Improving a DC motor

[AMPrentice]

Is it possible to improve a DC motor further with weight reductions
of certain components, changing proportions or quality of materials?

Eg.
expensive titanium shaft for long motors,
Shortening motor and increasing diameter,
(acting like a pancake motor)
highly insulated copper windings (like Baldor patented process),
higher purity copper windings,
Low Friction Tapered Roller Bearings?
Finned Armature housing for improved cooling?
Some king of alloy (spun alloy?) armature housing with fins for cooling?

[Lee Hart]

OHMyGod! wrote:

> Is it possible to improve a DC motor further with weight reductions
> of certain components, changing proportions or quality of materials?
>
> Eg.
> expensive titanium shaft for long motors,
> Shortening motor and increasing diameter,
> (acting like a pancake motor)
> highly insulated copper windings (like Baldor patented process),
> higher purity copper windings,
> Low Friction Tapered Roller Bearings?

Sure! Anything can be improved. Most of the ways to improve electric
motors are fairly well known, as well. The usual problem is cost. All
the things you describe can be done, but they increase the cost. Most
customers won't pay more to save energy or reduce motor weight.

--
Ring the bells that still can ring
Forget the perfect offering
There is a crack in everything
That's how the light gets in    --    Leonard Cohen
--
Lee A. Hart, 814 8th Ave N, Sartell MN 56377, leeahart_at_earthlink.net


_______________________________________________
For general EVDL support, see http://evdl.org/help/
For subscription options, see http://lists.sjsu.edu/mailman/listinfo/ev

[elecpulsar]

AMPrentice wrote:

Is it possible to improve a DC motor further with weight reductions
of certain components, changing proportions or quality of materials?

Eg.
expensive titanium shaft for long motors,
Shortening motor and increasing diameter,
(acting like a pancake motor)
highly insulated copper windings (like Baldor patented process),
higher purity copper windings,
Low Friction Tapered Roller Bearings?
Finned Armature housing for improved cooling?
Some king of alloy (spun alloy?) armature housing with fins for cooling?

A few years ago Nissan advertised a super inproved electric motor. It had the regular armarute and field winding,but had a armature (like a tube) OVER the fields. This armature had a shaft out the other end. Was suppose to have twice the power. Never heard anymore of it.

[Neon John]

On Wed, 16 Jul 2008 19:22:46 -0700 (PDT), OHMyGod! <darega@...> wrote:

>
>Is it possible to improve a DC motor further with weight reductions
>of certain components, changing proportions or quality of materials?
>

Yeah, but I think that many efforts would come under the category of polishing
a turd.

>Eg.
>expensive titanium shaft for long motors,

With a ton of lead onboard, why worry about a few pounds of shaft weight?

>Shortening motor and increasing diameter,
>(acting like a pancake motor)

If you're looking for low RPM, high torque, that would work.  That is
generally going backward in design, as the long, small diameter armature is
what lets traction motors operate over such a large RPM range.

Another aspect for a motor that spends much of its time accelerating is the
desire to minimize the flywheel effect.  Energy spent accelerating the
inertial mass is almost always lost during deceleration, the rare exception
being when the motor lugs down on a hill and loses speed.  Then the inertial
energy is returned to the drivetrain.  For a given mass, the smaller the
diameter of a rotating object, the lower the inertial loading.  That's why the
armature is long and narrow.

>highly insulated copper windings (like Baldor patented process),

Insulation takes up space and is a poor heat conductor compared to copper.  We
want the thinnest, highest temperature rated insulation possible.  Until we
move into the many hundreds of volts range, there is no need for exotic
insulation.

>higher purity copper windings,

Copper has to be high purity to have a low enough resistance to be electrical
wire in the first place.  That low oxygen, high purity stuff is just more
propaganda from the audiophool world.

>Low Friction Tapered Roller Bearings?

Ball bearings have the edge on rollers and rollers have the edge on tapered
roller bearings.  Tapered bearings are only used where there is high thrust
loading.  Rollers are used where very high radial force must be handled in a
small package and/or where there is very high impact loading such as in a
hammer mill.  We have neither of those requirements, nor are our speeds very
high so ordinary ball bearings are the best choice.

Any time you're looking for losses, look for heat.  You won't find any to
amount to anything in a ball bearing in good repair.  Some minor benefit might
be had by replacing grease with a low viscosity low shear liquid lube but that
definitely would be polishing a turd since the existing bearing barely rises
above ambient as it is.

I'm starting to see some audiophool-type bearing BS out there.  People selling
very expensive high tolerance bearings as some form of performance enhancer.
It's all rubbish.

Bearings come in several different grades that have to do with dimensional
tolerances, lateral play and similar parameters.  Having almost zero lateral
slop is vital in a machine tool spindle that is trying to hold
hundred-thousandth inch tolerances.  It is meaningless in a DC motor where
0.050" lateral movement would have little to no effect. Note that even the
cheapest bearings hold radial slop to less than 0.001".

