Anyone gotten their hands on a Trojan Trillium yet?

[john61ct]

I wouldn't be surprised if the warranty specifically excluded using them for propulsion.

Maybe even for marine use full stop.

And personally I'd wait at least a few years to get a decent sampling of user feedback.

[Yellowjacket]

Originally Posted by KTP

Blue Sea switches on a 72V system even though Blue Sea stuff really shouldn't even be used on a 48V battery bank.

Some Blue Sea switches are rated to 48 volts, and some of their
"C" series and UL-489 circuit breakers are rated to 80 volts DC.

[KTP]

Originally Posted by Yellowjacket

Some Blue Sea switches are rated to 48 volts, and some of their
"C" series and UL-489 circuit breakers are rated to 80 volts DC.

"rated to 48 volts" as some Blue Sea switches say doesn't mean they are approved for use on a 48V DC battery bank which can see float voltages as high as 60V during cold weather charging if the charger is temperature compensating.

[GordMay]

Originally Posted by john61ct

I wouldn't be surprised if the warranty specifically excluded using them for propulsion.
Maybe even for marine use full stop...

Trillium™ can be used in a variety of stationary and motive power applications.
➥ https://www.trojanbattery.com/trillium/
Note the Common Applications.

[Mike Banks]

When your batteries shut down it might be a good idea to have an emergency supply such as a jump-starter so you can use a radio.

[KTP]

Originally Posted by Mike Banks

When your batteries shut down it might be a good idea to have an emergency supply such as a jump-starter so you can use a radio.

Depending on the location, with 16 batteries, each individually capable of shutting themselves down and also waterproof, the odds of a total failure are exceedingly low.

Except for a lightning strike. Maybe keep one battery in a microwave ;-)

[Cpt Pat]

If all 16 batteries are connected in paralllel and designed to shut down at the same voltage threshold, then they will all shut down together in short succession. Maybe not in the same millisecond, but essentially all at the same time. (The higher the quality of the voltage sensing, the closer in time they all will shut down.)

I agree with Mike Banks: I have a small AGM in parallel with the bank to provide power once the LFPs (abruptly) drop off line. An entirely separate battery for the radios would be even better.

[Cpt Pat]

Originally Posted by GordMay

Trillium™ can be used in a variety of stationary and motive power applications.
➥ https://www.trojanbattery.com/trillium/
Note the Common Applications.

To take advantage of the cycle-life curve, the battery can't be operated within its full (0% to 100%) SoC range. I don't see how one can control the depth of charge/discharge without external management.

Looking at the user's guide (https://www.trojanbattery.com/pdf/Tr...UsersGuide.pdf), the high and low voltage disconnects are set at extreme limits. Both higher and lower, respectively, than most people use for LFPs.

The 110 AH battery disconnects when any cell voltage is higher than 3.8 volts (or assuming perfect top balance: all the cells are higher than 3.8 volts). The low voltage disconnect occurs when any cell voltage is below 2.3 volts, or the battery terminal voltage is less than 8.0 volts (I cannot logically reconcile how the battery terminal voltage could be less than 8 volts before any cell is less than 2.3 volts).

Controlling the depth of discharge is fairly easy with an external low voltage cutoff circuit. But, without using coulomb counting, I don't see how anyone can avoid charging to 100% SoC without manually monitoring and switching off charging during each recharge.

[john61ct]

Apparently open to CAN-bus commands.

But by default, the BMS should not be used to control charge termination.

Protective BMS functionality is there as a last-resort failsafe protection, for when your other protective gear fails, never used otherwise.

Adjustable LVCs, Essential circuits / loads set lower than Auxiliary is how you will maximize longevity and still have some reserve capacity when needed.

Obviously your charge sources need to be reliably regulated, IMO set to 3.45Vpc or so depending on C-rate

or if balancing is required, just above the "balance start" voltage setpoint, not sure if that's adjustable on these.

[Cpt Pat]

Originally Posted by john61ct

Apparently open to CAN-bus commands.

But by default, the BMS should not be used to control charge termination.

Thanks John. I wonder though how many users will use these batteries as "drop in" replacements for AGMs, without any external management, and then run the batteries to the internal protective limits. It appears that only the 92 AH model has CAN bus support.

It would have been more clear on my part to have expressed that concern up front.

