LiFePO4 and auto alternator concerns...

[DonPhillipe]

I am an early-on experimenter with lithium. Being in the electronics industry for a long time, I've seen my fair share of how new technology operates. No one knows anything about the latest iteration of what's new and the marketers capitalize on this and flood the market with equipment not everyone needs. Same with photography equipment over the years and what's being sold to "make you a great photographer".
The Victron video is what everyone rides on these days to sell B2B and I've seen enough people not using them that aren't having issues so that's what drives me personally to better figure out this archictecture. Unfortunatly the diverstity of equipment out there including how each alternator performs causes each implementation to be a different science project. The easy solution is to say "pay $300 for a B2B that produces 25A", then you say "but I'm already charging my Lead/Acid at 60A and I thought I was improving my charging profile while going lithium and that's not what I call improvement", then they say "well buy the $1200 Victron that charges at up to 80A" and I say "I don't want to buy the $1200 Victron combo" and we are then at a stalemate.

I am going to test my setup a little more and then make my decision.

 

[time2roll]

Short time lurker, first time poster... glad I found this thread

We've got a 2010 Class C on Ford E450 chassis. From what I've been able to determine, alternator is rated at 120-140A and internally regulated by the ECU.

Our "beginner" solar infrastructure is in place; 2x100W panels, 40A MPPT charge controller, 1500W sine wave inverter. It's been working flawlessly for a couple months now, but the flooded house batteries are definitely the weak link. I'm looking a replacing the two 100Ah "deep cycle" Wal Mart batteries with a DIY setup using four 304Ah EVE lithium cells, and want to make sure I have all our bases covered.

As it stands, we can charge the house batteries through shore power, generator, chassis via alternator, and solar although we are mostly using the latter two. We're looking to extend our boondocking capability. I try to use solar as much as possible, and give 'er a bump charge by firing up the rig every now and then when the clouds are out. I prefer to not run the genny, it's more of a nuisance, especially if we have neighbors. Don't wanna be "that guy" lol.

Our power needs are low; the most draw we have at any given time is about 15-20Ah. I've noticed the chassis/alternator regulates the charge it provides to the FLA; as house battery charge rises, alternator output diminishes. The microwave has been ditched for more cabinet space, the fridge runs off propane, and if I need to run the AC we can always fire up the generator. Mostly we chase 70 degrees though.

I *suspect* we'll be OK with the setup as-is and not need a DC to DC. However, I'm not well-versed when it comes to the house/chassis relationship, so reaching out for input from those who know more. Thoughts? Pitfalls? Am I full of it? LOL...

Edit: It's a 12V system and no plans or need to change at the moment.

I assume your existing system has a fuse to cut power if you create an overload current charging the LFP battery. A good tool is a DC clamp-on ammeter to measure what is actually going on and adjust accordingly.

 

[Rrrracer]

@time2roll, yes, all precautions are in place, and I have fulltime monitoring capability.

Think I'm gonna throw caution to the wind and go for it. Gonna be a couple weeks before the cells arrive and I can get everything set up. Hope to learn more from you guys during the downtime. Looking forward to hearing the results of your testing @DonPhillipe.

Unless someone changes my mind, I'll report back with my one-data-point evidence, yea or nay, once I am able to observe everything in operation. Fingers crossed and extinguisher close by

 

[DonPhillipe]

I bought 3 of those cheap (around $20 ea) Hall-Effect meters. They consist of little doughnuts of wound wire on one end of the lead wire and a meter on the other. The lead wire is not very long so you have to use a similar type (shielded/not) to extend. What you do is feed the positive of the device or battery cable through the doughnut to monitor the amp flow. I am placing one on the main alternator lead (routing lead through the dougnut), then the coach battery positive lead through another and then the cable of the coach house batteries through another( assumes you don't already have a "battery monitor" in-line with one of the house battery cables). That is what I am going to use to monitor my system through the process of lithium conversion. Each one of these doughnuts has a signal wire and I am going to extend for all 3 meter lead wires to the cab and I will mount them under the dash so I can monitor what's happening at least through the initial breakin of the new equipment. I am also going to place a "k type" (up to 2000F) sensor on the alternator itself with a remote digital gauge ($15 or so). I believe the shutdown for the diodes is 200C (for the Sprinter 150A alternator) and of course abort the test if I see this temp approaching or any amperage going out of spec. The Sprinter alternator is fused so that may or may not be a good thing. I am hoping that will provide a pretty clear picture of what is going on during the charging sequence.

