Would a solenoid work?

[Jonathan Montoya]

Hi, I've done a couple microcontroller projects, but otherwise I'm
pretty new to electronics. I recently bought an antique typewriter
that I'd like to automate for fun. My idea is that I could mount
solenoids under each key(which are essentially levers that push up on
some rods) and pull down with the same force that it would take press
key. I haven't measured how much force is necessary yet, and the
stroke is probably about an inch.

Before I started doing some expirements, I wanted to ask the experts
to see if this is at all feasable. Would solenoids be the best way to
do this? I picked them because they seemed like they'd better be able
to simulate the fast sharp action needed to get the antique key to
impress the ink on the paper. Will I be able to jam 50 or so of these
parallel and under the keys (given that they'll be mounted under a
desk or something like it), or will the size for the work needed
require more creative mounting? And, am I assuming correctly that I'd
have to make these myself, since they probably have speifications
different from what most solenoids are made for?

I appreciate any guidance I can get.

Thanks,

Lonlaz

 

[Tim Wescott]

On Mon, 07 Apr 2008 09:19:58 -0700, John Mianowski wrote:

On Apr 7, 10:37 am, Jonathan Montoya <lonlaza...@gmail.com> wrote:
Hi, I've done a couple microcontroller projects, but otherwise I'm
pretty new to electronics. I recently bought an antique typewriter
that I'd like to automate for fun. My idea is that I could mount
solenoids under each key(which are essentially levers that push up on
some rods) and pull down with the same force that it would take press
key. I haven't measured how much force is necessary yet, and the
stroke is probably about an inch.

Before I started doing some expirements, I wanted to ask the experts to
see if this is at all feasable. Would solenoids be the best way to do
this? I picked them because they seemed like they'd better be able to
simulate the fast sharp action needed to get the antique key to impress
the ink on the paper. Will I be able to jam 50 or so of these parallel
and under the keys (given that they'll be mounted under a desk or
something like it), or will the size for the work needed require more
creative mounting? And, am I assuming correctly that I'd have to make
these myself, since they probably have speifications different from
what most solenoids are made for?

I appreciate any guidance I can get.

Thanks,

Lonlaz

1 solenoid per key * how many keys?

You might be able to extend the stroke of a solenoid by using a lever.

Then again, I think I'd go with a pneumatic solution. Use a solenoid
valve to open/close air flow to a cylinder that presses the keys. It's
easy to find pneumatic cylinders with the right stroke & power
combination you'll need.

Instead of 1 "presser" per key, how about some smaller number that get
moved around using something similar to a X-Y table? It might be fun to
build 9 key pressers that emulate human fingers in the same finger-
to-key assignments used in normal touch typing (4 home keys on each
hand, plus a thumb on the space bar). This could consist of 2 moving
modules, with 4 pressers on the left & 5 on the right. I think that
would be fun to watch in action, & I wouldn't be surprised if somebody
else hadn't done it already. Make it programmable, so that any
character gets translated to a sequence of X-Y movements followed by a
press (2 presses if you need to hold down the Shift key) & you can use
it to type anything you want.

Is this going to be used on a manual or electronic typewriter? If
manual, you'll need to add a carriage return as well. I think
pneumatics would work fine for this as well, using a 3rd (fixed) unit.
If it's only supposed to work on an electronic typewriter, then just
pressing the "Return" with the RH "little finger" would do.

JM

Little robot hands, typing. What a cool thought.

--
Tim Wescott
Control systems and communications consulting
http://www.wescottdesign.com

Need to learn how to apply control theory in your embedded system?
"Applied Control Theory for Embedded Systems" by Tim Wescott
Elsevier/Newnes, http://www.wescottdesign.com/actfes/actfes.html

 

[Tim Wescott]

On Mon, 07 Apr 2008 08:37:03 -0700, Jonathan Montoya wrote:

Hi, I've done a couple microcontroller projects, but otherwise I'm
pretty new to electronics. I recently bought an antique typewriter that
I'd like to automate for fun. My idea is that I could mount solenoids
under each key(which are essentially levers that push up on some rods)
and pull down with the same force that it would take press key. I
haven't measured how much force is necessary yet, and the stroke is
probably about an inch.

