Thread Tools
Oct 10, 2009, 05:00 AM
Registered User
Discussion

Analysis of a REEM inadvertently used in a reverse power flow condition.


REEM “Reclaimed Externally Excited Motor”

The EBAY advertisement made no sense. All turn eight or something along those power lines.

The motor was manufactured by the DENSO Corporation in Tennessee. Some “genius” at Toyota connected the motor through an un-inverter in a reverse power flow condition.

It does not seem to have hurt the motor, but it does seem like some of the windings have gotten hot and the varnish has discolored or maybe oil got on the windings and left a residue.

The cost of the motor was $30.00 with an additional $14.00 for shipping.

I plan to put this motor in an RC airplane. The airplane will not have any propeller. I plan to use some type of mechanical transmission system connecting the motor to the tires on the landing gear. Power will be transmitted to the tires, and the airplane will be taxied a very long way. The wings will be retractable, and look suspiciously like arms with control surfaces that resemble hands. Once at takeoff speed, and simultaneously taxiing over a takeoff bump, the wings will be extended for a very short flight.

This thread is about the motor as a motor. We should all refrain from analyzing my aerodynamics decisions. I am just going to provide motor constants for this motor and people can design it into there own RC airplanes, or not, as they like.

Also, I want to support my country. The motor is made in Tennessee after all. So we should refrain from silly thoughts like installing permanent magnets because we all know that they come from China.

I specifically looked for one of the DENSO SC Alter… umm, motors because people here think that they are the “best”. It has rectangular windings and a high fill factor. The rating from Toyota was 100 amps.

The motor as received, weighed 12.5 lbs as measured on my high precision bathroom scale. After I un-installed the un-inverter, the motor weighed 11.9 lbs as measured by my high precision bathroom scale.

The lamination thickness is 0.35 mm I actually measured 0.345 mm but 0.35 seems to be a standard thickness so I am going with that.

I had read that this motor had 6 phases but that is not the case. The motor is a three phase machine, but both ends of each phase stick out the back. I think this is why some people thought it was 6 phases. The motor as used by Toyota was connected delta. I measured the motor wire to motor wire resistance as 0.0358 ohm. This puts the phase resistance at 0.0537 ohm, and the motor wire to motor wire resistance at 0.107 ohm for a wye terminated motor.

I measured the field winding at 2.47 ohm

The bearings have a lot of drag compared to any RC motor, but I would expect that considering the large sealed bearings.

Last thing for tonight is I powered up the field with 2 amps of current and you could certainly feel the hysteresis drag but there is no perceptible cogging.

I will post some pictures tomorrow.

In the future I hope to measure bearing and windage torque as a function of RPM, as well as the hysteresis drag torque and eddy current damping as a function of field current. That combined with the motors emf constant as a function of field current will allow us to construct good performance data for the motor under a variety of operating conditions.

Last last thing for tonight, is everyone needs to be aware that using this type of motor with our hobby ESC is very risky! If the motor looses the field current for any reason, the emf will go away. With the emf gone, it is the same as a stalled motor. If you are connected to a 48 volt source and you have 0.0345 ohm you will try and pull 1300 amps, and that means by by ESC. The power amplifier or speed control has to have current limiting, or at the least over current detection with a cutoff.

It is true that installing Chinese magnets will prevent that phenomena, but there are reasons why you might not want to in taxi operations that occur in lots of different conditions. Once we start producing efficiency plots we can look at that.

Dan

P.S. I am not saying anyone should use one of these. My curiosity was aroused while reading the transcripts of a mud wrestling competition here on RC groups, and it was worth $45 and a few hours of tinkering to get some real, or at least good honest attempts, at some real data.

Besides, after I fly my RC airplane, I plan to sell the motor on ebay for $180.
Sign up now
to remove ads between posts
Oct 11, 2009, 02:29 AM
Registered User
No new measurements today because I went hiking with the wife and some of her friends.

One observation though, is that there are two cooling fans. One is in the front and one is in the rear. Air is sucked in through the front and rear for cooling and is exhausted radially over the end turns of the windings. The fans are not symmetric and the motor is meant to run clockwise when viewed from the front, for maximum cooling.

I would like to get a few measurements with a thermister on the field winding, so I have an idea how much field current I can run without toasting things.