Now I am sure you have looked at the lithium battery charge curves and they are all flat as pancakes but in all my blog reading, most who are charging with alternators and who are actually monitoring the true charge amperage say there is a spike (or curve actually) at the front end for 10-15 minutes where the brunt of the potential alternator abuse occurs. One fellow used a small B2B that he switched in-line only for that onset period and then he threw a master switch to switch the B2B out and do a direct alternator charge until the end of the charge cycle where the charge amps petered out at their 9x% charge-level they reach at 14.2VDC charge voltage (from alternator). I think he had been using this setup for over a year. My theory is one could do the same with the proper resistor material and thus even optionally switch the high-amp resistor in/out with a solenoid which would short across the resistor in circuit once the alternator amps began to drop from the initial high-side. The high amp requirement would be maininly after a substantial depleting run-down of the batteries and at the first engine start up after that. This is what LION energy suggests (Costo supplier) by using a 12 gauge lead inline with the alternator but they don't say anything about a tedious solenoid type direct charge option as well.

My thoughts are also that if something like this will work practically , this will provide a great solution which will also inspire the B2B converter mfgs to come down to a more realistic price structure. For example with a 60-80A B2B unit (perhaps adjustable to alternator capacity) brought down in price to (perhaps) around $70, who would waste their time constructing the type of facility that I propose. However, once enough people get through the testing phase to produce a blue print for a small framework kit that contains base components (solenoid, resistor and maybe even an automatic cut on/off circuit for the resistor break out solenoid), then perhaps the B2B converter mfgs will stop gouging with these high prices and produce a functional unit other than the laughable 20-30A that most are buying today and thus provide an economical unit to make projects such as this unnecessary.

 

[DonPhillipe]

The person above my post who suggests a clamp meter - yes I don't mean to take away from that; I was just being lazy using the remote read outs (and one can always go overbudget trying to save LOL). Using the clamp meter would be the most economical way to test but would require one be light on the feet during the first dry run. Do be sure your clamp meter or your hall effect for that matter does DC current, as DC current measuring through hall-effect (clamp or doughnut) I believe is a realtively new feature (and I say based only on my collection of them from garage sales over they years and having done no actual research other then when buying mine I noticed not all worked with DC amps).

 

[Rrrracer]

A thermocouple on the alternator is a great idea. Love your plan, it's far more comprehensive than my simple house battery shunt/single gauge setup and will allow you to see every step of the process.

My plan is to not run the batteries down that far and keep the charge cycle between the knees heheh.

@DonPhillipe, would you mind providing links to the blogs you are following?

 

[Jasgeer]

The person above my post who suggests a clamp meter - yes I don't mean to take away from that; I was just being lazy using the remote read outs (and one can always go overbudget trying to save LOL). Using the clamp meter would be the most economical way to test but would require one be light on the feet during the first dry run. Do be sure your clamp meter or your hall effect for that matter does DC current, as DC current measuring through hall-effect (clamp or doughnut) I believe is a realtively new feature (and I say based only on my collection of them from garage sales over they years and having done no actual research other then when buying mine I noticed not all worked with DC amps).

Don, also being an EE type it seems that the 'right way' to control the charging of a lithium battery from an alternator is to open the field current so that the alternator never puts out current without having a load and perhaps to modulate it so that the output current drops as you reach the set point. It is my understanding that alternator regulators would be toast without a load (which normally the wet cell provides). I have read a '12V' Auto Alternators can easily output double that so any charge profile could be created with the proper controlling electronics. Sounds like a business opportunity to me to make a combination BMS for the Li battery and regulator for the proper voltage and current profile (all programmable). Load dump must be avoided by design.
I look forward to seeing your results.

 

[Jasgeer]

A thermocouple on the alternator is a great idea. Love your plan, it's far more comprehensive than my simple house battery shunt/single gauge setup and will allow you to see every step of the process.

My plan is to not run the batteries down that far and keep the charge cycle between the knees heheh.

@DonPhillipe, would you mind providing links to the blogs you are following?

Some Balmar alternators for marine use incorporate temperature sensors in the alternator housings.

 

[Zwy]

Bad idea. I am sure you are writing this ?. Surely this is not legal and you will soon electrocute the next mechanic who services your vehicle and possibly yourself.

Really? I gave out this idea some time ago as I installed a second 250a alternator on my truck and wanted to get around 100a or more to the camper battery. How would this be different than a factory installed inverter on a vehicle? Better tell Ford GM and others that it is illegal because you state it is.