Before I started doing some expirements, I wanted to ask the experts to
see if this is at all feasable. Would solenoids be the best way to do
this? I picked them because they seemed like they'd better be able to
simulate the fast sharp action needed to get the antique key to impress
the ink on the paper. Will I be able to jam 50 or so of these parallel
and under the keys (given that they'll be mounted under a desk or
something like it), or will the size for the work needed require more
creative mounting? And, am I assuming correctly that I'd have to make
these myself, since they probably have speifications different from what
most solenoids are made for?

I appreciate any guidance I can get.

Thanks,

Lonlaz

Manual typewriters take a lot of force, over that 1" stroke you mention.
That's going to take a _big_ solenoid. Yes you can do it, and you can
probably find one off the shelf, but you'll fill the underside of that
desk with all the solenoids.

The electric typewriters that I used to mess with had a motor that
rumbled away when the thing was on; I assume that the mechanism involved
a one-turn clutch to make the ball actually strike the paper.

--
Tim Wescott
Control systems and communications consulting
http://www.wescottdesign.com

Need to learn how to apply control theory in your embedded system?
"Applied Control Theory for Embedded Systems" by Tim Wescott
Elsevier/Newnes, http://www.wescottdesign.com/actfes/actfes.html

 

[John Mianowski]

On Apr 7, 10:37 am, Jonathan Montoya <lonlaza...@gmail.com> wrote:

Hi, I've done a couple microcontroller projects, but otherwise I'm
pretty new to electronics. I recently bought an antique typewriter
that I'd like to automate for fun. My idea is that I could mount
solenoids under each key(which are essentially levers that push up on
some rods) and pull down with the same force that it would take press
key. I haven't measured how much force is necessary yet, and the
stroke is probably about an inch.

Before I started doing some expirements, I wanted to ask the experts
to see if this is at all feasable. Would solenoids be the best way to
do this? I picked them because they seemed like they'd better be able
to simulate the fast sharp action needed to get the antique key to
impress the ink on the paper. Will I be able to jam 50 or so of these
parallel and under the keys (given that they'll be mounted under a
desk or something like it), or will the size for the work needed
require more creative mounting? And, am I assuming correctly that I'd
have to make these myself, since they probably have speifications
different from what most solenoids are made for?

I appreciate any guidance I can get.

Thanks,

Lonlaz

1 solenoid per key * how many keys?

You might be able to extend the stroke of a solenoid by using a lever.

Then again, I think I'd go with a pneumatic solution. Use a solenoid
valve to open/close air flow to a cylinder that presses the keys.
It's easy to find pneumatic cylinders with the right stroke & power
combination you'll need.

Instead of 1 "presser" per key, how about some smaller number that get
moved around using something similar to a X-Y table? It might be fun
to build 9 key pressers that emulate human fingers in the same finger-
to-key assignments used in normal touch typing (4 home keys on each
hand, plus a thumb on the space bar). This could consist of 2 moving
modules, with 4 pressers on the left & 5 on the right. I think that
would be fun to watch in action, & I wouldn't be surprised if somebody
else hadn't done it already. Make it programmable, so that any
character gets translated to a sequence of X-Y movements followed by a
press (2 presses if you need to hold down the Shift key) & you can use
it to type anything you want.

Is this going to be used on a manual or electronic typewriter? If
manual, you'll need to add a carriage return as well. I think
pneumatics would work fine for this as well, using a 3rd (fixed)
unit. If it's only supposed to work on an electronic typewriter, then
just pressing the "Return" with the RH "little finger" would do.

JM

 

[Mike Silva]

On Apr 7, 11:37 am, Jonathan Montoya <lonlaza...@gmail.com> wrote:

Hi, I've done a couple microcontroller projects, but otherwise I'm
pretty new to electronics.  I recently bought an antique typewriter
that I'd like to automate for fun.  My idea is that I could mount
solenoids under each key(which are essentially levers that push up on
some rods) and pull down with the same force that it would take press
key.  I haven't measured how much force is necessary yet, and the
stroke is probably about an inch.

Before I started doing some expirements, I wanted to ask the experts
to see if this is at all feasable.  Would solenoids be the best way to
do this? I picked them because they seemed like they'd better be able
to simulate the fast sharp action needed to get the antique key to
impress the ink on the paper.  Will I be able to jam 50 or so of these
parallel and under the keys (given that they'll be mounted under a
desk or something like it), or will the size for the work needed
require more creative mounting?  And, am I assuming correctly that I'd
have to make these myself, since they probably have speifications
different from what most solenoids are made for?

