Electric Porsche

This page will contain the current plan, and current state of the car. Please refer to the “Home” page for the progress and steps taken to get here.

 


Current state:

  • I haven’t updated this for awhile, everything was old news.
  • Heaters have been tested and just need to be installed permanently.
  • Battery box is half way there, just needs aluminium walls.
  • Tablet to be disassembled and custom mounted in the center console

What’s next?

  • EMW has created an arduino based gauge system that exceeds all of my early plans so it’s pretty easy I will be using their system!
  • The rear battery box is the new big project, under the hood things are coming together.
  • Battery testing and assembly is coming along, I have the last batch of batteries on order.

Parts on hand/ordered:

DC/DC converter modules for 14.4v system

Kats fluid heaters 2x 1500w

12v variable speed fluid pump for heating system

Netgain Warp 11-HV

Headway 38120S 10ah cells x12 for testing/12v replacement battery

Mes-Dea Electric vacuum pump

DC/DC converters for 90cell 96cell battery charger

Power supplies 85-264v for DC/DC battery charger

Material to prototype an aluminum battery box  Now a steel framed box with aluminium walls

Prototyping parts for the Arduino voltage monitoring system  EMW system to be used

Soliton1 300kw DC motor controller with Evnetics throttle pot

Clutchless Aluminum coupler

Compact steel adapter plate/tube

A123 20ah cells x120, more to follow final pack design is 288 cells (96S3P) remaining cells on order

Galaxy tab 7+ to get integrated into the center console for audio/navigation/gauges


The Plan!

Curb weight:  ≤3000lbs   ≤2900lbs

Name: “944DC” I might make an emblem to replace the stock 944 badge with 944DC or 944HV to represent the DC power system or Warp 11 “HV”.

√  Motor: Netgain Warp 11-HV ~300hp, ~300ft-lbs

√  Adapter plate: CNC machined aluminum Compact steel tube/plate with locating surfaces, no manual alignment necessary.

√  Coupler: Clutchless coupler from Charlie at evcouplerconnection.com. Taper lock with modified pulley

√  Controller: Evnetics Soliton1 300V, 1000A

√  Throttle: Evnetics hall effect throttle assembly controlled by the stock pedal & cable.

Batteries: Headway 38120S 3.2v 10ah cells in a 288V   A123 20ah cells in a 96S2P 96S3P 317V nominal pack with 1000A continuous peak continuous discharge .

Battery boxes:  Welded/riveted aluminum/Steel with heating system

BMS: Arduino based multichannel isolated voltage monitoring system.  Cell log 8’s

√  Charger: 90 96 isolated 1/4 brick DC/DC converters 22.6A @ 3.65v 23.5A @ 3.5v with 85-264v AC power supplies 3kw.  Manzanita Micro PFC-20 or Zivan NG3.

√  12v Battery: Headway 38120S cells in a 4s3p pack to replace the heavy stock battery.  A123 12v system battery.

√  DC/DC Converter: 1/4 brick modules with adjustable output 14.4v nominal 80A continuous.

√  Steering: Replace the power steering rack with a manual rack.  Depower the existing rack.

√  Mes-Dea electric vacuum pump for the power brake booster.

√  Heater: 3kw Fluid heater using the stock heater core and controls.

√  Air Conditioning: Stock A/C system powered by a dedicated electric motor or main motor tail shaft.

Gauges:

Speedometer – stock, no modifications required

Tachometer – driven by the controller

Temperature gauge – display the temperature of the heating system

Fuel guage – driven by the controller EMW gauge system will represent battery state of charge

Voltage gauge – monitor the stock 12v system

Fuel economy – I may use this as a representation of battery current draw (or fuel economy)

Oil pressure – I’m not sure what to do with this one yet, but if I print a new scale it might be possible to use it to represent pack voltage

Dummy lights will work in much the same way, check engine will be tied to the controller and display warnings or faults.


7 Replies to “Electric Porsche”

  1. Dont stop looking for weight to remove ! I pulled over 1,700lbs. out of my 928.
    A few places to look are sound insulation, doors,bumper shocks and inner bumper supports. These are just a few, so keep looking.

