Anyone using a LiFePO4 12 volt replacement battery ?

[sorka]

OK, can someone help me with the math? My car draws 70 watts when it's sleeping. The contactors kick in about 6 times a day to recharge the 12v battery. It's drawing 5.83 amps continuously and the contactors close every 4 hours or so to recharge it. If the car just sits there plugged in or not, I'm still drawing 70 watts from the 12v battery. That's 23.3 AH every 4 hours or about 140 AH per day or 51,100 ah / year.

Now let's say I'm driving 2 hours every single day so that's like 46,842 ah a year being drawn from the 12 volt battery. For a 30 AH battery, that would make it 1561 cycles in a year. How many cycles is this battery good for? I've seen anywhere from 500 to 1000 cycles quoted for lipo batteries.

I'm sure I'm missing something, so I'm just throwing out what comes to mind first and someone can tell me where I went wrong.

 

[ElectricLove]

@ElectricLove,
Looks like your battery is actually 30AH, for $400? That's impressive. It does look like a great fit for the Tesla.

How do you handle balance charging the cells? Have you verified the Tesla charges to 14.4V?

I have verified voltages of the 12V system during rest, short drives and extended road-trips; I have put about 25K miles on my car in the last 10 months doing testing and have also done extensive testing in a few other vehicles as well. Test set up is a DC clamp ammeter and a voltmeter, both are data-logging to a computer the entire time and both are connect to the 12V leads (ammeter is obviously clamped). I just wrote a long description of all the values but then realized I may want to consider that information somewhat "protected" as it is an asset to my business and a result of my sweat-equity... Anyways, suffice it to say, I've got the data and I've considered it in the chemistry selection...

The cells are balanced via BMS internal to the battery, there are also several layers of protection offered by our BMS as it is wired in series with the battery pack (so it has the ability to disconnect in fault scenarios).

My battery is indeed a 30Ah capacity (true capacity at C/5 rate); pricing is kept minimal because I run a "tight ship" and keep all my costs as small as possible, I do my research out of the house and on my own vehicle or friend's vehicles, I do all the work myself and I invest my time into it rather than throw money at it... This keeps my costs as low as possible for the batteries, as low as possible for the fulfillment process and as low as possible for the business on the whole! My margins may be considered "low" for the industry as well, but I'm not bothered by that because I am a TESLA driver and see the value of this product to myself and the industry on the whole and I see the value of it selling really well and eventually going to TESLA-HQ as higher than the value of having a bigger mark-up...

 

[ElectricLove]

OK, can someone help me with the math? My car draws 70 watts when it's sleeping. The contactors kick in about 6 times a day to recharge the 12v battery. It's drawing 5.83 amps continuously and the contactors close every 4 hours or so to recharge it. If the car just sits there plugged in or not, I'm still drawing 70 watts from the 12v battery. That's 23.3 AH every 4 hours or about 140 AH per day or 51,100 ah / year.

Now let's say I'm driving 2 hours every single day so that's like 46,842 ah a year being drawn from the 12 volt battery. For a 30 AH battery, that would make it 1561 cycles in a year. How many cycles is this battery good for? I've seen anywhere from 500 to 1000 cycles quoted for lipo batteries.

I'm sure I'm missing something, so I'm just throwing out what comes to mind first and someone can tell me where I went wrong.

Well, I'd say the best thing to do is to hook up a couple data-logging tools to your battery to get a more accurate representation of what is going on...

LiPo is very different from the chemistry in the BattMobile battery, which is LiFePO4... I know they sound similar and the Po and PO4 seem like they could be the same but Po=Polymer and PO4=Phosphate and also the Fe, which is Iron, is very important as it adds to the stability of the battery cell (LiFePO4 are well-known for being "safer" than others); this matters because LiFePO4 are much better at cycling than other chemistries, they last for several thousand cycles even if doing DEEP (100%) discharge cycles, which doesn't happen in these cars.

Well at least the CGI photoshopped battery is available

Haha, you are right, I added a photo of an Engineering Sample battery to my website listing but not the eBay one, I guess I should add it there... I wasn't happy with it b/c the ES batteries aren't as nice cosmetically as the finished production units will be, but its best I've got for now... Thank you for pointing this out for me!