I appreciate any guidance I can get.

Thanks,

Lonlaz

Wow, deja vu! I wanted to do this in about 1980, and I think I even
bought a load of surplus solenoids to do it, but never went beyond
that. Instead I bought a dot-matrix printing mechanism, built a
driver board and made a printer driven by my Z-80 CP/M system. Hint
on that one - buy lots of extra printhead pins to replace the ones you
break while debugging the software!

Theoretically it is certainly possible. If you use stiff wires or
rods between the keys and the solenoids that should make your physical
construction a lot easier. And make sure your power supply, drivers
and circuitry can all handle the current spikes (and resultant
noise). Use a short-timeout watchdog or gate the firing pulses with
an external signal (555?) to assure that you don't hold a solenoid
down too long by mistake. Make sure all your solenoid drivers power
up in the OFF state (imagine all 50 trying to fire on power-up!).

And be sure and post a video to youtube when you've got it
working.

 

[Jonathan Montoya]

On Apr 7, 11:19 am, John Mianowski <spamf...@skytex.net> wrote:

On Apr 7, 10:37 am, Jonathan Montoya <lonlaza...@gmail.com> wrote:





Hi, I've done a couple microcontroller projects, but otherwise I'm
pretty new to electronics.  I recently bought an antique typewriter
that I'd like to automate for fun.  My idea is that I could mount
solenoids under each key(which are essentially levers that push up on
some rods) and pull down with the same force that it would take press
key.  I haven't measured how much force is necessary yet, and the
stroke is probably about an inch.

snip
I appreciate any guidance I can get.

Thanks,

Lonlaz

1 solenoid per key * how many keys?

You might be able to extend the stroke of a solenoid by using a lever.

Then again, I think I'd go with a pneumatic solution.  Use a solenoid
valve to open/close air flow to a cylinder that presses the keys.
It's easy to find pneumatic cylinders with the right stroke & power
combination you'll need.

Instead of 1 "presser" per key, how about some smaller number that get
moved around using something similar to a X-Y table?  It might be fun
to build 9 key pressers that emulate human fingers in the same finger-
to-key assignments used in normal touch typing (4 home keys on each
hand, plus a thumb on the space bar).  This could consist of 2 moving
modules,  with 4 pressers on the left & 5 on the right.  I think that
would be fun to watch in action, & I wouldn't be surprised if somebody
else hadn't done it already.  Make it programmable, so that any
character gets translated to a sequence of X-Y movements followed by a
press (2 presses if you need to hold down the Shift key) & you can use
it to type anything you want.

Is this going to be used on a manual or electronic typewriter?  If
manual, you'll need to add a carriage return as well.  I think
pneumatics would work fine for this as well, using a 3rd (fixed)
unit.  If it's only supposed to work on an electronic typewriter, then
just pressing the "Return" with the RH "little finger" would do.

JM- Hide quoted text -

- Show quoted text -

Well, I should probably have expanded a bit more. Not only do I want
the keyboard to type on itself, I'd like to be able to type on it as
well.
I thougt I'd make a kind of a 'manual' word processor. You could type
a
page, it would store the text info, and you could print it back. I
wonder
if I can't adapt an X Y solution under the keyboard if the solenoids
are
too cumbersome.

I think it has 45 keys, I'll have to count again tonight.

Lonlaz

 

[Jonathan Montoya]

On Apr 7, 11:20 am, Mike Silva <snarflem...@yahoo.com> wrote:

On Apr 7, 11:37 am, Jonathan Montoya <lonlaza...@gmail.com> wrote:





Hi, I've done a couple microcontroller projects, but otherwise I'm
pretty new to electronics.  I recently bought an antique typewriter
that I'd like to automate for fun.  My idea is that I could mount
solenoids under each key(which are essentially levers that push up on
some rods) and pull down with the same force that it would take press
key.  I haven't measured how much force is necessary yet, and the
stroke is probably about an inch.

Before I started doing some expirements, I wanted to ask the experts
to see if this is at all feasable.  Would solenoids be the best way to
do this? I picked them because they seemed like they'd better be able
to simulate the fast sharp action needed to get the antique key to
impress the ink on the paper.  Will I be able to jam 50 or so of these
parallel and under the keys (given that they'll be mounted under a
desk or something like it), or will the size for the work needed
require more creative mounting?  And, am I assuming correctly that I'd
have to make these myself, since they probably have speifications
different from what most solenoids are made for?

