We have a 'track day' coming up in April, $30 for 3 pulls. I am thinking about taking our 03 down but I understand it is a single roller, Dynojet 224x. If the rear wheels are not moving, will the traction control come on and bugger up the results? If the rear wheel speed sensors are disconnected, with the car perform as before but with a code? BACKGROUND: Street testing has shown the Gen-3 won't product maximum engine power, 5200 rpm, until the speed is at least 76 mph. Also the acceleration curves suggest we are not getting full traction and power at low speeds. If this control law is implemented using the brake controller generated speed signal, it may be possible to trick the car into thinking it is traveling faster than 'in real life' and get more power and torque to the drive wheels a little sooner than a stock Prius. Thanks, Bob Wilson
It does. The car reaches 18 mph and no faster as the ABS tries to get the front and rear wheels in-sync. I need to return if I can figure out how to get the front and rear sensor tied together: AT BRAKE CONTROLLER AT SPEED SENSOR CONNECTORS Some sort of "Y" cable with a front female plug, front male socket, and rear male socket about 10-12 ft long. But a quick inspection of Ebay parts suggests this is not trivial: Rear wheel, speed sensor connector. Another view showing the three connectors. Front sensor uses a single pair, probably driven by the solenoid/magnet. I don't see either a Y-cable or some sort of modified facing connector making sense at this time. Either one is going to take about $100+ and probably 3-5 days of work. To make sense, there has to be a pay-off which I'll address later. DISABLE ABS Attempt to 'pedal dance' the ABS off didn't work. But my first attempt was on a slope and I hit the brake pedal midway. I'll try again also look at what miniVCI and Techstream might accomplish. Sad to say, I ran into a problem with Techstream. COMEDY OF ERRORS My plan was to record the OBD data using a Grahm miniscanner. But my original data recording system used an old, G4 Macintosh that recently had cooling fan failure. Regardless, USB-to-RS232 interfaces are fairly common, or so I thought. Office Depot had one for nearly $40. Best Buy and Walmart have them on their web site but not the local store. An independent Gigaparts had one for $20, a fair price. Cheaper are available but it would take time to arrive and this was Friday just before lunch. After lunch, I tested it with a MacBook that also has virtual PC: Installed on Windows XP and jumping pins 2 and 3, it worked fine with hyperterminal Installed on MacOS and jumping pins 2 and 3 works just fine with "screen", a built-in Unix utility However, I was also waiting for some stuff from Amazon to arrive which it didn't. I went back to the office and packed the Mac leaving my key-card badge in the computer and the USB-to-serial interface on the desk. Once I walked out of the office, it was (and is) safely in my office where I can't reach it until Monday. <SIGH> Still, I have miniVCI and Techstream. I connected it to the 03 Prius and WiFi connected it to a second MacBook. In a nice, warm home, I had full control and started working on getting Techstream ready to run. But this was also the first time I'd run it connected to the Internet. Techstream 10.x 'called home' and the 'evaluate' button stopped working. It 'bricked itself.' <SIGH> So I cloned another copy of a virtual machine with Windows and re-installed the Techstream 7.x that came with miniVCI. Then in the car with no internet access, I practiced getting it ready to record: MG1 rpm - with engine rpm, gives MG2 rpm and vehicle speed MG1 torque - divide by -28% gives the engine torque engine rpm - with engine torque, gives engine HP traction battery voltage - second power source current - times voltage gives watts in or out Good news, it mostly worked but the bad news were seeing random failures with pop-ups suggesting random memory smashed. Unfortunately, running it without stopping and restarting program does not clear the recording buffer. Worse, stopping and starting recording does not continue adding data to the buffer. Lesson learned, restart Techsteam for a single, recorded data file. One run, one file. Recording data starts with a pull-down menu that adjust the 'snapshot', data file recording buffer. Set the trigger to 'begin' and duration for 30 minutes. Select the controller; trim the data down to the variables of interest and; then start recording. When the data is captured, stop and save the file. The saved file, *.ENC, is <TAB> delimited text. Excel can open and load it without a problem. CONCLUSION All in all, a pretty disappointing time. Until I find a way to turn-off ABS, a single roller, dyno run makes no sense. If I can find a way with Techstream 7.x or the 'pedal-dance,' it makes sense to return for a dyno run. In the meanwhile, more practice with Techstream 7.x and get the Grahm miniscanner and serial interface working. Of the two, I prefer the miniscanner because it doesn't have the random memory smashes and it records data a little faster. The miniscanner captures two data points, MG1 torque and rpm, with