 

[ElectricLove]

OK, can someone help me with the math? My car draws 70 watts when it's sleeping. The contactors kick in about 6 times a day to recharge the 12v battery. It's drawing 5.83 amps continuously and the contactors close every 4 hours or so to recharge it. If the car just sits there plugged in or not, I'm still drawing 70 watts from the 12v battery. That's 23.3 AH every 4 hours or about 140 AH per day or 51,100 ah / year.

Now let's say I'm driving 2 hours every single day so that's like 46,842 ah a year being drawn from the 12 volt battery. For a 30 AH battery, that would make it 1561 cycles in a year. How many cycles is this battery good for? I've seen anywhere from 500 to 1000 cycles quoted for lipo batteries.

I'm sure I'm missing something, so I'm just throwing out what comes to mind first and someone can tell me where I went wrong.

Important to compare would be the spec-sheet on the C&D DCS-33RIT, which is the OEM battery; I've attached the datasheet for that battery (compare with DCS-33IT, it is the same). Similarly to LiFePO4 the cycle-life goes way up when doing smaller DOD cycles, the number I've found to be the most relevant is the 80% DOD value, with the AGM the 80% aligns with the way the TESLA treats the battery...

 

[sorka]

Well, I'd say the best thing to do is to hook up a couple data-logging tools to your battery to get a more accurate representation of what is going on...

So your answer is to tell me that I didn't accurately log my consumption? I have a very good Fluke data logger. I guarantee you it's more than accurate enough for this discussion.

 

[ElectricLove]

So your answer is to tell me that I didn't accurately log my consumption? I have a very good Fluke data logger. I guarantee you it's more than accurate enough for this discussion.

ok, sorry, it seemed inaccurate based on the data I've logged, we don't need to "compare data-dicks" haha... I root the cars that I work with as well so I can access the TESLA data-logging on top of my own equipment, so that does help some...

Do you by chance mind sharing some data with me? I'd be most grateful as I can't really ever have "enough", are you in an X or an S?

I didn't mean to sound like I was avoiding your question; your math makes sense, however from what I've learned the vehicle isn't cycling quite that much (1500 times per year); I think the missing link might be how much energy is being contributed to the 12V load by the DC/DC converter over the course of the year?

If you look at the data sheet for the C&D (OEM) battery you can see that at 80% DOD you are looking at 600 cycles, less at 100%, this means you would expect it to fail every 5 months or so based on your cycling analysis... But that isn't what we see in these vehicles, though sometimes it may feel like it, haha!

 

[sorka]

Similarly to LiFePO4 the cycle-life goes way up when doing smaller DOD cycles, the number I've found to be the most relevant is the 80% DOD value, with the AGM the 80% aligns with the way the TESLA treats the battery...

I agree with that. But the lipo dicharge voltage curve is far flatter than the lead acid. The onboard system will discharge the lipo almost completely before it detects enough voltage drop to kick in and close the contactors. If you somehow manage to trick the car into charging after only a 20% SOC draw, then the contactors will kick in far more frequently than they do now and accelerate wear.

Don't get me wrong. I think it's a cool idea, but I'm just wondering how you get around the fact that in my scenario, we're still drawing over 45000 ah a year out of the battery and I don't see how you prevent the battery from deep cycling or causing the car to cycle the contactors a whole ton more to keep the battery in a narrow SOC?

 

[sorka]

ok, sorry, it seemed inaccurate based on the data I've logged, we don't need to "compare data-dicks" haha... I root the cars that I work with as well so I can access the TESLA data-logging on top of my own equipment, so that does help some...

Do you by chance mind sharing some data with me? I'd be most grateful as I can't really ever have "enough", are you in an X or an S?

2015 P85D. Consumption rose from about 50 watts for the first 6 months and then climbed to about 70 watts bit after a software update(look for the vampire draw threads). Also, there are others on this forum that have done better than me at logging both current and voltage at the same time. I have to do one or the other(either inline for current) or parallel for voltage. I'll see what I can dig up. It's been over a year so I might have to repeat it.

