DC-DC back to start alternatives?

[hjohnson]

Originally Posted by donradcliffe

Suit yourself, but don't come here crying when your BMS has a high voltage cutoff and fries the alternator diodes. From the 614 manual

Many LiFePO4 batteries have a Battery Management System (BMS) that may disconnect the battery from the alternator as a protective action or when charging is complete. The regulator must be shut down before the battery is disconnected .Running an alternator without a battery will damage the alternator and may damage any attached system. This is doubly true if the battery can be disconnected during high current charging, causing a load dump. The load dump can easily cause a high voltage spike which will destroy the alternator’s rectifier, at minimum. This is not a warrantable failure. To reiterate: THE ALTERNATOR MUST BE SHUT DOWN BEFORE DISCONNECTING THE BATTERY. THE ONLY SAFE WAY TO SHUT DOWN THE ALTERNATOR IS TO TURN OFF THE REGULATOR.

Explain what you think is crazy inefficient.

You said it right there in the last sentence. You build your system so that the alternator is shutdown long before you ever get to the state where you’re going to see a disconnect. It’s not rocket science.

ABYC recommendations are that all BMSs should have a pre-warning signal before the actual disconnect event. You can use this to disconnect the brown wire on the 614, which will immediately kill the field Current on the alternator. If you set the voltage limits on the 614 conservatively, say a bulk of 14.3v and float of 13.8, the likelihood of an HVC disconnect is essentially zero. A HVC should never, ever happen in a properly designed system.

In my case, I have a Wakespeed that’s slaved to my BMS via CAN bus, so if the BMS isn’t happy, it will just command 0A into the battery, which has the same effect. If the BMS decides it wants to balance the cells, it will command a current limit of 1A, and let the balancer do its job.

I’ve been running my system now for 18 months and never once come close to a risk of a disconnect, and I run my batteries all the way up to 14.65 while aboard.

As far as my comment regarding efficiency, no dc-dc converter is perfectly efficient. Running at 30A, it’s going to be throwing a lot of heat, which is wasted energy. Plus then, his 100A alternator is effectively a 30A alternator. If you gang two together, then you risk pulling more power than the alternator can sustain, and will then potentially deplete the starter battery, or more likely cause the DC to DC to yoyo on and off after the starter is partially depleted.

The best way to handle this is to design things properly, use proper equipment.

[hjohnson]

Originally Posted by themorb

The whole concept of a lead acid starting battery seems like it should be obsolete by now -- the LiFePO batteries I ordered said they can do 800 amps for 5 seconds and 400 amps continuous. That should be enough to start my engine. And many boats install much more.

I think it mostly comes down to wanting simplicity in the most critical applications. Plus as others have said, Lead acid starting batteries are really cheap. On my boat, I’ve got a similar setup as you’re contemplating. The 18A orion charging a starter battery. The starter battery cost me all of $90 including the protective battery box it lives in in the bottom of my Laz. Cheap insurance.

[CaptainRivet]

best to trickle charge is to connect a small 30 or 50W solar panel to the start battery, will keep it always aflot doesn't matter if in marina or offshore.
that small 30-50W flexible panel fits on every boat and a cheap controller.

[MVDarlin]

Step up to AGM for the start bank. Charge rate limit similar to LiPO.

[CaptainRivet]

Originally Posted by MVDarlin

Step up to AGM for the start bank. Charge rate limit similar to LiPO.

AGM are the worst and most dangerous chemistry on boats and should be avoided at all costs. I tossed 2 times AGM overboard to avoid a fire on my old ketch that got a thermal runaway because one time the alternator regulator was faulty and 2nd time barnicals on rhe bowtruster nearly seizing it cause an overload on AGMs that caused a plate shortage.same in my cars, in 15 years 3 times thermal runaway...
Tossing overboard was only possible because i have a gas mask onboard with proper filters as gasing is poisoned and acid and a fire blanket that allows to wrap and touch them.

AGM are also much more vulnerable and cannot be resuced by refilling, good FLA are cheap and take a fair amount of abuse. You don't need the "advantage" of AGMs being deeper discharged, starting consumes 2-4AH in max...
I used refillable cheap FLA and 2 per engine instead one bigger, more redundancy...30AH LTO as starter are lying at friends and waiting for pickup as shipping to canaries not possible.

