Vernier caliper accuracy
 

Ed Huntress wrote:

Interesting. Why are pistons intentionally made elliptical?


Because the thicker sections, which are the boss areas for the wrist pins
(piston pins) expand with much more force than the thin sections. So the
pistons have a smaller diameter across the boss area.

Ah, I think I misunderstoood you. I thought you meant a piston shaped
like a beer barrel. Instead you mean a piston which is slightly
elliptical when viewed from the top or bottom?

I've heard of pistons where the top land is of a smaller diameter
(apparently it reduces wear because the lubrication is poorer at the top),
but I've not heard of an elliptical piston.


Almost all production automobile pistons made today are elliptical. In fact,
they're often elliptical with the major axis in one direction at the top of
the piston, and in the other direction at the bottom. The bottom ellipse is
for better sealing, to meet emission requirements. It has to do with
differential friction and heating between the neutral axis, which is
parallel to the crankshaft, versus the other axis.

When I was at Wasino we had some drawings from Ford that actually had three
different ellipses along their length, from top to bottom, and they had to
blend into each other.


I'm surprised it doesn't increase the wear on the sides due to the reduced
surface area, and I'd have thought it could allow the piston to vibrate in
an angular sense about the gudgeon pin. Any more information, Ed?


If you don't get an expert to chime in here, I'll see what I can dig up for
you. There is one guy who stops in here from time to time who is an engineer
for one of the world's top piston manufacturers; you won't get any better
info than that from him.

The thought I had about vibration only applies to a piston shaped like a
beer barrel. And if the piston expands to become a near-perfect cylinder
when it's heated, I can see why wear wouldn't be an issue either. Thanks!

Best wishes,

Chris

Vernier caliper accuracy
 

"Christopher Tidy" wrote in message
...
Ed Huntress wrote:

Interesting. Why are pistons intentionally made elliptical?


Because the thicker sections, which are the boss areas for the wrist pins
(piston pins) expand with much more force than the thin sections. So the
pistons have a smaller diameter across the boss area.

Ah, I think I misunderstoood you. I thought you meant a piston shaped like
a beer barrel. Instead you mean a piston which is slightly elliptical when
viewed from the top or bottom?

Yes. I wish I still had the 3D versions of those Ford programs that I
produced in Rhino, for our promotional material on cutting elliptical
pistons. I applied multipliers in Excel to the values in the CAD drawing
until you could actually see the shapes with the naked eye.

I don't want to confuse things, but they were shaped a lot lie beer barrels.
g As I mentioned, there were three different ellipses from top to bottom,
with major axes arranged differently, and the appearance from some angles
was that they were widest in the middle.


I've heard of pistons where the top land is of a smaller diameter
(apparently it reduces wear because the lubrication is poorer at the
top), but I've not heard of an elliptical piston.


Almost all production automobile pistons made today are elliptical. In
fact, they're often elliptical with the major axis in one direction at
the top of the piston, and in the other direction at the bottom. The
bottom ellipse is for better sealing, to meet emission requirements. It
has to do with differential friction and heating between the neutral
axis, which is parallel to the crankshaft, versus the other axis.

When I was at Wasino we had some drawings from Ford that actually had
three different ellipses along their length, from top to bottom, and they
had to blend into each other.


I'm surprised it doesn't increase the wear on the sides due to the
reduced surface area, and I'd have thought it could allow the piston to
vibrate in an angular sense about the gudgeon pin. Any more information,
Ed?


If you don't get an expert to chime in here, I'll see what I can dig up
for you. There is one guy who stops in here from time to time who is an
engineer for one of the world's top piston manufacturers; you won't get
any better info than that from him.

The thought I had about vibration only applies to a piston shaped like a
beer barrel. And if the piston expands to become a near-perfect cylinder
when it's heated, I can see why wear wouldn't be an issue either. Thanks!

Yes, that's more or less what happens.

--
Ed Huntress

Vernier caliper accuracy
 

On 2010-02-28, Christopher Tidy wrote:
DoN. Nichols wrote:

Hi Don,

Thanks for the description of the sine plate.

You're welcome.

Sorry I've taken a few
days to reply. Bad week!

A lot of that going around -- just from the weather. I hope
that is all that yours was.

Just briefly, what's a sine plate used for?

[ ... description of appearance and use snipped ... ]

Not what I was expecting! I had imagined a steel plate with a surface
shaped like a sine wave, sitting on a table (though what you'd use that
for, I don't know).

Actually -- there is something like that -- but a pair of them
go in a milling vise to support the workpiece. They are less prone to
topple over when the vise is loosened than some standard parallels, and
they can also be positioned to support workpieces with gaps near the
edges in some places and nearer the center in others.

I think that they are called "wave parallels". I've never owned
any, or had a chance to handle some belonging to someone else, but they
strike me as useful for production runs.

Thanks for the explanation. Some day I'll probably
need one!

If you see one used cheap -- go for it even before you need it.
I've gotten a lot of my tools that way against future needs.

Enjoy,
DoN.

--
Email: | Voice (all times): (703) 938-4564
(too) near Washington D.C. | http://www.d-and-d.com/dnichols/DoN.html
--- Black Holes are where God is dividing by zero ---

Vernier caliper accuracy
 

Ed Huntress wrote:

Ah, I think I misunderstoood you. I thought you meant a piston shaped like
a beer barrel. Instead you mean a piston which is slightly elliptical when
viewed from the top or bottom?


Yes. I wish I still had the 3D versions of those Ford programs that I
produced in Rhino, for our promotional material on cutting elliptical
pistons. I applied multipliers in Excel to the values in the CAD drawing
until you could actually see the shapes with the naked eye.

I don't want to confuse things, but they were shaped a lot lie beer barrels.
g As I mentioned, there were three different ellipses from top to bottom,
with major axes arranged differently, and the appearance from some angles
was that they were widest in the middle.

Got it. I can see the shape now. That's complicated!

The thought I had about vibration only applies to a piston shaped like a
beer barrel. And if the piston expands to become a near-perfect cylinder
when it's heated, I can see why wear wouldn't be an issue either. Thanks!


Yes, that's more or less what happens.

At least the concept sounds simpler :-)

Chris

Vernier caliper accuracy
 

Ed Huntress wrote:

Interesting. Why are pistons intentionally made elliptical?


Because the thicker sections, which are the boss areas for the wrist pins
(piston pins) expand with much more force than the thin sections. So the
pistons have a smaller diameter across the boss area.

By the way, Ed, how are elliptical pistons machined? Are they made using
a CNC mill, or a special lathe on which the tool can move in phase with
the rotation of the workpiece? Just curious...

Chris

Vernier caliper accuracy
 

"Christopher Tidy" wrote in message
...
Ed Huntress wrote:

Interesting. Why are pistons intentionally made elliptical?


Because the thicker sections, which are the boss areas for the wrist pins
(piston pins) expand with much more force than the thin sections. So the
pistons have a smaller diameter across the boss area.

By the way, Ed, how are elliptical pistons machined? Are they made using a
CNC mill, or a special lathe on which the tool can move in phase with the
rotation of the workpiece? Just curious...

Chris

A darned good question. I wish the piston man would step in here (Anthony?).
He's a world-class expert on this subject.

In high-volume production, they've traditionally been turned on special
lathes with cam-operated cross slides. That requires a cam for each piston
profile. Takisawa makes, or made, a programmable machine that, IIRC, uses
solenoid-actuated slides. It's slow. There also have been some piezoelectric
actuators. And that was all ten years ago. There may be something new.

I was involved with these things due to a joint project we had going at
Wasino lathes, where I worked in those days. We were using a
magnetostrictive actuator, made of the material used in submarine sonar. It
worked great but programming the hysteresis out of it was a nightmare. The
project was dropped about the time I left.

Anyway, the challenge has been to come up with a programmable system that
provides adequate thrust at high turning speeds. They're doing it, I think;
I just haven't kept up.

--
Ed Huntress

Vernier caliper accuracy
 

Ed Huntress wrote:

A darned good question. I wish the piston man would step in here (Anthony?).
He's a world-class expert on this subject.

In high-volume production, they've traditionally been turned on special
lathes with cam-operated cross slides. That requires a cam for each piston
profile. Takisawa makes, or made, a programmable machine that, IIRC, uses
solenoid-actuated slides. It's slow. There also have been some piezoelectric
actuators. And that was all ten years ago. There may be something new.