I appreciate any guidance I can get.

Thanks,

Lonlaz

Wow, deja vu!  I wanted to do this in about 1980, and I think I even
bought a load of surplus solenoids to do it, but never went beyond
that.  Instead I bought a dot-matrix printing mechanism, built a
driver board and made a printer driven by my Z-80 CP/M system.  Hint
on that one - buy lots of extra printhead pins to replace the ones you
break while debugging the software!

Theoretically it is certainly possible.  If you use stiff wires or
rods between the keys and the solenoids that should make your physical
construction a lot easier.  And make sure your power supply, drivers
and circuitry can all handle the current spikes (and resultant
noise).  Use a short-timeout watchdog or gate the firing pulses with
an external signal (555?) to assure that you don't hold a solenoid
down too long by mistake.  Make sure all your solenoid drivers power
up in the OFF state (imagine all 50 trying to fire on power-up!).

And be sure and post a video to youtube when you've got it
working. - Hide quoted text -

- Show quoted text -

Heh, look for it on You Tube in about 15 years.

Lonlaz

 

[Jonathan Montoya]

On Apr 7, 11:26 am, Tim Wescott <t...@seemywebsite.com> wrote:

On Mon, 07 Apr 2008 08:37:03 -0700, Jonathan Montoya wrote:
Hi, I've done a couple microcontroller projects, but otherwise I'm
pretty new to electronics.  I recently bought an antique typewriter that
I'd like to automate for fun.  My idea is that I could mount solenoids
under each key(which are essentially levers that push up on some rods)
and pull down with the same force that it would take press key.  I
haven't measured how much force is necessary yet, and the stroke is
probably about an inch.

Before I started doing some expirements, I wanted to ask the experts to
see if this is at all feasable.  Would solenoids be the best way to do
this? I picked them because they seemed like they'd better be able to
simulate the fast sharp action needed to get the antique key to impress
the ink on the paper.  Will I be able to jam 50 or so of these parallel
and under the keys (given that they'll be mounted under a desk or
something like it), or will the size for the work needed require more
creative mounting?  And, am I assuming correctly that I'd have to make
these myself, since they probably have speifications different from what
most solenoids are made for?

I appreciate any guidance I can get.

Thanks,

Lonlaz

Manual typewriters take a lot of force, over that 1" stroke you mention.  
That's going to take a _big_ solenoid.  Yes you can do it, and you can
probably find one off the shelf, but you'll fill the underside of that
desk with all the solenoids.

The electric typewriters that I used to mess with had a motor that
rumbled away when the thing was on; I assume that the mechanism involved
a one-turn clutch to make the ball actually strike the paper.

--
Tim Wescott
Control systems and communications consultinghttp://www.wescottdesign.com

Need to learn how to apply control theory in your embedded system?
"Applied Control Theory for Embedded Systems" by Tim Wescott
Elsevier/Newnes,http://www.wescottdesign.com/actfes/actfes.html- Hide quoted text -

- Show quoted text -

I don't want the whole apparatus to too large. I'll have to test the
force it takes
tonight.

Lonlaz

 

[Tim Wescott]

Jonathan Montoya wrote:

On Apr 7, 11:19 am, John Mianowski <spamf...@skytex.net> wrote:
On Apr 7, 10:37 am, Jonathan Montoya <lonlaza...@gmail.com> wrote:





Hi, I've done a couple microcontroller projects, but otherwise I'm
pretty new to electronics. I recently bought an antique typewriter
that I'd like to automate for fun. My idea is that I could mount
solenoids under each key(which are essentially levers that push up on
some rods) and pull down with the same force that it would take press
key. I haven't measured how much force is necessary yet, and the
stroke is probably about an inch.
snip
I appreciate any guidance I can get.
Thanks,
Lonlaz
1 solenoid per key * how many keys?

You might be able to extend the stroke of a solenoid by using a lever.

Then again, I think I'd go with a pneumatic solution. Use a solenoid
valve to open/close air flow to a cylinder that presses the keys.
It's easy to find pneumatic cylinders with the right stroke & power
combination you'll need.

