Replacing the 12V aux Battery

scotty2541 · Feb 21, 2015

Today I replaced my 12V battery in my Roadster, so here is a summary of what it took. With a life span of about 12 months, it's really annoying. It's also annoying to be standing in a parking lot trying to get into a car that cost more than my first house, and have it refuse to unlock for 2-4 minutes. All because of a cheap battery (actually, it's just that the battery is too weak to run the door latch motor).

Tesla wanted to charge about $400 to do it, and it only took a $70 battery and about an hour for a first time attempt. So this should be worth about $300 per hour to you. You're welcome :smile:

First thing is that the manual says to disconnect the charger before raising the car… I have no idea why, and I accidentally forgot to do that myself. It all went fine, but if the manual says to disconnect the charger, I'm not gonna contradict it.

Next, you need a battery. I walked into a local Kawasaki shop, and asked the parts guy for a battery for a 2011 Tesla roadster. He stared at me for a few seconds, and then said he would need a forklift. Payback for my attempt at humor.

I told him I needed a "TZ10" type battery because it had the size and Ah rating I needed. He had two, one was $65 US plus fees, taxes, etc. from "BikeMaster". The other one was $115. Since I didn't know how successful I would be, I chose the lower cost one. Both options were model BTZ10S.

As you will see, the dimensions are the most critical factor.

Here are a few shots of the battery size. The one I took out was an Interstate model FAYTX9. The depth was about 3.5 inches, the width was about 5.75 inches. The height was a little over 4 inches.

The old battery had a reading of 11.58 Volts. Which is why it wouldn't open the door.

Here is the overall view of where the battery is located. It is directly behind the right bumper, and you get to it from the right wheel well.

Here is how you get to it:
Jack up the car, and remove the right front wheel. If you can't figure this out without photos, then just give up here :smile:

At the front of the wheel well is a plastic panel, and there is a felt like cover that goes across the top toward the back of the wheel well. Almost everything is removed using a 10mm socket. There is also a small section in the felt cover that you can open by removing the 4 bolts, which allows access to the headlight. I recommend removing it, as it allows the felt to be more flexible. This is a quick shot of the panel, felt liner, and headlight cover.

There are bolts toward the bottom, and on the side of the plastic cover. Remove them, as well as the nuts that hold on the mud guard. The most difficult part here is getting the one bolt on the bottom of the guard, then peeling off the guard.

Once the bolts are removed, you can simply bend the felt cover back…

which will allow you to pull the plastic panel forward, push it left to get it off the bolts…

And the whole panel comes out.

Next, you see the battery. Get your 10mm socket, and remove the NEGATIVE CONNECTION FIRST, that is the black wire. Push the wire up, and out of the way, I wedged it in the felt cover.

You will see that the red wire is very difficult to get to while the battery is still mounted. So next get a smaller socket (I forgot… 3/8"?) and remove the bolts that hold the front bracket on. Then slide the battery out, and while holding it in your hand, you can detached the red wire. You can use the 10mm, or a Philips screwdriver.

The new battery has a small nut that goes into the battery terminal. Here is the old terminal.

Then you feed the screw through the wire connector, and screw it into the nut. Once the red wire is attached, you can reinstall the battery, and attach the bracket. In my case, it was a little tight fit. Once the bracket was put back, the battery is very snug and will not rattle around. Then attach the black wire, and you are ready to put it back together.

Reassembly is done by simply putting the plastic panel back on, and restoring the felt cover and the mud guard. In reverse order.

For reference, this is the box from the new battery, and the old battery.

The whole process took about an hour, including reopening the cover to see if I could find a bolt that was missing.

Hope this helps. And saves you some headache and money.

spaceballs · Feb 21, 2015

scotty2541 said:
... It's also annoying to be standing in a parking lot trying to get into a car that cost more than my first house, and have it refuse to unlock for 2-4 minutes. All because of a cheap battery (actually, it's just that the battery is too weak to run the door latch motor). ...

Did you open the charging port a few time to wake up the ESS, it then should have provided enough power 12v power to let you unlock the car.

scotty2541 · Feb 22, 2015

spaceballs said:
Did you open the charging port a few time to wake up the ESS, it then should have provided enough power 12v power to let you unlock the car.

Yes, had been told that by Tesla. I have tried that a few times. Sometimes the door will be able to open immediately, sometimes not. I would almost call it random (except as a computer programmer, I know it's not, I just don't know the conditions... or they may be internal conditions that I cannot perceive).