OK, going to sleep at a reasonable time tonight.
Oct 13, 2009, 03:06 AM
.................
Dylwad's Avatar
Good stuff, maybe a bit too serious for Bob though

I thought Denso was Japan?
Oct 14, 2009, 01:11 AM
Dave North
timocharis's Avatar
Quote:
Originally Posted by Dylwad
I thought Denso was Japan?
DENSO Manufacturing Tennessee
www.densocorp-na-dmtn.com
Location:
1720 Robert C. Jackson Drive
Maryville, TN 37801-3748

Founded:
July 20, 1988

Products/Business Activities:
Manufactures starters, alternators, instrument clusters and center displays, automotive electronic products, including electronic control units, airbag sensors and controls and keyless systems

Key Management:
- Yoshihiro "Van" Saka, president
- Masayuki "Mick" Yokoyama, executive vice president
- Tomoyasu "Tommy" Nakamura, executive vice president
- Jack Helmboldt, senior vice president, Body Electronics Divisions

Employees:
2,512 (as of March 31, 2009)

Ownership
100% DENSO International America, Inc.
Oct 14, 2009, 02:31 AM
Registered User
Quote:
Originally Posted by Dylwad
Good stuff, maybe a bit too serious for Bob though
Shhh, be very very quiet!

Quote:
Originally Posted by timocharis
DENSO Manufacturing Tennessee
Ownership
100% DENSO International America, Inc.
Wow, who would have thought! I thought it was Japanese owned with a manufacturing plant in the USA.

I measured the winding inductance today. For a wye terminated motor with one phase left open, as would be the case when driven six step ESC. I measured 80 uH. I am going to redo the measurement because it just seems way too small. I must have messed something up.

It will be a little while before I get the performance plots up. I will be able to measure the emf constants easy enough, but I need to make a flywheel of known inertia so I can measure the loss torques. I am essentially going to make an inertia dyno to measure the loss torque as a function of rpm and field strength.

I might just hook this thing up to an ESC and grab some Io / RPM numbers at a couple of voltages and field strengths to get some initial estimates.
Oct 14, 2009, 12:58 PM
Dave North
timocharis's Avatar
Quote:
Originally Posted by IneptOne
Wow, who would have thought! I thought it was Japanese owned with a manufacturing plant in the USA.
With executives named "Van" and "Mick" and "Tommy?" Come on, pal...


Dave
Oct 17, 2009, 10:03 PM
Registered User
I finally had a chance to measure the voltage constants for the Denso alternator / motor. It is going to be a while more before I can measure all the bearing and core losses. One thing is for sure this thing can be a torque monster.

Even without measuring the core losses I can tell you that 3700W or 5HP at 2400 RPM is possible.

The voltage “constant” Ke seem to be a fairly linear function of field current up to 2 amps. By 2.5 amps it is starting to saturate. I am leery of running a higher field current until I can get a thermister on the field windings.

Km for this motor is 0.5 for a field current of 2.0 A. For comparison the big torque monster Scorpion S5545 has a Km of 0.37
Last edited by IneptOne; Nov 02, 2009 at 02:16 PM. Reason: Corrected Graph and Km
Oct 30, 2009, 12:27 AM
Registered User
OK. I know I have not been posting any info about the Denso SC alternator, but I have not been totally idle. I wrote a program that generates efficiency maps in preparation for all the data to be generated.

For development purposes I used a Scorpion 2208-X. The –X is because these plots are independent of winding assuming constant fill factor. This plot is useful in helping decide what Kv to wind for. You pick a power level you want, and then you follow the power curve to a nice efficient RPM and torque combination for the particular motor.

On this plot is overlaid the static power curve for an APC 6-4E and the curve for 30MPH

I still have lots to do on the program but it makes pretty pictures so I thought I would share what I have been up to.

You can be sure I will be plotting the Denso-SC Alternator!
Oct 31, 2009, 03:34 AM
Registered User
I added an efficiency scale to the efficiency map output. Also I forgot to mention that the brighter region to the lower left is OK for continuous operation while the darker shaded region is for intermittent operation only. The black line from the lower left corner is the peak efficiency line. From the intersection of a power output curve and the peak efficiency curve you can drop down to the x axis to read the RPM for maximum efficiency for the given power level. I also added a marker at 600g of static thrust.
Nov 02, 2009, 03:18 AM
Registered User
Ok then. I finally made some measurements of the Denso Alternator. I generated an efficiency map for a field current of 2.0 A. The bearings are really killing the efficiency at lower power levels. If I was going to use this thing on a bike, I would seriously look at some lower friction bearings.

The peak efficiency is not bad 87.4% at 3.3 KW output power. One thing to note is the maximum output power on the plot is based on the power dissipation that occurs at 100A which is the rating for the alternator. I also let the power dissipation climb to 150% at 8000 RPM because the fans provide extra cooling. I guess I am saying take the maximum continuous power rating with a grain of salt. I see ratings on some model motors that are using a lot higher dissipation / surface area.