I did a LOT of work in this area when I was preparing small displacement
racing engines (50 and 100cc engines that turned 20,000 RPM or more).  I set
up a bearing dyno to measure the drag at high speed.  

I could not measure statistical differences in the torque required to turn the
bearing bog-standard pillow block bearing (cleaned of grease and given a very
light oil lube, of course) and the best available at the time.  

Modern ceramic bearings might be a bit better - they're what enabled
turbochargers to go to ball bearings - but their shock loading performance
isn't nearly as good.  Knock the motor shaft while installing it, cram the
tranny into gear or whatnot and knock a little divot out of a ceramic ball and
you've just put a big hole in your wallet.

The improvement, if any, would be tiny.  Your effort and money would be much
better spent, say, getting the motor-transmission coupling run-out down to
zero.

There are LOTS of things that can be done if you're willing to pay for them.
The motors we use are designed to produce decent power, have decent lives and
be very cheap.

Were I to design a traction motor from scratch where cost is secondary, I
would concentrate on efficiency.  That is, reducing the production of heat.
Two main areas.  1) I^2R losses in all the conductors, 2) reduction in
temperature of all the conductors since copper has a fairly high positive
temperature coefficient of resistance.  Another area that might return minor
benefits would be using higher quality iron in the armature to reduce eddy
current losses.  The losses are fairly low as it is so this might also be
polishing a turd.

Here's what I'd look at.  Not all of these may have a significant return on
investment.  An engineering study would have to be done to evaluate each item.
These are things that I'd LOOK at, not necessarily do. For the pedants out
there.

Increasing the frame size for a given armature size.  This allows one to put
more copper in the field, reducing its resistance and thus heat losses.  It
also allows more air passages which enhances the cooling, minimizing the
temperature rise and accompanying increase in copper's resistance.

Winding the field using oval copper tubing through which coolant (including
air) is passed.  This further minimizes the temperature rise and resistance
rise.

Use the best magnetic steel available in the armature.

Manufacture the armature and field poles to the highest precision practical,
reduce the run-out to zero and reduce the field-armature air gap to the
absolute minimum.  Generally speaking, the smaller the air gap, the fewer
amp-turns needed in the field and thus the conductors can be larger,
minimizing heat production.

Make the armature winding out of tubing and arrange things mechanically so
that centrifugal force would move cooling air through the tubes.  In effect,
make each winding into a centrifugal fan.

Increase the length of the commutator.  This will be a biggie.  Brushes get
hot, both from the traction current flowing through them and from the short
circuit inter-bar current if the timing/neutralization isn't just right. Brush
resistance is a compromise between traction current heating and short circuit
heating.

With more brush area on the commutator and less specific current, a higher
resistance brush compound could be used.  The lower specific current (current
per square) reduces the traction current losses while the higher resistance
material reduces inter-bar current.

Interpoles, or as they're also known, commutating poles.  This removes the
problem of the brush neutral point shifting with current.  Set 'em and forget
'em.

A hybrid PM/wound field.  Several potential benefits.  If the PM supplies the
field necessary to run the highest anticipated RPM and the wound field
supplies the rest for lower speed operation, the wound field can be smaller,
made from larger conductor and dissipate less heat.  Plus the permanent
minimum field prevents run-away and armature explosions if the motor suddenly
becomes unloaded.


A few other areas that are more exotic could make some differences.

Cast the armature in thermally conductive, aerodynamically smooth polymer.
This reduces turbulence and air drag on the armature.  For high speed
operation, this is a biggie.  This is already done commercially on larger
motors and motors that are designed to run immersed in a liquid (think:
electric gas pump on a car).

Wind the coils using silver.  It has both a conductivity and a temperature
coefficient advantage over copper.  ChaChing! :-)

Make the motor hermetic and pressurize it with hydrogen.  This is a standard
technique for very large generators and motors.  Hydrogen has many benefits.
Its thermal conductivity is much higher than air.  It is much less dense,
leading to lower drag.  It has a much higher speed of sound than air,
eliminating possible concerns about high velocity losses at high armature
speeds.  It's an excellent insulator.  This will allow higher voltages with
smaller diameter armatures and commutators. It is inert to the materials that
a motor is made from, prohibiting oxidation damage.  This is important both to
insulation life and to minimizing damage during a flash-over.

A very large turbogenerator with a peripheral speed near supersonic (in air)
has vastly less drag in hydrogen, even when pressurized to 75psi, a standard
pressure.  A large turbogenerator rotor will spin for hours after the steam is
cut off if condenser vacuum is maintained.  Venting the hydrogen and replacing
it with atmospheric pressure air causes a dramatic increase in drag and fairly
rapidly slows the rotor.  This is only done when the generator is to be
entered for maintenance, of course, because a LOT of hydrogen is involved.

