12V battery dying but no warning?

[tga]

So I think my 4 year old 12V is nearly dead, but the odd thing is I have had no warnings. The last couple of nights I heard the HV contactor click when I took the dog out for his late night walk. This morning I paid more attention, and noticed the contactor is clicking once/minute. I pulled the nose cone (2014) and put a voltmeter on the jumper terminals, and found some interesting behavior:

• Voltage drops to ~12.4V, contactors close, coolant pumps come on
• Battery charges up to 14.5V in one minute; pumps shut off and contactors open
• Battery drops to 12.4V in another minute, and the cycle repeats

The odd thing is, I've gotten no "replace 12V" warnings, even though my battery is clearly down to maybe 1% of its nominal 33Ah. Anyone seen 12V failures without warning?

To make matters more annoying, there's no way to request mobile service to come out to replace it. "Replace 12V battery" isn't a choice in the app's service menu, so I select "something else", which apparently means "I have nothing better to do than drive 5 hours round trip to the 'local' service center" since that's all it offers...

 

[murphyS90D]

12.4 volts is a 75% charge.
12.7 volts is a 100% charge.
Anything over 12.7 volts is a charging voltage.

 

[tga]

12.4 volts is a 75% charge.
12.7 volts is a 100% charge.
Anything over 12.7 volts is a charging voltage.

No argument with your numbers, but it shouldn't be discharging from 100% to 75% in 60 seconds. If the battery was good, that would be 8Ah in 1/60th of an hour, or about a 500A draw. It should be cycling 2-3 times/day, not 720x (30/hr * 24 hrs).

 

[ShawnA]

Hi @tga ,

I have been tracking some peculiar charging behavior lately on my 2018 X and my wife's 2015 S.
It started in early December... (Both cars have new 12 volt batteries)
Turns on charger (HPWC) like you are seeing for 1 minute or less...
No kW added to car...
It repeats every 10 or fifteen minutes.

While we are trying to figure out what is happening, I unplug the cars after charging completes
to desired mileage or percentage.
It seems to stop the excessive cycling of the charging equipment.

I hope we figure it out soon...

Good luck,

Shawn

PS - Look at some of the similar threads below for more information.

 

[murphyS90D]

No argument with your numbers, but it shouldn't be discharging from 100% to 75% in 60 seconds. If the battery was good, that would be 8Ah in 1/60th of an hour, or about a 500A draw. It should be cycling 2-3 times/day, not 720x (30/hr * 24 hrs).

I went and checked my slightly over 5 year old battery. The CCA is down to 188. The CCA meter says the battery is bad. And yes from a full charge it drops down very quickly. I'm now losing 3% in 24 hours from the HVB. They must have modified the software to prop up the 12 volt battery at the expense of the HVB. I've never seen a 12 volt battery warning. I'll set up the computerized voltmeter as soon as I'm sure the car can stay parked for 24 hours to see what the charge count per 24 hours is now. When the car was new it was 4 times a day.

 

[tga]

I'll set up the computerized voltmeter

Tell me more. This sounds like a gadget I must have!

 

[Wol747]

There are several threads here and elsewhere on the constant on/off of the 12v charging.
I'm convinced it's due to the car program not allowing it to go below its nominal voltage, so constantly winding up all the system to stick just a few electrons into it. I would have expected the design to allow a reasonably substantial drop before it went into charge mode. Just waking up the charging system must draw quite a lot of energy quite apart from the constant wear on contactors etc.

 

[murphyS90D]

When I got my car in 2016 the SOC of the 12 volt battery was allowed to drop to 50% before it was recharged. That happened 4 times in a 24 hour period while the car was parked in my garage. Yesterday I waited until the car stopped charging the battery and watched the voltage drop very quickly. I connected my battery charger and charged the battery to 100%. I then waited until the voltage dropped to 12.7 volts. I then measured the CCA and the CCA tester indicated the battery was bad.

The computerized voltmeter is a Radio Shack product. It connects to a Windows computer via USB. I don't know who made it for Radio Shack.
Here is one on Amazon that looks like it is similar.

https://www.amazon.com/BTMETER-Mult...04&sprefix=multimeter+with+usb,aps,164&sr=8-5

Search for a multimeter with a USB interface.

I'll see if I can find the trace from 2016 and add it to this thread if I can find it.

 

[murphyS90D]

It seems my memory has a problem. In 2016 the 12 volt battery was being charged 3 times a day.
Here is the trace from May 7, 2016.
A reading was taken every second so 86,400 seconds is 24 hours.

I started a run this morning so tomorrow morning I will have a new trace to upload.

 

[croman]

I just had my 12v replaced. This is the second replacement for my car. The first replacement, 8 months after delivery, was preceded with a warning. I was on a road trip ( eclipse) and the warning was unnerving but roadside assistance confirmed battery would be fine until service and it was.

Latest had no warning and service alerted me that my MCU2 issues were related to a failed 12V.

 

[tga]

I have been tracking some peculiar charging behavior lately on my 2018 X and my wife's 2015 S.
It started in early December... (Both cars have new 12 volt batteries)
Turns on charger (HPWC) like you are seeing for 1 minute or less...
No kW added to car...
It repeats every 10 or fifteen minutes.

After you mentioned this, I did check AC draw in the powerwall section of the Tesla app, my Sense history, and watching the J1772 station - there's no AC power draw when this occurs, so we have something different going on. Also, I unplugged the car and no change in behavior.

