DIY Electric Car Forums banner

41 - 60 of 124 Posts

·
Registered
Joined
·
243 Posts
Great project - I look forward to following your progress. Since you're using the Tesla drivetrain, why not use Tesla cells too? They are easy to find on EBay, and relatively cheap. They can be reconfigured to double their voltage (so 48v per module instead of 24), allowing you achieve input voltage with only half the modules. It would still cost more than the Volt cells, but would probably double your range. They're also much lighter than the Volt's, giving you better acceleration.
 

·
Registered
Joined
·
141 Posts
Discussion Starter #42
The reconfigured Tesla modules could work. I'd certainly like to see some feedback on actual usage. The biggest challenge is space and weight distribution. 7-8 modules mean taking up engine bay and the DeLoreans frunk. Even with good weight distribution, battery placement is likely to raise the center of gravity to a greater height.
 

·
Registered
Joined
·
243 Posts
I've got 7 of the 48v (12s) modules, but I won't be running them in my car until September. If be happy to let you know how it goes then.

If they can be modified to 48v (12s) then maybe they can be modified to 96v (24s) too, then you'd only need 3-4 of them. 4 modules would give you over 21 kWh and weigh just over 200 lbs.

Something to think about anyway. Pm me if you'd like me to put you in touch with the guy that modified them for me.
 

·
Registered
Joined
·
141 Posts
Discussion Starter #44
Quick update on the TesLorean project...

Lots of powder coating on rear suspension parts and calipers. Getting closer to putting the wheel back on and starting the body/frame separation. Surprising what you can achieve with a toaster over, besides just toast.

Got some new Tesla parts 1) Tesla PTC Heater (electric air heater), and 2) Tesla HVJB . The design previously called for the same coolant loop that cooled/heated the drive unit & batteries to also heat the cabin, but after further study the heat from the drive unit / batteries would not show up fast enough to aid in defrosting the windows. This removes the need for a coolant loop into the airbox, so the coolant loops can stay with the batteries and drive unit. The Delorean airbox will need to be modified to accommodate the PTC heater.


Jeff
 

Attachments

·
Registered
Joined
·
141 Posts
Discussion Starter #45
Since the Delorean is a 35 year old car, rather than just convert it to electric, it is begs to do more... renew components (rubber bushings), upgrade original parts (bolts subject to premature failure), and improve on the design in preparation for E demands.

Key changes...
- poly bushing and inconel bolt on the trailing arm
- new bushings throughout
- rebuilt caliper and upgraded lines
- adjustable shocks

Still rebuilding the parking brake.
 

Attachments

·
Registered
Joined
·
141 Posts
Discussion Starter #46
In preparation for frame-body separation

I've been running down through the checklist to separate the DeLorean body from the frame. Got everything done apart from getting the wheels back on and the stick shift lever freed.

The stainless is a skin that sits on top of a fiberglass 'tub' (you can see it here as the black structure), which then mounts on top of the frame. The suspension, engine, and transmission are all mounted to the frame (the parts epoxy coated in grey).

Once I get the body and frame separated, I'll remove the engine and transmission, and then start positioning the Tesla small rear motor - to measure for drive axles and mounts.
 

Attachments

·
Registered
Joined
·
141 Posts
Discussion Starter #49
Tesla steering wheel control buttons - DECODED

Some background... Tesla Model S steering column is a Mercedes Benz unit. The control buttons in the steering wheel pad talk LIN (via the clock spring) on three wires back to the circuit on the main column. The main column then talks CAN to the rest of the car.

To start with I tried (without success) to wake up the steering column (with CAN messages) and get it to send CAN frames when the steering pad or control stalks were used. The only buttons that sent CAN were the steering column position adjuster and the P'ark button. Even so, activating any of these sent out a stream of CAN messages, but for every CAN frame containing useful data, there were 100s of frames with nothing useful (3 second long stream of crap frames).

I decided to interface directly with the (19200 baud) LIN messages coming from the steering pad controls. LIN is not as easily approachable as CAN, but I eventually got it to work. The steering wheel pad and control buttons can be removed and disconnected from the clock spring.

I used...

1) LIN-BUS (from skpang, via CopperHill Technologies in the US (http://copperhilltech.com/lin-bus-breakout-board/)
- Essentially just protects the Arduino from the 12v LIN bus. You supply it with 12v, it powers up the LIN connection, and then it outputs a TX/RX 5v connection for the Arduino. Finally got it going after figuring out that CS (chip select) needed to be 12v to activate the chip.

2) Modified code from Andrew Stone on Github
(https://github.com/gandrewstone/LIN)
- Andrews LIN code was very helpful, but I did need to back out his demo code specifically designed to talk to two LIN-enabled multicolored LEDs. My new main code prompts LIN-slave chips on the steering pad buttons for a response. LIN is a prompt and response network, so the Arduino stands in for the Master node. (Note: I tried using the LIN-BUS code by zapta on github, but I was unable to get it to be reliable for this application. It attempts, by using interrupts and clever timing, to make a digital pin on the Arduino act like a serial RX pin.)