I have also found that opening the trunk sometimes wakes it up as well. Emphasis on sometimes.

The worst was standing in a parking lot, waiting over 2 minutes... And looking like an idiot trying to flap the charge port door over and over, then open and close the trunk, etc... Finally plummeting to the ground an using the key lock under the drivers door.

Really nice drawing that attention to oneself... and to the car.

spaceballs · Feb 22, 2015

scotty2541 said:
Yes, had been told that by Tesla. I have tried that a few times. Sometimes the door will be able to open immediately, sometimes not. I would almost call it random (except as a computer programmer, I know it's not, I just don't know the conditions... or they may be internal conditions that I cannot perceive). ...

I used to have that issue too where it seemed random, because I actually don't have a 12v battery ever connected I've got it modded now to work without it.
Turns out all you have to do is wait long enough for the water pump to start turn on and you should be good to go. But if you press the button to unlock the car before the car is awake the car firmware will get stuck in a confused state on the door status (computer programmer also). And it has a timeout for x-time (like 30 seconds), so if you press the door open button it will reset the timeout.. i.e. reason why it super annoying if your in a hurry, as if you keep pressing the door open button before the 30 second timeout, the car will never ever let you in! It's like punishing you for being impatient, you just have to be like Zen man.. and just wait for the car to fully wake before tapping asking it to open

(which is not really that long, I think it like 4-8 seconds)

gregd · Feb 22, 2015

Has anyone needing a new 12V battery noticed what the JVC Display claims for the 12V value (left-hand circle)?

I've owned the car for about 2 months. I noticed mine read 12v most of the time, then started dipping down over the past month, jumping between 12.0 and 11.9, 11.8, and even lower. Now the car tells me that it needs to have the 12v battery serviced (code ID 428). With the car "on", we measured the voltage at the "accessory" outlet, and it was healthy at a bit over 13v, as one would expect, so the SC thinks that 1) yes, the battery needs replacing, and 2) the radio must have a bad ground somewhere, and not a serious problem.

I'm wondering if there is a common spot where a weak ground or other loose wire might account for both symptoms? Now many 12v batteries are really dead, vs improperly charged?

Mark77a · Feb 23, 2015

ACE scotty !
Its illustrated 'how-to's' like your that make TMC such a great community.

(Just think, in time we'll be sharing ESS Battery updates

..... here's hoping we can see the Workshop manuals)

slcasner · Feb 23, 2015

One aspect of the 1.5 that I really appreciate is the lack of a 12V battery. In all the other EVs I've driven since 1998 (EV1, Ranger EV, 2003 RAV4-EV), the 12V auxiliary battery has consistently been the most likely point of failure.

scotty2541 · Feb 23, 2015

gregd said:
Has anyone needing a new 12V battery noticed what the JVC Display claims for the 12V value (left-hand circle)?

I've owned the car for about 2 months. I noticed mine read 12v most of the time, then started dipping down over the past month, jumping between 12.0 and 11.9, 11.8, and even lower. Now the car tells me that it needs to have the 12v battery serviced (code ID 428). With the car "on", we measured the voltage at the "accessory" outlet, and it was healthy at a bit over 13v, as one would expect, so the SC thinks that 1) yes, the battery needs replacing, and 2) the radio must have a bad ground somewhere, and not a serious problem.

I'm wondering if there is a common spot where a weak ground or other loose wire might account for both symptoms? Now many 12v batteries are really dead, vs improperly charged?

Actually, I think most of them are dead. Mine was at 11.5V, when a Lead-Acid should be about 13V when full. I think they are not properly charging them, like a true alternator does in a real car (or motorcycle in this case). If they are not charging it with a voltage decently above the cell level, there won't be any charging flow, and the battery life is shortened. One thing about lead acid batteries is that they are unforgiving about being deep cycled.

Others have said, and they may be correct, that the battery is just too small, and leaving the car in a parking lot, relying only on that little battery to run the radio standby, car door lock receiver, and who knows what else. So the battery becomes too deeply discharged when it's just sitting there. And again, same result, those batteries are unforgiving.

Someone suggested a LiIon battery to do the same thing. That would be more tolerant of going deeper into discharge. But they are less forgiving about being overcharged.... So either way, oh well.

-Scotty

- - - Updated - - -

Mark77a said:
ACE scotty !
Its illustrated 'how-to's' like your that make TMC such a great community.

