the factory. In fact, they won't even trip from 200-640 amps until it has been going on for over 32 seconds. So they must not think these things are as delicate as you do.

I already have two batteries in my house bank for 400 amps of continuous power and my starter motor is on a separate lead acid. My consideration of a capacitor is for extra insurance on top of all that. Thanks for your concern though. I advise you to ask questions before just assuming someone else has done everything wrong and without thought.

[SOLAR SUPPORT]

Originally Posted by CaptainRivet

To measure the surge and spike you need a oscilloscope, forget what you measured with a current clamp.

The D1-30 as well as D2 have a 2000W starter, so average is 180A but surge is around 450-500A for some ms and then gets to about 300A for another 1/2 sec, thats what your clamp meter measured.

Forget paralleling, just throw the lead out and connect it to your service, BMS and cells can handle that. Put LCV at 3.0V and you have always enough AH left to start that little engine. Wanna do better get another 4 cell and BMS and raise bank to 600AH, so you have 3 batteries and eg can start the engine, for all 3 parallel its piece of cake.

By paralleling you stressing the LFP additionally as current will flow from LFP to lead and starter b because when lead drops in voltage the LFP is supporting and at the same time deliver the juice for the starter.
2nd throw the megafuses out and replace with class T or NH fuses, in a LFP bank short they will be arced=useless.
For LFP main battery fuse you MUST use clas T or NH, nothing else

The peak current drawn from the LA crank battery & Lifepo4 combo may reach levels of 500-600A for milliseconds. Considering my total Bms capacities, this is not a problem. Since Lifepo4 and LA batteries shoulder the high starting current together, the load of Lifepo4 battery BMS is reduced.

My Lifepo4 service and LA crank batteries are separated by the battery isolator. I find it unnecessary to relocate the engine battery and connect it to the Lifepo4 bank. In order not to distract from the topic, I will not go into the advantages of isolating Lifepo4 and LA battery groups with battery isolator. I did not have any problems with the mega fuses. I may consider replacing them with T fuses in the future.

[Phil_Fr17]

Originally Posted by CaptainRivet

6xyinglong LTO cells 40AH b-grade ypu can get for 350Euro which have around 35AH and can deliver 350A const and 700A for 10sek. Add a 5 A heltec for 30Euro. They have 30000cycles so >50year lifespan. Means do it right ones and forget about them. A for that 400Euro all in is cheap.

I looked up what this device was and didn't know. It has to be said that I don't do developing any more and wouldn't be comfortable with it. Is there any data or typical set-ups for using it?
Interesting, in any case.
And I completely agree with you about the performance and capabilities of LTOs. As for the price of LTOs, I was stuck on the price of 45 Ah Yinlong cells, which cost over $87 excluding tax and shipping... Hence my slightly biased comparison with lead-acid batteries.
But I also think that lead batteries are no longer really in the running when there are batteries to replace.

[CaptainRivet]

Originally Posted by SOLAR SUPPORT

The peak current drawn from the LA crank battery & Lifepo4 combo may reach levels of 500-600A for milliseconds. Considering my total Bms capacities, this is not a problem. Since Lifepo4 and LA batteries shoulder the high starting current together, the load of Lifepo4 battery BMS is reduced.

My Lifepo4 service and LA crank batteries are separated by the battery isolator. I find it unnecessary to relocate the engine battery and connect it to the Lifepo4 bank. In order not to distract from the topic, I will not go into the advantages of isolating Lifepo4 and LA battery groups with battery isolator. I did not have any problems with the mega fuses. I may consider replacing them with T fuses in the future.

Current is like water, it takes the way of the lowest resistance and thats the lifepo4 so you can be sure that 500-600A amp comes to 95% from the Lifo4 but yes your 2 BMS parallel can handle that. With an oszi you can see that.

You connected both batteries to starter...so cables are already there, just disconnect the lead as its contraproductive, keep it with a 1/2 switch as backup. By the way how did you isolate them from each other???
If that would help, it would be done long time ago by a lot of people but it isn't....we tested that 15 years ago in competition car stereo paralleling lead and LFP (with a lot lab equipement like oszi) the results where as i told you, short high peaks come only from LFP and when big loads and lead sags the LFP need to deliver the load and support the sagging lead means double stressed.

[CaptainRivet]

Originally Posted by CapsizedVeteran

That's not my settings, that's how they're programmed from the factory. In fact, they won't even trip from 200-640 amps until it has been going on for over 32 seconds. So they must not think these things are as delicate as you do.

I already have two batteries in my house bank for 400 amps of continuous power and my starter motor is on a separate lead acid. My consideration of a capacitor is for extra insurance on top of all that. Thanks for your concern though. I advise you to ask questions before just assuming someone else has done everything wrong and without thought.

Can you explain where and how exactly you wanna use this capacitor and how big....

Andi from offgrid garage tested all this BMS, JBD,JK.. and most of them hardly made the rating. To be able to deliver this high loads the mosfets must be matched otherwise and that happens here one mosfet gets higher load then the other and thats the mosfet that dies first...BMS still works but with less mosfets that have even a higher load... and for the price JBD sells its not possible to use matched mosfets as they cost more then the whole sales price of the BMS.
Chinese testing well from > 20years dealing with LFP i would be conservative and assume they can do 2/3 max. 3/4 of there rating if you get know "quality" like JK or JBD.
The MuellerJK BMS 250A, a special build by JK for Mueller energy in Australia and the base is their 300A rated BMS but with matched Mosfets...and mueller rates it 250A after testing it heavily....
Jk and JBD come from the same chinese factory with different branding and wholesaler company is then programming them. Thats why the same bt work on both.