I was involved with these things due to a joint project we had going at
Wasino lathes, where I worked in those days. We were using a
magnetostrictive actuator, made of the material used in submarine sonar. It
worked great but programming the hysteresis out of it was a nightmare. The
project was dropped about the time I left.

Anyway, the challenge has been to come up with a programmable system that
provides adequate thrust at high turning speeds. They're doing it, I think;
I just haven't kept up.

Thanks for the information. Sorry I've been slow to reply; I've been
away. I worked myself as part of a team developing electrostatic
actuators a few years back. The actuators consisted of a sheet of a
rubbery material with conductive coatings on each side. When subjected
to a high DC voltage, the conductive coatings could be made to attract
or repel, squeezing or stretching the material (the actuation direction
was perpendicular to the electric field, as this gave a greater
movement). One idea was to use these actuators to move a car wing
mirror, although I'm unsure if a prototype was made as I'd left the
laboratory by then. But the actuators we had at that time were
definitely too flexible to be used for moving a lathe tool.

Best wishes,

Chris

Vernier caliper accuracy
 

"Christopher Tidy" wrote in message
...
Ed Huntress wrote:

A darned good question. I wish the piston man would step in here
(Anthony?). He's a world-class expert on this subject.

In high-volume production, they've traditionally been turned on special
lathes with cam-operated cross slides. That requires a cam for each
piston profile. Takisawa makes, or made, a programmable machine that,
IIRC, uses solenoid-actuated slides. It's slow. There also have been some
piezoelectric actuators. And that was all ten years ago. There may be
something new.

I was involved with these things due to a joint project we had going at
Wasino lathes, where I worked in those days. We were using a
magnetostrictive actuator, made of the material used in submarine sonar.
It worked great but programming the hysteresis out of it was a nightmare.
The project was dropped about the time I left.

Anyway, the challenge has been to come up with a programmable system that
provides adequate thrust at high turning speeds. They're doing it, I
think; I just haven't kept up.

Thanks for the information. Sorry I've been slow to reply; I've been away.
I worked myself as part of a team developing electrostatic actuators a few
years back. The actuators consisted of a sheet of a rubbery material with
conductive coatings on each side. When subjected to a high DC voltage, the
conductive coatings could be made to attract or repel, squeezing or
stretching the material (the actuation direction was perpendicular to the
electric field, as this gave a greater movement). One idea was to use
these actuators to move a car wing mirror, although I'm unsure if a
prototype was made as I'd left the laboratory by then. But the actuators
we had at that time were definitely too flexible to be used for moving a
lathe tool.

It requires a lot of force. The material we used, Terfenol-D, is one of the
few that can deliver the force with good speed (several thousand cycles per
second with high force; up to 20 kHz at lower force levels, IIRC).
Piezoelectrics, which are the other option, are used in stacks to get the
required travel, but they're a little fragile for the application:

http://en.wikipedia.org/wiki/Magnetostriction

Terfenol is the material used in advanced sonar detectors. We were first
looking at it as a counter-vibratory attachment, but my boss got the idea
that it would work for turning elliptical pistons. We had a demo machine at
IMTS 2000, I think, and it attracted a lot of attention. We could get it to
work extremely well and with high accuracy and repeatability. But if you
make a small change in the program, magnetic hysteresis would complicate the
programming, making the whole thing problematic.

--
Ed Huntress

Vernier caliper accuracy
 

"Ed Huntress" wrote in message
...

"Christopher Tidy" wrote in message
...
Ed Huntress wrote:

A darned good question. I wish the piston man would step in here
(Anthony?). He's a world-class expert on this subject.

In high-volume production, they've traditionally been turned on special
lathes with cam-operated cross slides. That requires a cam for each
piston profile. Takisawa makes, or made, a programmable machine that,
IIRC, uses solenoid-actuated slides. It's slow. There also have been
some piezoelectric actuators. And that was all ten years ago. There may
be something new.

I was involved with these things due to a joint project we had going at
Wasino lathes, where I worked in those days. We were using a
magnetostrictive actuator, made of the material used in submarine sonar.
It worked great but programming the hysteresis out of it was a
nightmare. The project was dropped about the time I left.

Anyway, the challenge has been to come up with a programmable system
that provides adequate thrust at high turning speeds. They're doing it,
I think; I just haven't kept up.

Thanks for the information. Sorry I've been slow to reply; I've been
away. I worked myself as part of a team developing electrostatic
actuators a few years back. The actuators consisted of a sheet of a
rubbery material with conductive coatings on each side. When subjected to
a high DC voltage, the conductive coatings could be made to attract or
repel, squeezing or stretching the material (the actuation direction was
perpendicular to the electric field, as this gave a greater movement).
One idea was to use these actuators to move a car wing mirror, although
I'm unsure if a prototype was made as I'd left the laboratory by then.
But the actuators we had at that time were definitely too flexible to be
used for moving a lathe tool.

It requires a lot of force. The material we used, Terfenol-D, is one of
the few that can deliver the force with good speed (several thousand
cycles per second with high force; up to 20 kHz at lower force levels,
IIRC). Piezoelectrics, which are the other option, are used in stacks to
get the required travel, but they're a little fragile for the application:

http://en.wikipedia.org/wiki/Magnetostriction

Terfenol is the material used in advanced sonar detectors. We were first
looking at it as a counter-vibratory attachment, but my boss got the idea
that it would work for turning elliptical pistons. We had a demo machine
at IMTS 2000, I think, and it attracted a lot of attention. We could get
it to work extremely well and with high accuracy and repeatability. But if
you make a small change in the program, magnetic hysteresis would
complicate the programming, making the whole thing problematic.

--
Ed Huntress


I worked on a product using the Polaroid Piso distance sensors. 1980's.
Problem was the ringing in the piso unit required a separate sender and a
receiver for up close sensing, which is what I was trying to do. Figured
out was not a big enough market to continue on the idea. As to fast, heavy
duty, accurate movers, Voice coils work great. Use a linear transformer as
the sensor. Was the way we build disk drives in the 1970's early 80's. The
magnet on a CDC 200 megabyte drive probably weighs 25-30 pounds total.

Vernier caliper accuracy
 

"Bill McKee" wrote in message
m...

"Ed Huntress" wrote in message
...

"Christopher Tidy" wrote in message
...
Ed Huntress wrote:

A darned good question. I wish the piston man would step in here
(Anthony?). He's a world-class expert on this subject.

In high-volume production, they've traditionally been turned on special
lathes with cam-operated cross slides. That requires a cam for each
piston profile. Takisawa makes, or made, a programmable machine that,
IIRC, uses solenoid-actuated slides. It's slow. There also have been
some piezoelectric actuators. And that was all ten years ago. There may
be something new.

I was involved with these things due to a joint project we had going at
Wasino lathes, where I worked in those days. We were using a
magnetostrictive actuator, made of the material used in submarine
sonar. It worked great but programming the hysteresis out of it was a
nightmare. The project was dropped about the time I left.

Anyway, the challenge has been to come up with a programmable system
that provides adequate thrust at high turning speeds. They're doing it,
I think; I just haven't kept up.

Thanks for the information. Sorry I've been slow to reply; I've been
away. I worked myself as part of a team developing electrostatic
actuators a few years back. The actuators consisted of a sheet of a
rubbery material with conductive coatings on each side. When subjected
to a high DC voltage, the conductive coatings could be made to attract
or repel, squeezing or stretching the material (the actuation direction
was perpendicular to the electric field, as this gave a greater
movement). One idea was to use these actuators to move a car wing
mirror, although I'm unsure if a prototype was made as I'd left the
laboratory by then. But the actuators we had at that time were
definitely too flexible to be used for moving a lathe tool.

It requires a lot of force. The material we used, Terfenol-D, is one of
the few that can deliver the force with good speed (several thousand
cycles per second with high force; up to 20 kHz at lower force levels,
IIRC). Piezoelectrics, which are the other option, are used in stacks to
get the required travel, but they're a little fragile for the
application:

http://en.wikipedia.org/wiki/Magnetostriction

Terfenol is the material used in advanced sonar detectors. We were first
looking at it as a counter-vibratory attachment, but my boss got the idea
that it would work for turning elliptical pistons. We had a demo machine
at IMTS 2000, I think, and it attracted a lot of attention. We could get
it to work extremely well and with high accuracy and repeatability. But
if you make a small change in the program, magnetic hysteresis would
complicate the programming, making the whole thing problematic.