The computerized voltmeter is a Radio Shack product. It connects to a Windows computer via USB. I don't know who made it for Radio Shack.
Here is one on Amazon that looks like it is similar.

https://www.amazon.com/BTMETER-Multimeter-Auto-Ranging-Resistance-Measuring/dp/B07C238GTP/ref=sr_1_5?crid=10J1KR701SO02&dchild=1&keywords=multimeter+with+usb+interface&qid=1619950904&sprefix=multimeter+with+usb,aps,164&sr=8-5

Search for a multimeter with a USB interface.

Thanks for the link. One of the Amazon reviews mentioned that it works with Sigrok if you select "digitek-dt4000zc"; I forgot to look at the Sigrok site. Lots of Sigrok-compatible DMM's listed here - Supported hardware - sigrok

It seems my memory has a problem. In 2016 the 12 volt battery was being charged 3 times a day.
Here is the trace from May 7, 2016.
A reading was taken every second so 86,400 seconds is 24 hours.

So it looks like ~2 hours of charging and ~6hrs of discharging, which is roughly what I remembered.

Latest had no warning and service alerted me that my MCU2 issues were related to a failed 12V.

I didn't want to complicate my original post, but I'm seeing the car occasionally "freak out", disable traction control, ABS, regen, and power brake and steering assist, only to return a few minutes later. I thought I might be chasing a bad ground, ABS wheel speed sensor, etc. At first, my next step was to pull the frunk liner and check/clean the ground studs, but replacing the battery is now an obvious first step.

I was a bit concerned about excess wear and tear on the HV contactors cycling a few hundred times a day, but this thread seems to imply it's not really an issue - HV Contactor Longevity I suppose I could disconnect the 12V and unplug the "fireman's disconnect" if I really wanted to stop it.

I did email mobile service and ask them to "take over" my SC appointment. I thought about ordering a Gruber 12V and DIY after watching the Gruber replacement video (doesn't seem like the nightmare everyone makes it out to be), but with $70 UPS ground shipping added in ($210 total) it's not much cheaper than paying Tesla to do it.

 

[Hookemhorns]

I replaced my 12V after 3 years. No indication of any problems, but I'm guessing if I replace it every 3 years I will buy 1 more battery over the life of the car than if I wait for it to start to fail and then I'll be in a hurry. Cheap insurance against something that needs fixing quickly when it starts to fail.

 

[murphyS90D]

The new battery scan is done.

It's charging more than it is discharging.
At the end of charge it drops directly to 12.5 volts.
That is about 83% SOC.
It is resuming the charge at 50% but the sharp drop to 50% is strange.

 

[tga]

Argh. Short answer, it's Teslafi's once-per-minute polling. Not sure what changed; I had that all working correctly for a long time.

More info:
Mobile service came out and replaced the 12V battery for me yesterday. Old one (the car's second, replaced in 2016) was a DCS-33UNCR, as is the new one. He verified the 12V battery settings were already set correctly for the DCS-33UNCR. Whereas the old battery dropped to 12.4V one minute after the HV contactors opened, the new one holds 12.7V for at least 10-15 min before I stopped checking. He was surprised that the battery would die without warning, but confirmed that my other issues (random loss/return of steering/brake assist, regen, traction control and ABS, etc) could be caused by unstable 12V power, possibly causing the gateway to reboot. We'll see what happens.

But the contactor cycling remained. Disable remote access in the car, it stops and the car sleeps. Re-enable remote access and the contactors start cycling again. Delete TeslaFi's access token, and the cycling stops and the car sleeps.

I wouldn't be surprised if the constant cycling and resulting "charge bump" to the 12V would mess with the car's ability to detect a failing 12V battery. Clearly, the battery was not in good shape if it fell to 12.4V 1 minute into a discharge phase. From @murphyS90D's 2016 chart above, it takes around 5.5hrs for his good battery to drop to 12.4V.

 

[ShawnA]

Hi All,

I would be interested in observations of this Model S 85 D 2015 with MCU2.
My Model X has similar charging patterns when they are plugged in...
It looks like too often to me.

I cannot figure out anything that may be running to pull down the 12 volt battery so quickly.
The 12 volt battery was replaced August 2020 at the same time the MCU2 was added...

@murphyS90D ?

Thank You,

Shawn

 

[murphyS90D]

All of my traces were done with the car not plugged in and the car off.
If the car is on then the DC to DC converter is on and the battery charging can't be separated from the car's need for 12 volt power. The DC to DC converter provides the 12 volts when the car is on.

 

[2101Guy]

So..was the Teslafi app working as intended? If yes, wasnt letting the car ever go to a deep sleep, no?
I keep hearing not to so great things about the various 3rd party apps..

 

[murphyS90D]

I know nothing about Teslafi. The only app I used, Visible Tesla, no longer works.

 

[ShawnA]

Hi @murphyS90D ,

This is what my X looks like sleeping. Sentry off, Cabin overheat off, Summon standby off...
It was sleeping unplugged and the traction battery still showed 80% - Same as when unplugged at
8:30 last night.
I don't think I have ever seen it below 12.24 volts but the battery monitor sends warnings this low.
The spike was when I woke it up to drive this morning.

Shawn

 

[trekkie]

If it helps troubleshoot any, I had a phantom drain and noticed the contacts clicking one night while grilling. I felt silly standing there with a stop watch but it was pretty consistent, single click, 1m45s triple click, then 2m later, repeat.

After taking my 2019 X in it was pointed at the standby power supply had an issue. initially they thought maybe the 12v battery was dying because of the constant recharging of it, but it was pointed at that instead. It's about a 4-5 hour worth of replacement, sounds like they might need to drop the battery to get to it. They said it had abnormally low output and was present in the logs. Hope that helps.