Measuring Prius thrust from MG1/MG2 torque
 

It looks like using just MG1 torque, MG2 torque, and the gear ratios, we can figure out the torque to the wheels and vehicle thrust. This works because the ICE torque is matched by MG1 torque so the ICE torque is not needed. Dividing the final wheel torque by the wheel radius provides a direct readout of the thrust matching the vehicle drag.

My gear ratio sources are:
http://prius.ecrostech.com/original/...plitDevice.htm
http://www.autoshop101.com/forms/Hybrid02.pdf

1) -2.6 * MG1 : MG1 torque ratio to ring
2) (-2.6 * MG1) + MG2 : sum of torques
3) 3.905 * ((-2.6 * MG1) + MG2) : torque to axle ratio
4) (3.905 * ((-2.6 * MG1) + MG2)) / .2921 : N from tire radius

The obvious advantage of this approach, if I haven't made a mistake, is a nearly direct readout of vehicle thrust and by implication drag and inertial load. With this formula, I've measured 193 N at a cruise control maintained, 25 mph (11.17 m/sec) at a temperature of 75F/24C over a distance of 10 miles/16 km on a 1.1 mile closed loop (1,315 samples.) But my measured thrust is lower than the NHW11 drag formula:

193 ~= 242 N : NHW11 drag formula 190 + (0.42*(V*V))

I'm running Sumitomo T4s at 51 psi and the rear wheels are metal tab aligned to 0.03 toe and equal 1.1 camber. Also, I'm using lower friction Type WS in the transaxle and the vehicle was throughly warmed up. I plan to take out my rear wheel tabs and repeat the test prior to getting the alignment checked and swapping in the EZ Shims 75800. This will let me quantify the toe and camber effects.

So what happened to the ICE torque? It only exists because MG1 provides the counter torque. Because of this, we only need MG1 to account for it.

Bob Wilson

ps. As part of auditing this formula, I plotted the wheel thrust over the laps and included MG1 torque as a refernce:


http://hiwaay.net/%7Ebzwilson/prius/pri_model_010.jpg
Note how when the ICE kicks in under cruise control, it substantially increases the thrust to return to the cruise control set speed. This happens when local rises in the track trigger the ICE to start, resume speed and then run until a downgrade allows the ICE to shutdown.

Re: Measuring Prius thrust from MG1/MG2 torque
 

Why not find a level road, shift to N at some speed and start coasting, measuring speed at time intervals. Then see how well it curve fits to the solution of the diffyQ v'=-k1-k2V^2. Then, of course, P=FV.

Re: Measuring Prius thrust from MG1/MG2 torque
 

Most who have tried that approach called "coast down" have found the results too unreliable. Worse, an absolutely flat, 'open road' is a hard to find resource in my area. You'll notice when I went to validate the data I used a large number of samples, 1.3k, which nicely evened out the variable elements. Those with money use a dynometer for reproducible results.

I have two other closed loop, 'test tracks' I can use that have greater altitude changes. The proof will be getting similar values from the other two tracks at the same temperature for the same distance, at the same speed.

Bob Wilson

Re: Measuring Prius thrust from MG1/MG2 torque
 

Great thinking Bob. A DIY dynamometer and plenty of fuel-saving ideas to test with it. For a nearly useless comparison, last week's large earthquake in China released 1.4 X 10^16 Newton-meters, according to the US Geological survey.

It would be neat if you could simultaneously log with a sensitive accelerometer, to distinguish drag from this factor. You just happen to have one of those on the bench, right?

DAS

Re: Measuring Prius thrust from MG1/MG2 torque
 

MG2 rpm ... <grins>

Altitude remains the 'hard problem' due to GPS variability. I'm good for a couple of meters but only if I get bunches and bunches of readings but that is coming for some routes.

Bob Wilson

Re: Measuring Prius thrust from MG1/MG2 torque
 

Funny, the coast down gives me pretty good results. Lots of flat roads around here. Plus, the v^2 component comes on sooner and stronger in my FEH than a Prius I guess. Also, I am relying on PIDs to get the motor and generator speeds and torque. The speeds are accurate, but I'm not so sure about the M/G torques, although they seem reasonable. The engine torque PID gives huge torque readings, which are clearly incorrect.
CC will only engage at 30PH and up, and when the ICE kicks on it will stay on indefinitely absent driver intervention.

Re: Measuring Prius thrust from MG1/MG2 torque
 

This free topo tool was recently brought to my attention:

http://www.toporoute.com/routePlanner.html

I have not tried to assess its accuracy or ease of use at all, because its coverage does not include my country of current residence. Say no more, wink wink nudge nudge.

DAS

Re: Measuring Prius thrust from MG1/MG2 torque
 

A question of this newbe:

The Prius torque of MG1 is rather high compared to normal gasoline eingines (150Nm typically here in europe). Due to the transmission at the wheels it seems to be about 600Nm max. I have not calculated it yet but I can imagine that the torque at the wheels of a normal gearbox car will be much higher, or am I overlooking something?

Re: Measuring Prius thrust from MG1/MG2 torque
 

Hi Rob,

Welcome to GreenHybrid.com and an excellent question:
I'll give more details later but the summary is:

• MG1 applies 28%, counter torque to the internal combustion engine (ICE) torque during normal operation
• I have a lot of Graham scanner data of typical MG1 torque values they range over a wide range depending upon the ICE rpm and power output
• You are correct that the wheel torque is in the 600 Nm range
• Another approach is to use mass, tire diameter, and acceleration to calculate the wheel Nm

Sorry these are not metric units but Google can lookup the conversion factors and calculate the metric units: 2,700 pound weight, 918 tire revolutions per mile, and maximum acceleration:
http://hiwaay.net/%7Ebzwilson/prius/pri_100mph.jpg
I've got to do some other tasks right now but I'll try to update this note later with metric units and some ICE and MG1 data.

Understand that this data is for an NHW11, the 2001-03 Prius. The current model is much faster and a little heavier. BTW, some of the USA hot-rod sites allow you to insert the USA units and read out the torque.

Bob Wilson

Re: Measuring Prius thrust from MG1/MG2 torque
 

The max torque of the Prius at the wheels seem to be about 600Nm.
When I make a simple calc for an average european family car:
- max torque at crankshaft about 150Nm
- total reduction in 1st gear about 1:12
This gives about 150x12=1800Nm at the wheels.
(I hope I dont overlook something)
So the thrust of the Prius is relatively low.
Could this be the reason that it's not allowed to tug a caravan or so?