    Have fun,
    Jeff McCabe

  2. Hi,

    I’m taking a similar journey by replacing a bulk charger with many individual isolated chargers. I see a couple problems with your setup as shown, though. You have heavy connections running to the ends of the pack, with only small wires running to the tap points, and heavy wires series connecting the chargers. There are two serious problems with this setup:
    -If a bus bar (pack series connection) fails, you will get the entire pack current flowing through the series charger connection, which could start a fire.
    -The DC-DC converters can’t sink current, only source it, so if a cell reaches 100% charge sooner than the rest, it will continue to be (over)charged by the rest of the chargers connected in series. The charger connected to that cell will no longer be supplying current, but it won’t do anything until its output protection zener turns on from an overvoltage condition, probably destroying both the overcharged cell and the charger.
    It would be better to give each charger a (fused) 2-lead individual connection to the cell it charges, and have the bus bars be the only series connection in the pack.

    Good luck!

    1. You are correct about the wire sizes, the pack shown was used for bench testing to prove the concept, which it did. The final version will use larger wires for all connections. You are also 100% correct about a buss bar failure causing more current in the (red) wires as shown. After extensive testing the situation you describe about overcharging is not actually correct in practice. Although the dc/dc converter does not sink any current, like you said the fully charged cell actually gets minimal current from the remaining still charging cells. I did this on purpose to test that exact situation, I used 1 almost fully charged set, two sets around 60-70% SOC and one set at about 30% SOC, I then monitored the voltage of the fully charged group to see if they would overcharge. During the peak charging (highest current which was also the highest voltage measured) the charged group went up to 3.55V (from the 3.5V set point) I performed this test twice, moving the location of the discharged and full cells. Going with the system you proposed would be less efficient, and involve many more connections. If the cells are in balance then 99% of the current flows in the large wires as shown in my setup, the only time current flows through the red wires is when two adjacent cell groups are at different voltage/current levels. This is what occurs in the test above since the groups around the charged cell will continue to charge with minimal (but still some) current flowing through the already charged group.

      I appreciate the comments though, I had the same concerns and questions that you raised before I began. I did thorough testing to make sure everything works safely and reliably, I measured voltage and current throughout the whole system in all different types of charging states. The system simply brings all of the cells up to 3.5V reliably every single time.
      (Just wait till I make the post about high current discharge testing, the setup looks a bit freaky even though it is very solid and reliable.)

  3. Sounds good. You’re right that my system will be less efficient- I also have a series diode, as my power supplies draw 30-50 mA from their output when shut off (from the optoisolator and an indicator LED) which would eventually drain the battery. Between the fuse, the diode, and the fairly long runs of 18ga wire, I’m only about 85% efficient. I’m convinced the efficiency loss is worth it for gains in safety and reliability, but experience will tell.

    1. Just make sure the 18awg wire is large enough, my first test used 18awg for all of the wires (the red ones are 18awg) and the dc/dc converters can put out much more than that size wire can handle long term. If you charge at more than a couple of amps I would up size the wire. I did also consider the diode, it’s a good idea however my dc/dc converters couldn’t get the voltage high enough at full power to overcome the diode drop. To do that I would have been better off getting 5V output dc/dc converters and trimming down. My “leakage” current is about 6-7ma when the system is not charging. This works out to around 1.5 years to discharge the pack, I’m planning on using it as a daily driver so the impact of the opto-isolator and dc/dc converter itself is minimal. What type of voltage monitoring are you going to use (since you mention an opto-isolator I assume you have one)? Your setup needs to instantly tell if a fuse blew while charging or you would probably over discharge the cell(s) and destroy it/them. My basic theory is keep the high power section as absolutely simple as possible, complexity invokes more points of failure, my dc/dc converters have what appears (in testing) to be good built in protection for over temp, over voltage, over current, so they will do what’s necessary to protect themselves, I just need to consider a fuse as close to the battery as possible to protect against a physical problem like a shorted or broken wire. I will be posting the almost final version of the PCB soon, it will hold 16 dc/dc converters, opto-isolators, multiplexer, temp sensor, and a 2 stage charge voltage setup.
      Thanks again for the comments.

  4. Hi,

    You adventure are a stepping stone for me and all other new to EV to follow. If you have time Sir, can you kindly do a write up with pictures too. 😀 I’ll be bookmarking your blog.

    Keep us posted.

Leave a Reply

Your email address will not be published. Required fields are marked *