If a motor is designed from the ground up to be hermetic like AC compressors
are, with a bolted, gasketed and/or O-ringed end-plate, glass conductor
penetrations and otherwise welded construction AND with a high quality
floating carbon/ceramic seal (as commonly used on car AC compressors), an
initial, say, 100 PSI hydrogen load could be considered a near-lifetime load.
The work done on high pressure Sterling engine seals can be applied here.

One might worry about hydrogen being flammable (it's not in the absence of
oxygen) and might think that helium might be better.  Helium has similar (but
not as good) thermal and aero properties but all of those are offset by one
major problem.  

Being monatomic, the smallest elemental particle that exists in nature (the
hydrogen atom is smaller but it's diatomic and the molecule is much larger),
helium "goes where no gas has gone before".  Through pores that stop any other
gas.  It even diffuses through many intact materials.  It'll diffuse through
the glass walls of vacuum tubes, making them gassy.  That's why when we helium
leak test scintillation radiation monitors, the scintillator/photomultiplier
tube is removed.  The test gas will gas that photomultiplier tube right up.

Plus, pure helium is expensive and must be purchased.  If hydrogen cooling
caught on, the tiny bit that might be needed to replenish what leaked could be
made right in the car with a small electrolyzer.

Of course, the single biggest thing one can do to improve a DC motor is, tada,
get rid of the commutator and brushes.  That is, make it a BLDC.  Most (all?)
current BLDCs use permanent magnets for the field.  That eliminates brushes
and slip rings.  In the process we lose that amps-squared torque relationship
of a series motor.  

Nothing says that an electronically commutated motor couldn't have a wound
rotor AND a wound field, either series or shunt (SEPEX).  The rotor would have
3 or more slip-rings.  In return for the added complexity, one could have
whatever torque profile he desired.  Linear, current squared or anything
in-between. Whether the added complication would be worth the increased torque
would have to be determined with calculations and would have to be weighed
against the cost/benefit of a controller with voltage boost.

A significant problems with BLDCs and PM steppers (essentially the same thing)
is that the counter-emf generated as the rotor's field cuts the stator coils
makes forcing current at high speed difficult.  High forcing voltage is
necessary.  Field weakening, as in a conventional SEPEX motor, might pay big
dividends in terms of the usable speed range for a given battery voltage.  A
buck/boost circuit in the controller that steps the pack voltage up might
achieve the same results, perhaps at a higher cost.

All sorts of things can be done if one is willing to pay the price.  Question
is, how much demand is there for a $5,000 or $10,000 motor?  The squirrel cage
induction motor design being used in almost all AC drives is a good
compromise.  Not as efficient as a BLDC but capable of very wide speed ranges
at modest cost.  The ultimate motor may be a liquid cooled, hydrogen
pressurized induction motor.

John
--
John De Armond
See my website for my current email address
http://www.neon-john.com
http://www.johndearmond.com <-- best little blog on the net!
Tellico Plains, Occupied TN
Save a tree, kill a beaver


_______________________________________________
For general EVDL support, see http://evdl.org/help/
For subscription options, see http://lists.sjsu.edu/mailman/listinfo/ev

[Tom Watson-4]

Hi Neon John

I found your mention of sealing the motor and using hydrogen to increase thermal conduction, reduce internal drag and high velocity losses very interesting! All these things are great and apparently already proven design enhancements too... (on large scale generators that is)

I can also see the possibility of using the time you're hooked up to the grid charging and preheating your cabin, to pre cool your hydrogen to start getting some reduction in resistance in the coils as you lower the temperature. No need to go so low as as to superconduct! (The Temperature Coefficient of Copper (near room temperature) is +0.393 percent per degree C. This means if the temperature increases 1°C the resistance will increase 0.393%.)(taken from Cirrius site)

http://www.cirris.com/testing/temperature/copper.html)

reduction starts from room temperature I believe. Any precooling that is done will enhance the whole system.

One thing you didn't mention was a manufacturer could use the "spin" that is being put out there for hydrogen fuelcell vehicles in a positive way for ev's.  

"Making the hydrogen enabled vehicle today at the price of a regular vehicle?"
or

"Making the hydrogen enabled vehicle today at 1/2 the projected price car manufacturers are spouting for foolcell vehicles"

LOL, That would burn their conniving big oil asses, wouldn't it?