3) I also used an Arduino Mega (with additional hardware serial ports - other than the one used to talk back to the PC via the USB port). The LIN-BUS is connected to the Serial1 pins 18,19 on the Arduino.

The Arduino code acts as a Master node in the 'steering pad' LIN network. It sends out a request with the ID 0X3D, and gets 8 bytes (1 ID byte and 7 data bytes) in response back from the LIN slaves in the steering pad. Depending on which buttons are pressed / wheels rolled, the bytes of the response are modified.

My TesLorean goal in all this is to replace the stock DeLorean steering wheel with the Tesla wheel. I will also be using the Tesla stalks for PNDR and indicators/beams/wipers. The LIN messages from the steering pad will be combined with the stalk control signals to produce TesLorean CAN messages for the main computer (which will link to the motor controller, battery BMS, charger, etc).

=============================
LIN Network Responses

ID + 7 bytes

Baseline
=======
Prompt 0x12 Receive ID 0x00 and 00 00 00 00 00 00 00
(while the steering pad responds to prompt id 0x12, returned data is always 00,...,00
Prompt 0X3D Receive ID 0x00 and 00 00 82 80 00 00 00

Horn ::
Prompt 0X3D Receive ID 0x00 and 00 01 82 80 00 00 00

Voice & Right Up ::
Prompt 0X3D Receive ID 0x08 and 00 00 82 80 00 00 00

Menu Back & Right Down ::
Prompt 0X3D Receive ID 0x20 and 00 00 82 80 00 00 00

Scroll Wheel Right Press ::
Prompt 0X3D Receive ID 0x10 and 00 00 82 80 00 00 00

Scroll Wheel Right Down ::
Prompt 0X3D Receive ID 0x00 and 3C 00 82 80 00 00 00 (down 1 click)
Prompt 0X3D Receive ID 0x00 and 38 00 82 80 00 00 00 (down 2 clicks)
Prompt 0X3D Receive ID 0x00 and 34 00 82 80 00 00 00 (down 3 clicks)
...
Prompt 0X3D Receive ID 0x00 and 20 00 82 80 00 00 00 (down N clicks)

Scroll Wheel Right Up ::
Prompt 0X3D Receive ID 0x00 and 04 00 82 80 00 00 00 (up 1 click)
Prompt 0X3D Receive ID 0x00 and 08 00 82 80 00 00 00 (up 2 clicks)
...

Forward & Left + ::
Prompt 0X3D Receive ID 0x01 and 00 00 82 80 00 00 00

Back & Left - ::
Prompt 0X3D Receive ID 0x04 and 00 00 82 80 00 00 00

Scroll Wheel Left Press ::
Prompt 0X3D Receive ID 0x02 and 00 00 82 80 00 00 00

Scroll Wheel Left Down ::
Prompt 0X3D Receive ID 0xC0 and 03 00 82 80 00 00 00 (down 1 click)
Prompt 0X3D Receive ID 0x80 and 03 00 82 80 00 00 00 (down 2 clicks)
Prompt 0X3D Receive ID 0x40 and 03 00 82 80 00 00 00 (down 3 clicks)
...
Prompt 0X3D Receive ID 0xC0 and 02 00 82 80 00 00 00 (down N click)
Prompt 0X3D Receive ID 0x80 and 02 00 82 80 00 00 00 (down N+ clicks)
Prompt 0X3D Receive ID 0x40 and 02 00 82 80 00 00 00 (down N++ clicks)

Scroll Wheel Left Up ::
Prompt 0X3D Receive ID 0x40 and 00 00 82 80 00 00 00 (up 1 click)
Prompt 0X3D Receive ID 0x80 and 00 00 82 80 00 00 00 (up 2 clicks)
Prompt 0X3D Receive ID 0xC0 and 00 00 82 80 00 00 00 (up 3 clicks)
...
Prompt 0X3D Receive ID 0x40 and 01 00 82 80 00 00 00 (up N clicks)
Prompt 0X3D Receive ID 0x80 and 01 00 82 80 00 00 00 (up N+ clicks)
Prompt 0X3D Receive ID 0xC0 and 01 00 82 80 00 00 00 (up N++ clicks)
 

Attachments

·
Registered
Joined
·
141 Posts
Discussion Starter #51 (Edited)
The TesLorean - Brusa NLG513 Charger

I got the Brusa NLG513 charger setup and running (but funky window size problem).

IF ANYONE HAS... a charge profile for a 2014 Chevy Spark A123 battery pack, I'd be very interested!!!