Instead of 1 "presser" per key, how about some smaller number that get
moved around using something similar to a X-Y table? It might be fun
to build 9 key pressers that emulate human fingers in the same finger-
to-key assignments used in normal touch typing (4 home keys on each
hand, plus a thumb on the space bar). This could consist of 2 moving
modules, with 4 pressers on the left & 5 on the right. I think that
would be fun to watch in action, & I wouldn't be surprised if somebody
else hadn't done it already. Make it programmable, so that any
character gets translated to a sequence of X-Y movements followed by a
press (2 presses if you need to hold down the Shift key) & you can use
it to type anything you want.

Is this going to be used on a manual or electronic typewriter? If
manual, you'll need to add a carriage return as well. I think
pneumatics would work fine for this as well, using a 3rd (fixed)
unit. If it's only supposed to work on an electronic typewriter, then
just pressing the "Return" with the RH "little finger" would do.

JM- Hide quoted text -

- Show quoted text -

Well, I should probably have expanded a bit more. Not only do I want
the keyboard to type on itself, I'd like to be able to type on it as
well.
I thougt I'd make a kind of a 'manual' word processor. You could type
a
page, it would store the text info, and you could print it back. I
wonder
if I can't adapt an X Y solution under the keyboard if the solenoids
are
too cumbersome.

I think it has 45 keys, I'll have to count again tonight.

Lonlaz

So put the button-pusher in a box that can register its position to the
typewriter, so you can easily lift it away.

I still vote for robot hands that look like hands, probably because I'm
not the one who has to make it work...

--

Tim Wescott
Wescott Design Services
http://www.wescottdesign.com

Do you need to implement control loops in software?
"Applied Control Theory for Embedded Systems" gives you just what it says.
See details at http://www.wescottdesign.com/actfes/actfes.html

 

[whit3rd]

On Apr 7, 8:37 am, Jonathan Montoya <lonlaza...@gmail.com> wrote:

Hi, I've done a couple microcontroller projects, but otherwise I'm
pretty new to electronics.  I recently bought an antique typewriter
that I'd like to automate for fun.  My idea is that I could mount
solenoids... 50 or so

Yep, couild work. I'd aim for 15 solenoids (a 7 x 7 button array and
a single for 'shift' key.

Each key would be connected by tensile wire to a T bar, one
branch of the T to the row solenoid, one to the column solenoid.
Fire two solenoids, and one row of 7 buttons gets the
wire pulled taut (but not quite depressing any key) while one column
of 7 buttons gets the same treatment. The single key that
is both in the targeted row AND the targeted column gets
full depression.

I seem to recall a Selectric interface that was simply and literally
a plate of solenoids clamped over the keys. Ick.

 

[Jonathan Montoya]

On Apr 7, 6:29 pm, whit3rd <whit...@gmail.com> wrote:

On Apr 7, 8:37 am, Jonathan Montoya <lonlaza...@gmail.com> wrote:

Hi, I've done a couple microcontroller projects, but otherwise I'm
pretty new to electronics.  I recently bought an antique typewriter
that I'd like to automate for fun.  My idea is that I could mount
solenoids... 50 or so

Yep, couild work.  I'd aim for 15 solenoids (a 7 x 7 button array and
a single for 'shift' key.

Each key would be connected by tensile wire to a T bar, one
branch of the T to the row solenoid, one to the column solenoid.
Fire two solenoids, and one row of 7 buttons gets the
wire pulled taut (but not quite depressing any key) while one column
of 7 buttons gets the same treatment.  The single key that
is both in the targeted row AND the targeted column gets
full depression.

I seem to recall a Selectric interface that was simply and literally
a plate of solenoids clamped over the keys.  Ick.

I had to think about this for a bit to invison it, it looks somewhat
like an inverted marionette. I like the idea of reducing solenoids,
though it seems like the solenoids involved have to work a bit harder
than a solenoid-per key solution. I'm guessing the trade off ends up
being advantageous anyway.

The only possible problem I can see with this at the moment is
all the partially pressed keys causing the hammers to stick.

I did end up measuring the force necessariy using highly
sciency means, by pressing down with my finger on a scale
approximately with the same force needed to press a key. It ended
up being somewhere between 1 1/2 to 2 1/2 lbs of initial force.
I think anything over 3 might be damaging. The travel needed
is .75". Although if I grid it out I can split the distance
between the solenoids (I believe).

Thanks.

