How much power the SDU really makes

Topics concerning the Tesla front and rear drive unit drop-in board
Post Reply
User avatar
StefSonic7EV
Posts: 6
Joined: Sat Oct 24, 2020 9:00 am

How much power the SDU really makes

Post by StefSonic7EV »

hi all, this question was bothering me for some time. What is your experience, anybody dyno'ed your car with the SDU?

A friend of mine created this app RaceDroid Pro, coupled with 10Hz GPS (Garmin Glo) I have done some tests.

850kg car's weight (with driver), I went for 10% drivetrain loses and 0.45 Cd x 2m2 frontal area.

Result below. What I'm getting is around 190 horsepower. Close to 200 when battery with high SoC, 170 horsepower when closing 30% SoC, much less below 30%.
What I don't understand is why only 190 hp. I'm using Torque app and OBDII device that connects to the Orion BMS. It shows 500 amps current and 370 volts on the peak. That should be giving 185kW so 250 hp. I will measure current and voltage on inverter and on BMS app directly, maybe that is voltage sag...
Attachments
Screenshot_20201025-154854_RaceDroid Pro.jpg
Screenshot_20201025-154731_RaceDroid Pro.jpg
collin80
Posts: 110
Joined: Sun Aug 30, 2020 3:28 pm
Location: United States, Michigan
Been thanked: 4 times
Contact:

Re: How much power the SDU really makes

Post by collin80 »

This might be due to the way an AC motor works. You're taking the peak amperage and voltage from the DC side. 500A * 370V does equal 185kw and that is about 250HP. However, AC motors need sine wave drive (not strictly true but sort of) so the inverter must chop up the DC voltage and create fluctuating voltage. If the voltage were a true sine wave then the effective AC voltage given an input DC voltage is DC / sqrt(2). This yields about 260V AC. 500 * 260 is MUCH closer to the power you're seeing (175HP). However, things are not so simple since the motor is three phase. I actually don't remember the equation for power conversion of DC to three phase. But, I think the fact that the drivetrain is AC while you are measuring DC might explain why the readings look different. You need to know the actual AC amps and volts as reported by the inverter to get a true idea of the output power.
Isaac96
Posts: 656
Joined: Sat Oct 05, 2019 6:50 pm
Location: Northern California, USA
Been thanked: 1 time
Contact:

Re: How much power the SDU really makes

Post by Isaac96 »

collin80 wrote: Sun Oct 25, 2020 4:48 pm This might be due to the way an AC motor works. You're taking the peak amperage and voltage from the DC side. 500A * 370V does equal 185kw and that is about 250HP. However, AC motors need sine wave drive (not strictly true but sort of) so the inverter must chop up the DC voltage and create fluctuating voltage. If the voltage were a true sine wave then the effective AC voltage given an input DC voltage is DC / sqrt(2). This yields about 260V AC. 500 * 260 is MUCH closer to the power you're seeing (175HP). However, things are not so simple since the motor is three phase. I actually don't remember the equation for power conversion of DC to three phase. But, I think the fact that the drivetrain is AC while you are measuring DC might explain why the readings look different. You need to know the actual AC amps and volts as reported by the inverter to get a true idea of the output power.
Well... I'd say not quite. P = VI. So 500A DC at 370V DC on the input means 700A at 260V AC output. Power in must equal power out unless you've got a resistive controller.

OP, I don't know why your power levels are lower when measured by GPS. Your drag may be higher than you have calculated, or the 10Hz GPS might be too slow. Can that app use wheel speed sensors or the phone's accelerometer? Either of those might give more accurate results.
User avatar
StefSonic7EV
Posts: 6
Joined: Sat Oct 24, 2020 9:00 am

Re: How much power the SDU really makes

Post by StefSonic7EV »

Let me draw the udc/idc at the inverter to get the accurate reading. Hard to argue with P=VI. I guess minus ~5% as inverter and motor have some losses. But not 40kW.

App works with 10Hz GPS really well for ICE cars, close to dyno numbers. Guess I need to go on a dyno to calibrate for electric powertrain.



