Let the hacking begin... (Model S parts on the bench)

[kennybobby]

...Well that is certainly an interesting point, do the 60's use more of the voltage range, or have I misinterpreted the data (I worry more about the latter as this stuff is quite a way from my area of expertise.)

Do you have any 100% charged data yet? It looks like from your data we have seen

79% ~ 4.09 V
89% ~ 4.179 V

lots of M85 data but not so much M60, lola's sc data put 93% at the charging CV knee with about 24 cells >= 4.19 V. Your cells look evenly balanced.

 

[llavalle]

Ok, this MS has the center console installed, but the cubby/shelf is gone, right?

What if there is no console ... then remove the cubby, bend the TDC cable to the
right, and put the TD Connector through some large hole?

And, how does one get the two rear clips of the cubby back in place, please?

I will try again today, but suggestions for the no-console case welcomed.
Thanks

This is my car. I had just removed the console and cubby - I was looking for the connector. I did reinstall both a couple of minutes later.

As for the rear clips : it takes patience Went to the floor of the car with a lamp and carefully aligned the clips in the holes then firmly pushed up and forward at the same time.

 

[apacheguy]

See this post by wk: Let the hacking begin... (Model S parts on the bench) - Page 43

wk said Max Battery Power increases cooling targets to 50/52ºC, not sure why this screenshot shows 52/55ºC.

This screen is from an S85. It does not have Max Power mode. Even more odd...

 

[llavalle]

OT, but I had to double take there!

Still OT but what exactly did you see before figuring out it was my boot?

 

[smac]

Do you have any 100% charged data yet? It looks like from your data we have seen

79% ~ 4.09 V
89% ~ 4.179 V

lots of M85 data but not so much M60, lola's sc data put 93% at the charging CV knee with about 24 cells >= 4.19 V. Your cells look evenly balanced.

Nope not yet, I generally do < 9 miles each way on my commute, so I'd end up leaving the car at a fairly high SOC. As soon as I get the opportunity, I'm more than happy to do it to fill out the "community dataset".

Seems we have lots of 85 packs investigated (and most effort gone in to tools for those). I hope by adding some data from a rarer pack it helps the broader understanding.

- - - Updated - - -

Still OT but what exactly did you see before figuring out it was my boot?

I didn't. It was just funny as I was rocking a very similar footwear / jeans combo when I saw the thread

 

[Ingineer]

Ok, this MS has the center console installed, but the cubby/shelf is gone, right?

What if there is no console ... then remove the cubby, bend the TDC cable to the
right, and put the TD Connector through some large hole?

And, how does one get the two rear clips of the cubby back in place, please?

I will try again today, but suggestions for the no-console case welcomed.
Thanks

If you made the mistake of not following the instructions and did more than just unsnap the front of the cubby, you are now going to have to take the yacht floor (or carpet if an older S) out of the bowling alley, and on some cars this is a PITA.

Reach your fingers up behind the rear top of the yacht floor and pull towards the rear of the car and try to work the yacht floor piece out. There is a little black piece of plastic that snaps in on either side at rear of the cubby space that must come out in order to re-install the cubby. This piece pulls out straight towards the rear of the car to remove it. So this also means the yacht floor must come out first. Be extremely careful with the front of the yacht floor where it meets the console. If you have the version with rubber all around then you are much less likely to damage something. If not, and the yacht floor has no rubber surround, then be careful not to damage the corners and/or the rear of the console (where the USB ports are, etc). If you have an older S with only carpet, then it's super easy.

I made a couple of hook tools out of thin tempered strips of sheet metal I found in my junk stash that makes this procedure much easier, but again, just go slow and be careful.

Warning to those that have not fully removed the cubby, DO NOT! Just unsnap the front only and let it swing down. If you remove it, you have a lot of tricky expletive-laden work to do in order to get it back together!

 

[smac]

If you made the mistake of not following the instructions and did more than just unsnap the front of the cubby, you are now going to have to take the yacht floor (or carpet if an older S) out of the bowling alley, and on some cars this is a PITA.

