Re: Direct Drive Revisited

[Todd H <toddh@execpc.com>]

Robert McDonald wrote:

> Hi Todd,
>
> There have been many interesting discussions regarding the use of auto
> trannies in the past, if you search the archives you will come up with
> some useful info.  The basic run down is that autos without the torque
> converter are very efficient.  An auto with a locking torque converter
> is almost as efficient as a manual.  Regen in drive is not possible due
> to the sprag clutch.  So if regen is implemented with an auto you need
> to shift down to 2nd or 3rd or whatever is below drive to get regen.
>

Torque converters, sprag clutches that's mechanical stuff aren't they??  :-)
I'm an EE. You lost me.  I can make an micro sing, but a transmission is
another story!

>
> I am from Brisbane, Australia and here DC motors and controllers cost an
> arm and a leg.  AC induction motors are dirt cheap (I just just picked
> up two EV sized motors for US$67 each!  BRAND NEW!  It was admittedly
> from an auction though. :-) )
>
> I am an EE student and I have my final year thesis to do next year, so I
> will be building a 3phase AC induction motor controller.  I have been
> researching it for about 6 months and am starting to get my head around
> all the considerations involved.
>

Lets talk!!  I have designed production 1/8th HP PM PWM motor controls in the
past.  As I also stated to Lee Hart, I work at Miller Electric = welding
equipment.  High freq switching welders are very similar to EV controllers.
400 V bus input to 1 to 1000 Amps output with >1000 A/msec response.  I do
not design the switching supplies, but I know a lot about them and know who
to ask.  I specialize in microprocessors & software & real time control of
the welding process.  Just talking about this project at work, I have gotten
a couple of others interested.

>
> Your conversion path depends entirely on your priorities.  If you want
> to spend the least amount of time and money to get a reasonable EV, then
> leave the auto in and convert to DC.  If the 1990 ranger has a locking
> TC then you should not lose much range over a manual conversion.  If it
> does not have a locking TC you'd lose about 15%-20% range over an
> identical manual conversion (unless you remove the TC altogether)

> Depending on the type of tranny in the rangers, you may have to alter
> shift positions and TC lockup rpm.  If it is uP controlled (likely for a
> 1990 model) then you'd have to find someone knowledgeable on this
> particular transmission to help you out.  If the TC is removed, an
> auxiliary pump is needed to provide hydraulic pressure to the clutches
> in the planetary gearset inside the auto.

I will look into it.  But I'd like having something simple rather then duct
taped together!

>
> Direct drive is EXPENSIVE!  It can be done very successfully but it
> takes much bigger motors, controllers and battery packs than are
> required for an EV with a trans.
>
> AC induction is also EXPENSIVE!  But if you build your own controller it
> would not have to cost all that much more than DC and you would likely
> end up with a more efficient vehicle, hence better range and
> performance.  I am hoping that my controller and US$67 motor combo will
> cost less than the Australian price of an ADC 9" and curtis.  That price
> is about A$5000.
>
> Finding a manual tranny for the ranger and hoiking the auto would also
> be simpler than designing an AC controller or doing direct drive.

> There are a bunch of us on this list who would really like to come up
> with an economical AC drive for EVs, a while ago we set up a sub-list
> for this purpose, however it has been lying pretty dormant due to
> everyone being busy with work/uni/etc.  I will be going ahead with this
> project for certain, it would be great to have another EE mind to help
> out!

Sure!!  Let's keep up the dialog.

>
> There is a great book on just this topic: "Motor Control Electronics
> Handbook" by Richard Valentine ISBN:0-07-066810-8
>

I'll get it.

>
> Another advantage to AC is inherant regen, whereas it is complex to
> implement DC regen.

I was thinking of a boost converter.  That is not that complex.  What am I
missing?

>
> The address for the semi-dormant AC motor control discussion group is:
> http://www.deja.com/~acmotcon/j.xp?j=acmotcon
>
> If you would like to join, send me an email and I'll join you up.
>

Sure - Thanks!

>
> Cheers,
> Rob.
>
> Todd H wrote:
> >
> > Tom and everyone,
> >
> > I'm very interested in your suggestions.  I have a 1990 Ford Ranger 4x4
> > automatic whose engine is near death.  I have done one EV in the past
> > (w/ a high school - dune buggy).  I'm an electrical engineer w/ a
> > retired & handy father.  I'm investigating my options.
> >
> > My main question is what to do with the automatic transmission?
> > 1) Remove it & direct drive?
> > 2) Install a manual transmission if possible?
> > 3) Use as is?
> > 4) Go to a fancy motor (AC???)?
> > Note:  I could design my own controller (except for the time required -
> > I am comfortable with the task & have access to big IGBTs and power
> > conversion experts, and have designed PM motor controls in the past.)
> >
> > I would welcome any suggestions I can get.  I need to understand this
> > fully before I start.
> >
> > Thanks all
> > Confused in Appleton, Wisconsin
> > Todd Holverson
> > toddh@execpc.com
> >
> > Thomas Shay wrote:
> >
> > >   The recent  discussion about direct drive for an EV
> > > with two dc motors got me to thinking about it.
> > >
> > >   There's no need to use high voltage like 288 volts
> > > unless one wants to go racing.  And it's not
> > > necessary to switch the motors between series
> > > and parallel.  The motors could be left in
> > > parallel.
> > >
> > >    Consider my Ranger pickup.  I'm using it for an
> > > example because it's familiar to me.  Two ADC
> > > 9-inch motors could be coupled end-to-end and
> > > coupled to the drive shaft with no transmission.
> > >   Let's consider at least for now than the present
> > > 4.10 rear axle ratio is kept, and a 120 volt
> > > battery pack were used.  A controller with
> > > 1000 motor amps limit and 500  battery
> > > amps limit should be adequate.
> > >    Now let's consider how this setup would run.
> > > At takeoff and speeds under about 25 mph,
> > > the controller would limit total motor current to
> > > 1000 amps total (500 per motor)  The truck
> > > would accelerate just as it does now in 2nd
> > > gear with 500 amps to the single motor and the
> > > transmission providing about 2:1 torque
> > > multiplication.   At about 25 to 30 mph the
> > > battery current would reach its limit of 500
> > > amps while the two motors would still be
> > > getting 500 amps each.  As the speed is
> > > further increased the battery current would be
> > > limited to 500 amps and the motor currents would
> > > decrease until at about 60 mph, the motor currents
> > > would be about 250 amps each.  Beyond this speed
> > > motor EMF would limit the current and the battery
> > > current and motor currents would drop off rapidly.
> > > Top speed would be by my estimate about 75 mph.
> > >
> > > Battery voltage, controller current limits, tire size
> > > and axle gear ratio could be varied to meet ones
> > > needs or wishes.
> > >
> > >   In  my never humble opinion the setup described
> > > could work great on level or nearly level terrain and
> > > no need or desire for brisk acceleration.  The
> > > maximum torque is limited to twice that of a single motor
> > > which isn't enough for brisk takeoffs or steep hills at
> > > reasonable motor currents.  It takes about 4 times the torque
> > > of a single ADC 9-inch motor for spirited acceleration and
> > > steep hill climbing with a heavy lump like a Ford Ranger.
> > >
> > > Tom Shay