1. Attachments are working again! Check out this thread for more details and to report any other bugs.

2001-2003 EV only mode possible? Replaced stock batt with lithium

Discussion in 'Generation 1 Prius Discussion' started by plasmaticosine, Nov 15, 2015.

Tags:
  1. plasmaticosine

    plasmaticosine New Member

    Joined:
    Nov 9, 2015
    4
    1
    0
    Location:
    Vancouver
    Vehicle:
    2001 Prius
    Model:
    One
    I see there is quite a bit of documentation on making a gen2 prius EV only mode, but nothing on the gen1... Has anyone heard of anyone doing this mod? It seems like the jist of it is pulling #27 in H14 to ground momentarily. (wont let me post links, you can google "prius ev only mod" to find them easily).

    I just want to see if behind the dash is similar... it would be re-assuring prior to complete disassembly. I replaced my stock battery with a 84s 25Ah 306V lithium battery (which matches the stock battery HVC and LVC pretty well), and I have been wanting to do this mod for a while to take advantage of the beefed up battery.

    For those interested, I tricked the stock BMS by soldering 1kohm resistors in series (19 of them) and have the BMS feel between the resistors, so it always things the batt is OK.

    I also need to find a way to charge the battery.. Eventually :) I have a multi-meter on my dash so I can watch the voltage, its kinda fun gunning it or braking hard to see the full voltage swing. Its definitely more 'stiff' than the original battery.
     

    Attached Files:

    #1 plasmaticosine, Nov 15, 2015
  2. vskid3

    vskid3 Active Member

    Joined:
    Jun 23, 2013
    773
    228
    0
    Location:
    Texas
    Vehicle:
    2005 Prius
    More info on the lithium battery, please.

    Part of what made the EV mode button so easy in the Gen2 was that it was designed for it already, the US just didn't get the actual button.
     
    #2 vskid3, Nov 15, 2015
  3. plasmaticosine

    plasmaticosine New Member

    Joined:
    Nov 9, 2015
    4
    1
    0
    Location:
    Vancouver
    Vehicle:
    2001 Prius
    Model:
    One
    I bought a crashed 2011 zero motorcycle on craigslist, I thought I could fix it although it was too mangled in the end so I ended up re-purposing the cell boxes.

    Yeah I hear you on that one, although I'm sure there is ghetto ways of getting around it. I heard somewhere you can fool the ecu into thinking the engine is really hot, so it only turns on when necessary. I also think that if a charger was added, it wouldn't use much gas at all, since it would see an excessive voltage (354V charged vs 306V nominal)

    I also have a raspberry pi with a CAN shield, I might try messing around with that eventually. Seems like a time sink
     
    #3 plasmaticosine, Nov 15, 2015
    vskid3 likes this.
  4. bwilson4web

    bwilson4web BMW i3 and Model 3

    Joined:
    Nov 25, 2005
    27,665
    15,664
    0
    Location:
    Huntsville AL
    Vehicle:
    2018 Tesla Model 3
    Model:
    Prime Plus
    An interesting approach. The photo appears to show three, module assemblies. Do you have any technical specs on the LiON battery modules?

    Since you are spoofing the battery controller, you could rig up a variable voltage supply to the sense resistors and based upon your reading of the LiON SOC, adjust the voltage to the resistor network to trigger charging. This sounds like an excellent task for the Rasberry PI. One challenge being how to monitor the LiON SOC to avoid overcharging.

    As for disabling the ICE, I've run our NHW11 out of gas multiple times and the car continues to run on traction battery power with a lot of error lights. I would probably just open the fuel-pump line to replicate running out of gas. I'll have to check but it may be on its own fuse. The problem is the 'three start' lock out.

    If you run out of gas and try to start the engine three times, the car sets a latch to prevent future starts until the controller is reset (software or power OFF/ON.) I have not investigated this because I was just interested in draining the tank.

    There is also a 'speed limit' built-in to the control laws. If the car is rolling faster than 42 mph, the engine has to run regardless of SOC to keep MG1 rpm under 6,000 rpm. You can see this effect coasting down a tall, steep hill in "N" with the engine running. Once you reach 42 mph, the engine will idle faster even though no power flows to the vehicle just to keep MG1 rpm under control. The concern is if driving faster than 42 mph on EV, MG1 may try to spin the engine and that would waste charge.