Good info in the YouTube video from Damien Maguire on Brusa connection (note: Pin 15 is ground for the RS232 on the AMPseal connector)

I did run into the Windows not releasing the COM1 port properly (or ChargeStar not recognizing correctly that it was released). To get it 'Released' the first time - I used Windows Hyperterminal and connected to COM1 and then exited so that COM1 was released on exit. ChargeStar doesn't release COM1 (correctly - or windows ignores it), so you need to open-close-open ChargeStar when you next want to connect.

I did not have to hook the charger up to Mains power (120-240 volts) to get the charger to respond to RS232. I just applied the GND on pin 1 and +12v on Pins 2 and 3. Pin 3 is PON (power on) which the manual says can be used for programming if the mains is not connected.

When I read from or write to the Brusa, once the operation is completed the application window resizes (to about 1/2 height). Nothing I did could get it to resize - which is necessary to see the input/output data at the bottom half of the app. I think I can work around it by loading/saving profiles between reading/writing to the Brusa.

Jeff
 

Attachments

·
Registered
Joined
·
106 Posts
Are you planning to implement an automated precharging circuit for the NLG513 charger, as recommended by Metric Mind? I was not aware of this requirement for these chargers until recently, and would prefer to have a completely disconnected traction pack except when charging and discharging. I would think that most people would have a manual safety disconnect switch plus contactors in place to disconnect the pack, and the only way to avoid an automatic precharging circuit is to have the charger permanently connected to the pack circuit before these devices.
 

·
Registered
Joined
·
141 Posts
Discussion Starter #53
A123 Battery Pack from 2014 Chevy Spark

Starting the teardown of the 2014 Chevy Spark A123 Battery Pack.

Some features...
- Pack composed of 4 modules, two in front of the rear axle and two behind. Battery box tunnel straddles the rear axle.
- Coolant plate under the front two modules and another under the rear two modules.
- Temp sensors on the coolant lines and what may be a small heater
- Contactors in the neck
- BMS modules appear to be on either end of the battery modules.
 

Attachments

·
Registered
Joined
·
141 Posts
Discussion Starter #54
2014 Chevy Spark Module Voltages

There are four identical A123 battery modules in the 2014 Chevy Spark EV. I measured the voltage on each module and got 92.4 - exactly the nominal module voltage, for a pack voltage of 370v.

Jeff
 

·
Registered
Joined
·
141 Posts
Discussion Starter #55
The TesLorean - Battery Pack Control

I recently completed a test to control the contactors in the A123 battery pack from a 2014 Chevy Spark.

[Caution: Please do not work with high voltage batteries without sufficient knowledge, precautions, safety protection and equipment. In this demo the pack is closed but HV is still present external to the pack. This can be very dangerous.]

demo here

The Spark has four connectors (technically 5 including a body ground connection). Two are HV (drive unit - large, charger - small) and two are LV (contactors control, data links).

X358 (LV connector found in the neck)
1 BLK High voltage interlock loop low reference
2 VIO High voltage interlock loop signal
5 BLK Body Ground - Relay pin 6 (or 2) on all Contractors
Note: Means that all Relay pins are 12v to activate contactors
7 VIO/GRY EV PCM Batt (-) relay to battery negative contactor
8 GRY/BLU EV PCM Batt (+) relay to multifunction Contactor
9 WHT/BRN EV PCM to Pre-Charge Contactor
10 BRN/GRN EV PCM Batt 1 (+) relay to battery positive contactor
11 WHT/BRN EV PCM Batt (-) relay to batt Charging Sys Neg Contactor
12 YEL/VIO EV PCM Batt (+) relay to battery Charging Sys Pos Contactor
14 VIO Auxiliary heater control
Note: May switch on the auxiliary coolant heater

I wanted to be able to use the OEM contactors in the pack to connect the battery to the charger. The 'neck' of the battery pack contains a pre-charger circuit and contactors for both the battery to driveline connection, and for battery to charger. Connector X358 has lines that set the appropriate contactors.

To connect the battery to the charger HV connection I did the following...

Sequence 1
X358-1 to Ground (pack ground)
X358-2 to 12v (pack 12v)
X358-5 to Ground (acts as the ground for all the contactors in the neck)

Sequence 2
X358-9 to 12v (pre-charge contactor)
X358-11 to 12v (charger negative contactor)
X358-8 to 12v (multifunction contactor)
Pause
X358-12 to 12v (charger positive contactor)
X358-9 to Open (pre-charge contactor)

Sequence 3
X358-12 to Open (charger positive contactor)
X358-8 to Open (multifunction contactor)
X358-11 to Open (charger negative contactor)

Sequence 4
X358-5 to Open (contactors' ground)
X358-2 to Open (12v to pack)
X358-1 to Open (ground for pack)

This will make the connection from the Battery Pack to the Charger, another set of contactors (not using the pre-charger circuit) enables the battery connection to the drive unit (power distribution) using pins 7 and 10 listed above.