Trojan offers no guidance on external management, and states in fact:

6.5. PREFERRED CHARGING PROFILES
The recommended charging approach is to use an appropriate I-E charging profile with a voltage setting of 14.4 to 14.8 VDC per battery (28.8 to 29.6 V for the U1). Many AGM or GEL charging algorithms that use an I-E profile will charge a well-balanced bank of Trillium batteries acceptably.

CHARGING
6.1. BACKGROUND
Trillium batteries are frequently used in applications where lead-acid batteries have previously been utilized...

(emphasis mine)

Users who allow the batteries to bang up against both the high and low voltage internal limits may come to grief on cycle life. Unless the cells have some hitherto unknown chemistry, they won't last long.

[john61ct]

"acceptably" is a great weasel-word there!

Unfortunately when 99.9% of the prospective buyers know very little, demand is for convenience, ease of use, no learning required, no changes to infrastructure

that is what the market will supply.

Even here many say things like, as long as it lasts X years that's fine.

So be it, have to acknowledge reality.

[Cpt Pat]

So true!

I'd be surprised to see more than a few hundred cycles without external management controls.

Here is an excellent paper: https://marinehowto.com/lifepo4-batteries-on-boats/

It's long, so skip to the germane section titled: DECIPHERING CHINESE MARKETING

Or, in context here, it could be retitled: Deciphering Chinese-influenced Trojan Marketing.

[john61ct]

Well that's more pessimistic than I think is warranted.

The fact that these "semi drop-ins" do have some comms to outside systems is a big plus IMO. Also aren't the cell-level balance leads accessible, at least for monitoring?

Yes, skepticism of marketing claims should be the default, Trojan hyperbole has grown extreme in recent years. This is basically a K2 product anyway, true not the greatest rep out there.

But IMO 2-3000 cycles should be easily possible with a reasonably self-educated and diligent owner.

Proper care is possible without everything being automated, and if desired, basic protective infrastructure would not cost the world.

[Cpt Pat]

Reference: https://marinehowto.com/lifepo4-batteries-on-boats/

The paper linked also covers a point I've tried to make here on this forum over and over: When determining the stop-charging threshold -- VOLTAGE MEANS NOTHING! Your 4S pack can be fully charged at 13.6 volts, if charging at low current, and continuing to charge without determining how many amp hours were taken out and then returned through charging will result in a damaged battery. It appears the Trojan battery is only detecting cell voltage for its high-voltage charge cutoff which, without other external management, will ruin the battery. Especially when using a dumb charger that runs at full absorption voltage for several hours, or an alternator that keeps chugging away at 14.4 volts indefinitely. And a charger that floats at any voltage above 13.6 volts will slowly cook the life out of an LFP battery -- without ever hitting the Trillium 15.2 volt (3.80 volts X 4) charge cutoff threshold.


My conclusion: The Trillium battery is definitely not a viable drop-in replacement for AGM batteries. Not without external management added by the user.

[Cpt Pat]

Originally Posted by john61ct

Well that's more pessimistic than I think is warranted.

The fact that these "semi drop-ins" do have some comms to outside systems is a big plus IMO. Also aren't the cell-level balance leads accessible, at least for monitoring?

In the case of the 110 AH battery, the answers are: no and no. Only the 92 AH battery has CAN bus support. Unless you want to crack the case open and add your own wires, the 110 AH battery has only two terminals to the outside world: the positive and negative terminals.

4.13 CAN BUS
[Only] The Trillium Group 24 battery [92 AH] is equipped with a CAN bus interface to allow communication with other CAN-based devices.
https://www.trojanbattery.com/pdf/Tr...UsersGuide.pdf

And besides, my point is the user still needs external management circuitry. It's not a "pull AGM out, drop in Trillium, tighten the terminal lugs, and forget it" implementation -- which I sadly suspect many people will misinterpret (lead on by Trojan). If you have to implement all that extra support circuitry, why not just use prismatics -- and save yourself several hundred dollars, plus have the flexibility of replacing an individual cell that may go bad in the future?

I've also found that physically rotating the cell positions in the pack about once a year keeps their capacity balanced because the cells in the center of the pack run a little hotter (less exposed heat dissipating surface area) and therefore age a little faster. You'll never have that flexibility with the Trillium due to its cell format.