--
Ed Huntress


I worked on a product using the Polaroid Piso distance sensors. 1980's.
Problem was the ringing in the piso unit required a separate sender and a
receiver for up close sensing, which is what I was trying to do. Figured
out was not a big enough market to continue on the idea. As to fast,
heavy duty, accurate movers, Voice coils work great. Use a linear
transformer as the sensor. Was the way we build disk drives in the 1970's
early 80's. The magnet on a CDC 200 megabyte drive probably weighs 25-30
pounds total.

There were some reasons that solenoid-type actuators weren't ideal. I think
it was the response rate with the weight of the actuator required for the
job, but it's been too long for me to trust my memory.

I see that Takisawa is still making their line of oval-piston machines (TPS
Series). Maybe there's something on their website about the actuator:

http://www.takisawa.co.jp/-e/index-e02.htm

--
Ed Huntress

Vernier caliper accuracy
 

On Thu, 11 Mar 2010 22:07:12 -0500, "Ed Huntress"
wrote:


"Bill McKee" wrote in message
om...

"Ed Huntress" wrote in message
...

"Christopher Tidy" wrote in message
...
Ed Huntress wrote:

A darned good question. I wish the piston man would step in here
(Anthony?). He's a world-class expert on this subject.

In high-volume production, they've traditionally been turned on special
lathes with cam-operated cross slides. That requires a cam for each
piston profile. Takisawa makes, or made, a programmable machine that,
IIRC, uses solenoid-actuated slides. It's slow. There also have been
some piezoelectric actuators. And that was all ten years ago. There may
be something new.

I was involved with these things due to a joint project we had going at
Wasino lathes, where I worked in those days. We were using a
magnetostrictive actuator, made of the material used in submarine
sonar. It worked great but programming the hysteresis out of it was a
nightmare. The project was dropped about the time I left.

Anyway, the challenge has been to come up with a programmable system
that provides adequate thrust at high turning speeds. They're doing it,
I think; I just haven't kept up.

Thanks for the information. Sorry I've been slow to reply; I've been
away. I worked myself as part of a team developing electrostatic
actuators a few years back. The actuators consisted of a sheet of a
rubbery material with conductive coatings on each side. When subjected
to a high DC voltage, the conductive coatings could be made to attract
or repel, squeezing or stretching the material (the actuation direction
was perpendicular to the electric field, as this gave a greater
movement). One idea was to use these actuators to move a car wing
mirror, although I'm unsure if a prototype was made as I'd left the
laboratory by then. But the actuators we had at that time were
definitely too flexible to be used for moving a lathe tool.

It requires a lot of force. The material we used, Terfenol-D, is one of
the few that can deliver the force with good speed (several thousand
cycles per second with high force; up to 20 kHz at lower force levels,
IIRC). Piezoelectrics, which are the other option, are used in stacks to
get the required travel, but they're a little fragile for the
application:

http://en.wikipedia.org/wiki/Magnetostriction

Terfenol is the material used in advanced sonar detectors. We were first
looking at it as a counter-vibratory attachment, but my boss got the idea
that it would work for turning elliptical pistons. We had a demo machine
at IMTS 2000, I think, and it attracted a lot of attention. We could get
it to work extremely well and with high accuracy and repeatability. But
if you make a small change in the program, magnetic hysteresis would
complicate the programming, making the whole thing problematic.

--
Ed Huntress


I worked on a product using the Polaroid Piso distance sensors. 1980's.
Problem was the ringing in the piso unit required a separate sender and a
receiver for up close sensing, which is what I was trying to do. Figured
out was not a big enough market to continue on the idea. As to fast,
heavy duty, accurate movers, Voice coils work great. Use a linear
transformer as the sensor. Was the way we build disk drives in the 1970's
early 80's. The magnet on a CDC 200 megabyte drive probably weighs 25-30
pounds total.

There were some reasons that solenoid-type actuators weren't ideal. I think
it was the response rate with the weight of the actuator required for the
job, but it's been too long for me to trust my memory.

A voice coil actuator is not a solenoid in the usual sense. Voice
coil actuators can exhibit fairly linear force vs current over a fair
range of motion, while solenoid force is a highly nonlinear function
of current and distance from closed. They tend to "snap shut" when
actuated, are best-suited for two-position situations like valve
actuators. A voice coil actuator requires a magnet while a solenoid
does not.

Vernier caliper accuracy (machining elliptical pistons)
 

In article ,
"Ed Huntress" wrote:

"Bill McKee" wrote in message
m...

"Ed Huntress" wrote in message
...

"Bill McKee" wrote in message
m...

"Ed Huntress" wrote in message
...

"Christopher Tidy" wrote in message
...
Ed Huntress wrote:

A darned good question. I wish the piston man would step in here
(Anthony?). He's a world-class expert on this subject.

In high-volume production, they've traditionally been turned on
special lathes with cam-operated cross slides. That requires a cam
for each piston profile. Takisawa makes, or made, a programmable
machine that, IIRC, uses solenoid-actuated slides. It's slow. There
also have been some piezoelectric actuators. And that was all ten
years ago. There may be something new.

I was involved with these things due to a joint project we had going
at Wasino lathes, where I worked in those days. We were using a
magnetostrictive actuator, made of the material used in submarine
sonar. It worked great but programming the hysteresis out of it was a
nightmare. The project was dropped about the time I left.

Anyway, the challenge has been to come up with a programmable system
that provides adequate thrust at high turning speeds. They're doing
it, I think; I just haven't kept up.

Thanks for the information. Sorry I've been slow to reply; I've been
away. I worked myself as part of a team developing electrostatic
actuators a few years back. The actuators consisted of a sheet of a
rubbery material with conductive coatings on each side. When subjected
to a high DC voltage, the conductive coatings could be made to attract
or repel, squeezing or stretching the material (the actuation
direction was perpendicular to the electric field, as this gave a
greater movement). One idea was to use these actuators to move a car
wing mirror, although I'm unsure if a prototype was made as I'd left
the laboratory by then. But the actuators we had at that time were
definitely too flexible to be used for moving a lathe tool.

It requires a lot of force. The material we used, Terfenol-D, is one of
the few that can deliver the force with good speed (several thousand
cycles per second with high force; up to 20 kHz at lower force levels,
IIRC). Piezoelectrics, which are the other option, are used in stacks
to get the required travel, but they're a little fragile for the
application:

http://en.wikipedia.org/wiki/Magnetostriction

Terfenol is the material used in advanced sonar detectors. We were
first looking at it as a counter-vibratory attachment, but my boss got
the idea that it would work for turning elliptical pistons. We had a
demo machine at IMTS 2000, I think, and it attracted a lot of
attention. We could get it to work extremely well and with high
accuracy and repeatability. But if you make a small change in the
program, magnetic hysteresis would complicate the programming, making
the whole thing problematic.

--
Ed Huntress


I worked on a product using the Polaroid Piso distance sensors. 1980's.
Problem was the ringing in the piso unit required a separate sender and
a receiver for up close sensing, which is what I was trying to do.
Figured out was not a big enough market to continue on the idea. As to
fast, heavy duty, accurate movers, Voice coils work great. Use a linear
transformer as the sensor. Was the way we build disk drives in the
1970's early 80's. The magnet on a CDC 200 megabyte drive probably
weighs 25-30 pounds total.

There were some reasons that solenoid-type actuators weren't ideal. I
think it was the response rate with the weight of the actuator required
for the job, but it's been too long for me to trust my memory.

I see that Takisawa is still making their line of oval-piston machines
(TPS Series). Maybe there's something on their website about the
actuator:

http://www.takisawa.co.jp/-e/index-e02.htm

--
Ed Huntress


Voice coils and solenoids are totally different beasts.

Right. Don reminded me of that. But as I said to him, the biggest car
manufacturers and lathe builders have all tried everything they can think
of. We're talking about a manufacturing issue that's worth many tens of
millions, if not hundreds of millions, of dollars. There are an awful lot of
pistons made and, until the EPA relented, even Homelite was exploring our
machines for making pistons for string-trimmer engines. All of the car
manufacturers, all over the world, are caught up in it. They want
programmable machines.

