Vampire drain - I'm seeing a pattern...

[FraserC]

I have read a lot of posts on VD, and I too am seeing up to 5kWh per day loss from my Gen. 2 R1T. But today I happened to notice that there has been a pattern to the exact amounts of loss per sleep cycle - despite total amount of loss per cycle the losses are always multiples of 0.22 kWh (though this number is rounded up from 0.217). See here for an example:


I have not seen anyone post about this, and I have been speculating on where these losses are going and finding no answers. But finding this (data above from yesteray) I went back and found this pattern to last month, and unless the behavior cycle was interrupted by driving or charging, the precise loss increment of 0.217 kWh is followed. What do we think of this? The VD losses are being managed with good precision, so logically there must be just one highly consistent drain source... I'm a mech eng so no good at electrical systems analysis, any sparky's out there with time and curiosity to investigate?

Sponsored

 

[Doc]

Don't know if this helps explain things or not. This came from the [Banned Site]...so giving credit where credit is due.

Yes my friends it can be done, or at least the effects of VD significantly reduced:



TLDR: The root problem is that the 12v battery is vastly undersized, in combination with the Gen1 electrical design requiring the Rivian's computers to be fired up to run the High Voltage -> 12v recharger, and those computers then eat around 300w of power. This is why you see 1-2% SOC loss per day when your R1 is unplugged, it's the 12v discharge/recharge cycle that's killing it.

Nifty graph from Electrafi merged with DC datalogger output showing how the R1's puny 12v battery cycles over 24 hours:

Several ways to fix it then described in the video above:

• Hook up a 12v battery tender, and boom your R1 will basically never wake up (without otherwise being perturbed)
• If you're on the move off the grid, a battery tender + supplemental 'camp' battery does the same.
• If you're off the grid but mostly stationary, add solar to feed the camp battery system indefinitely.
• Also, vampire drain is reduced by approximately 1/3rd by upgrading to an Ohmmu 12v LFP battery, delivering more time between charge cycles and also faster recharging than the OEM lead-acid AGM 12v battery.

And then lastly, beyond vampire drain, how can you help keep your R1 asleep when off the grid?
Turn off the AC, turn off proximity locking, put it into camp mode (power 'off' of course), and then be mindful about opening and closing doors. Opening any door wakes a Rivian up, as does closing (most) of them, and then it takes 5-10 minutes generally to fall back asleep.
LEAVING doors or windows open is just fine, it's the state change from closed to open or vise versa that does the damage.

 

[FraserC]

Don't know if this helps explain things or not.

Yes my friends it can be done, or at least the effects of VD significantly reduced:


TLDR: The root problem is that the 12v battery is vastly undersized, in combination with the Gen1 electrical design requiring the Rivian's computers to be fired up to run the High Voltage -> 12v recharger, and those computers then eat around 300w of power. This is why you see 1-2% SOC loss per day when your R1 is unplugged, it's the 12v discharge/recharge cycle that's killing it.

Nifty graph from Electrafi merged with DC datalogger output showing how the R1's puny 12v battery cycles over 24 hours:

Several ways to fix it then described in the video above:

• Hook up a 12v battery tender, and boom your R1 will basically never wake up (without otherwise being perturbed)
• If you're on the move off the grid, a battery tender + supplemental 'camp' battery does the same.
• If you're off the grid but mostly stationary, add solar to feed the camp battery system indefinitely.
• Also, vampire drain is reduced by approximately 1/3rd by upgrading to an Ohmmu 12v LFP battery, delivering more time between charge cycles and also faster recharging than the OEM lead-acid AGM 12v battery.

And then lastly, beyond vampire drain, how can you help keep your R1 asleep when off the grid?
Turn off the AC, turn off proximity locking, put it into camp mode (power 'off' of course), and then be mindful about opening and closing doors. Opening any door wakes a Rivian up, as does closing (most) of them, and then it takes 5-10 minutes generally to fall back asleep.
LEAVING doors or windows open is just fine, it's the state change from closed to open or vise versa that does the damage.

I've been doing the 12V charging on a regular cycle, and sometimes this gives me a few days of zero VD loss, but then the cycle continues. This said, I can't see a logical connection between 12V charge demand and very precise power drain from the big battery, can you? And the magnitude of VD loss (up to 5kWh) cannot possibly be used to keep that small 12V battery supplied, even with computer activity. Sometimes I get a total of 25kWh loss in a week, you think this is all attributable to 12V battery losses? (serious question!), as well as the perfectly proportioned demand increments - per my post?

