Hot-rodding the Toyota MGR

[thingstodo]

Originally Posted by e*clipse

With a dyno, we still need a HUGE load dump. Pump water? Heat water?

That's why I suggested 2 MGRs and 2 controllers. The first is in motor mode, the second in regen or generator mode. Connect to the same battery pack. If you are removing 50 kw and regen of 40, you only need to 'top up' 10 kw to keep the system running.

Or you could run for a while just on battery.

One thought I had was to just build the car in my case with just the front motor. Build the controller so is could supply about 50% more than the rated power ( 75 kW?? ) Measure everything really well, and go do some slow drag races. F=ma. Step the controller's power limit up each run, and allow a LONG cool-down time.

And monitor the heck out of everything? Sounds like a workable plan.

url=http://www.powerstream.com/Wire_Size.htm]American Wire Gauge table and AWG Electrical Current Load Limits with skin depth frequencies and wire breaking strength[/url]
for 0.032, AWG20 wire: 10.15Ohms/1000ft 11A max chassis 1.5A max power 27 kHz
for 0.0285, AWG21 wire: 12.8Ohms/1000ft 9A max chassis 1.2A max power 33 kHz

I'm assuming "power" wiring is continuous, "chassis" wiring is for short-term bursts. I'm assuming for motor use, they're using someplace in between because these wires are bundled and can't dissipate heat like a single strand.

I'm not sure about the 'power' column. I've never seen anyone use currents that low for the cables involved. For example, #16 lists 22A for chassis and 3.7A for power. That's your house wiring. It'll take 15A all day, every day. #14 is used for outside plugs - 32A chassis, 5.9 for power.

[e*clipse]

d'Oh!

Of course! Since the BEMF is one of the main "problems" With the BLDC motor, you could probably get away with one controller and one 3 phase rectifier. Hmmmm, I wonder about stability - would there be any wierd resonances when feeding something back to itself like that.

Regarding the "power" column, you're right - I'm having an impossible time corroborating any of those numbers. Another company gives the "enclosed" value of roughly 50%:
AWG20: diameter: 0.032" 10.15Ohms/kft 11.0A max 7.5A enclosed
Wire-Gauge Ampacity - Transwiki

Wikipedia had a great NEC chart, unfortunately no ampacity numbers below AWG14:
American wire gauge - Wikipedia, the free encyclopedia

I looked to see what ORNL did with the Prius tests. One thing that was VERY interesting is the 2010 Prius motor used the same number and gauge of wire per phase:
Number of wires in parallel: 12 (chart on pg. 46 of 2010 Prius evaluation)
Wire size, AWG: 20 (chart on pg. 46 of 2010 Prius evaluation)

They then did a huge test on the magnets, and then set up the motor on a dyno with a bunch of thermocouples. The locked rotor test shows the torque with current and electrical position. This is graphed on pages 60 and 61, I'll list the peak results:
75A 120 degrees (electrical) 80 Nm
100A 135 degrees (electrical) 110 Nm
125A 135 degrees (electrical) 130 Nm
150A 135 degrees (electrical) 155 Nm
200A 135 degrees (electrical) 195 Nm
250A 135 degrees (electrical) 215 Nm

These locked rotor tests were done with a DC input current for just enough time to measure the torque. Similar tests were done with the Camry and Lexus LS 600h at much higher current levels (note- they have different wiring).

Note that the rotor and stator dimensions are different from the MGR we're using, BUT the wiring is the same. Thus, we can't accurately use the torque numbers, BUT consider how much MORE current (250A vs 108A) they used for the test.

- E*clipse

Originally Posted by thingstodo

That's why I suggested 2 MGRs and 2 controllers. The first is in motor mode, the second in regen or generator mode. Connect to the same battery pack. If you are removing 50 kw and regen of 40, you only need to 'top up' 10 kw to keep the system running.

Or you could run for a while just on battery.



And monitor the heck out of everything? Sounds like a workable plan.