Reach your fingers up behind the rear top of the yacht floor and pull towards the rear of the car and try to work the yacht floor piece out. There is a little black piece of plastic that snaps in on either side at rear of the cubby space that must come out in order to re-install the cubby. This piece pulls out straight towards the rear of the car to remove it. So this also means the yacht floor must come out first. Be extremely careful with the front of the yacht floor where it meets the console. If you have the version with rubber all around then you are much less likely to damage something. If not, and the yacht floor has no rubber surround, then be careful not to damage the corners and/or the rear of the console (where the USB ports are, etc). If you have an older S with only carpet, then it's super easy.

I made a couple of hook tools out of thin tempered strips of sheet metal I found in my junk stash that makes this procedure much easier, but again, just go slow and be careful.

Warning to those that have not fully removed the cubby, DO NOT! Just unsnap the front only and let it swing down. If you remove it, you have a lot of tricky expletive-laden work to do in order to get it back together!

Has there been a design change? My cubby has two plastic tangs at the back, and two corresponding slots in the back of the central storage area. Underneath the slots is a lip which supports the cubby and acts as a guide. It's super simple to install / remove my cubby, but I have neither yacht flooring or premium console. I didn't need to take out any carpet though.

I'll admit to not trying the side panel, and will give that a try. This seems the most universal solution!

 

[garygid]

The "little black piece of plastic that snaps in on either side at rear of the cubby space
that must come out in order to re-install..." ...
is this a piece that extends from the left side to the right side?

After the top of the rug is pulled toward the rear of the car, and the "black piece" is removed by pulling it
toward the rear of the car, the cubby would go into position and up, then the black piece back in, and
finally the top of the rug back in place?

If so, I will try to get the black piece out. Rather sorry that I did not understand (or remember)
the instruction details. When it is all back together, I still do not see where the connector could
come out the right side to get accessed in the passenger footspace.

However, since I am trying to make the TDC accessible, there is some good in my blunder.
with the back of the cubby down about 3/4 inch too far (but the front of the cubby snapped
in place), the cubby is usable, looks reasonable, and there is a narrow v-shaped space for
the cable to/from the TDC (or the TDC to OBD adapter) to enter/exit the space behind the
cubby. Also, there might be enough additional room above the "saggy-down" cubby to
hide both the adapter and the ELM-type dongle.

As a minimum for the TM-Spy, the CAN3 (H and L) would go to the OBD pins 6 (H) and 14 (L),
Ground will connect to OBD pins 5 and 4, and "12v" will connect the "always On power" to OBD pin 16.

Hopefully, I will have the necessary parts "soon", perhaps within a week.

 

[pupik]

OK So I've pulled some logs and just had a quick mess through some stuff.

First thing was to have a poke around the lifetime stats. Some of it is quite interesting!

- Battery total miles is 13,708 miles however my car only has 12,291 miles on the clock (weird!!)
- Lifetime discharge is at 6,539 kWh (vs. 6,977kWh charged). This puts total Wh/mi at 530Wh/mi vs my lifetime average on the display which is 350Wh/mi.

Message ID 0x0382 has the following data: 4D 72 67 9D 76 0 70 24
Which if my code is right equates to my 60 having 59.8kWh of battery (So pretty much as advertised, unlike the 77kWh figure we've seen reported from the 85kWh packs )
Interestingly all the remaining figures tally with 77-78% of the 59.8kWh figure which is what the dash was reading. Energy buffer seems set at 2.8kWh.

I did a 0-60 run, but it certainly won't be as impressive as the PxxD cars Will look to pull the graph up later.

Excuse me, how you acquired 59.8kWh?
I tried insert your data from CAN into my SW and I think you have a mistake for all further calculations on the forum for S60 :wink:

so, 59.8kWh is probably typing error?

 

[zdre]

The power train bus is easily a few thousand frames per second. It's likely not the laptop but the CAN logger that isn't catching them all.

It did turn out to be my crappy netbook not keeping up... python was maxing out one of the threads, so it was able to capture only 300fps. My real laptop was able to capture 1400+fps. It did not look like it was missing any.

Interesting. This is from 0x102? I'd definitely be interested in what the whole array from 0x6F2 shows for your pack... (CAN IDs from memory...)