If Takisawa is still selling their TPS machines, it seems likely that no
one has succeeded in building anything better. And they all have some
engineer or another who knows how a voice coil works. g

See US Patents 5,085,109 and 5,313,694. The first uses a hydraulic servo
system, the second uses a servo motor of some kind. Lots of other details are
given. Referenced and referencing patents may also be instructive.

Joe Gwinn

Vernier caliper accuracy
 

Don Foreman wrote:
On Thu, 11 Mar 2010 22:07:12 -0500, "Ed Huntress"
wrote:

"Bill McKee" wrote in message
m...
"Ed Huntress" wrote in message
...
"Christopher Tidy" wrote in message
...
Ed Huntress wrote:

A darned good question. I wish the piston man would step in here
(Anthony?). He's a world-class expert on this subject.

In high-volume production, they've traditionally been turned on special
lathes with cam-operated cross slides. That requires a cam for each
piston profile. Takisawa makes, or made, a programmable machine that,
IIRC, uses solenoid-actuated slides. It's slow. There also have been
some piezoelectric actuators. And that was all ten years ago. There may
be something new.

I was involved with these things due to a joint project we had going at
Wasino lathes, where I worked in those days. We were using a
magnetostrictive actuator, made of the material used in submarine
sonar. It worked great but programming the hysteresis out of it was a
nightmare. The project was dropped about the time I left.

Anyway, the challenge has been to come up with a programmable system
that provides adequate thrust at high turning speeds. They're doing it,
I think; I just haven't kept up.
Thanks for the information. Sorry I've been slow to reply; I've been
away. I worked myself as part of a team developing electrostatic
actuators a few years back. The actuators consisted of a sheet of a
rubbery material with conductive coatings on each side. When subjected
to a high DC voltage, the conductive coatings could be made to attract
or repel, squeezing or stretching the material (the actuation direction
was perpendicular to the electric field, as this gave a greater
movement). One idea was to use these actuators to move a car wing
mirror, although I'm unsure if a prototype was made as I'd left the
laboratory by then. But the actuators we had at that time were
definitely too flexible to be used for moving a lathe tool.
It requires a lot of force. The material we used, Terfenol-D, is one of
the few that can deliver the force with good speed (several thousand
cycles per second with high force; up to 20 kHz at lower force levels,
IIRC). Piezoelectrics, which are the other option, are used in stacks to
get the required travel, but they're a little fragile for the
application:

http://en.wikipedia.org/wiki/Magnetostriction

Terfenol is the material used in advanced sonar detectors. We were first
looking at it as a counter-vibratory attachment, but my boss got the idea
that it would work for turning elliptical pistons. We had a demo machine
at IMTS 2000, I think, and it attracted a lot of attention. We could get
it to work extremely well and with high accuracy and repeatability. But
if you make a small change in the program, magnetic hysteresis would
complicate the programming, making the whole thing problematic.

--
Ed Huntress

I worked on a product using the Polaroid Piso distance sensors. 1980's.
Problem was the ringing in the piso unit required a separate sender and a
receiver for up close sensing, which is what I was trying to do. Figured
out was not a big enough market to continue on the idea. As to fast,
heavy duty, accurate movers, Voice coils work great. Use a linear
transformer as the sensor. Was the way we build disk drives in the 1970's
early 80's. The magnet on a CDC 200 megabyte drive probably weighs 25-30
pounds total.
There were some reasons that solenoid-type actuators weren't ideal. I think
it was the response rate with the weight of the actuator required for the
job, but it's been too long for me to trust my memory.

A voice coil actuator is not a solenoid in the usual sense. Voice
coil actuators can exhibit fairly linear force vs current over a fair
range of motion, while solenoid force is a highly nonlinear function
of current and distance from closed. They tend to "snap shut" when
actuated, are best-suited for two-position situations like valve
actuators. A voice coil actuator requires a magnet while a solenoid
does not.

Voice coils tend to be limp, long travel things. They're great for
machines that need to completely isolate one part from another, but
that's the opposite of what you want in a lathe.

You may be able to do it with a voice coil driving a lever, with a
flexure at the other end and the tool holder a lot closer to the flexure
than the coil. It'd be one involved set of work to get it working, and
to make it fast enough to keep up with the spindle.

It'd be a fun development project, for sure.

--
Tim Wescott
Control system and signal processing consulting
www.wescottdesign.com

Vernier caliper accuracy
 

Ed Huntress wrote:
"Christopher Tidy" wrote in message
-- snip --

It would really help to have a good engine man chime in here. There are
several around; try a new thread with a title like "Measuring engine bores"
or something like that. You may drag one up.

Many of us can help you with handling gages but engine cylinders have other
issues, like taper and ovality, etc. If it's an old engine, the pistons may
be cylindrical. If it's a newer one, they're probably elliptical ("oval").
And they're tapered along their lengths, too.

That sounds like what I can reasonably expect to come off of _my_ lathe!
And not because that's what I wanted, either!

--
Tim Wescott
Control system and signal processing consulting
www.wescottdesign.com

Vernier caliper accuracy
 

On Fri, 12 Mar 2010 00:23:53 -0500, "Ed Huntress"
wrote:


"Don Foreman" wrote in message
.. .
On Thu, 11 Mar 2010 22:07:12 -0500, "Ed Huntress"
wrote:


"Bill McKee" wrote in message
news:iuWdnXlu0JJoNgTWnZ2dnUVZ_jOdnZ2d@earthlink .com...

"Ed Huntress" wrote in message
...

"Christopher Tidy" wrote in message
...
Ed Huntress wrote:

A darned good question. I wish the piston man would step in here
(Anthony?). He's a world-class expert on this subject.

In high-volume production, they've traditionally been turned on
special
lathes with cam-operated cross slides. That requires a cam for each
piston profile. Takisawa makes, or made, a programmable machine that,
IIRC, uses solenoid-actuated slides. It's slow. There also have been
some piezoelectric actuators. And that was all ten years ago. There
may
be something new.

I was involved with these things due to a joint project we had going
at
Wasino lathes, where I worked in those days. We were using a
magnetostrictive actuator, made of the material used in submarine
sonar. It worked great but programming the hysteresis out of it was a
nightmare. The project was dropped about the time I left.

Anyway, the challenge has been to come up with a programmable system
that provides adequate thrust at high turning speeds. They're doing
it,
I think; I just haven't kept up.

Thanks for the information. Sorry I've been slow to reply; I've been
away. I worked myself as part of a team developing electrostatic
actuators a few years back. The actuators consisted of a sheet of a
rubbery material with conductive coatings on each side. When subjected
to a high DC voltage, the conductive coatings could be made to attract
or repel, squeezing or stretching the material (the actuation
direction
was perpendicular to the electric field, as this gave a greater
movement). One idea was to use these actuators to move a car wing
mirror, although I'm unsure if a prototype was made as I'd left the
laboratory by then. But the actuators we had at that time were
definitely too flexible to be used for moving a lathe tool.

It requires a lot of force. The material we used, Terfenol-D, is one of
the few that can deliver the force with good speed (several thousand
cycles per second with high force; up to 20 kHz at lower force levels,
IIRC). Piezoelectrics, which are the other option, are used in stacks
to
get the required travel, but they're a little fragile for the
application:

http://en.wikipedia.org/wiki/Magnetostriction

Terfenol is the material used in advanced sonar detectors. We were
first
looking at it as a counter-vibratory attachment, but my boss got the
idea
that it would work for turning elliptical pistons. We had a demo
machine
at IMTS 2000, I think, and it attracted a lot of attention. We could
get
it to work extremely well and with high accuracy and repeatability. But
if you make a small change in the program, magnetic hysteresis would
complicate the programming, making the whole thing problematic.

--
Ed Huntress


I worked on a product using the Polaroid Piso distance sensors. 1980's.
Problem was the ringing in the piso unit required a separate sender and
a
receiver for up close sensing, which is what I was trying to do.
Figured
out was not a big enough market to continue on the idea. As to fast,
heavy duty, accurate movers, Voice coils work great. Use a linear
transformer as the sensor. Was the way we build disk drives in the
1970's
early 80's. The magnet on a CDC 200 megabyte drive probably weighs
25-30
pounds total.