(Just think, in time we'll be sharing ESS Battery updates ..... here's hoping we can see the Workshop manuals)

Thanks Mark.
I used the John's post to replace my Alpine, so I wanted to "pay it forward" as they say, helping the next guy who wants to take it on.

I almost didn't post it, as I spent over an hour writing it, only to lose the entire composition due to an 'auto-logout'.
I was furious.
That should be fixed.
(even my bank could write a web site that pops up a box telling me I'm about to be auto-logged off, and asks if I want to stay on)

gregd · Feb 24, 2015

scotty2541 said:
Actually, I think most of them are dead. Mine was at 11.5V, when a Lead-Acid should be about 13V when full. I think they are not properly charging them, like a true alternator does in a real car (or motorcycle in this case). If they are not charging it with a voltage decently above the cell level, there won't be any charging flow, and the battery life is shortened. One thing about lead acid batteries is that they are unforgiving about being deep cycled.

Others have said, and they may be correct, that the battery is just too small, and leaving the car in a parking lot, relying only on that little battery to run the radio standby, car door lock receiver, and who knows what else. So the battery becomes too deeply discharged when it's just sitting there. And again, same result, those batteries are unforgiving.

Someone suggested a LiIon battery to do the same thing. That would be more tolerant of going deeper into discharge. But they are less forgiving about being overcharged.... So either way, oh well.

-Scotty

- - - Updated - - -

The puzzle here is electrical, not chemical (battery state). If the car is On, and the Accessory outlet is reading 13.something volts, then the overall "12v" subsystem should be at a similar voltage. The JVC display is reading the overall "12v" subsystem voltage, so it should also read 13.something volts, not 11.something. It never reads the charge state of the battery since it's off when the car is off (i.e. during the time the little battery is engaged).

If the JVC display is significantly different than the Accessory outlet, there must be a wiring problem somehwere. Is it possible that this issue is also responsible for the apparent lack of charge in the battery, resulting in an incomplete charge, and/or an early battery failure? If so, where should we look?

The service center guy I talked to thought perhaps the battery failure might be pulling the voltage down, but our reading of the accessory outlet proved that was not the case. The car is going into the shop for the battery swap in about an hour, so we'll see what they find. I'm going to ask them to install a pair of wires to the new battery, so I can get to it electrically after the car is re-assembled.

spaceballs · Feb 24, 2015

With so many people saying they are having Model S or Roadster 12v battery issues, I'm tempted to make my own 12v LifePo4 battery with custom BMS and add in some neat Wifi tricks and sell it as a kit.

1) Prevent over discharge, over current, over voltage.
2) Be able to report back it's SOC/Volts/Amps/Temp over Wifi.
3) Disconnect itself when it hits X% SOC, Volts, Amps, or temp limits are hit, but you can wake it back up and it up (via your phones wifi).
4) Data logging it's stats i.e. Volts/Amps/Temp.
5) Able to alert you (audio pizo speaker and/or wifi) if x% SOC is reached, or calculated amp hour is below a certain amount (battery too old).
etc..

Botbldr45 · Feb 24, 2015

This thread should be bronzed, placed in a special spot and never deleted!!!!!
Good reference and great share.....................Thank You!!
I WILL use this, 'im sure.

gregd · Feb 25, 2015

spaceballs said:
With so many people saying they are having Model S or Roadster 12v battery issues, I'm tempted to make my own 12v LifePo4 battery with custom BMS and add in some neat Wifi tricks and sell it as a kit.

1) Prevent over discharge, over current, over voltage.
2) Be able to report back it's SOC/Volts/Amps/Temp over Wifi.
3) Disconnect itself when it hits X% SOC, Volts, Amps, or temp limits are hit, but you can wake it back up and it up (via your phones wifi).
4) Data logging it's stats i.e. Volts/Amps/Temp.
5) Able to alert you (audio pizo speaker and/or wifi) if x% SOC is reached, or calculated amp hour is below a certain amount (battery too old).
etc..

I was thinking along the same lines, though not quite so fancy. I figured I at least needed access to the 12v system for jump starting the car if it goes flat, or perhaps to add aome capacity (parallel another SLA battery). So when the service center replaced mine earlier this week I brought some wire and connectors, and asked that they bring it up to where the charge fuse is, under the hood. They did a very nice job (Thanks, Rocklin!); see below. The connector is an Anderson "Power Pole", often used in Ham radio and emergency services these days; sort of a standard for powering things. The black rectangle above and to the right is the existing 12v battery charging fuse. The new wire snakes along the route of that lead, down to the battery, where the leads are mashed into the connectors along with the main cables under the screws.