I'm looking to understand why the Rivian operating system is seemingly deliberately sending a lot of power somewhere, so precise that it has to be by design...

 

[Yossarian]

Thanks for the info @Doc, very helpful stuff. I have a couple of additional questions about VD mitigation that I hope you, or others more knowledgeable than I am, may be able to help with.

Regarding the use of a 12v battery charger, what capacity is needed? Is something like a 1-1/4 amp Battery Tender sufficient, or do you need something more robust like a 5 amp or larger charger?
When attaching the charger, should it be directly to the 12v battery itself, or can it be through an alternative like the 7-pin towing receptacle or the OBD port (assuming the Rivian has one)?

With respect to solar charging, does the controller need to be one that regulates or matches current flow to the battery (PWM, MPPT), or is a simple shunt a better choice? If the former, would either PWM or MPPT be OK, or is one better (or worse) than the other?

Regarding an Ohmmu (or other) lithium battery, does using it as a replacement for the OEM battery raise any warranty issues, as in possible voiding?

Finally, love your avatar. We have two Welshies, so are also very familiar with sharing space with spaniels.

 

[FraserC]

Guys, there are TONS of old threads on 12V battery drain... I was hoping that this thread might not get buried in 'same old' repeats rather than yielding 'new ideas'! I believe this is called hijacking? ?

I'm looking for reasons the VD has a pattern, which looks too incremental for random 12V charge cycles.... Unless of course someone can suggest why that could logically be the cause! ?

 

[mkhuffman]

Guys, there are TONS of old threads on 12V battery drain... I was hoping that this thread might not get buried in 'same old' repeats rather than yielding 'new ideas'! I believe this is called hijacking? ?

I'm looking for reasons the VD has a pattern, which looks too incremental for random 12V charge cycles.... Unless of course someone can suggest why that could logically be the cause! ?

Why do you think the 12v charge cycles are random?

It has been established in this thread (and the ultra long vampire drain thread) that maintaining the LVB with the HVB is the root cause of most (if not all) of the vampire drain people have experienced. It seems to me a slowly draining LVB due to a consistent load will result in a consistent response from the HVB system, resulting in what you are measuring.

 

[fxstein]

I have read a lot of posts on VD, and I too am seeing up to 5kWh per day loss from my Gen. 2 R1T. But today I happened to notice that there has been a pattern to the exact amounts of loss per sleep cycle - despite total amount of loss per cycle the losses are always multiples of 0.22 kWh (though this number is rounded up from 0.217). See here for an example:


I have not seen anyone post about this, and I have been speculating on where these losses are going and finding no answers. But finding this (data above from yesteray) I went back and found this pattern to last month, and unless the behavior cycle was interrupted by driving or charging, the precise loss increment of 0.217 kWh is followed. What do we think of this? The VD losses are being managed with good precision, so logically there must be just one highly consistent drain source... I'm a mech eng so no good at electrical systems analysis, any sparky's out there with time and curiosity to investigate?

Regarding the steps you are seeing: this is a simple math/precision thing that is happening to any app or service that consumes the Rivian API.

The battery SoC sensor that Rivian exposes has a resolution of 0.1% or 1/1000th. If you multiply this with the size of the capacity of the battery pack you get the approximate step size.
For example if the SoC is at exactly 69% the API actually reports 69.0%, the next update when discharging will be 68.9% then 68.8%. Every 0.1% fraction is the lowest resolution. It used to be fairly constant in capacity until 47.01 and 51.0x started modifying the capacity rating based on currently unverified/unknown factors (only theories exist). But if you multiply out the current battery capacity and SoC you get approx kWh of stored energy and you can track that. The increments will always be a step or multiple of 0.1% SoC times the capacity of the battery pack.

 

[FraserC]

I have read a lot of posts on VD, and I too am seeing up to 5kWh per day loss from my Gen. 2 R1T. But today I happened to notice that there has been a pattern to the exact amounts of loss per sleep cycle - despite total amount of loss per cycle the losses are always multiples of 0.22 kWh (though this number is rounded up from 0.217). See here for an example:


I have not seen anyone post about this, and I have been speculating on where these losses are going and finding no answers. But finding this (data above from yesteray) I went back and found this pattern to last month, and unless the behavior cycle was interrupted by driving or charging, the precise loss increment of 0.217 kWh is followed. What do we think of this? The VD losses are being managed with good precision, so logically there must be just one highly consistent drain source... I'm a mech eng so no good at electrical systems analysis, any sparky's out there with time and curiosity to investigate?