Here are my 0x6F2 logs. I have not had time to work on my parser unfortunately.
6f2,0,177,241,108,12,27,207,198
6f2,1,177,241,108,220,26,187,198
6f2,2,175,241,107,236,26,199,198
6f2,3,178,49,109,76,27,211,198
6f2,4,180,49,109,44,27,207,198
6f2,5,178,113,109,60,27,219,198
6f2,6,159,49,104,188,25,127,198
6f2,7,158,241,103,28,27,191,198
6f2,8,174,241,107,252,26,203,198
6f2,9,177,113,109,44,27,211,198
6f2,10,177,241,109,12,27,211,198
6f2,11,178,49,109,76,27,227,198
6f2,12,184,49,110,124,27,227,198
6f2,13,184,49,110,108,27,235,198
6f2,14,181,241,109,124,27,231,198
6f2,15,173,49,108,12,27,203,198
6f2,16,175,113,108,60,26,143,198
6f2,17,160,49,104,204,25,139,198
6f2,18,178,113,109,76,27,207,198
6f2,19,179,49,109,140,25,103,198
6f2,20,151,113,101,124,25,103,198
6f2,24,84,66,151,160,37,252,9
6f2,25,93,130,160,112,38,16,10
6f2,26,93,130,158,160,37,196,9
6f2,27,47,2,151,224,36,0,9
6f2,28,103,194,155,112,37,200,9
6f2,29,103,130,164,240,37,232,9
6f2,30,101,130,156,144,36,144,9

 

[smac]

Excuse me, how you acquired 59.8kWh?
I tried insert your data from CAN into my SW and I think you have a mistake for all further calculations on the forum for S60 :wink:

so, 59.8kWh is probably typing error?

Good catch, thanks for the QA. It must have been a brain to keyboard interface error (Apologies for any confusion)

You are correct from that frame it is 58.9. I just stepped the code and my calc is correct, it seems it's my typing that needs more work in this instance :redface:

More strangely though as I went to double check the results, I grabbed a different log file and ran this through (one of the static logs I was using to find out cell voltages). This is where I really confused myself...

45 2A 98 E0 81 0 70 20 => which (and please correct me if I'm wrong) goes to 89.1kWh

This log and the first one, which was taken whilst doing a 0-60 run, are under two weeks apart. So I was VERY surprised to see such a large change. The biggest thing I can think of is that the cell temperatures are different between the two logs. The driving one had the cell temps around 16c, the static had them around 5c. I wonder if there is more to this statistic than meets the eye ?!

 

[pupik]

Good catch, thanks for the QA. It must have been a brain to keyboard interface error (Apologies for any confusion)

You are correct from that frame it is 58.9. I just stepped the code and my calc is correct, it seems it's my typing that needs more work in this instance :redface:

More strangely though as I went to double check the results, I grabbed a different log file and ran this through (one of the static logs I was using to find out cell voltages). This is where I really confused myself...

45 2A 98 E0 81 0 70 20 => which (and please correct me if I'm wrong) goes to 89.1kWh

This log and the first one, which was taken whilst doing a 0-60 run, are under two weeks apart. So I was VERY surprised to see such a large change. The biggest thing I can think of is that the cell temperatures are different between the two logs. The driving one had the cell temps around 16c, the static had them around 5c. I wonder if there is more to this statistic than meets the eye ?!

89.1 is wrong

45 2A 98 E0 81 00 70 20 = 58.1kWh
hex 45 = dec 69
hex 2A = dec 42
58.1 = (69 + ((42 & 3) << 8)) * 0.1; ( all the values in a decimal form )
2 = 42 & 3
512 = 2 << 8
581 = 69 + 512
58.1 = 581 * 0.1

:smile: attention:
- on signed and unsigned type
- byte0 is left
- for hexa je writing 0x..