I'm not sure about the 'power' column. I've never seen anyone use currents that low for the cables involved. For example, #16 lists 22A for chassis and 3.7A for power. That's your house wiring. It'll take 15A all day, every day. #14 is used for outside plugs - 32A chassis, 5.9 for power.

[freebeard]

Originally Posted by thingstodo

*MUCH* simpler than what I was thinking. I LIKE IT!

I think I'd still use 2 MGRs and 2 controllers if I had the choice, with the sprockets and the chains.

I'm thinking you'd still need the intermediate shaft, unless you lock the differentials. Maybe.

Originally Posted by e*clipse

I know that's no help for ***engineering*** stuff, but it would help for my multiple motor design.

“If we knew what we were doing, it wouldn't be called research” ― Albert Einstein

[Cobb]

Wow, careful not to wipe the charge from the magnet.

Originally Posted by e*clipse

75A 120 degrees (electrical) 80 Nm
100A 135 degrees (electrical) 110 Nm
125A 135 degrees (electrical) 130 Nm
150A 135 degrees (electrical) 155 Nm
200A 135 degrees (electrical) 195 Nm
250A 135 degrees (electrical) 215 Nm

These locked rotor tests were done with a DC input current for just enough time to measure the torque.

[e*clipse]

Exactly!

I think that's why ORNL did the coercivity tests first. Although some of their tests involved sawing things in half, so they probably weren't as concerned about breaking stuff as we are.

The other thing that got me is that 250A/ 12 wires = 20A through a 0.032" diameter wire! This is more than 2X as much as the maximum NEC ampacity rating. I'd like to know how long they left the current on to do the tests. I mean, if they did the tests in 15 seconds, that's easily enough time to max out the speed this motor will push a car.

They did note that these currents were only possible for a very short amount of time. They specifically warned that the wires were bundled within materials that couldn't dissipate the heat very well.

- E*clipse

Originally Posted by Cobb

Wow, careful not to wipe the charge from the magnet.

[e*clipse]

Oh, and an another thought (which relates to the hot-rodding philosophy )

The graph of the Prius motor's torque vs current is pretty linear. Proportionally more current produces proportionally more torque.

The power equation is linear as well...

So, we may not know the exact numbers for the MGR we're using, BUT we do know it's construction is the similar.

See below: We may be doing well to ACTUALLY USE IT at the published specs...

Edit - I did some double checking on ORNL's tests vs Toyota's published specs. I find this very interesting: ORNL merely tested it a LITTLE BIT beyond the motor's published data! Seriously, the torque rating of the Prius motor is 207 Nm!

So, it looks (to me ) like Toyota pushed it hard for some marketing specs, but I seriously doubt the controller lets people drive it like that.

[cRiPpLe_rOoStEr]

I'd be more concerned about the cooling. Wondering how expensive it would be to convert the casing for water-cooling...

[e*clipse]

Originally Posted by cRiPpLe_rOoStEr

I'd be more concerned about the cooling. Wondering how expensive it would be to convert the casing for water-cooling...

Totally agree!
In fact, one of my immediate thoughts when I got one was WTF - no cooling??
Then, I started to see that the oil was used for cooling. In fact in one of the ORNL tests they pointed out how it is really good that oil spray is used to cool the stator windings. At that point, I was just dissappointed that they made no provisions for an external oil cooler. I mean, my Honda Insight uses part of the radiator as an external oil cooler for the CVT transmission.

Regarding adding water cooling, I think it would be tough. The motor casing is very lumpy - its shape accurately replicates whatever is at that place - bearings, bolt holes, the motor stator, etc.

If you have ideas for adding cooling to a lumpy shape - I'd love to hear them.

[e*clipse]

Hey freebeard!

Any news from the salt??

- E*clipse

[MPaulHolmes]

I'm in Arizona. I really need to check that place out. It probably isn't that far of a drive. It could be murder on my high blood pressure though, if I decide to eat the ground.