Tom


--- On Thu, 7/17/08, Neon John <jgd@...> wrote:

> Make the motor hermetic and pressurize it with hydrogen.
> This is a standard
> technique for very large generators and motors.  Hydrogen
> has many benefits.
> Its thermal conductivity is much higher than air.  It is
> much less dense,
> leading to lower drag.  It has a much higher speed of sound
> than air,
> eliminating possible concerns about high velocity losses at
> high armature
> speeds.  It's an excellent insulator.  This will allow
> higher voltages with
> smaller diameter armatures and commutators. It is inert to
> the materials that
> a motor is made from, prohibiting oxidation damage.  This
> is important both to
> insulation life and to minimizing damage during a
> flash-over.
>
> A very large turbogenerator with a peripheral speed near
> supersonic (in air)
> has vastly less drag in hydrogen, even when pressurized to
> 75psi, a standard
> pressure.  A large turbogenerator rotor will spin for hours
> after the steam is
> cut off if condenser vacuum is maintained.  Venting the
> hydrogen and replacing
> it with atmospheric pressure air causes a dramatic increase
> in drag and fairly
> rapidly slows the rotor.  This is only done when the
> generator is to be
> entered for maintenance, of course, because a LOT of
> hydrogen is involved.
>
> If a motor is designed from the ground up to be hermetic
> like AC compressors
> are, with a bolted, gasketed and/or O-ringed end-plate,
> glass conductor
> penetrations and otherwise welded construction AND with a
> high quality
> floating carbon/ceramic seal (as commonly used on car AC
> compressors), an
> initial, say, 100 PSI hydrogen load could be considered a
> near-lifetime load.
> The work done on high pressure Sterling engine seals can be
> applied here.
>
> One might worry about hydrogen being flammable (it's
> not in the absence of
> oxygen) and might think that helium might be better.
> Helium has similar (but
> not as good) thermal and aero properties but all of those
> are offset by one
> major problem.  
>
> Being monatomic, the smallest elemental particle that
> exists in nature (the
> hydrogen atom is smaller but it's diatomic and the
> molecule is much larger),
> helium "goes where no gas has gone before".
> Through pores that stop any other
> gas.  It even diffuses through many intact materials.
> It'll diffuse through
> the glass walls of vacuum tubes, making them gassy.
> That's why when we helium
> leak test scintillation radiation monitors, the
> scintillator/photomultiplier
> tube is removed.  The test gas will gas that
> photomultiplier tube right up.
>
> Plus, pure helium is expensive and must be purchased.  If
> hydrogen cooling
> caught on, the tiny bit that might be needed to replenish
> what leaked could be
> made right in the car with a small electrolyzer.

      __________________________________________________________________
Yahoo! Canada Toolbar: Search from anywhere on the web, and bookmark your favourite sites. Download it now at
http://ca.toolbar.yahoo.com.


_______________________________________________
For general EVDL support, see http://evdl.org/help/
For subscription options, see http://lists.sjsu.edu/mailman/listinfo/ev

[AMPrentice]

Thank you John for the eye opening and brain neuron stimulating experience.
The only issue I have is remove the gearbox altogether and connect directly
to the drive-shafts that turn the wheels so rpm is same as a hub motor.
Can lowering its rpm range allow affordable efficiency to be designed into a DC?

Since the DC motor produces most of it torque down low and heat is an issue
as are other issues at higher rpm, then it could go wider in diameter and
shorter in length. This will allow more copper on the windings where they could
act as fans also and the shorter shaft length will reduce part of the weight issue.
I believe cars dont need to do more than 70-80mph and torque is the most important
factor for stop and start driving as well as climbing hills and even cruising as the hp
is hardly needed once in the whole thing is in motion.

Another aspect is what if the LifePo4 batteries are used or better and overall weight
is an issue? then wouldnt a shorter titanium shaft and any other weight saving be of
use and cheaper/easier to apply in a large diameter short length DC motor?

If 1200rpm is all that is required to turn a wheel to 80mph then a large diameter short
length DC motor could possibly rev to 2500rpm for safety and also be used with a
well geared differential and multiple speed transmission to move bigger heavier vehicles.

Neon John wrote:

On Wed, 16 Jul 2008 19:22:46 -0700 (PDT), OHMyGod! <[e-mail]> wrote:
>Is it possible to improve a DC motor further with weight reductions
>of certain components, changing proportions or quality of materials?
Yeah, but I think that many efforts would come under the category of polishing
a turd.
>Eg.
>expensive titanium shaft for long motors,
With a ton of lead onboard, why worry about a few pounds of shaft weight?
>Shortening motor and increasing diameter,
>(acting like a pancake motor)
If you're looking for low RPM, high torque, that would work.  That is
generally going backward in design, as the long, small diameter armature is
what lets traction motors operate over such a large RPM range.
Another aspect for a motor that spends much of its time accelerating is the
desire to minimize the flywheel effect.  Energy spent accelerating the
inertial mass is almost always lost during deceleration, the rare exception
being when the motor lugs down on a hill and loses speed.  Then the inertial
energy is returned to the drivetrain.  For a given mass, the smaller the
diameter of a rotating object, the lower the inertial loading.  That's why the
armature is long and narrow.
>highly insulated copper windings (like Baldor patented process),
Insulation takes up space and is a poor heat conductor compared to copper.  We
want the thinnest, highest temperature rated insulation possible.  Until we
move into the many hundreds of volts range, there is no need for exotic
insulation.
>higher purity copper windings,
Copper has to be high purity to have a low enough resistance to be electrical
wire in the first place.  That low oxygen, high purity stuff is just more
propaganda from the audiophool world.
>Low Friction Tapered Roller Bearings?
Ball bearings have the edge on rollers and rollers have the edge on tapered
roller bearings.  Tapered bearings are only used where there is high thrust
loading.  Rollers are used where very high radial force must be handled in a
small package and/or where there is very high impact loading such as in a
hammer mill.  We have neither of those requirements, nor are our speeds very
high so ordinary ball bearings are the best choice.
Any time you're looking for losses, look for heat.  You won't find any to
amount to anything in a ball bearing in good repair.  Some minor benefit might
be had by replacing grease with a low viscosity low shear liquid lube but that
definitely would be polishing a turd since the existing bearing barely rises
above ambient as it is.
I'm starting to see some audiophool-type bearing BS out there.  People selling
very expensive high tolerance bearings as some form of performance enhancer.
It's all rubbish.
Bearings come in several different grades that have to do with dimensional
tolerances, lateral play and similar parameters.  Having almost zero lateral
slop is vital in a machine tool spindle that is trying to hold
hundred-thousandth inch tolerances.  It is meaningless in a DC motor where
0.050" lateral movement would have little to no effect. Note that even the
cheapest bearings hold radial slop to less than 0.001".
I did a LOT of work in this area when I was preparing small displacement
racing engines (50 and 100cc engines that turned 20,000 RPM or more).  I set
up a bearing dyno to measure the drag at high speed.  
I could not measure statistical differences in the torque required to turn the
bearing bog-standard pillow block bearing (cleaned of grease and given a very
light oil lube, of course) and the best available at the time.  
Modern ceramic bearings might be a bit better - they're what enabled
turbochargers to go to ball bearings - but their shock loading performance
isn't nearly as good.  Knock the motor shaft while installing it, cram the
tranny into gear or whatnot and knock a little divot out of a ceramic ball and
you've just put a big hole in your wallet.
The improvement, if any, would be tiny.  Your effort and money would be much
better spent, say, getting the motor-transmission coupling run-out down to
zero.
There are LOTS of things that can be done if you're willing to pay for them.
The motors we use are designed to produce decent power, have decent lives and
be very cheap.
Were I to design a traction motor from scratch where cost is secondary, I
would concentrate on efficiency.  That is, reducing the production of heat.
Two main areas.  1) I^2R losses in all the conductors, 2) reduction in
temperature of all the conductors since copper has a fairly high positive
temperature coefficient of resistance.  Another area that might return minor
benefits would be using higher quality iron in the armature to reduce eddy
current losses.  The losses are fairly low as it is so this might also be
polishing a turd.
Here's what I'd look at.  Not all of these may have a significant return on
investment.  An engineering study would have to be done to evaluate each item.
These are things that I'd LOOK at, not necessarily do. For the pedants out
there.
Increasing the frame size for a given armature size.  This allows one to put
more copper in the field, reducing its resistance and thus heat losses.  It
also allows more air passages which enhances the cooling, minimizing the
temperature rise and accompanying increase in copper's resistance.
Winding the field using oval copper tubing through which coolant (including
air) is passed.  This further minimizes the temperature rise and resistance
rise.
Use the best magnetic steel available in the armature.
Manufacture the armature and field poles to the highest precision practical,
reduce the run-out to zero and reduce the field-armature air gap to the
absolute minimum.  Generally speaking, the smaller the air gap, the fewer
amp-turns needed in the field and thus the conductors can be larger,
minimizing heat production.
Make the armature winding out of tubing and arrange things mechanically so
that centrifugal force would move cooling air through the tubes.  In effect,
make each winding into a centrifugal fan.
Increase the length of the commutator.  This will be a biggie.  Brushes get
hot, both from the traction current flowing through them and from the short