There is writing for 32bit C++


unsigned short int byte0 = 0x45;
unsigned short int byte1 = 0x2A;
float nominalFullPackEnergy = (byte0 + ((byte1 & 0x03) << 8)) * 0.1;

 

[smac]

89.1 is wrong

Argh!!!! I really must double check my posts :redface::redface::redface:. Brainfutz meant I typed in the file name (89 was the % SOC of the log!) I'm obviously going mad


My C# code is similar (same) as your C++:

var nominalFullPackEnergy = (data[0] + ((data[1] & 0x03) << 8)) * 0.1;


Surprise, surprise, I too get 58.1!!

I guess the salient point is 58.1 < 58.9 from less than two weeks ago.

 

[pupik]

Argh!!!! I really must double check my posts :redface::redface::redface:. Brainfutz meant I typed in the file name (89 was the % SOC of the log!) I'm obviously going mad


My C# code is similar (same) as your C++:

var nominalFullPackEnergy = (data[0] + ((data[1] & 0x03) << 8)) * 0.1;


Surprise, surprise, I too get 58.1!!

I guess the salient point is 58.1 < 58.9 from less than two weeks ago.

small wobble for parameter nominalFullPackEnergy is normal. My standard is +/- 0.2kWh
come back to you 58.9kWh soon, do not worry :smile:

 

[zdre]

Here are my 0x6F2 logs. I have not had time to work on my parser unfortunately.
6f2,0,177,241,108,12,27,207,198
6f2,1,177,241,108,220,26,187,198
6f2,2,175,241,107,236,26,199,198
6f2,3,178,49,109,76,27,211,198
...

parsed the data...
1 3.879 3.880 3.879 3.880 3.879 3.880
2 3.878 3.878 3.879 3.879 3.878 3.879
3 3.880 3.880 3.880 3.880 3.880 3.880
4 3.880 3.880 3.880 3.881 3.880 3.881
5 3.874 3.874 3.873 3.874 3.874 3.874
6 3.879 3.879 3.878 3.879 3.879 3.880
7 3.879 3.881 3.880 3.880 3.879 3.881
8 3.879 3.880 3.880 3.880 3.880 3.882
9 3.882 3.882 3.881 3.882 3.882 3.882
10 3.881 3.882 3.881 3.881 3.881 3.882
11 3.878 3.879 3.879 3.880 3.879 3.879
12 3.875 3.875 3.874 3.874 3.873 3.875
13 3.880 3.881 3.880 3.880 3.880 3.880
14 3.872 3.872 3.871 3.871 3.871 3.872
Total: 325.8 V

 

[pupik]

It did turn out to be my crappy netbook not keeping up... python was maxing out one of the threads, so it was able to capture only 300fps. My real laptop was able to capture 1400+fps. It did not look like it was missing any.

Here are my 0x6F2 logs. I have not had time to work on my parser unfortunately.

Message ID 0x0382 please
thx :smile:

 

[zdre]

Message ID 0x0382 please
thx :smile:

From earlier post (different logging session though):

Here are some of my numbers when fully charged:
Nominal Energy Pack Full: 56.6 kWh
Nominal Energy Remaining: 39.2 kWh
Expected Energy Remaining: 40.8 kWh
Ideal Energy Remaining: 39.2 kWh
Energy (Bricking?) Buffer: 1.4 kWh
Pack Voltage: 336.50
Battery Odometer: 34172
SoC UI: 69%

 

[kennybobby]

parsed the data...
Total: 325.8 V

i didn't have a chance to calculate the voltages from the previous data yet, but from your parse it looks like your pack is in great shape.

 

[pupik]

smac, Thanks for posting that data it allowed me to understand the mapping and update TM-Spy so it now correctly displays voltage and temp data taken from a real Tesla.

I am impressed with how well the battery pack is balanced. Temps also show the difference between inlet and outlet temps.

Notice that the temperature at the module 9 is conversely
but for my car is the same :scared: ( but it's not always ) :crying:
I thought at first that anomaly is only in my car...
some idea?

 

[pupik]

Here are some of my numbers when fully charged:
Nominal Energy Pack Full: 56.6 kWh
Energy (Bricking?) Buffer: 1.4 kWh
I am not sure what the difference between Expected and Ideal energy remaining.

Only my calculations ... :wink:
58 = 56.6 + 1.4
96.6% = 58/60

81.6 = 77.6 + 4
96.6% = 81.6/85