one read, an atomic read. Reading one before and the second after the engine rpm gives more data than the Techstream and closely coupled to the important engine rpm. PAY-OFF Testing with the 2010 suggests in the first 3+ seconds the Prius control laws limit the amount of engine power generated which also limits the torque to the wheels. Since there is a speed signal generated from the brake controller fanned to the other ECUs, it may be possible to spoof this signal to indicate the car is traveling faster than actual. This might allow the control laws to add more power and accelerate faster. But how can we tell if that is the case? The solution is to measure maximum acceleration on a steep upgrade and downgrade while recording the data. If the engine power is a function of speed regardless of slope, we should be able to see it. In which case, the only approach that makes sense is to shim-in the spoof circuit at the brake controller. If I decide this makes sense, I'll also add the ABS defeat logic with a cabin enable/disable that would allow dyno verification of the effect. The pay-off is we might get our Prius to accelerate even faster from a standing stop. This is very important safety feature if three lanes soon merge into two or one and the other drivers are unskilled. Best to get in front and them sort it out among themselves behind the the Prius. . . . Yes Sir, Officer. I didn't want to risk an accidents. Bob Wilson ps. I will attempt to de-install the 10.x Techstream and re-install to see if I can regain that function. However, the 'poison pill' may still remain. If no luck, I'll put 7.x on that virtual machine and 10.x on the new one. The 10.x already paid for itself by letting me reprogram a key-fob.
Ok, I checked TechStream and could not find a way to disable the ABS. The utilities appear to handle various ABS alignment and bleeding the brakes but nothing to disable it. I also tried the 'pedal dance' and no luck. So that means I'll have to fabricate two front sensor splitters to run to the rear wheels: Need rear wheel sensor-side, connectors, two Front wheel sensor-side, connectors, two Front wheel plug, connectors, two Shielded cable Everything shows them as being passive, inductors so splitting should work. The problem is all of these parts from the 'usual suspects' are running $50-100 each. So I'm looking at nearly $300-600. I'll hunt around and see if I can find a salvage donor. The other alternative is to modify the wire harness with water-tight, connectors. This may be more affordable. FYI, the Graham miniscanner is working just fine with the USB-to-serial adapter. Bob Wilson
So this is a work in progress but my first cut at 2001-03 Prius performance: 92-94 hp net power after 70 mph ICE power increases by rpm which is speed dependent by the transaxle gear ratios ICE power limited by MG1 torque which is nearly constant for each run First run after forced charge has traction battery energy at end of run Second run without forced charge runs out of battery energy Without some way to increase MG1 torque, we can not get more power from the engine. This is probably the same limitation of all Prius. Regardless, there is no trick to getting more power unless we can increase MG1 torque and the requested power from the engine. It is only indirectly, speed related because the engine can spin faster as the car speed increases. But there is a trick that would work. Put smaller diameter tires on the drive wheels and the transaxle gears will let the engine turn faster at slower speeds. This will increase the engine power across the board and give faster acceleration. Even if we spoof the wheel encoders to simulate faster wheel speed, the MG1 torque limit and transaxle gears will still limit engine power regardless of the power request to the engine. Bob Wilson
Hopefully this will help explain what is going on: I've pulled out the maximum acceleration data and reset the time stamps to start at zero. For each launch, the car was held by the brake and then released to have the engine already running and avoid traction battery load. Significant to note: MG1 torque is nearly constant after 15 seconds - if we could figure out a way to trick the car into more MG1 torque and speeding up the engine, we could see faster acceleration. ICE rpm is gear ratio and MG1 rpm drivem - Graham Davies gave us "MG1 = 3.6 * ICE - 2.6 * MG2" which means the maximum ICE rpm is driven by MG1 and vehicle speed, MG2. So in theory, MG1 speed and torque might be reprogrammed to get more ICE power. Max ICE rpm happens at 70 mph - but this power profile is controlled by MG1. FIrst run had a force-charged, traction battery so it could provide 25 hp all the way to the end. Without a forced charge, the second run saw the traction battery fall off in the last 30 seconds. The peak, net power was ~94 hp in the first run and 92 hp in the second run. Notice the peak traction battery power does not occur with the peak engine power. BTW, this car is due for spark plugs, air filter, and oil change. Also, it has been more than two months since filling the tires and I've done nothing to lighten the weight. Bob Wilson