circuit inter-bar current if the timing/neutralization isn't just right. Brush
resistance is a compromise between traction current heating and short circuit
heating.
With more brush area on the commutator and less specific current, a higher
resistance brush compound could be used.  The lower specific current (current
per square) reduces the traction current losses while the higher resistance
material reduces inter-bar current.
Interpoles, or as they're also known, commutating poles.  This removes the
problem of the brush neutral point shifting with current.  Set 'em and forget
'em.
A hybrid PM/wound field.  Several potential benefits.  If the PM supplies the
field necessary to run the highest anticipated RPM and the wound field
supplies the rest for lower speed operation, the wound field can be smaller,
made from larger conductor and dissipate less heat.  Plus the permanent
minimum field prevents run-away and armature explosions if the motor suddenly
becomes unloaded.
A few other areas that are more exotic could make some differences.
Cast the armature in thermally conductive, aerodynamically smooth polymer.
This reduces turbulence and air drag on the armature.  For high speed
operation, this is a biggie.  This is already done commercially on larger
motors and motors that are designed to run immersed in a liquid (think:
electric gas pump on a car).
Wind the coils using silver.  It has both a conductivity and a temperature
coefficient advantage over copper.  ChaChing! :-)
Make the motor hermetic and pressurize it with hydrogen.  This is a standard
technique for very large generators and motors.  Hydrogen has many benefits.
Its thermal conductivity is much higher than air.  It is much less dense,
leading to lower drag.  It has a much higher speed of sound than air,
eliminating possible concerns about high velocity losses at high armature
speeds.  It's an excellent insulator.  This will allow higher voltages with
smaller diameter armatures and commutators. It is inert to the materials that
a motor is made from, prohibiting oxidation damage.  This is important both to
insulation life and to minimizing damage during a flash-over.
A very large turbogenerator with a peripheral speed near supersonic (in air)
has vastly less drag in hydrogen, even when pressurized to 75psi, a standard
pressure.  A large turbogenerator rotor will spin for hours after the steam is
cut off if condenser vacuum is maintained.  Venting the hydrogen and replacing
it with atmospheric pressure air causes a dramatic increase in drag and fairly
rapidly slows the rotor.  This is only done when the generator is to be
entered for maintenance, of course, because a LOT of hydrogen is involved.
If a motor is designed from the ground up to be hermetic like AC compressors
are, with a bolted, gasketed and/or O-ringed end-plate, glass conductor
penetrations and otherwise welded construction AND with a high quality
floating carbon/ceramic seal (as commonly used on car AC compressors), an
initial, say, 100 PSI hydrogen load could be considered a near-lifetime load.
The work done on high pressure Sterling engine seals can be applied here.
One might worry about hydrogen being flammable (it's not in the absence of
oxygen) and might think that helium might be better.  Helium has similar (but
not as good) thermal and aero properties but all of those are offset by one
major problem.  
Being monatomic, the smallest elemental particle that exists in nature (the
hydrogen atom is smaller but it's diatomic and the molecule is much larger),
helium "goes where no gas has gone before".  Through pores that stop any other
gas.  It even diffuses through many intact materials.  It'll diffuse through
the glass walls of vacuum tubes, making them gassy.  That's why when we helium
leak test scintillation radiation monitors, the scintillator/photomultiplier
tube is removed.  The test gas will gas that photomultiplier tube right up.
Plus, pure helium is expensive and must be purchased.  If hydrogen cooling
caught on, the tiny bit that might be needed to replenish what leaked could be
made right in the car with a small electrolyzer.
Of course, the single biggest thing one can do to improve a DC motor is, tada,
get rid of the commutator and brushes.  That is, make it a BLDC.  Most (all?)
current BLDCs use permanent magnets for the field.  That eliminates brushes
and slip rings.  In the process we lose that amps-squared torque relationship
of a series motor.  
Nothing says that an electronically commutated motor couldn't have a wound
rotor AND a wound field, either series or shunt (SEPEX).  The rotor would have
3 or more slip-rings.  In return for the added complexity, one could have
whatever torque profile he desired.  Linear, current squared or anything
in-between. Whether the added complication would be worth the increased torque
would have to be determined with calculations and would have to be weighed
against the cost/benefit of a controller with voltage boost.
A significant problems with BLDCs and PM steppers (essentially the same thing)
is that the counter-emf generated as the rotor's field cuts the stator coils
makes forcing current at high speed difficult.  High forcing voltage is
necessary.  Field weakening, as in a conventional SEPEX motor, might pay big
dividends in terms of the usable speed range for a given battery voltage.  A
buck/boost circuit in the controller that steps the pack voltage up might
achieve the same results, perhaps at a higher cost.
All sorts of things can be done if one is willing to pay the price.  Question
is, how much demand is there for a $5,000 or $10,000 motor?  The squirrel cage
induction motor design being used in almost all AC drives is a good
compromise.  Not as efficient as a BLDC but capable of very wide speed ranges
at modest cost.  The ultimate motor may be a liquid cooled, hydrogen
pressurized induction motor.
John
--
John De Armond
See my website for my current email address
http://www.neon-john.com
http://www.johndearmond.com <-- best little blog on the net!
Tellico Plains, Occupied TN
Save a tree, kill a beaver