There were some reasons that solenoid-type actuators weren't ideal. I
think
it was the response rate with the weight of the actuator required for the
job, but it's been too long for me to trust my memory.

A voice coil actuator is not a solenoid in the usual sense. Voice
coil actuators can exhibit fairly linear force vs current over a fair
range of motion, while solenoid force is a highly nonlinear function
of current and distance from closed. They tend to "snap shut" when
actuated, are best-suited for two-position situations like valve
actuators. A voice coil actuator requires a magnet while a solenoid
does not.

Right, I forgot about how they work. But rest assured that the piston
manufacturers have tried about everything, and that combination of
programmability, adequate force, and speeds high enough for productive
turning of aluminum pistons has been a son of a gun to accomplish. All sorts
of electromagnetic devices have been tried.

Right. Voice coil actuators are quick but they're not stiff or
high-force devices. That said, the shakers used to vibration-test
aerospace hardware to 20 G's and beyond were voice coil actuated.
Several feet in diameter, water cooled. Helluva ride!

Vernier caliper accuracy (machining elliptical pistons)
 

"Joseph Gwinn" wrote in message
...
In article ,
"Ed Huntress" wrote:

"Bill McKee" wrote in message
m...

"Ed Huntress" wrote in message
...

"Bill McKee" wrote in message
m...

"Ed Huntress" wrote in message
...

"Christopher Tidy" wrote in message
...
Ed Huntress wrote:

A darned good question. I wish the piston man would step in here
(Anthony?). He's a world-class expert on this subject.

In high-volume production, they've traditionally been turned on
special lathes with cam-operated cross slides. That requires a cam
for each piston profile. Takisawa makes, or made, a programmable
machine that, IIRC, uses solenoid-actuated slides. It's slow.
There
also have been some piezoelectric actuators. And that was all ten
years ago. There may be something new.

I was involved with these things due to a joint project we had
going
at Wasino lathes, where I worked in those days. We were using a
magnetostrictive actuator, made of the material used in submarine
sonar. It worked great but programming the hysteresis out of it
was a
nightmare. The project was dropped about the time I left.

Anyway, the challenge has been to come up with a programmable
system
that provides adequate thrust at high turning speeds. They're
doing
it, I think; I just haven't kept up.

Thanks for the information. Sorry I've been slow to reply; I've
been
away. I worked myself as part of a team developing electrostatic
actuators a few years back. The actuators consisted of a sheet of a
rubbery material with conductive coatings on each side. When
subjected
to a high DC voltage, the conductive coatings could be made to
attract
or repel, squeezing or stretching the material (the actuation
direction was perpendicular to the electric field, as this gave a
greater movement). One idea was to use these actuators to move a
car
wing mirror, although I'm unsure if a prototype was made as I'd
left
the laboratory by then. But the actuators we had at that time were
definitely too flexible to be used for moving a lathe tool.

It requires a lot of force. The material we used, Terfenol-D, is one
of
the few that can deliver the force with good speed (several thousand
cycles per second with high force; up to 20 kHz at lower force
levels,
IIRC). Piezoelectrics, which are the other option, are used in
stacks
to get the required travel, but they're a little fragile for the
application:

http://en.wikipedia.org/wiki/Magnetostriction

Terfenol is the material used in advanced sonar detectors. We were
first looking at it as a counter-vibratory attachment, but my boss
got
the idea that it would work for turning elliptical pistons. We had a
demo machine at IMTS 2000, I think, and it attracted a lot of
attention. We could get it to work extremely well and with high
accuracy and repeatability. But if you make a small change in the
program, magnetic hysteresis would complicate the programming,
making
the whole thing problematic.

--
Ed Huntress


I worked on a product using the Polaroid Piso distance sensors.
1980's.
Problem was the ringing in the piso unit required a separate sender
and
a receiver for up close sensing, which is what I was trying to do.
Figured out was not a big enough market to continue on the idea. As
to
fast, heavy duty, accurate movers, Voice coils work great. Use a
linear
transformer as the sensor. Was the way we build disk drives in the
1970's early 80's. The magnet on a CDC 200 megabyte drive probably
weighs 25-30 pounds total.

There were some reasons that solenoid-type actuators weren't ideal. I
think it was the response rate with the weight of the actuator
required
for the job, but it's been too long for me to trust my memory.

I see that Takisawa is still making their line of oval-piston machines
(TPS Series). Maybe there's something on their website about the
actuator:

http://www.takisawa.co.jp/-e/index-e02.htm

--
Ed Huntress


Voice coils and solenoids are totally different beasts.

Right. Don reminded me of that. But as I said to him, the biggest car
manufacturers and lathe builders have all tried everything they can think
of. We're talking about a manufacturing issue that's worth many tens of
millions, if not hundreds of millions, of dollars. There are an awful lot
of
pistons made and, until the EPA relented, even Homelite was exploring our
machines for making pistons for string-trimmer engines. All of the car
manufacturers, all over the world, are caught up in it. They want
programmable machines.

If Takisawa is still selling their TPS machines, it seems likely that no
one has succeeded in building anything better. And they all have some
engineer or another who knows how a voice coil works. g

See US Patents 5,085,109 and 5,313,694. The first uses a hydraulic servo
system, the second uses a servo motor of some kind. Lots of other details
are
given. Referenced and referencing patents may also be instructive.

Joe Gwinn

There are many ideas and many patents for doing this. What's in short supply
is solutions that actually work.

--
Ed Huntress

Vernier caliper accuracy
 

"Tim Wescott" wrote in message
...
Ed Huntress wrote:
"Christopher Tidy" wrote in message
-- snip --

It would really help to have a good engine man chime in here. There are
several around; try a new thread with a title like "Measuring engine
bores" or something like that. You may drag one up.

Many of us can help you with handling gages but engine cylinders have
other issues, like taper and ovality, etc. If it's an old engine, the
pistons may be cylindrical. If it's a newer one, they're probably
elliptical ("oval"). And they're tapered along their lengths, too.

That sounds like what I can reasonably expect to come off of _my_ lathe!
And not because that's what I wanted, either!

--
Tim Wescott
Control system and signal processing consulting
www.wescottdesign.com

Maybe that's how they came up with the first elliptical piston. g

--
Ed Huntress

Vernier caliper accuracy (machining elliptical pistons)
 

In article ,
"Ed Huntress" wrote:

"Joseph Gwinn" wrote in message
...
In article ,
"Ed Huntress" wrote:

"Bill McKee" wrote in message
m...

"Ed Huntress" wrote in message
...

"Bill McKee" wrote in message
m...

"Ed Huntress" wrote in message
...

"Christopher Tidy" wrote in message
...
Ed Huntress wrote:

A darned good question. I wish the piston man would step in here
(Anthony?). He's a world-class expert on this subject.

In high-volume production, they've traditionally been turned on
special lathes with cam-operated cross slides. That requires a cam
for each piston profile. Takisawa makes, or made, a programmable
machine that, IIRC, uses solenoid-actuated slides. It's slow.
There
also have been some piezoelectric actuators. And that was all ten
years ago. There may be something new.

I was involved with these things due to a joint project we had
going
at Wasino lathes, where I worked in those days. We were using a
magnetostrictive actuator, made of the material used in submarine
sonar. It worked great but programming the hysteresis out of it
was a
nightmare. The project was dropped about the time I left.

Anyway, the challenge has been to come up with a programmable
system
that provides adequate thrust at high turning speeds. They're
doing
it, I think; I just haven't kept up.

Thanks for the information. Sorry I've been slow to reply; I've
been
away. I worked myself as part of a team developing electrostatic
actuators a few years back. The actuators consisted of a sheet of a
rubbery material with conductive coatings on each side. When
subjected
to a high DC voltage, the conductive coatings could be made to
attract
or repel, squeezing or stretching the material (the actuation
direction was perpendicular to the electric field, as this gave a
greater movement). One idea was to use these actuators to move a
car
wing mirror, although I'm unsure if a prototype was made as I'd
left
the laboratory by then. But the actuators we had at that time were
definitely too flexible to be used for moving a lathe tool.