In order to electrocute someone as you state, a few things would have to occur. One, the inverter has to be powered up. I don't know anyone that would use this method of moving large amounts of watts to a camper that would leave the inverter powered up when the vehicle isn't running for the simple fact it would leave the battery dead over time. Switching it off manually or even automatically can be accomplished several ways, whether thru a remote switch connected to a toggle or triggered off ignition power. Or, ignition power triggers a continuous duty solenoid that feeds the DC power to the inverter.

Second, all wiring should be clearly marked with orange bands ( usually orange electrical tape wound spirally) to indicate it is high voltage AC, this is no different than the use of orange marked circuits in hybrid and electric vehicles. 120V AC is not even close to the 400V plus used on hybrid and electric.

Third, someone has to intentionally intrude upon the circuit. Best practice is to use proper connectors designed for AC power connection such as one like this. https://www.amazon.com/Kohree-Power.../B07JZ44ZKZ/ref=psdc_6359402011_t2_B08CKYWK7L

Now if your mechanic doesn't know the difference between what an AC outlet looks like compared to a DC outlet rated for 12V to 36V, well, you should probably find a better mechanic.

 

[Zwy]

Greetings. I am planning on replacing the 100 amp hr AGM house battery in my 2000 VW camper. I have a single remaining concern with the variety of Lithium batteries that advertise Plug and Play status. Because the BMS in a good lithium battery will shut the battery off when fully charged, what happens to the alternator circuit? It is my long held belief that in boats and cars/vans that one does not ever want to experience an open circuit with regard to the alternator. I do have a 160 watt solar panel on the camper, ( AGM) and it is connected to the alternator charging circuit. I have changed nothing in that circuit.

I would appreciate input regarding this issue. Most cars come with some form of lead acid battery and the charge tappers off but does not stop and lead acid batteries are reasonably forgiving and accept a trickle charge.

One well known Lithium provider simply said that this issue should be taken up with folks that manufacture 12 volt alternators... I think there must be better answers out there from knowledgable battery folks.

Any educational tips here are welcome.

How big is the wire from alternator to house batteries? That determines how much voltage drop will be in the circuit. It also regulates how high the charging voltage at the battery will be.

Now if VW put 4/0 cable from the alternator to house batteries, you just might need to regulate the load and charging profile. LOL

Personally, I would use the BMS as it really is designed to be, final protection against overcharge/discharge and cell monitoring plus discharge rate. I would prefer some type of charge control such as B2B or external alternator regulator so parameters can be tailored to the battery chemistry if the battery cable from alternator to the house batteries would not limit the charge current due to excessive voltage drop. The cheapest method and one that is simple is simply the smaller gauge fused wire from alternator to house batteries that might not completely charge the house battery but would still allow some charging.

Do you understand what/why?

 

[DonPhillipe]

I continue to test my setup. I have a small camper on a T1N Sprinter chassis. The alternator shipped with it is a 150A. For 5 years I've been using 2 6V GC2 batteries from Duracell that were supposedly extended to 230Ah (over the lower priced Costco bottom of the line that delivered 210Ah). Of course that leaves 115Ah to work with because with lead/acid the warnings are to never go lower than 50% SOC. As I've added more living quarters fans and electronics over the years, I find I only have only 24 hrs of reserve zero-sunlight days and at the moment a worst case scenario (furnace all night in 20F) I now get down to the upper 5x% range by dawn on each of these extreme condition days, so I decided to look at adding more battery power. So I ordered the "on a shoestring" LiFePo4 four-cell-set that was noted elsewhere here and after 3 months they finally arrived. I conducted a few tests and decided to order a second set since each battery set of 4 for a total of 8 can fit in my vehicle battery case with a little metal re-work.

I was worried initially about a DIY install and thus altering my system to use a 280aH bargain LiFePo4 setup. I ordered and finally received the batteries last year and only recently the newly released JBD 200A BMS. At the same time I hoped to avoid the B2B charger route because first of all it defeated one of the major bullet points and that is charging a system at at 1C rate. Every model for less than $300 USD was only showing outputs in the 40A range which was a downgrade for me, since I frequently saw at least 65A pouring into my lead/acid old-school system, for me it would be a major downgrade in the charging department.

Big shocker came from the tests I ran. I found that my two existing lead/acid 6v GC2's in series actually put a notable higher amp load on my alternator than did a single bank of 280aH LiFePo4 batteries. Otherwise the tests revealed that I could take a rish and actually charge up to 560aH with the two 280aH banks in parallel but I would somehow have to wait for the overall load to stabilize (once the chassis battery was fully charged) and this would mean that I would need a switch to toggle the LiFePo4's out of the charging circuit and not charge them for about an hour after I started the engine (allow the chassis battery to top out) so I'd have to add a toggle switch to disable the chassis/house-leisure bridging solenoid to break house batteries out of the charging circuit until the chassis battery had come up to 0A draw. This would mean also that while charging 560A I could not run the Heater or AC. Still the alternator would be supplying a constant 140A out of the 150A available with added suffering no AC or heater, so this is a concern and thus the reason I plan to wait and simple swithch in one bank OR the other initially while not thinking about bridging them constantly until once I upgrade to a 200A alternator.