_______________________________________________
For general EVDL support, see http://evdl.org/help/
For subscription options, see http://lists.sjsu.edu/mailman/listinfo/ev

[cowtown]

> All sorts of things can be done if one is willing to pay the price.  Question
> is, how much demand is there for a $5,000 or $10,000 motor?  The  
> squirrel cage
> induction motor design being used in almost all AC drives is a good
> compromise.  Not as efficient as a BLDC but capable of very wide speed ranges
> at modest cost.  The ultimate motor may be a liquid cooled, hydrogen
> pressurized induction motor.

A WarP13 runs $4900 (http://www.evsource.com/tls_warp13.php):

Technical Information:

     * Maximum operating voltage 170V, recommended voltage 150-160V
     * HP = 43.7 @ 72 volts DC (452.9 amps)
     * 135 Ft. pounds torque @ 72 volts DC (452.9 amps)
     * Weight, approx. 300+ pounds
     * 5500 RPM
     * Exceeds class "H" insulation standard for temperatures
     * Large style brushes, factory preset at over 90% contact
     * Drive and tail shafts keyed - pilot bearing hole with 5/16 - 18  
mounting hole
     * Double ended shaft - Turbo 400 spline

I'll bet if you could beat those specs, you could ask a few $K more.  
The Zilla fit to maximize the WarP13 costs >$5K - how much for a  
controller for yours?


_______________________________________________
For general EVDL support, see http://evdl.org/help/
For subscription options, see http://lists.sjsu.edu/mailman/listinfo/ev

[James Massey]

G'day All

At 02:59 PM 17/07/2008, Lee Hart wrote:
>OHMyGod! wrote:
> > Is it possible to improve a DC motor further with weight reductions
> > of certain components, changing proportions or quality of materials?
> ><snip>
>
>Sure! Anything can be improved. Most of the ways to improve electric
>motors are fairly well known, as well. The usual problem is cost. All
>the things you describe can be done, but they increase the cost. Most
>customers won't pay more to save energy or reduce motor weight.

IIRC NASA bought $500 each Maxon motors for the Mars rovers, then
spent something like 500k improving them, as a significant saving
over building them from scratch.

I've often thought what you'd get with a silver wound motor (and
suitably massive brushes or IGBTs depending on DC or AC).

As Lee says, anything can be improved... depending on what you
consider to be an improvement!

Regards

[Technik] James


_______________________________________________
For general EVDL support, see http://evdl.org/help/
For subscription options, see http://lists.sjsu.edu/mailman/listinfo/ev

[Tom Parker-12]

On Thu, 2008-07-17 at 13:26 -0400, Neon John wrote:

> Any time you're looking for losses, look for heat.  You won't find any to
> amount to anything in a ball bearing in good repair.  Some minor benefit might
> be had by replacing grease with a low viscosity low shear liquid lube but that
> definitely would be polishing a turd since the existing bearing barely rises
> above ambient as it is.

My Siemens 1LH5118 motor has a splined hole instead of a stick out
shaft. This hole gets quite warm inside after about 30 seconds at 5500
rpm no load.

I haven't got the motor into a car yet, maybe it will get better with a
few hours of use. I'm not planning to take it out of the car once
installed to find out though.


_______________________________________________
For general EVDL support, see http://evdl.org/help/
For subscription options, see http://lists.sjsu.edu/mailman/listinfo/ev

[EVstuff]

30 Seconds ??
That doesn't sound right.
Check for excessive "no load" current.
Tom Meyers

)SNIP(
> My Siemens 1LH5118 motor has a splined hole instead of a stick out
> shaft. This hole gets quite warm inside after about 30 seconds at 5500
> rpm no load.


_______________________________________________
For general EVDL support, see http://evdl.org/help/
For subscription options, see http://lists.sjsu.edu/mailman/listinfo/ev

[Neon John]

On Thu, 17 Jul 2008 21:17:01 -0700, cowtown@... wrote:

>> All sorts of things can be done if one is willing to pay the price.  Question
>> is, how much demand is there for a $5,000 or $10,000 motor?  The  
>> squirrel cage
>> induction motor design being used in almost all AC drives is a good
>> compromise.  Not as efficient as a BLDC but capable of very wide speed ranges
>> at modest cost.  The ultimate motor may be a liquid cooled, hydrogen
>> pressurized induction motor.
>
>A WarP13 runs $4900 (http://www.evsource.com/tls_warp13.php):
>
>Technical Information:
>
>     * Maximum operating voltage 170V, recommended voltage 150-160V
>     * HP = 43.7 @ 72 volts DC (452.9 amps)

Damn, that's a hell of a "EV sucker surcharge"!  I just can't get my mind
around the sucker surcharge that "EV parts" carry.  I'm used to thinking about
industrial pricing when I see DC motors and such.