Now that it's accessible, I've done some voltage measurements on the new battery. With the car On, and letting it stabilize for about 5 minutes, the voltage reads 13.72v. This is with the key in the "on" position, in Park, and with the radio on (low volume). Driver's door closed, and the car NOT plugged in, ready to drive off. I turned the "ignition" off, removed the key, and saw the voltage drop to 13.69. The pump was still running at that point. Let it sit for a few minutes after the pump shut down, and the reading is still a healthy 13.68v, so that presumably is the 100% SOC for my particular battery. The JVC display, by the way, read "12.2v" while the car was on (up from 11.7 before the battery replacement), so the mysteries of why there is a difference between the measured battery voltage and the JVC display, and the JVC display before and after the replacement, remain. Both sets should be the same, in my opinion, given good wiring and a properly calibrated JVC unit (which is likely a part of the problem!).

Greg.

Doug_G · Feb 25, 2015

Most people don't realize that "12V" lead-acid batteries are actually almost dead when their open circuit voltage is 12V. A fully charged normal battery has an open circuit voltage of about 13.8V.

scotty2541 · Mar 5, 2015

gregd said:
The puzzle here is electrical, not chemical (battery state). If the car is On, and the Accessory outlet is reading 13.something volts, then the overall "12v" subsystem should be at a similar voltage. The JVC display is reading the overall "12v" subsystem voltage, so it should also read 13.something volts, not 11.something. It never reads the charge state of the battery since it's off when the car is off (i.e. during the time the little battery is engaged).

If the JVC display is significantly different than the Accessory outlet, there must be a wiring problem somehwere. Is it possible that this issue is also responsible for the apparent lack of charge in the battery, resulting in an incomplete charge, and/or an early battery failure? If so, where should we look?

The service center guy I talked to thought perhaps the battery failure might be pulling the voltage down, but our reading of the accessory outlet proved that was not the case. The car is going into the shop for the battery swap in about an hour, so we'll see what they find. I'm going to ask them to install a pair of wires to the new battery, so I can get to it electrically after the car is re-assembled.

I don't know what the JVC display actually shows. The mains running the charging voltage when the car is on, or the aux battery when the rest of the system is off...? Can the JVC display even be on when the car 12V main from the system is off?
And I am assuming this is in service mode, since I have never seen a voltage measurement on that screen.

Well, the point here is that the main system (ESS? Whatever that stands for) keeps the voltage at 13.x volts, which also changes the battery. So measuring the voltage anywhere will give you what the main (charging) voltage is. Turn it all off, and the battery is suddenly running on it's own, and if it is near dead, it will likely drop from 13V, to or below 12V rather quickly (minutes or even seconds). The battery in good condition should be over 13V when it's on its own, as Doug_S points out.

Turn the car on, the system knows the battery is low, and hits you with that fault message. But it goes away, because once the main power is back on and running at 13V, there is no way to see that the battery is low (the two systems are now connected, as it were).

At least, that's the way I see it. I didn't measure the voltage on the wires when the aux battery wasn't present, just the battery itself. I should have, as well as planned for better pictures before I started. But I didn't know how successful I would be in this endeavor.

-Scott

- - - Updated - - -

Botbldr45 said:
This thread should be bronzed, placed in a special spot and never deleted!!!!!
Good reference and great share.....................Thank You!!
I WILL use this, 'im sure.

Thanks. For inflating my ego.

- - - Updated - - -

gregd said:
I was thinking along the same lines, though not quite so fancy. I figured I at least needed access to the 12v system for jump starting the car if it goes flat, or perhaps to add aome capacity (parallel another SLA battery). So when the service center replaced mine earlier this week I brought some wire and connectors, and asked that they bring it up to where the charge fuse is, under the hood. They did a very nice job (Thanks, Rocklin!); see below. The connector is an Anderson "Power Pole", often used in Ham radio and emergency services these days; sort of a standard for powering things. The black rectangle above and to the right is the existing 12v battery charging fuse. The new wire snakes along the route of that lead, down to the battery, where the leads are mashed into the connectors along with the main cables under the screws.