Lonlaz

 

[Tim Wescott]

On Tue, 08 Apr 2008 05:54:27 -0700, Jonathan Montoya wrote:

On Apr 7, 6:29 pm, whit3rd <whit...@gmail.com> wrote:
On Apr 7, 8:37 am, Jonathan Montoya <lonlaza...@gmail.com> wrote:

Hi, I've done a couple microcontroller projects, but otherwise I'm
pretty new to electronics.  I recently bought an antique typewriter
that I'd like to automate for fun.  My idea is that I could mount
solenoids... 50 or so

Yep, couild work.  I'd aim for 15 solenoids (a 7 x 7 button array and a
single for 'shift' key.

Each key would be connected by tensile wire to a T bar, one branch of
the T to the row solenoid, one to the column solenoid. Fire two
solenoids, and one row of 7 buttons gets the wire pulled taut (but not
quite depressing any key) while one column of 7 buttons gets the same
treatment.  The single key that is both in the targeted row AND the
targeted column gets full depression.

I seem to recall a Selectric interface that was simply and literally a
plate of solenoids clamped over the keys.  Ick.

I had to think about this for a bit to invison it, it looks somewhat
like an inverted marionette. I like the idea of reducing solenoids,
though it seems like the solenoids involved have to work a bit harder
than a solenoid-per key solution. I'm guessing the trade off ends up
being advantageous anyway.

The only possible problem I can see with this at the moment is all the
partially pressed keys causing the hammers to stick.

I did end up measuring the force necessariy using highly sciency means,
by pressing down with my finger on a scale approximately with the same
force needed to press a key. It ended up being somewhere between 1 1/2
to 2 1/2 lbs of initial force. I think anything over 3 might be
damaging. The travel needed is .75". Although if I grid it out I can
split the distance between the solenoids (I believe).

Thanks.

Lonlaz

The internals of an IBM Selectric boil down to six or seven bars that are
pulled in various combinations, then a 'print' bar is pulled and the
character types. Oddly enough, the bars have a 1:1 correspondence with
the bits in EBCDIC - I can't _imagine_ how that could be.

So you can modify a Selectric into a printer with a few solenoids
underneath on a plate, and an interface box, and still use it as a
typewriter.

--
Tim Wescott
Control systems and communications consulting
http://www.wescottdesign.com

Need to learn how to apply control theory in your embedded system?
"Applied Control Theory for Embedded Systems" by Tim Wescott
Elsevier/Newnes, http://www.wescottdesign.com/actfes/actfes.html

 

[Michael Black]

On Tue, 8 Apr 2008, Jonathan Montoya wrote:

Unfortunately it won't be that easy,
I have a typewriter that I have in mind, a LC Smith Super Speed
purchased from a Thrift store for $15. Still mulling over a
design. I am going to wind up a solenoid, just to find out how
boring an ineffective it is and convince myself to just buy some.

You should get one of those toy typewriters, if they still make

them, that had a knob that you had to turn to select the letter,
and then a single something to press on to actually type the later.
It was never that much slower than when I first tried to use a real
typewriter.

Then you'd use a stepper motor to rotate the letter selector, and
then only a single solenoid to get it to print that letter.

I bet they don't make those anymore, likely they don't even have
toy typewriters. It's so much easier to make things in electronics
that I imagine all you can get now are toy computers for your juvenile
typing needs.

Michael

 

[Tim Wescott]

On Tue, 08 Apr 2008 12:47:54 -0700, Jonathan Montoya wrote:

On Apr 8, 10:59 am, Tim Wescott <t...@seemywebsite.com> wrote:
On Tue, 08 Apr 2008 05:54:27 -0700, Jonathan Montoya wrote:
On Apr 7, 6:29 pm, whit3rd <whit...@gmail.com> wrote:
On Apr 7, 8:37 am, Jonathan Montoya <lonlaza...@gmail.com> wrote:

Hi, I've done a couple microcontroller projects, but otherwise I'm
pretty new to electronics.  I recently bought an antique
typewriter that I'd like to automate for fun.  My idea is that I
could mount solenoids... 50 or so

Yep, couild work.  I'd aim for 15 solenoids (a 7 x 7 button array
and a single for 'shift' key.

Each key would be connected by tensile wire to a T bar, one branch
of the T to the row solenoid, one to the column solenoid. Fire two
solenoids, and one row of 7 buttons gets the wire pulled taut (but
not quite depressing any key) while one column of 7 buttons gets the
same treatment.  The single key that is both in the targeted row AND
the targeted column gets full depression.