It requires a lot of force. The material we used, Terfenol-D, is one
of
the few that can deliver the force with good speed (several thousand
cycles per second with high force; up to 20 kHz at lower force
levels,
IIRC). Piezoelectrics, which are the other option, are used in
stacks
to get the required travel, but they're a little fragile for the
application:

http://en.wikipedia.org/wiki/Magnetostriction

Terfenol is the material used in advanced sonar detectors. We were
first looking at it as a counter-vibratory attachment, but my boss
got
the idea that it would work for turning elliptical pistons. We had a
demo machine at IMTS 2000, I think, and it attracted a lot of
attention. We could get it to work extremely well and with high
accuracy and repeatability. But if you make a small change in the
program, magnetic hysteresis would complicate the programming,
making
the whole thing problematic.

--
Ed Huntress


I worked on a product using the Polaroid Piso distance sensors.
1980's.
Problem was the ringing in the piso unit required a separate sender
and
a receiver for up close sensing, which is what I was trying to do.
Figured out was not a big enough market to continue on the idea. As
to
fast, heavy duty, accurate movers, Voice coils work great. Use a
linear
transformer as the sensor. Was the way we build disk drives in the
1970's early 80's. The magnet on a CDC 200 megabyte drive probably
weighs 25-30 pounds total.

There were some reasons that solenoid-type actuators weren't ideal. I
think it was the response rate with the weight of the actuator
required
for the job, but it's been too long for me to trust my memory.

I see that Takisawa is still making their line of oval-piston machines
(TPS Series). Maybe there's something on their website about the
actuator:

http://www.takisawa.co.jp/-e/index-e02.htm

--
Ed Huntress


Voice coils and solenoids are totally different beasts.

Right. Don reminded me of that. But as I said to him, the biggest car
manufacturers and lathe builders have all tried everything they can think
of. We're talking about a manufacturing issue that's worth many tens of
millions, if not hundreds of millions, of dollars. There are an awful lot
of
pistons made and, until the EPA relented, even Homelite was exploring our
machines for making pistons for string-trimmer engines. All of the car
manufacturers, all over the world, are caught up in it. They want
programmable machines.

If Takisawa is still selling their TPS machines, it seems likely that no
one has succeeded in building anything better. And they all have some
engineer or another who knows how a voice coil works. g

See US Patents 5,085,109 and 5,313,694. The first uses a hydraulic servo
system, the second uses a servo motor of some kind. Lots of other details
are
given. Referenced and referencing patents may also be instructive.

Joe Gwinn

There are many ideas and many patents for doing this. What's in short supply
is solutions that actually work.

These are Takisawa's patents.

Joe Gwinn

Vernier caliper accuracy (machining elliptical pistons)
 

"Joseph Gwinn" wrote in message
...
In article ,
"Ed Huntress" wrote:

"Joseph Gwinn" wrote in message
...
In article ,
"Ed Huntress" wrote:

"Bill McKee" wrote in message
m...

"Ed Huntress" wrote in message
...

"Bill McKee" wrote in message
m...

"Ed Huntress" wrote in message
...

"Christopher Tidy" wrote in message
...
Ed Huntress wrote:

A darned good question. I wish the piston man would step in
here
(Anthony?). He's a world-class expert on this subject.

In high-volume production, they've traditionally been turned on
special lathes with cam-operated cross slides. That requires a
cam
for each piston profile. Takisawa makes, or made, a
programmable
machine that, IIRC, uses solenoid-actuated slides. It's slow.
There
also have been some piezoelectric actuators. And that was all
ten
years ago. There may be something new.

I was involved with these things due to a joint project we had
going
at Wasino lathes, where I worked in those days. We were using a
magnetostrictive actuator, made of the material used in
submarine
sonar. It worked great but programming the hysteresis out of it
was a
nightmare. The project was dropped about the time I left.

Anyway, the challenge has been to come up with a programmable
system
that provides adequate thrust at high turning speeds. They're
doing
it, I think; I just haven't kept up.

Thanks for the information. Sorry I've been slow to reply; I've
been
away. I worked myself as part of a team developing electrostatic
actuators a few years back. The actuators consisted of a sheet
of a
rubbery material with conductive coatings on each side. When
subjected
to a high DC voltage, the conductive coatings could be made to
attract
or repel, squeezing or stretching the material (the actuation
direction was perpendicular to the electric field, as this gave
a
greater movement). One idea was to use these actuators to move a
car
wing mirror, although I'm unsure if a prototype was made as I'd
left
the laboratory by then. But the actuators we had at that time
were
definitely too flexible to be used for moving a lathe tool.

It requires a lot of force. The material we used, Terfenol-D, is
one
of
the few that can deliver the force with good speed (several
thousand
cycles per second with high force; up to 20 kHz at lower force
levels,
IIRC). Piezoelectrics, which are the other option, are used in
stacks
to get the required travel, but they're a little fragile for the
application:

http://en.wikipedia.org/wiki/Magnetostriction

Terfenol is the material used in advanced sonar detectors. We
were
first looking at it as a counter-vibratory attachment, but my
boss
got
the idea that it would work for turning elliptical pistons. We
had a
demo machine at IMTS 2000, I think, and it attracted a lot of
attention. We could get it to work extremely well and with high
accuracy and repeatability. But if you make a small change in the
program, magnetic hysteresis would complicate the programming,
making
the whole thing problematic.

--
Ed Huntress


I worked on a product using the Polaroid Piso distance sensors.
1980's.
Problem was the ringing in the piso unit required a separate
sender
and
a receiver for up close sensing, which is what I was trying to do.
Figured out was not a big enough market to continue on the idea.
As
to
fast, heavy duty, accurate movers, Voice coils work great. Use a
linear
transformer as the sensor. Was the way we build disk drives in
the
1970's early 80's. The magnet on a CDC 200 megabyte drive
probably
weighs 25-30 pounds total.

There were some reasons that solenoid-type actuators weren't ideal.
I
think it was the response rate with the weight of the actuator
required
for the job, but it's been too long for me to trust my memory.

I see that Takisawa is still making their line of oval-piston
machines
(TPS Series). Maybe there's something on their website about the
actuator:

http://www.takisawa.co.jp/-e/index-e02.htm

--
Ed Huntress


Voice coils and solenoids are totally different beasts.

Right. Don reminded me of that. But as I said to him, the biggest car
manufacturers and lathe builders have all tried everything they can
think
of. We're talking about a manufacturing issue that's worth many tens
of
millions, if not hundreds of millions, of dollars. There are an awful
lot
of
pistons made and, until the EPA relented, even Homelite was exploring
our
machines for making pistons for string-trimmer engines. All of the car
manufacturers, all over the world, are caught up in it. They want
programmable machines.

If Takisawa is still selling their TPS machines, it seems likely that
no
one has succeeded in building anything better. And they all have some
engineer or another who knows how a voice coil works. g

See US Patents 5,085,109 and 5,313,694. The first uses a hydraulic
servo
system, the second uses a servo motor of some kind. Lots of other
details
are
given. Referenced and referencing patents may also be instructive.

Joe Gwinn

There are many ideas and many patents for doing this. What's in short
supply
is solutions that actually work.

These are Takisawa's patents.

Joe Gwinn

Oh. g

--
Ed Huntress

Vernier caliper accuracy
 

"Tim Wescott" wrote in message
...
Don Foreman wrote:
On Thu, 11 Mar 2010 22:07:12 -0500, "Ed Huntress"
wrote:

"Bill McKee" wrote in message
m...
"Ed Huntress" wrote in message
...
"Christopher Tidy" wrote in message
...
Ed Huntress wrote:

A darned good question. I wish the piston man would step in here
(Anthony?). He's a world-class expert on this subject.

In high-volume production, they've traditionally been turned on
special lathes with cam-operated cross slides. That requires a cam
for each piston profile. Takisawa makes, or made, a programmable
machine that, IIRC, uses solenoid-actuated slides. It's slow. There
also have been some piezoelectric actuators. And that was all ten
years ago. There may be something new.

I was involved with these things due to a joint project we had going
at Wasino lathes, where I worked in those days. We were using a
magnetostrictive actuator, made of the material used in submarine
sonar. It worked great but programming the hysteresis out of it was
a nightmare. The project was dropped about the time I left.