I saw your post this morning so I called down to the motor shop and asked my
friend, the owner, what a traction motor with those kinds of specs should
cost.  He said that off the top of his head, it would run $1000-1500.   That
matched my estimate.  I called him because he's been in the business for >50
years and has owned the shop since 1975.  Of course, a motor that he'd sell
would come painted basic black and wouldn't have a kewl name.

OK, so let me re-state, figuring in the "EV sucker surcharge".  Is there any
demand for a $10-20,000 DC motor?

John
--
John De Armond
See my website for my current email address
http://www.neon-john.com
http://www.johndearmond.com <-- best little blog on the net!
Tellico Plains, Occupied TN
Unable to locate Coffee -- Operator Halted!


_______________________________________________
For general EVDL support, see http://evdl.org/help/
For subscription options, see http://lists.sjsu.edu/mailman/listinfo/ev

[AMPrentice]

Even myself a beginner in EVs found out those 13 and 11 inch logs
with the ¨red herring¨ paint job are heavy and overpriced lumps of steel
you see them in good second hand condition for 300$US sometimes
on fleebay.

Back on the subject if 1200rpm is all that is required to turn a wheel to 80mph
and DC motors have plenty of torque at low rpm isnt it then possible a large
diameter short length DC motor could be more beneficial than the current ones?

If a direct driven wheel from driveshafts from such a motor is possible wouldnt it
last longer and have better efficiency with more copper being used to act as a fan?
Even then it could possibly rev to 2500rpm and also be used with a well geared
differential and multiple speed transmission to move bigger heavier vehicles.

This kind of efficiency/improvement could be achieved just by changing its proportions.
After all why do we need to rev to 5000 rpm when its not a car engine or should be
treated and designed as one. Same thing with diesel engines in LeMans lots of torque
low rpm, high efficiency and durability no?

Neon John wrote:

On Thu, 17 Jul 2008 21:17:01 -0700, [e-mail] wrote:

>> All sorts of things can be done if one is willing to pay the price.  Question
>> is, how much demand is there for a $5,000 or $10,000 motor?  The  
>> squirrel cage
>> induction motor design being used in almost all AC drives is a good
>> compromise.  Not as efficient as a BLDC but capable of very wide speed ranges
>> at modest cost.  The ultimate motor may be a liquid cooled, hydrogen
>> pressurized induction motor.
>
>A WarP13 runs $4900 (http://www.evsource.com/tls_warp13.php):
>
>Technical Information:
>
>     * Maximum operating voltage 170V, recommended voltage 150-160V
>     * HP = 43.7 @ 72 volts DC (452.9 amps)

Damn, that's a hell of a "EV sucker surcharge"!  I just can't get my mind
around the sucker surcharge that "EV parts" carry.  I'm used to thinking about
industrial pricing when I see DC motors and such.

I saw your post this morning so I called down to the motor shop and asked my
friend, the owner, what a traction motor with those kinds of specs should
cost.  He said that off the top of his head, it would run $1000-1500.   That
matched my estimate.  I called him because he's been in the business for >50
years and has owned the shop since 1975.  Of course, a motor that he'd sell
would come painted basic black and wouldn't have a kewl name.

OK, so let me re-state, figuring in the "EV sucker surcharge".  Is there any
demand for a $10-20,000 DC motor?

John
--
John De Armond
See my website for my current email address
http://www.neon-john.com
http://www.johndearmond.com <-- best little blog on the net!
Tellico Plains, Occupied TN
Unable to locate Coffee -- Operator Halted!


_______________________________________________
For general EVDL support, see http://evdl.org/help/
For subscription options, see http://lists.sjsu.edu/mailman/listinfo/ev

[Lee Hart]

>> A WarP13 runs $4900...

Neon John wrote:
> I saw your post this morning so I called down to the motor shop and
> asked my friend, the owner, what a traction motor with those kinds of
> specs should cost.  He said that off the top of his head, it would
> run $1000-1500.   That matched my estimate.  I called him because
> he's been in the business for >50 years and has owned the shop since
> 1975.  Of course, a motor that he'd sell would come painted basic
> black and wouldn't have a kewl name.

Then tell him he needs to get into the EV motor business!

Right now, good motors are very hard to get. GE won't sell unless you
want thousands. ADC only sells reluctantly through a very few dealers,
and has a bad attitude (go way kid, ya bother me). Netgain has high
prices because they are severely backlogged and can't meet demand.
--
Ring the bells that still can ring
Forget the perfect offering
There is a crack in everything
That's how the light gets in    --    Leonard Cohen
--
Lee A. Hart, 814 8th Ave N, Sartell MN 56377, leeahart_at_earthlink.net


_______________________________________________
For general EVDL support, see http://evdl.org/help/
For subscription options, see http://lists.sjsu.edu/mailman/listinfo/ev