[nuku34]

How about a separate small alternator for starting battery charging only? That's what I have. Its cheap and easy to fashion a bracket, and most engines have crankshaft pulleys with multiple belt grooves. The small alternator will use its internal regulator and be wired directly to the start battery.

[s/v Jedi]

I used to have three alternators and yes, it used to be the best method to have a dedicated alternator for the start battery.

But this whole setup is obsolete and I removed and disposed of all three alternators, removed 100lbs of brackets and pulleys, then replaced the oem pulleys with serpentine pulleys and put back a Balmar 170XT as the single alternator for charging the start battery and supplying 12V power to whatever needs it, llus a pair of Orions that charge the house battery.

All serious house battery charging is done by solar. Diesel is over the hill, they will probably limit what you are allowed to buy in the future. Put the effort into solar production.

[Ballsnall]

Originally Posted by themorb

Question is what is the simplest way to charge the start battery? Just to replace a few amps lost each time starting... A $300 Victron DC-DC seems a bit overkill maybe? Is there a cheaper decent DC-DC trickle charger? What about a 120v trickle charger from my inverter? Not efficient but does it matter

Simple, most cost effective solution is to install a $40 VSR which should be capable or 40 to 60 amps. A Victron DC-DC charger is a great but expensive option for a start battery, a trickle charger would work fine and also has the benefit of applying a battery maintenance voltage so if there are extended periods between engine use then a trickle charger is probably ideal.

[AEM]

I decided not to change batteries to lithium and use regular Ca+ service batteries 540ah and 100ah lead for the motor. I removed the diode splitter, the alternator charges only the service batteries, regulator of solar/wind have outputs for 2 battery banks, with the priority of the motor battery. The starter battery is charged only from solar panels, it is enough to compensate the seldom start of the engine.

[rgleason]

I agree with hjohnson and s/v jedi.
Hjohnson, what bms and batteries are you using?

[hjohnson]

Originally Posted by rgleason

I agree with hjohnson and s/v jedi.
Hjohnson, what bms and batteries are you using?

I’m using a REC ABMS with 8 230Ah EVE cells in a 2p4s configuration. I believe you and I had a long back and forth this over on the DIY Solar Forums.

Interestingly, I actually did a bit of further research into this, and discovered that as long as the CAN bus is functional, the WS500 is watching for “0x35A” messages from the BMS. If the BMS goes into a warning state, or alarm state, the WS500 will immediately kill the field on the alternator and go into its own inhibited state. If the BMS then chooses to open the contactor, it’s no harm/no foul as the alternator is already shutdown.

If it loses CAN communications completely, the WS500 will go into a limp mode, with a preconfigured voltage limit. I have that set for 13.2v which should be safe.

[rgleason]

hjohnson Yes we did have a long dialogue on DIYsolar.

"discovered that as long as the CAN bus is functional, the WS500 is watching for “0x35A” messages from the BMS. If the BMS goes into a warning state, or alarm state, the WS500 will immediately kill the field on the alternator and go into its own inhibited state."

Very interesting. So as long as canbus works, The relay would not be needed. Is the relay just a backup then, or is its purpose to disconnect tbe charge bus affirmatively?

[hjohnson]

Originally Posted by rgleason

hjohnson Yes we did have a long dialogue on DIYsolar.

"discovered that as long as the CAN bus is functional, the WS500 is watching for “0x35A” messages from the BMS. If the BMS goes into a warning state, or alarm state, the WS500 will immediately kill the field on the alternator and go into its own inhibited state."

Very interesting. So as long as canbus works, The relay would not be needed. Is the relay just a backup then, or is its purpose to disconnect tbe charge bus affirmatively?

Nah, I still have the main relay there, but it means I'm even less worried about an errant disconnect doing damage to my alternator/power system, as if the BMS does decide it needs to disconnect the charging sources, it will first signal its intentions over CAN, which will cause the wakespeed to kill the field before the disconnect happens (and may, in fact, avoid the disconnect).

I have my CAN wired BMS->Wakespeed->Cerbo->N2k bridge so if that CAN Bus goes kaput, I'll know about it pretty quickly.