    Do you have a laptop with Windows XP that can support a miniVCI? This would give you the maximum amount of engineering data. Alternatively, contact me by PM (i.e., start a conversation) as I can loan a Graham miniscanner that can send up to six, OBD data values over an RS-232 port. The miniscanner is limited to ~50 metrics while the miniVCI reads from all of the controllers.

    Bob Wilson
     
    #4 bwilson4web, Nov 16, 2015
  5. plasmaticosine

    plasmaticosine New Member

    Joined:
    Nov 9, 2015
    4
    1
    0
    Location:
    Vancouver
    Vehicle:
    2001 Prius
    Model:
    One
    Here's the battery spec, I cant post links since this is my third post lol...
    Google "2011 zero farasis datasheet", its a wordpress, the first thing that comes up for me

    Hmm I have a windows XP virtual machine, it could work. That's crazy that the graham mini-scanner reads all of the controllers, I thought what you could get was it for an OBD2 port. I just ordered one of those bluetooth OBD2 readers from ebay for $10 to work with the torque app. I get some misfires accelerating after turning so I was going to debug. Just googling the graham scanner now...

    Hmm I would rather try and trick the computer rather than disabling the gas pump... Good idea though :) I'll send you a PM Bob
     
    #5 plasmaticosine, Nov 17, 2015
  6. bwilson4web

    bwilson4web BMW i3 and Model 3

    Joined:
    Nov 25, 2005
    27,665
    15,664
    0
    Location:
    Huntsville AL
    Vehicle:
    2018 Tesla Model 3
    Model:
    Prime Plus
    Thanks,

    I found the datasheet. My NHW11 tests indicated the following limits:
    • maximum discharge - 70 A.
    • maximum charge (during braking) - 50 A. (exothermic in NiMH batteries)
    The Prius control laws monitor the traction battery during brake-regeneration and if the voltage and current exceeds some values, 80% SOC or 50 A., the brake controller engages more of the hydraulic pads. But I've not made a study to understand these limits.

    I'm going to suggest spoofing the voltage resistor network to manage the charge and discharge characteristics to your LiON pack. Higher spoof voltages, >60% SOC, should limit current and voltage into the pack. Lower spoof voltages, < 60%, will trigger charging behavior. There is a chart here that shows some idle limits:
    Prius - UPS Project

    [​IMG]
    Also:
    [​IMG]

    Bob Wilson
     
    #6 bwilson4web, Nov 17, 2015
  7. plasmaticosine

    plasmaticosine New Member

    Joined:
    Nov 9, 2015
    4
    1
    0
    Location:
    Vancouver
    Vehicle:
    2001 Prius
    Model:
    One
    Thanks for the hysteresis pics/links! Yeah, the voltage spoofing device I think would be best to have hysteresis, it would have to be microprocessor controlled IMO, so it charges to a decent amount, then shuts off for a while until necessary.

    Voltage spoofing device... or a HV battery charger. Both would be ideal (charger just for short trips, but voltage spoofer to utilize the entire SOC of the lithium pack instead of its top 50% would be awesome). The spoofer would be an odd ball thing to design, I can't think of any off the shelf cheapish parts you could use. The only thing I could think of is:
    Search on ebay "600w dc dc". It has an adjustable 12-80V output and can be run off 12V, and its isolated.

    Since the change I have gotten worse gas mileage, before averaging with 5.1L/100km and now its 6.2L/100km. It's pissed off though... red triangle. I get the P3030 "Battery voltage detective line snapped", it knows something is up. I checked it and nothing has unsoldered.... I'm running 19, 1k resistors in series. The whole board gets very warm, although I might need to run a lower impedance (thus higher power) to make the error go away. I was scoping the voltage drop across the resistors, and it actually loads each cell periodically! Pretty sophisticated. Eventually the ideal setup will have a relay which turns on the resistor network when the key is in the ignition, prior to starting the car so it will not always be draining the HV battery (and getting hot). At the moment I have a switch that runs up to the drivers seat so I manually turn it on/off.

    Also, the current sensor (hall based) has a 0V and +24V input (power) and its output is nominally 12V (when not accellerating or braking, it sways less or more depending on current flow). I was thinking of making a resistor bridge so that it uses 120% of the current it thinks its using... thoughts? I see there are fuses in the HV system, although I don't think +20% would kill anything. 70*1.2=84A, an extra few kW :)
     
    #7 plasmaticosine, Nov 18, 2015
(You must log in or sign up to post here.)