Isaac96 wrote: Sun Oct 25, 2020 5:14 pm
collin80 wrote: Sun Oct 25, 2020 4:48 pm This might be due to the way an AC motor works. You're taking the peak amperage and voltage from the DC side. 500A * 370V does equal 185kw and that is about 250HP. However, AC motors need sine wave drive (not strictly true but sort of) so the inverter must chop up the DC voltage and create fluctuating voltage. If the voltage were a true sine wave then the effective AC voltage given an input DC voltage is DC / sqrt(2). This yields about 260V AC. 500 * 260 is MUCH closer to the power you're seeing (175HP). However, things are not so simple since the motor is three phase. I actually don't remember the equation for power conversion of DC to three phase. But, I think the fact that the drivetrain is AC while you are measuring DC might explain why the readings look different. You need to know the actual AC amps and volts as reported by the inverter to get a true idea of the output power.
Well... I'd say not quite. P = VI. So 500A DC at 370V DC on the input means 700A at 260V AC output. Power in must equal power out unless you've got a resistive controller.

OP, I don't know why your power levels are lower when measured by GPS. Your drag may be higher than you have calculated, or the 10Hz GPS might be too slow. Can that app use wheel speed sensors or the phone's accelerometer? Either of those might give more accurate results.
Isaac96
Posts: 656
Joined: Sat Oct 05, 2019 6:50 pm
Location: Northern California, USA
Been thanked: 1 time
Contact:

Re: How much power the SDU really makes

Post by Isaac96 »

Idc at inverter is not accurate (calculated from the AC currents), udc is generally good though.
slow67
Posts: 151
Joined: Thu Jan 30, 2020 6:52 pm
Location: Texas
Been thanked: 1 time

Re: How much power the SDU really makes

Post by slow67 »

It probably is voltage sag. How fast does the BMS report voltage and current?
User avatar
ManuFromParis
Posts: 31
Joined: Sat Feb 15, 2020 6:34 pm
Location: Near Paris

Re: How much power the SDU really makes

Post by ManuFromParis »

I don't know yet, because we haven't rolled it on a bench, but behind the steering wheel the human feeling is much higher than the OEM Porsche 911 SC which has 200hp and around 220Nm torque.

I guess it's linked to the battery pack characteristics as well (total V, discharge perfs, etc...)
911 SC with Tesla SDU & i3 pack
jon volk
Posts: 572
Joined: Wed Apr 10, 2019 7:47 pm
Location: Connecticut
Been thanked: 2 times

Re: How much power the SDU really makes

Post by jon volk »

I guess it comes down to how confident you are in the voltage/current sensor data accuracy.
In terms of power, calculations based on weight and acceleration over a known distance (i.e. 1/4 mile trap speed) are, to me, the foolproof standard. Particularly since you will achieve pretty consistent trap speeds regardless of the launch. Sensors can have offset errors, dynos can be manipulated in the software with a slew of correction factors, but you cant change the physics that it takes x power to accelerate a mass over a set distance.

If i had a dollar for every 3500lb car that claimed to dyno 1000 hp made dismal pass at 135mph, Id be retired.
Formerly 92 E30 BMW Cabrio with Tesla power
User avatar
StefSonic7EV
Posts: 6
Joined: Sat Oct 24, 2020 9:00 am

Re: How much power the SDU really makes

Post by StefSonic7EV »

I have done some extra testing to figure it out. As Isaac96 was pointing out, the readout from the idc at the inverter was not accurate (BMS data on the left):
Screenshot 2020-10-31 at 11.26.43.png
Udc looks to be correct.

I tend to think that BMS readout is correct. For this run Race Droid Pro has shown 187 hp.
Screenshot_20201031-101317_RaceDroid Pro.jpg
Screenshot_20201031-171501_RaceDroid Pro.jpg
Report based on the BMS log:
Screenshot 2020-10-31 at 17.08.27.png
shows the current at the peak power moment was 450 amps. Voltage sag is massive though, dropped from 374 to 345. So 155kW peak power, 208hp.
Since Race Droid showed 187 hp, BMS logged power 208hp minus 5% for inverter/motor loses gives 198 hp. Even if the error of the Race Droid and BMS measurement is only 5%, we are in the ballpark.

So clearly looks like voltage sag is the answer on why I get the power of 155kW instead of factory 210kW.

Anybody with stronger battery pack could post some logs? I wonder if with a better pack the motor can really handle 210kW.
Sparky 61
Posts: 8
Joined: Tue Oct 15, 2019 3:28 am
Location: AUSTRALIA

Re: How much power the SDU really makes

Post by Sparky 61 »

Hi after twelve months final have the the little red race car ready for this years pilgrimage to Speed Week Down Under, interested in the set up for the controller, are you using the standard set up or have you changed any of the parameters to get more out of the motor, I'm looking for the highest top speed I can achieve.
Post Reply