Sure splitting diode for charge lfp and lead in parallel is a must.

[freshalaska]

What’s LTO?

[CapsizedVeteran]

Originally Posted by CaptainRivet

Can you explain where and how exactly you wanna use this capacitor and how big....

Andi from offgrid garage tested all this BMS, JBD,JK.. and most of them hardly made the rating. To be able to deliver this high loads the mosfets must be matched otherwise and that happens here one mosfet gets higher load then the other and thats the mosfet that dies first...BMS still works but with less mosfets that have even a higher load... and for the price JBD sells its not possible to use matched mosfets as they cost more then the whole sales price of the BMS.
Chinese testing well from > 20years dealing with LFP i would be conservative and assume they can do 2/3 max. 3/4 of there rating if you get know "quality" like JK or JBD.
The MuellerJK BMS 250A, a special build by JK for Mueller energy in Australia and the base is their 300A rated BMS but with matched Mosfets...and mueller rates it 250A after testing it heavily....
Jk and JBD come from the same chinese factory with different branding and wholesaler company is then programming them. Thats why the same bt work on both.

Sure splitting diode for charge lfp and lead in parallel is a must.

I'm thinking something along the lines of a 5 farad car audio capacitor hooked up to the main positive and negative bus bars would serve well to support large inrush currents and help clamp down on transient voltage spikes from inductive loads. While my starter is on a separate lead acid my windlass is on the house bank. I'm definitely not sure if it would work as intended but it's cheap enough to try and checking the results with an oscilloscope would be easy enough.

[CaptainRivet]

Originally Posted by CapsizedVeteran

I'm thinking something along the lines of a 5 farad car audio capacitor hooked up to the main positive and negative bus bars would serve well to support large inrush currents and help clamp down on transient voltage spikes from inductive loads. While my starter is on a separate lead acid my windlass is on the house bank. I'm definitely not sure if it would work as intended but it's cheap enough to try and checking the results with an oscilloscope would be easy enough.

I would just use the money and get a high quality DC2DC 12V to 12V converter like the victron orion-TR 30/12/12 (NOT the TR-smart, thats the charger!) Which is around 160Euro but you often get them unused=new for a bargain as people order that and think its the charger which costs double. Paid 80Euro for mine.

The orion goes in between your main busbar and senstive electronics like chartplotter, radar, radio,ap computer. I have connected all life dependent must have electronics to its output ( and plotter+radar i consider to be)
The orion does 3 things:
1) it galvanically isolates the output completely from the input, means no spike,inrushes or alike, total clean power.
You can even add a small LTO or AGM motorbike battery as buffer battery, lifepo4 is not good as that buffer is aways at high SOC 90-100%. Then it acts as top up charger in this case.
2) it stabilizes the fluctuating input voltage 9-18V into a stabilized adjustable output voltage. So you can compensate voltage drop and even raise it higher. I have it at 14.8V, that compensates voltage drop and electronics get min 13.8V and max 14.4V, independent of the input voltage down to 9V. so optimum parameters to work.
3) it also has a remote on/off which i streer with the BMS and it is the last load that get cut off by BMS in LCV. I have a manual override switch so in emergency case i can choose to have it on and run below LCV=10V eg if BMS switches off and i need the autopilot running to have 2 or 5min more time to switch to backup batterie.

Like this all that cares about spike and surge is protected. The inrush current an empty 5F capacitor creates could easily kill the FETs in the BMS...just one issue with it.
We had them with lead but with LFP noone uses them in car audio as the LFP is as fast as cap, or fast enough...

[CapsizedVeteran]

Originally Posted by CaptainRivet

I would just use the money and get a high quality DC2DC 12V to 12V converter like the victron orion-TR 30/12/12 (NOT the TR-smart, thats the charger!) Which is around 160Euro but you often get them unused=new for a bargain as people order that and think its the charger which costs double. Paid 80Euro for mine.

The orion goes in between your main busbar and senstive electronics like chartplotter, radar, radio,ap computer. I have connected all life dependent must have electronics to its output ( and plotter+radar i consider to be)
The orion does 3 things:
1) it galvanically isolates the output completely from the input, means no spike,inrushes or alike, total clean power.
You can even add a small LTO or AGM motorbike battery as buffer battery, lifepo4 is not good as that buffer is aways at high SOC 90-100%
2) it stabilizes the fluctuating input voltage 9-18V into a stabilized adjustable output voltage. So you can compensate voltage drop and even raise it higher. I have it at 14.8V, that compensates voltage drop and electronics get min 13.8V and max 14.4V, independent of the input voltage down to 9V. so optimum parameters to work.
3) it also has a remote on/off which i streer with the BMS and it is the last load that get cut off by BMS in LCV. I have a manual override switch so in emergency case i can choose to have it on and run below LCV=10V eg if BMS switches off and i need the autopilot running to have 2 or 5min more time to switch to backup batterie.

Like this all that cares about spike and surge is protected. The inrush current an empty 5F capacitor creates could easily kill the FETs in the BMS...just one issue with it.
We had them with lead but with LFP noone uses them in car audio as the LFP is as fast as cap, or fast enough...

None of what you said has anything to do with the mosfets in the BMS handling transient currents from inductive loads.