 

[ElectricLove]

I agree with that. But the lipo dicharge voltage curve is far flatter than the lead acid. The onboard system will discharge the lipo almost completely before it detects enough voltage drop to kick in and close the contactors. If you somehow manage to trick the car into charging after only a 20% SOC draw, then the contactors will kick in far more frequently than they do now and accelerate wear.

Don't get me wrong. I think it's a cool idea, but I'm just wondering how you get around the fact that in my scenario, we're still drawing over 45000 ah a year out of the battery and I don't see how you prevent the battery from deep cycling or causing the car to cycle the contactors a whole ton more to keep the battery in a narrow SOC?

I like your questions and thought process!

You should look at LiFePO4 discharge voltage curves though, be careful with LiPo since it is not the same chemistry; they do have similar curves though.

What I did was first collect a lot of data and find the min/max voltages that the vehicle uses to "manage" the AGM battery installed; I used these min/max values to establish the design of the battery and I maintained as one of my requirements that it should only discharge about 50% DOD within those values. It took quite a bit of data collection to figure out if this would work or not and I've learned that it will.

 

[ejames]

I like your questions and thought process!

You should look at LiFePO4 discharge voltage curves though, be careful with LiPo since it is not the same chemistry; they do have similar curves though.

What I did was first collect a lot of data and find the min/max voltages that the vehicle uses to "manage" the AGM battery installed; I used these min/max values to establish the design of the battery and I maintained as one of my requirements that it should only discharge about 50% DOD within those values. It took quite a bit of data collection to figure out if this would work or not and I've learned that it will.

Thank you all for the excellent discussion . learned alot . Jim

 

[FlatSix911]

Well, I'd say the best thing to do is to hook up a couple data-logging tools to your battery to get a more accurate representation of what is going on... LiPo is very different from the chemistry in the BattMobile battery, which is LiFePO4... I know they sound similar and the Po and PO4 seem like they could be the same but Po=Polymer and PO4=Phosphate and also the Fe, which is Iron, is very important as it adds to the stability of the battery cell (LiFePO4 are well-known for being "safer" than others); this matters because LiFePO4 are much better at cycling than other chemistries, they last for several thousand cycles even if doing DEEP (100%) discharge cycles, which doesn't happen in these cars.
Haha, you are right, I added a photo of an Engineering Sample battery to my website listing but not the eBay one, I guess I should add it there... I wasn't happy with it b/c the ES batteries aren't as nice cosmetically as the finished production units will be, but its best I've got for now... Thank you for pointing this out for me!

Keep up the good work!
I am interested in your LiFePO4 battery... are you considering a group buy for TMC members?

 

[ElectricLove]

I hadn't thought about that but I would definitely do it! I love TMC, I went to TMC-Connect this year and had a lot of fun!

So with a "group-buy" do I establish pricing at various thresholds (order volume; like "If 5, X price. If 10, Y price. If 20, Z price")? I haven't done one so I'm not sure what is most effective...

 

[theslimshadyist]

This may be a dumb question but I didn't know there was a 12V battery in my MS. What is this for and why would the car even need when it has a 100kWh battery pack in it?

 

[FlatSix911]

I hadn't thought about that but I would definitely do it! I love TMC, I went to TMC-Connect this year and had a lot of fun!

So with a "group-buy" do I establish pricing at various thresholds (order volume; like "If 5, X price. If 10, Y price. If 20, Z price")? I haven't done one so I'm not sure what is most effective...

Excellent. I have participated in a few group-buys for the local Porsche club and they all work the same way.
Higher order volume = lower price. If we have 5 orders, 10% discount ... 25 orders, 30% discount, etc ...

 

[thimel]

This may be a dumb question but I didn't know there was a 12V battery in my MS. What is this for and why would the car even need when it has a 100kWh battery pack in it?

12 volt battery is needed to run the computers that check the interlocks to make sure it is safe to close the contactors to check connect the HV battery to everything.

 

[theslimshadyist]

I didn't know that so thank you.