Active Cooling Cycles (While Charging) MYSTERY?!

[DeedWest]

Hey everyone. I currently have two 2.5 Sports under my roof - VIN 1414 and VIN 1462. I've noticed something rather spooky recently when studying the active cooling cycles while both cars are charging. I was hoping some of you guys could weigh in.

THE ISSUE - 1414's cooling cycles (while charging) are nearly 2.5X LONGER than 1462's.

As you know, the Roadster performs a series of active cooling cycles when charging on 240V (when the very loud HVAC fans in the front of the car start screaming). This is a very important aspect, as it brings the battery temperature down into healthier ranges after a long day of driving, or time spent in higher ambient temperatures. For me, here in Texas, this is one of the main reasons I keep my Roadsters plugged in all the time when not in use. The Roadster seems to perform these periodically while charging. The first one always starts about 30 seconds after a charging session is initiated (unless charging in Performance Mode, but who does that anyway?). I have not timed the distance between cooling cycles, as those seem to greatly vary based on ambient temps, initial temps of battery when initiating charge, distance to full in Standard Mode, etc.

I've had VIN 1414 for nearly a year now, and have never thought twice about the cooling times. When I get in my car (after a night of charging), even in the summertime, the battery is healthy at around 26C to 29C - even after sitting for an entire day in a hot garage. However, when I purchased 1462, I noticed that this car in particular doesn't seem to have even HALF of the duration of 1414's cooling cycles during charging. I'm stumped. So, I started to test them both. I tested both in the same day, in relatively the same ambient temperatures, same SOC (initiating charge at about 50% SOC) and same amperage (240V @ 32A) using my Foundry Mobile Connector.

VIN 1414 STATS -
Build Date - August 2011
Current Firmware - 4.7.2
Version - 2011 2.5 Sport
Battery CAC - 129.77
Battery - Remanufactured ESS (from Tesla)
Mileage - 6,363
Charging Rate (Amps) - 32A
Ambient Garage Temp - 32C

Start of Charge SOC - 48%
Start of Charge Battery Temp (C) - 36C
Temp After 5 Cooling Cycles - 25C

Cooling Cycles Log for VIN 1414 -
Cycle 1 - 3 minutes 28 seconds
Cycle 2 - 2 minutes 49 seconds
Cycle 3 - 2 minutes 37 seconds
Cycle 4 - 1 minute 57 seconds
Cycle 5 - 1 minute 38 seconds


VIN 1462 STATS -
Build Date - November 2011
Current Firmware - 4.7.2
Version - 2011 2.5 Sport
Battery CAC - 151.18
Battery - Original Factory ESS
Mileage - 40,154
Charging Rate (Amps) - 32A
Ambient Garage Temp - 32C

Start of Charge SOC - 52%
Start of Charge Battery Temp (C) - 37C
Temp After 5 Cooling Cycles - 32C

Cooling Cycles Log for VIN 1462 -
Cycle 1 - 1 minute 28 seconds
Cycle 2 - 1 minute 18 seconds
Cycle 3 - 1 minute 24 seconds
Cycle 4 - 1 minute 9 seconds
Cycle 5 - 1 minute 12 seconds

I only did five cycles per car to measure the duration of each one. It's obvious that 1414 is cooling for considerably longer at first, until the battery gets healthier, then it seems to taper off in duration. However, for 1462, the cycles are completely predictable - regardless of temperature.

The cars have identical firmware and every number in the Firmware screen of the Diagnostic Menu matches (apart from the IMEI number). The only thing that makes sense at this point, is that 1462 has an original ESS, while 1414 has a Remanufactured ESS from Tesla (installed in December 2018). Do ESS have their own internal firmware that could be controlling the cooling cycle durations?

Any insight would be greatly appreciated!

- T

 

[gregd]

So, two wild guesses.

One is that the HVAC systems on the two cars have different efficiencies, with 1414's perhaps being low in refrigerant, or maybe the condenser fins are clogged.

Other is that the battery on 1414 has a significantly lower CAC, so it's internal resistance is higher, which means it heats up faster than the other car.

Third guess is just an observation, that each of our cars has a unique personality. So it may just be that way.

 

[DeedWest]

So, two wild guesses.

One is that the HVAC systems on the two cars have different efficiencies, with 1414's perhaps being low in refrigerant, or maybe the condenser fins are clogged.