Anyway, the challenge has been to come up with a programmable system
that provides adequate thrust at high turning speeds. They're doing
it, I think; I just haven't kept up.
Thanks for the information. Sorry I've been slow to reply; I've been
away. I worked myself as part of a team developing electrostatic
actuators a few years back. The actuators consisted of a sheet of a
rubbery material with conductive coatings on each side. When
subjected to a high DC voltage, the conductive coatings could be made
to attract or repel, squeezing or stretching the material (the
actuation direction was perpendicular to the electric field, as this
gave a greater movement). One idea was to use these actuators to move
a car wing mirror, although I'm unsure if a prototype was made as I'd
left the laboratory by then. But the actuators we had at that time
were definitely too flexible to be used for moving a lathe tool.
It requires a lot of force. The material we used, Terfenol-D, is one
of the few that can deliver the force with good speed (several
thousand cycles per second with high force; up to 20 kHz at lower
force levels, IIRC). Piezoelectrics, which are the other option, are
used in stacks to get the required travel, but they're a little
fragile for the application:

http://en.wikipedia.org/wiki/Magnetostriction

Terfenol is the material used in advanced sonar detectors. We were
first looking at it as a counter-vibratory attachment, but my boss got
the idea that it would work for turning elliptical pistons. We had a
demo machine at IMTS 2000, I think, and it attracted a lot of
attention. We could get it to work extremely well and with high
accuracy and repeatability. But if you make a small change in the
program, magnetic hysteresis would complicate the programming, making
the whole thing problematic.

--
Ed Huntress

I worked on a product using the Polaroid Piso distance sensors.
1980's. Problem was the ringing in the piso unit required a separate
sender and a receiver for up close sensing, which is what I was trying
to do. Figured out was not a big enough market to continue on the
idea. As to fast, heavy duty, accurate movers, Voice coils work great.
Use a linear transformer as the sensor. Was the way we build disk
drives in the 1970's early 80's. The magnet on a CDC 200 megabyte
drive probably weighs 25-30 pounds total.
There were some reasons that solenoid-type actuators weren't ideal. I
think it was the response rate with the weight of the actuator required
for the job, but it's been too long for me to trust my memory.

A voice coil actuator is not a solenoid in the usual sense. Voice
coil actuators can exhibit fairly linear force vs current over a fair
range of motion, while solenoid force is a highly nonlinear function
of current and distance from closed. They tend to "snap shut" when
actuated, are best-suited for two-position situations like valve
actuators. A voice coil actuator requires a magnet while a solenoid
does not.

Voice coils tend to be limp, long travel things. They're great for
machines that need to completely isolate one part from another, but that's
the opposite of what you want in a lathe.

You may be able to do it with a voice coil driving a lever, with a flexure
at the other end and the tool holder a lot closer to the flexure than the
coil. It'd be one involved set of work to get it working, and to make it
fast enough to keep up with the spindle.

It'd be a fun development project, for sure.

--
Tim Wescott
Control system and signal processing consulting
www.wescottdesign.com

Voice coils are can also be short travel and very precise. I designed disk
drives and head actuators buy and large are voice coils. The positioning
track to track can be short or far and keeping the head on track is very
precise.

Vernier caliper accuracy
 

On 2010-03-12, Bill McKee wrote:

"Tim Wescott" wrote in message
...

[ ... ]

Voice coils tend to be limp, long travel things. They're great for
machines that need to completely isolate one part from another, but that's
the opposite of what you want in a lathe.

You may be able to do it with a voice coil driving a lever, with a flexure
at the other end and the tool holder a lot closer to the flexure than the
coil. It'd be one involved set of work to get it working, and to make it
fast enough to keep up with the spindle.

[ ... ]

Voice coils are can also be short travel and very precise. I designed disk
drives and head actuators buy and large are voice coils. The positioning
track to track can be short or far and keeping the head on track is very
precise.

They are precise only because they get feedback from the heads
and "servo tracks" written on one surface (at least the ones in sealed
hard drives, though I think that the old 5MB removable platter ones used
an an optical sensor to determine position, because in those days every
surface was precious.

Enjoy,
DoN.

--
Email: | Voice (all times): (703) 938-4564
(too) near Washington D.C. | http://www.d-and-d.com/dnichols/DoN.html
--- Black Holes are where God is dividing by zero ---

Vernier caliper accuracy
 

"DoN. Nichols" wrote in message
...
On 2010-03-12, Bill McKee wrote:

"Tim Wescott" wrote in message
...

[ ... ]

Voice coils tend to be limp, long travel things. They're great for
machines that need to completely isolate one part from another, but
that's
the opposite of what you want in a lathe.

You may be able to do it with a voice coil driving a lever, with a
flexure
at the other end and the tool holder a lot closer to the flexure than
the
coil. It'd be one involved set of work to get it working, and to make
it
fast enough to keep up with the spindle.

[ ... ]

Voice coils are can also be short travel and very precise. I designed
disk
drives and head actuators buy and large are voice coils. The positioning
track to track can be short or far and keeping the head on track is very
precise.

They are precise only because they get feedback from the heads
and "servo tracks" written on one surface (at least the ones in sealed
hard drives, though I think that the old 5MB removable platter ones used
an an optical sensor to determine position, because in those days every
surface was precious.

Enjoy,
DoN.

--


They also used linear transformers. The CDC 2 mb cartrige disk was this
way.

Vernier caliper accuracy
 

Bill McKee wrote:
"Tim Wescott" wrote in message
...
Don Foreman wrote:
On Thu, 11 Mar 2010 22:07:12 -0500, "Ed Huntress"
wrote:

"Bill McKee" wrote in message
m...
"Ed Huntress" wrote in message
...
"Christopher Tidy" wrote in message
...
Ed Huntress wrote:

A darned good question. I wish the piston man would step in here
(Anthony?). He's a world-class expert on this subject.

In high-volume production, they've traditionally been turned on
special lathes with cam-operated cross slides. That requires a cam
for each piston profile. Takisawa makes, or made, a programmable
machine that, IIRC, uses solenoid-actuated slides. It's slow. There
also have been some piezoelectric actuators. And that was all ten
years ago. There may be something new.

I was involved with these things due to a joint project we had going
at Wasino lathes, where I worked in those days. We were using a
magnetostrictive actuator, made of the material used in submarine
sonar. It worked great but programming the hysteresis out of it was
a nightmare. The project was dropped about the time I left.

Anyway, the challenge has been to come up with a programmable system
that provides adequate thrust at high turning speeds. They're doing
it, I think; I just haven't kept up.
Thanks for the information. Sorry I've been slow to reply; I've been
away. I worked myself as part of a team developing electrostatic
actuators a few years back. The actuators consisted of a sheet of a
rubbery material with conductive coatings on each side. When
subjected to a high DC voltage, the conductive coatings could be made
to attract or repel, squeezing or stretching the material (the
actuation direction was perpendicular to the electric field, as this
gave a greater movement). One idea was to use these actuators to move
a car wing mirror, although I'm unsure if a prototype was made as I'd
left the laboratory by then. But the actuators we had at that time
were definitely too flexible to be used for moving a lathe tool.
It requires a lot of force. The material we used, Terfenol-D, is one
of the few that can deliver the force with good speed (several
thousand cycles per second with high force; up to 20 kHz at lower
force levels, IIRC). Piezoelectrics, which are the other option, are
used in stacks to get the required travel, but they're a little
fragile for the application:

http://en.wikipedia.org/wiki/Magnetostriction

Terfenol is the material used in advanced sonar detectors. We were
first looking at it as a counter-vibratory attachment, but my boss got
the idea that it would work for turning elliptical pistons. We had a
demo machine at IMTS 2000, I think, and it attracted a lot of
attention. We could get it to work extremely well and with high
accuracy and repeatability. But if you make a small change in the
program, magnetic hysteresis would complicate the programming, making
the whole thing problematic.

--
Ed Huntress

I worked on a product using the Polaroid Piso distance sensors.
1980's. Problem was the ringing in the piso unit required a separate
sender and a receiver for up close sensing, which is what I was trying
to do. Figured out was not a big enough market to continue on the
idea. As to fast, heavy duty, accurate movers, Voice coils work great.
Use a linear transformer as the sensor. Was the way we build disk
drives in the 1970's early 80's. The magnet on a CDC 200 megabyte
drive probably weighs 25-30 pounds total.
There were some reasons that solenoid-type actuators weren't ideal. I
think it was the response rate with the weight of the actuator required
for the job, but it's been too long for me to trust my memory.
A voice coil actuator is not a solenoid in the usual sense. Voice
coil actuators can exhibit fairly linear force vs current over a fair
range of motion, while solenoid force is a highly nonlinear function
of current and distance from closed. They tend to "snap shut" when
actuated, are best-suited for two-position situations like valve
actuators. A voice coil actuator requires a magnet while a solenoid
does not.
Voice coils tend to be limp, long travel things. They're great for
machines that need to completely isolate one part from another, but that's
the opposite of what you want in a lathe.