Guys, I apologize - I previewed my post and the image I inserted had been attached, but then it vanished!! The meaning of my post was all but blown away by the lack of the image... Here goes with an an attempt to restore:

 

[FraserC]

Doc, Yossarian, sorry for rude response before noticing my thread was buggered!

fxstein, mkhuffman; Now my thread contains the all important image showing precise increments of loss, what do you think about the drain rates? Do you think that these extremely precise loss values could be 12V charge cycles? Varying by up to 10:1 amounts from hour to hour? Why the huge variability, but precise incremental, highly controlled charge cycles being demanded for 12V charging?

 

[fxstein]

It's most likely multiples of 0.1% of your batteries capacity... (see my explanation above)

 

[FraserC]

Hmmm, it is a percentage, 0.159% in fact.. Not a logical way to account for... Maybe it's in increments of 5274 Joules? That's at least a round number! Just kidding of course! You have given me a rational reason for some way that Rivian and/or ElectraFi choose to account for power use increments, thanks for that fxstein!!

 

[fxstein]

… or what ElectraFi thinks the capacity is. They do their own math, but also only get 0.1% increments of SoC from Rivian.

 

[mkhuffman]

Doc, Yossarian, sorry for rude response before noticing my thread was buggered!

fxstein, mkhuffman; Now my thread contains the all important image showing precise increments of loss, what do you think about the drain rates? Do you think that these extremely precise loss values could be 12V charge cycles? Varying by up to 10:1 amounts from hour to hour? Why the huge variability, but precise incremental, highly controlled charge cycles being demanded for 12V charging?

… or what ElectraFi thinks the capacity is. They do their own math, but also only get 0.1% increments of SoC from Rivian.

I guess you are using data from ElectraFi?

The BMS estimates the Entergy to Empty (EtE), and it is only an estimate. It is not nearly as accurate as a gas tank with a floater that measures how much fuel is remaining. So it is not surprising to me that the BMS reports increments that are mathematically consistent. It is an estimate, after all.

And the BMS is constantly evaluating the voltage of the pack to determine if a specific voltage is equivalent to a specific SoC. This is why it is important to charge to 100% and discharge to 10% (maybe lower if you feel lucky) every once and a while - so the BMS can calibrate and determine what 100% SoC is equivalent to in HVB voltage. It is very complex.

I expect a few 10% to 100% to 10% to 100% cycles will change what you are seeing. IMO it is just the way Rivian designed the BMS. Nothing more than that.

 

[mikehmb]

Don't know if this helps explain things or not. This came from the [Banned Site]...so giving credit where credit is due.

Yes my friends it can be done, or at least the effects of VD significantly reduced:



TLDR: The root problem is that the 12v battery is vastly undersized, in combination with the Gen1 electrical design requiring the Rivian's computers to be fired up to run the High Voltage -> 12v recharger, and those computers then eat around 300w of power. This is why you see 1-2% SOC loss per day when your R1 is unplugged, it's the 12v discharge/recharge cycle that's killing it.

Nifty graph from Electrafi merged with DC datalogger output showing how the R1's puny 12v battery cycles over 24 hours:

Several ways to fix it then described in the video above:

• Hook up a 12v battery tender, and boom your R1 will basically never wake up (without otherwise being perturbed)
• If you're on the move off the grid, a battery tender + supplemental 'camp' battery does the same.
• If you're off the grid but mostly stationary, add solar to feed the camp battery system indefinitely.
• Also, vampire drain is reduced by approximately 1/3rd by upgrading to an Ohmmu 12v LFP battery, delivering more time between charge cycles and also faster recharging than the OEM lead-acid AGM 12v battery.

And then lastly, beyond vampire drain, how can you help keep your R1 asleep when off the grid?
Turn off the AC, turn off proximity locking, put it into camp mode (power 'off' of course), and then be mindful about opening and closing doors. Opening any door wakes a Rivian up, as does closing (most) of them, and then it takes 5-10 minutes generally to fall back asleep.
LEAVING doors or windows open is just fine, it's the state change from closed to open or vise versa that does the damage.

Have you been able to determine the amount of draw on a battery tender to measure effective losses without having to wake up the HV system?

I'd love to see a SBS (12V tended; 12V untended and HV takes up the slack) but no longer have a tender to check with.

 

[TerryInHouston]

FYI - DRAIN - I was in Europe for a month on vacation - watching my truck online thru the trip. 150 miles of charged went to almost zero. In the middle I turned off my security video and notice that did slow the drain a bit.....

Now - surprisingly - soon as I got back I got a call from Rivian wanted to replace my 12 V...

Coincidence?

Sponsored