Jeff
 

·
Registered
Joined
·
4 Posts
Very cool project! I'm looking forward to seeing what you'll do with the battery pack. I love the idea of splitting it between front and rear to improve weight distribution, but it seems to me that this could considerably complicate the design in certain aspects.
Subscribed!
 

·
Registered
Joined
·
141 Posts
Discussion Starter #57
Very cool project! I'm looking forward to seeing what you'll do with the battery pack. I love the idea of splitting it between front and rear to improve weight distribution, but it seems to me that this could considerably complicate the design in certain aspects.
Subscribed!
I will be splitting the battery pack into two parts (which it essentially already is), but they will both continue to sit at the rear of the DeLorean - one half in front of the rear axle and the other half behind the rear axle. Splitting the pack is just to accommodate the available space in the DeLorean engine bay.

In the Chevy Spark 2 (of 4) modules sit in front of the rear axle, and 2 (of 4) modules sit behind the rear axle. Between the two halves are...

a) coolant flow lines
b) HV lines
c) BMS lines (rear submodules to the front master module)

In the DeLorean the two halves will be about 2 ft further apart than they are in the Chevy Spark. I will be keeping the two halves/frame/case intact and just running longer coolant, HV, and BMS lines between the two modules. I've almost reverse engineered enough about the packs operation that my plan is to use it in OEM mode, i.e. leave it 'functionally intact'.

Jeff
 

·
Registered
Joined
·
4 Posts
I will be splitting the battery pack into two parts (which it essentially already is), but they will both continue to sit at the rear of the DeLorean - one half in front of the rear axle and the other half behind the rear axle. Splitting the pack is just to accommodate the available space in the DeLorean engine bay.

In the Chevy Spark 2 (of 4) modules sit in front of the rear axle, and 2 (of 4) modules sit behind the rear axle. Between the two halves are...

a) coolant flow lines
b) HV lines
c) BMS lines (rear submodules to the front master module)

In the DeLorean the two halves will be about 2 ft further apart than they are in the Chevy Spark. I will be keeping the two halves/frame/case intact and just running longer coolant, HV, and BMS lines between the two modules. I've almost reverse engineered enough about the packs operation that my plan is to use it in OEM mode, i.e. leave it 'functionally intact'.

Jeff
Sounds like a solid plan. With the two halves fairly far apart, would you have separate contactors and fuses for each half?
My background is with the Formula SAE Electric competition, where the competition rules require that if separate battery packs/containers are used, each one must have 2 contactors and 1 main fuse. I think that all of the competition rules are overkill just to make things extra safe, but I'm not sure what the standard practice is in the real world.
 

·
Registered
Joined
·
141 Posts
Discussion Starter #59
The coolant pumps out of the Tesla (2015 Model S 70D) are VariMax Intercooler Pumps. Listed for "C4 Corvette 1985-1996" by Lingenfelter Engineering.

They have four control lines...
+12v and GND, PWM and Signal.
PWM is a 5v, 2Hz signal (2 cycles per second)
(switching on for 0.25 secs, and off for 0.25 secs will set speed to 50%)
Signal is PWM-like and indicates the pump speed. **I controlled it with a simple 5v digital pin on an Arduino.

"- Target flow rate 720 LPH @ 70 kPa
- Inlet / Outlet connection: 19 MM Barb
- Motor syle: Brushless
- Operating voltage: 8-16 VDC
- Maximum amp draw: 7.3 Amp with RSDS Software"

It has soft start which means that on applying 12v or adjusting the PWM signal it slowly speeds up or slows down as necessary.

Jeff
 

·
Registered
Joined
·
141 Posts
Discussion Starter #60
The Tesla has a number of coolant diverter valves, one of which is 4 way : TMN 6007370-00-B (two ins, switched between two outs), and others are 3 way : TMN 6007384-00-B (one in, switching between two outs).

The 4-port valve is used to connect the coolant flow into 1 continuous large loop or separate it into 2 smaller loops. Useful for heating/cooling just the battery unit - distinct from whatever the temp of the driveunit. The 3-oprt is used to bypass the radiator and another to bypass the coolant chiller.

The valves are listed as PWM controlled, but this is not correct. There are four wires, +12v and Gnd, Signal and Control. Rather than being PWM, Control should switch between Ground and +12v to switch flow from one outlet to the other (for both the 3 and 4 port valves). Signal indicates value position - but need not be connected for valve control purposes.

3-Port
TMN 6007384-00-B
Electrical Actuated 3/4" 3-port
TMN G9361-0R010
H42M-9000-000

4-Port
TMN 6007370-00-B
Electrical actuated 3/4" 4-port
Invensys 15B16
Motor actuator (12v)
127-00033-001 (90)
H42M-8000-000
 
41 - 60 of 124 Posts
Top