Other is that the battery on 1414 has a significantly lower CAC, so it's internal resistance is higher, which means it heats up faster than the other car.

Third guess is just an observation, that each of our cars has a unique personality. So it may just be that way.

Thank you for your response! I was thinking it could have something to do with the HVAC as well. 1414 recently had its HVAC system replaced and refrigerant recharged. Could that be why 1414 is allowing itself to cool the battery for longer durations? Could it be that 1462 deserves an HVAC servicing?

As for the CAC, while I do agree in theory it should have increased resistance, I have not noticed any differences in how quickly either car's ESS temps increase (and I am very OCD about keeping track of temperatures). As I noted above, both charge cycles were started around the same temperature. Now, I could see there being something in the firmware that tells the car when the CAC is lower over time, to increase cooling cycles while charging.

The interesting part to me, is that with just logging five cooling cycles on each car, 1414's temperatures decreased by 11C - and 1462's decreased by just 5C. That's a huge difference.

I do also agree with the "different personalities" part. When I used to have my 2.0 VIN 523, there were random times when the car would active cool for 4 minutes straight when charging.

 

[markwj]

I too see very different cooling cycle behaviour in different roadsters (I looked into this in some detail back in the day of working on COOLDOWN for OVMS). Even triggering a cycle in software is more voodoo than science. A few guesses/comments:

1. I suspect that the whole reason for the 'cycles' thing is to avoid rapid cooling, and the expansion/contraction that would cause within the PEM. Another reason is to avoid condensation within the ESS. Presumably Tesla considers these worse than the high temperature thing.
   
   
2. What concerns us most about the temperature is the continued operation of that little coolant pump, and the inability of the car to sleep with a hot battery (not good for long term wear). However, I don't think the Tesla engineers prioritise this at all.
   
   
3. Cooling cycles appear to be purely time based, and do not account for variances in efficiency of the HVAC system.
   
   
4. While we look at ESS temperature as one number, the car has multiple (dozens) of temperature sensors in the ESS. It knows min and max temperatures, which sheets (and parts of sheets) are hot, temperatures of the coolant going in and out, and a host of other things (such as humidity, dewpoint, etc).

In general, I don't think our 'simplistic' approach of looking at that one temperature value matches the algorithms in the car at all.

 

[im4uttx]

I agree with @markwj ...The full algorithm for cycling is much more complicated than one or two numbers we see to try to correlate between two separate roadster systems. All interesting data points more likely proving how little is really known about the software control system, and least from we "owners" ...

 

[ML Auto]

Check the A/C pressures and temps against each other to see if it is the A/C system or not.

 

[DeedWest]

Hilariously enough, I think my car KNEW I was talking about it. After a drive last night, 1462's first active cooling cycle went for a whopping TWO MINUTES AND 28 SECONDS.

This officially makes zero sense. But, great idea @MLAUTO - I'll take a photo of the Diagnostic Menu of both cars while active cooling and repost.

 

[hmbprius]

HI newbie follow-on question, I only had the front fans turn on once so far (and didn't notice how long) but how long should the back fan run? there are 2 noises, the little yellow fluid circulating that turns on when ever anything turns on in the car (from doors opening to starting to charge...I think of it as the "hello, welcome back mom, are we going for another fun ride" greeting from my blue baby...to another back noise that is more fan like (not the front ones) that comes on when charging at 240 sometimes and sometimes when charging at 120 for most of the time charging. Can someone reassure me that is normal?

 

[X.l.r.8]

I call it the take off sequence. Put it in range charge and it’s automatic always. Its normal charge it’s intermittent

 

[DeedWest]

HI newbie follow-on question, I only had the front fans turn on once so far (and didn't notice how long) but how long should the back fan run? there are 2 noises, the little yellow fluid circulating that turns on when ever anything turns on in the car (from doors opening to starting to charge...I think of it as the "hello, welcome back mom, are we going for another fun ride" greeting from my blue baby...to another back noise that is more fan like (not the front ones) that comes on when charging at 240 sometimes and sometimes when charging at 120 for most of the time charging. Can someone reassure me that is normal?

The Coolant is always running when the car is woken up - caused by opening/closing trunk, doors, touching the VDS, etc.
It is also always running if the battery temp is above 31C.

As for the rear fans - those are the PEM fans and are completely normal while charging, as well as for a short amount of time after a drive while they're cooling the PEM down.

 

[hmbprius]

Thanks !