(Note: All tests run at engine idle speed of 700RPM and with belt gear up factor, the alternator shaft is likely turning at around 1400 or slightly faster)
(Note: The "it's a set up" Victron alternator burn-up sales-demo video has their alternator turning at 500 RPM shaft speed)

SPRINTER 150A ALTERNATOR
(no house battery connected - only stock chassis AGM battery and auto accessories)

- first 5 minutes after start:
alternator produces 120A
40C (diode bank temp)
cab electronics with running lamps 30A; up to 45-50A with AC on high

- next 30 m minutes after start
alternator drops to 80A and continues to deplete to constant match 30-55A chassis load
60C (diode bank temp)
cab electronics remain constant

- next hour after start
alternator drops to 30A (up to 50A with AC/heater) and remains constant
65C (diode bank temp)
cab electronics remain constant and are the only alternator load


SPRINTER 150A ALTERNATOR - COMBO 230aH dual 6V GC2 lead/acid in series
(external lead/acid connected to charging system through 200A solenoid)

- first 5 minutes after start:
alternator produces 165A
45C (diode bank temp)
charging amps into dual 6V lead/acid - 65A
cab electronics with running lamps 30A; up to 45-50A with AC on high

- next 30 m minutes after start
alternator drops to 90A (105A with AC/heater) and continues to deplete as SOC grows
80C (diode bank temp)
charging amps into dual 6V lead/acid - 58A
cab electronics remain constant

- next hour after start
alternator drops to 80A (95A with AC/heater) and remains constant
85C (diode bank)
charging amps into dual 6V lead/acid - 50A
cab electronics remain constant and are the only alternator load


SPRINTER 150A ALTERNATOR - 4 LiFePo4 @ 280aH in series
(external LiFePo4 connected to charging system through 200A solenoid)

- first 5 minutes after start:
alternator produces 165A
45C (diode bank temp)
charging amps into 280aH LiFePo4 bank - 55A
cab electronics with running lamps 30A; up to 45-50A with AC on high

- next 30 m minutes after start
alternator drops to 85A (100A with AC/heater) and remains constant until bank fully charged
80C (diode bank temp)
charging amps into 280aH LiFePo4 bank - 55A
cab electronics remain constant

- next hour after start
alternator remains at 85A (100A with AC/heater) and remains constant until bank fully charged
85C (diode bank)
charging amps into 280aH LiFePo4 bank - 55A
cab electronics with running lamps 30A; up to 45-50A with AC on high

Note: For all scenarios where diode temperature reaches 80C or above, increasing engine RPM to 1200RPM will cool
the diode bank down to 65C - remember alternator turns double speed of engine RPM)

My own conclusion for my own needs:
B2B is not needed for 280aH LiFePo4 and below this amount of aH (chg rate 55A)
B2B is not needed for 560aH LiFePo4 and below this amount of aH if a 200A alternator replaces the 150A (chg rate 110A)

 

[DonPhillipe]

And another observation is I do understand that a B2B flies with the old adage "if you've got the money honey, they've got the time" because a very expensive B2B could be used (presumably) to pull the full capacity of an alternator and charge a bank that has a 1C charge rate at closer to the 1C rate (mine is 280A!) or at least over the re-proportioned V/A step up to sap the alternator for all it had to offer and (presumably) changing every moment as needed, but that assumes the expensive B2Bs are even capable of being that intelligent.

Still this was not my goal, however and I am reaching a happy medium with my system and with a 110A to be placed in the future on a 200A alternator. It provides a good charging rate yet at a fraction of the 1C advertised rate, but at a much higher amp output than the "affordable" B2B's in today's market deliver, which currently (pun) seems to be in the 20-40A charge rate range. It also avoids the over-sold market of the low-end B2B that camper owners are buying up like crazy because of this one highly deceptive youtube video.

For me, the alternator provides a sizable sum of the charging power I do significantly depend upon and it doesn't make sense to salute and say "yes sir" to Victron while turning away from all the power the 150A Sprinter alternator has available which simply waiting to be called up most of the time. (With the chassis drawing a 30-40A output that means I am letting all that 120A go to waste the majority of the time and thus why would I ever add a 20A-40A B2B when my extra amps varies from about 40-120A all day long.) So simply chunking the idea of the B2B that everyone salivates over for such a low profile setup as I have, without the B2B this automatically doubles or triples the charging rate I could utilize from any 20-40A output delivered from an overpriced piece in the under $300 USD range on the market today.