Now that it's accessible, I've done some voltage measurements on the new battery. With the car On, and letting it stabilize for about 5 minutes, the voltage reads 13.72v. This is with the key in the "on" position, in Park, and with the radio on (low volume). Driver's door closed, and the car NOT plugged in, ready to drive off. I turned the "ignition" off, removed the key, and saw the voltage drop to 13.69. The pump was still running at that point. Let it sit for a few minutes after the pump shut down, and the reading is still a healthy 13.68v, so that presumably is the 100% SOC for my particular battery. The JVC display, by the way, read "12.2v" while the car was on (up from 11.7 before the battery replacement), so the mysteries of why there is a difference between the measured battery voltage and the JVC display, and the JVC display before and after the replacement, remain. Both sets should be the same, in my opinion, given good wiring and a properly calibrated JVC unit (which is likely a part of the problem!).

Greg.

One open question... Is there a shunt to measure current, and protect the system. Logically there should be. And with a battery that can deliver over 100CCA, the voltage drop across the shunt it could easily exceed a volt depending. That could be the difference between the test points and the JVC. And whether it's higher or lower would depend on the current direction.

More speculation on my part

I'd love to have a schematic on it.

-Scott

markwj · Mar 5, 2015

spaceballs said:
I used to have that issue too where it seemed random, because I actually don't have a 12v battery ever connected I've got it modded now to work without it.
Turns out all you have to do is wait long enough for the water pump to start turn on and you should be good to go. But if you press the button to unlock the car before the car is awake the car firmware will get stuck in a confused state on the door status (computer programmer also). And it has a timeout for x-time (like 30 seconds), so if you press the door open button it will reset the timeout.. i.e. reason why it super annoying if your in a hurry, as if you keep pressing the door open button before the 30 second timeout, the car will never ever let you in! It's like punishing you for being impatient, you just have to be like Zen man.. and just wait for the car to fully wake before tapping asking it to open (which is not really that long, I think it like 4-8 seconds)

Yep. You can also do this by just tapping the door handle as well. Trick is to wait for the car to awake and that little pump in the rear right corner to start trickling. st that point, 12V is coming from the main ESS via it's converter.

markwj · Mar 5, 2015

One other comment is to be aware that in the 2.x cars there are two separate 12V power sources (as well as a few switched variants).

One comes from the 12V battery in the front right wheel arch. This provides emergency power in the event the main ESS fails. Things like brake lights, hazard flashers, doors, etc.
The other comes from the 12V DC converter in the main ESS in the trunk/boot. This provides the power for most everything else (including radio, etc).

Some systems can operate from either source.

And a final note on the power you see on the AUX power socket: This comes when the car is awake (typically driving, charging, cooling, heating, etc) from the 12V converter. You can tell when the car is awake because the little coolant pump in the rear right corner of the PEM will trickle coolant continually. Normally, if you lock the car and walk away, after 5 minutes or so the AUX power will be cut. But if the car is too hot/cold, that won't necessarily happen and AUX power can stay on.

spaceballs · Mar 5, 2015

markwj said:
Yep. You can also do this by just tapping the door handle as well. Trick is to wait for the car to awake and that little pump in the rear right corner to start trickling. st that point, 12V is coming from the main ESS via it's converter.

btw there is another deep sleep state where the car can go into where it will not wake even if you tap the door handle. At that point only way to wake the car is opening and the charge door.

Dave EV · Mar 5, 2015

Doug_G said:
Most people don't realize that "12V" lead-acid batteries are actually almost dead when their open circuit voltage is 12V. A fully charged normal battery has an open circuit voltage of about 13.8V.

That's quite high.

At no-load a fully charged 12V lead-acid is around 12.7V. Put any load on the battery and it will be less, so you can only measure this with the battery disconnected from the car.

The float voltage (the voltage you'd hold a 12V battery at to keep it fully charged) is around 13.5-13.7V.

markwj · Mar 5, 2015

spaceballs said:
btw there is another deep sleep state where the car can go into where it will not wake even if you tap the door handle. At that point only way to wake the car is opening and the charge door.

It is like a reset for the VMS.

spaceballs · Mar 5, 2015

markwj said:
It is like a reset for the VMS.

I believe the Linux BusyBox isn't actually reset, as that takes a lot longer to boot up vs just opening the charge port.

Replacing the 12V aux Battery

Member

Member

Member

Member

Active Member

Member

Active Member

Member

Active Member

Member

Member

Active Member

Attachments

Lead Moderator

Member

Asia Pacific

Asia Pacific

Member

Active Member

Asia Pacific

Member

Similar threads