You may be able to do it with a voice coil driving a lever, with a flexure
at the other end and the tool holder a lot closer to the flexure than the
coil. It'd be one involved set of work to get it working, and to make it
fast enough to keep up with the spindle.

It'd be a fun development project, for sure.

--
Tim Wescott
Control system and signal processing consulting
www.wescottdesign.com

Voice coils are can also be short travel and very precise. I designed disk
drives and head actuators buy and large are voice coils. The positioning
track to track can be short or far and keeping the head on track is very
precise.


But a voice coil _by itself_ is infinitely floppy. Cut the power -- or
just supply it with a constant current -- and it displays zero spring
rate, zero damping rate, just a constant force.

It's just what you want for a disk drive -- speedy, no hysteresis,
rigidity doesn't matter much because the whole assembly is mechanically
quiet, etc.

(take apart a hard drive today and you'll find a voice coil, with some
Really Strong rare earth magnets providing a field)

But this is quite the opposite of what you need on a lathe. To get the
rigidity from the voice coil itself you need an exceedingly high
bandwidth on your control loop, and to develop lots of force you need a
physically large coil with lots of inductance -- which is going to cause
all sorts of electrical difficulties.

Not having tried this I couldn't say for sure, but _having_ worked on
voice coil actuated control loops that shove a 30 pound gimbal around, I
can say that you're probably going to be at least two orders of
magnitude short of nirvana with just a voice coil, sensor, and 'the
usual' electronics.

--
Tim Wescott
Control system and signal processing consulting
www.wescottdesign.com

Vernier caliper accuracy
 

DoN. Nichols wrote:
On 2010-03-12, Bill McKee wrote:
"Tim Wescott" wrote in message
...

[ ... ]

Voice coils tend to be limp, long travel things. They're great for
machines that need to completely isolate one part from another, but that's
the opposite of what you want in a lathe.

You may be able to do it with a voice coil driving a lever, with a flexure
at the other end and the tool holder a lot closer to the flexure than the
coil. It'd be one involved set of work to get it working, and to make it
fast enough to keep up with the spindle.

[ ... ]

Voice coils are can also be short travel and very precise. I designed disk
drives and head actuators buy and large are voice coils. The positioning
track to track can be short or far and keeping the head on track is very
precise.

They are precise only because they get feedback from the heads
and "servo tracks" written on one surface (at least the ones in sealed
hard drives, though I think that the old 5MB removable platter ones used
an an optical sensor to determine position, because in those days every
surface was precious.

The are also precise because they're not responsible for holding a
cutting tool in position when the darn thing just wants to chatter!

Lathe == more rigidity is better.
Voice coil == infinitely compliant.

--
Tim Wescott
Control system and signal processing consulting
www.wescottdesign.com

Vernier caliper accuracy (machining elliptical pistons)
 

Joseph Gwinn wrote:

If Takisawa is still selling their TPS machines, it seems likely that no
one has succeeded in building anything better. And they all have some
engineer or another who knows how a voice coil works. g


See US Patents 5,085,109 and 5,313,694. The first uses a hydraulic servo
system, the second uses a servo motor of some kind. Lots of other details are
given. Referenced and referencing patents may also be instructive.

Thanks for the links to the patents. Those systems are more like what I
was imagining. I was surprised that the magnetostrictive actuators Ed
mentioned can give a great enough distance of travel. To what degree are
pistons typically elliptical, relative to their diameter? Is it visible
with the naked eye?

Best wishes,

Chris

Vernier caliper accuracy (machining elliptical pistons)
 

In article ,
Christopher Tidy wrote:

Joseph Gwinn wrote:

If Takisawa is still selling their TPS machines, it seems likely that no
one has succeeded in building anything better. And they all have some
engineer or another who knows how a voice coil works. g


See US Patents 5,085,109 and 5,313,694. The first uses a hydraulic servo
system, the second uses a servo motor of some kind. Lots of other details
are given. Referenced and referencing patents may also be instructive.

Thanks for the links to the patents. Those systems are more like what I
was imagining. I was surprised that the magnetostrictive actuators Ed
mentioned can give a great enough distance of travel. To what degree are
pistons typically elliptical, relative to their diameter? Is it visible
with the naked eye?

I don't really know how big a deviation from circular is needed, but I'd guess
that it isn't much, and isn't visible.

Magnetostrictive actuators generate very high force levels, so one can use
levers to trade force for stroke. The length of a Terfenol-D rod changes by
about one part per thousand as one goes from zero to 2,000 gauss field.

Joe Gwinn

Vernier caliper accuracy (machining elliptical pistons)
 

Joseph Gwinn wrote:

I don't really know how big a deviation from circular is needed, but I'd guess
that it isn't much, and isn't visible.

Ah, right. I was thinking of a lathe that can turn parts that are
noticeably elliptical to the naked eye. I imagine it's simpler to design
a machine if the workpiece only needs to be slightly elliptical.

Chris

Vernier caliper accuracy (machining elliptical pistons)
 

"Christopher Tidy" wrote in message
...
Joseph Gwinn wrote:

I don't really know how big a deviation from circular is needed, but I'd
guess that it isn't much, and isn't visible.

Ah, right. I was thinking of a lathe that can turn parts that are
noticeably elliptical to the naked eye. I imagine it's simpler to design a
machine if the workpiece only needs to be slightly elliptical.

Chris

Typical ovality runs around 0.010 to 0.040 inches, depending upon the piston
size and purpose. It can be more or less. It could be enough to see, if you
took two identical pistons and laid them side-by-side on a surface plate,
and oriented them 90 deg. to each other in rotation.

This is for modern cast pistons. Forged pistons are made with an alloy that
has a significantly higher coefficient of thermal expansion. You might want
to check on diesel engine pistons to see what their range is.

I haven't looked, but I expect that you'll find good information about them
at the SAE's website. If you're Googling, look for "cam-ground pistons."

--
Ed Huntress

Vernier caliper accuracy (machining elliptical pistons)
 

Ed Huntress wrote:

Typical ovality runs around 0.010 to 0.040 inches, depending upon the piston
size and purpose. It can be more or less. It could be enough to see, if you
took two identical pistons and laid them side-by-side on a surface plate,
and oriented them 90 deg. to each other in rotation.

Hard to see then. Not quite beer barrel proportions.

This is for modern cast pistons. Forged pistons are made with an alloy that
has a significantly higher coefficient of thermal expansion. You might want
to check on diesel engine pistons to see what their range is.

I haven't looked, but I expect that you'll find good information about them
at the SAE's website. If you're Googling, look for "cam-ground pistons."

Found a good link he
http://www.autospeednet.com/sites/3d...viewterm/3332/

Thanks for the knowledge you've shared in this thread!

Best wishes,

Chris

Vernier caliper accuracy (machining elliptical pistons)
 

"Christopher Tidy" wrote in message
...
Ed Huntress wrote:

Typical ovality runs around 0.010 to 0.040 inches, depending upon the
piston size and purpose. It can be more or less. It could be enough to
see, if you took two identical pistons and laid them side-by-side on a
surface plate, and oriented them 90 deg. to each other in rotation.

Hard to see then. Not quite beer barrel proportions.

This is for modern cast pistons. Forged pistons are made with an alloy
that has a significantly higher coefficient of thermal expansion. You
might want to check on diesel engine pistons to see what their range is.

I haven't looked, but I expect that you'll find good information about
them at the SAE's website. If you're Googling, look for "cam-ground
pistons."

Found a good link he
http://www.autospeednet.com/sites/3d...viewterm/3332/

That's good. Remember, as I mentioned with the Ford pistons, the latest and
greatest can have several different profiles from crown to skirt.


Thanks for the knowledge you've shared in this thread!

I only wish the *real* expert here was around. He's an engineer with a
piston company you would recognize, but which I'm sworn not to name. g He
knows all the subleties, including ones he can't tell *us* about.

Anyway, 'glad you found it useful.

--
Ed Huntress