 

[Rrrracer]

We've been running our 280aH lithium in parallel with the starter battery and charging all via alternator for four months now, no issues whatsoever. It's great to have charging power on tap on when solar is not an option. It's also wonderful to arrive at our destination with the house batteries fully topped off and ready to go.

 

[DonPhillipe]

at our destination with the house batteries fully topped off

I am afraid I am going to have to report you to Victron ;-)

 

[DonPhillipe]

Another interesting observation when conducting the two tests that compare the alternator-charging of a dual bank of 6V 230aH lead/acid vs. a bank of 4 280aH 3.2V LiFePo4, I observed that although the lead/acids at a 60% SOC showed 65A into them, the remote voltage reading at the lead/acid battery terminals was 14.2V. With the LiFePo4 bank at a 34% SOC showing 55A into them, the voltage at the battery was 13.8 rather than 14.2. This must have been from the larger internal resistance of the lead/acid.

 

[Zwy]

Have you checked your starter battery? The reason I ask is the 120A after startup with just the starter battery does raise some questions. That is about 80A going to recharge the battery and really there should have been faster voltage recovery just after the load (starter) stopped. The starter battery may have some sulfation and the reserve capacity is less as a result. This can be checked with a carbon pile load tester if you have access to one and know how to use it based upon temperature. Note also how fast the voltage recovers after the load is shut off, this reading is important and many miss it as it also shows reserve capacity. A quick recovery close to fully charged voltage of 12.6V indicates a healthy battery, a slow return or it doesn't recover close to 12.6V indicates low reserve capacity.

 

[DonPhillipe]

Battery is less than a year old but "as cheap as one can buy" Costco so it may not be an optimum performer. Also on my test days I neglected to state that the chassis battery test each time had been initiated "raw" after the vehicle has been parked for a week or more. Sprinters for some reason have phantom drains on the chassis battery so the battery would have some level of depletion. Also this model Sprinter runs with the headlamps lit and the test was with no heater/AC fan on which meant whatever number one sees on the overall alternator output, that 30A static chassis-load must be subtracted to get the current that was flowing into the chassis battery (raise to 45 if the heater/AC fan is on high). The Sprinter has a small fuse panel attached via a significant metal bus bar attached dirrectly to the chassis battery positive post that distributes all the power and includes the alternator charging cable so it is not easy to physically measure what current is entering or exiting the chassis battery "directly" but one can subtract the value of 30A during my tests (heater fan off) because this is the constant value seen coming from the alternator once all the other loads go to zero. Of course a test harness could be built but I did not go that far.

(With the house/leisure battery engauged, one would add the 30A chassis static load plus the amp value going into the house/leisure battery and subtract the sum of those from the overall alternator amp output to come up with what amps are going into the chassis battery.)

 

[Zwy]

Cheap batteries are always interesting when it comes to load tests, they never die but fall short on load tests. Pretty amazing. I've seen 10 yr old cheap batteries like Walmart or Autozone that will still hold a charge but if you have to crank for more than 5 seconds the battery is dead but still holds a recharge.

 

[eXodus]

My own conclusion for my own needs:
B2B is not needed for 280aH LiFePo4 and below this amount of aH (chg rate 55A)
B2B is not needed for 560aH LiFePo4 and below this amount of aH if a 200A alternator replaces the 150A (chg rate 110A)

I've been saying that for years, that B2B are not necessary to protect batteries for many applications.

Actually LFP charge worse with an Alternator, depending on the chemistry they never get 100%,

My 450AH bank - never draws more then 100A from a 180A Alternator. (but I got 25Ft of cable and use the frame as return, negative)
The usually charge is somewhere around 60-80A. It's actually pretty slow, considering what the alternator could provide.
I mean I need to drive around 5 hours to get the bank recharged

B2B would increase the charge rate and the load on the alternator in my opinion and when sized incorrectly contribute to alternator burnout.

 

[eXodus]

Note: For all scenarios where diode temperature reaches 80C or above, increasing engine RPM to 1200RPM will cool
the diode bank down to 65C - remember alternator turns double speed of engine RPM)

No idea what Mercedes used in their alternators, but even 80C is not an issues usually, (sine the engine blocks runs at 90C) most Diodes are specified to 135-150C

Just with rising temperature you have to de-Rate the amps.