Lastly the plot is not currently taking into account thermal effects on resistance. I would want to measure some temperatures vs dissipation and fan RPM before I take a stab at that. But be aware the efficiency is probably a little worse than the plot shows at the highest power levels.

The plot is taking into account the power required to power the field coil.

Oh, I almost forgot. The plot for Kt and Kv as a function of field current had an error in it. I accidentally clicked on the wrong cell and the values were off by a factor of two. I fixed it and replaced the picture in the original post.

I also provided a plot that is a vertical slice through the efficiency map showing 2200 RPM. This is where the motor peak efficiency is for 1000W output power. The 1000W power level is of interest to some e-bike people.
Nov 04, 2009, 12:17 AM
Registered User
Here is a plot at 4800 RPM
Nov 04, 2009, 01:43 AM
Registered User
Here are some efficiency maps at a few different field strengths. I plotted them all with the same ranges so they can be compared directly. Notice that the maximum output power does not change all that much but the RPM for maximum output power changes drastically. The variable field lets you gear the top end for a very high RPM with lower field. This implies a higher gear ratio. Now at low speed, the field can be cranked up for more motor torque. This maximum motor torque combined with the higher gear ratio gives better acceleration than can be had if the motor were only used at maximum field and geared for the resultant lower max motor RPM.

A second benefit is that the motor efficiency for any power level can be optimized for the operating RPM. You use high field currents when you need max power for acceleration, but once up to speed and you need a fraction of the power for cruising, the motor would be very inefficient if the field were left at the maximum value. If the field is reduced, the core losses drop for a given RPM. Look at the 2A efficiency map at 4800RPM. At 3kW output power the efficiency is in the mid 80s, but say you drop back to 500W for cruising, the efficiency drops to around 65% or so. Now look at the 1A efficiency map, 4800RPM and 500W output power. The efficiency for this case is around 78%. A map of field current vs RPM and Power Level can be created for maximum efficiency. The variable field motor has a much wider operating area than a permanent magnet motor! The downside is more complicated controls for optimum operation, less reliability because of the field brushes and of coarse the power required for the field. The field power is really only an issue at the lower power levels. At the highest output powers the field power is insignificant.

I mentioned that the bearing friction is really hurting this motor at the low power levels. It is also hurting efficiency at low field currents. I have included an efficiency map with the bearings at half of there current drag value, for the 0.5A field case. You can see the efficiency really improves. There is less dramatic improvement at the higher field currents.

Now I am not advocating this “motor” for our model airplanes, but you can see that the variable field concept can better match the motor to the prop over a wider operating area. For instance, instead of a constant speed prop like on a full scale airplane, you can have a variable Kv motor to keep the prop and motor in the sweet spot, despite changes in airspeed or altitude.
Last edited by IneptOne; Nov 05, 2009 at 04:33 PM.
Nov 04, 2009, 09:46 AM
Registered User
Your results seem similar to the characteristics of a wound rotor ac induction motor.
Nov 04, 2009, 06:33 PM
Dieselized User
gkamysz's Avatar
What software are you using to generate those plot graphics?

Greg
Nov 05, 2009, 01:10 AM
Registered User
Quote:
Originally Posted by salty9
Your results seem similar to the characteristics of a wound rotor ac induction motor.
Not exactly the same, but similar in that the motor can be optimized for low speed operation of high speed.

It is funny you should mention induction motors, as the Wrightspeed and Tesla electric cars, both use induction motors because the partial load efficiency is higher “as well as cost savings from the magnets” The induction motors are not wound rotor machines, but the control comes from inverter algorithms that carefully adjust slip speed to provide the correct rotor current.

The Alternator Motor achieves the variable magnetic field directly, and uses simpler control methods. The downside is the requirement of brushes and small additional circuitry to control the filed current. The Alternator Motor is easier to use for a hobbyist until the required inverters become more common place for traction applications.


Quick Reply
Message:

Thread Tools

Similar Threads
Category Thread Thread Starter Forum Replies Last Post
Can a Marine/RV battery be used in a car? Ben74 Power Systems 14 Feb 02, 2018 10:04 PM
Which Hacker should be used in "E" Turbax supersluf Electric Ducted Fan Jet Talk 25 Nov 20, 2006 10:02 PM
Foam used in Wattage Hawk? Depron? or? Balin Foamies (Kits) 0 Apr 15, 2002 01:51 PM
ESC 600 speed used in 400 speed dancer Power Systems 6 Oct 17, 2001 04:43 PM
Can a folding prop be used in pusher config? MtnGoat Power Systems 2 Jul 26, 2001 10:44 PM