Pause in the charging?

[arnolddeleon]

I haven't studied it closely yet but it looks like sometimes there is pause in the power draw when charging. I have a power monitor that logs the amount of power being drawn every minute. Here is an example of the pause. I chose this one because it also shows the draw later while the car is just sitting there:

http://flickr.com/gp/arnolddeleon/j894K3

arnold

 

[Dsinned]

This has been well documented here (mostly by me) as I have observed the same thing many times myself. From your posted graph, I can safely say you "scheduled" your charge completion time for 9:00am even though it was still doing a (low level) charge for ~10 minutes longer. This is EXACTLY what happens during all my scheduled charges; 10 minutes after the end time is always what happens.

This pause period followed by a brief low level charge of nearly 30 minutes is believed to be some sort of "cell balancing" operation. I do not know if that is accurate. If you just leave your car plugged in indefinitely, the same thing will happen again each day at the same time, but why?

Unfortunately, Toyota has not seen fit to shed any light on this strange behavior since the car's production release, nor is any of this documented in any information about the car that I have ever read. That in itself is rather peculiar and disppointing to me as an owner. Sadly, after all this time, NOBODY at Toyota (or Tesla) is letting on that they know anything about this apparent time gap (dead time) in the scheduled charging operation of the RAV4 EV. However, I can assure you this anomoly has been brought to the attention of their dealers and Toyota corporate customer support, by many different owners, from almost the very beginning of this car's lifecycle. This is definitely NO SECRET!

Furthermore, Toyota's failure to resolve this issue, not to mention their apparent refusal to even acknowledge its existence, has been most displeasing to many of their customers.

 

[arnolddeleon]

Sorry I wasn't clear enough, there are actually two pauses. I had read about the "early end" of charging followed by the low power "charging" or "balancing". The other pause I was curious about is the one that happened roughly around 4:40 am in the example I posted. It doesn't happen every time. It also doesn't happen at the same amount of from the beginning of the charge. Does anyone else see it or it is being caused by some external event?

arnold

 

[Dsinned]

That appears to be a "glitch" in your monitoring system, or possibly a sample point taken at the exact same time as some sort of power line anomoly such as a very brief AC surge created by something else turning on or off in your house, or perhaps a spurious event within your power line monitoring system itself. I see these same kind of random "glitches" from time to time as well. I don't think they are related to the RAV4 EV or its charging system.

 

[arnolddeleon]

It could be a data collection problem however I didn't see a random dip with my 2002 RAV4-EV that used the exact same circuit with the exact same circuit monitor. The original RAV4-EV had a well known pause it its charging cycle in the first few minutes of charging. That's the only dip that shows up in my logs when it was using the circuit. None of the LEAF charging sessions I looked at show any dips either (although I didn't do an exhaustive search). My other suspicion is the car/charger is responding to a grid event.

arnold

 

[miimura]

Along the lines of glitches, I'm wondering if these energy monitoring systems are integrating some higher frequency data collection to report an energy amount, or just reporting a single power reading that is supposed to be representative of the time period.

 

[Dsinned]

The OP stated the graph represents a AC power point sample taken every minute. That is the same for my home energy monitoring system's graphically displayed results. However, in my system, the actually sampling rate is about every 6 to 10 seconds, and I have a multiple channel monitoring system. So, there are multiple sensors transmitting data samples asynchronously at the same frequency. When the receiver sees a collision from two or more sensors transmitting at the same time, this could result in a dropout (null entry) in their respective data bases. This in turn may appear in the graph of one particular sensor's output as a "glitch".

Just my two cents here . . .

 

[TonyWilliams]

When the cells are charged, or discharged, one cell will reach the target voltage threshold somewhere in the pack.

If charging, the charge will stop, and the peak cells will be discharged to somewhat match the lowest ones. Then, the pack needs another charge to bring all the cells back up to the target voltage, and then the process may do it again.

 

[Ampster]

When the cells are charged, or discharged, one cell will reach the target voltage threshold somewhere in the pack.

If charging, the charge will stop, and the peak cells will be discharged to somewhat match the lowest ones. Then, the pack needs another charge to bring all the cells back up to the target voltage, and then the process may do it again.

To clarify in my mind, when you say cells I presume you mean the first group of paralleled cells (brick) in the pack.

I don't know the details of the Tesla Battery management system but it sounds like it may use a combination of passive and active cell balancing.

Passive cell balancing is when the cells that are high are bled down to the target voltage of the other cells. Active cell balancing may take place during a low amperage charge phase when the high cells are shunted by a resistor or other mechanism and the other cells continue to receive a small charge to bring them up to the target voltage. This could explain the lower amperage draw during the final phase of charging. A possible explanation of the voltage glitch might be that the charger is turned off momentarily so that the battery management system can take a "no load" measurement of cell voltage to decide which cells need to be balanced.
The above theory is only hypothetical based on battery management systems in the DIY marketplace. The details of the Tesla system are proprietary and I don't know for a fact if they use the above methods of balancing.

 

[Dsinned]

Ampster, from what I have seen the numerous times I have monitored my RAV4 EV while doing scheduled charges, cell balancing seems like the "active" type.

But, I seldom see any power dropouts in the charge being applied, other than the long pause period - sometimes hours long - just before the final phase of low power charging. This latent charge always appears to be at a constant power level of ~700W (~240Vac @ 3A) for just under 30 minutes. Power fluctuations are almost insignificant; at least not enough to be easily detectable in my home energy monitoring system at one minute graphic intervals. And, this same pattern reoccurs the next day exactly the same as the day before ( i.e. at a constant low power level of ~700W for slightly less than 30 minutes) while my car is left continuously plugged in, and totally undisturbed, for two (or more) consecutive 24 hour timer scheduled charging cycles.