I seem to recall a Selectric interface that was simply and literally
a plate of solenoids clamped over the keys.  Ick.

I had to think about this for a bit to invison it, it looks somewhat
like an inverted marionette. I like the idea of reducing solenoids,
though it seems like the solenoids involved have to work a bit harder
than a solenoid-per key solution.  I'm guessing the trade off ends up
being advantageous anyway.

The only possible problem I can see with this at the moment is all
the partially pressed keys causing the hammers to stick.

I did end up measuring the force necessariy using highly sciency
means, by pressing down with my finger on a scale approximately with
the same force needed to press a key.  It ended up being somewhere
between 1 1/2 to 2 1/2 lbs of initial force. I think anything over 3
might be damaging.  The travel needed is .75". Although if I grid it
out I can split the distance between the solenoids (I believe).

Thanks.

Lonlaz

The internals of an IBM Selectric boil down to six or seven bars that
are pulled in various combinations, then a 'print' bar is pulled and
the character types.  Oddly enough, the bars have a 1:1 correspondence
with the bits in EBCDIC - I can't _imagine_ how that could be.

So you can modify a Selectric into a printer with a few solenoids
underneath on a plate, and an interface box, and still use it as a
typewriter.

--
Tim Wescott
Control systems and communications
consultinghttp://www.wescottdesign.com

Need to learn how to apply control theory in your embedded system?
"Applied Control Theory for Embedded Systems" by Tim Wescott
Elsevier/Newnes,http://www.wescottdesign.com/actfes/actfes.html- Hide
quoted text -

- Show quoted text -

*gag* I am a COBOL programmer by profession, I do electronics to get
*away* from EBCDIC.

Unfortunately it won't be that easy,
I have a typewriter that I have in mind, a LC Smith Super Speed
purchased from a Thrift store for $15. Still mulling over a design. I
am going to wind up a solenoid, just to find out how boring an
ineffective it is and convince myself to just buy some.

Lonlaz

Look at the surplus places to see just how cheap solenoids can be if
you're not picky about mounting features. Herbach & Rademan, All
Electronics and M.P. Jones & Associates all come to mind.

Winding your own will be _very_ tedious, and you won't get the same force
out of them that an off-the-shelf one will until you've _really_ learned
a lot about it.

--
Tim Wescott
Control systems and communications consulting
http://www.wescottdesign.com

Need to learn how to apply control theory in your embedded system?
"Applied Control Theory for Embedded Systems" by Tim Wescott
Elsevier/Newnes, http://www.wescottdesign.com/actfes/actfes.html

 

[Jonathan Montoya]

On Apr 8, 10:59 am, Tim Wescott <t...@seemywebsite.com> wrote:

On Tue, 08 Apr 2008 05:54:27 -0700, Jonathan Montoya wrote:
On Apr 7, 6:29 pm, whit3rd <whit...@gmail.com> wrote:
On Apr 7, 8:37 am, Jonathan Montoya <lonlaza...@gmail.com> wrote:

Hi, I've done a couple microcontroller projects, but otherwise I'm
pretty new to electronics.  I recently bought an antique typewriter
that I'd like to automate for fun.  My idea is that I could mount
solenoids... 50 or so

Yep, couild work.  I'd aim for 15 solenoids (a 7 x 7 button array and a
single for 'shift' key.

Each key would be connected by tensile wire to a T bar, one branch of
the T to the row solenoid, one to the column solenoid. Fire two
solenoids, and one row of 7 buttons gets the wire pulled taut (but not
quite depressing any key) while one column of 7 buttons gets the same
treatment.  The single key that is both in the targeted row AND the
targeted column gets full depression.

I seem to recall a Selectric interface that was simply and literally a
plate of solenoids clamped over the keys.  Ick.

I had to think about this for a bit to invison it, it looks somewhat
like an inverted marionette. I like the idea of reducing solenoids,
though it seems like the solenoids involved have to work a bit harder
than a solenoid-per key solution.  I'm guessing the trade off ends up
being advantageous anyway.

The only possible problem I can see with this at the moment is all the
partially pressed keys causing the hammers to stick.

I did end up measuring the force necessariy using highly sciency means,
by pressing down with my finger on a scale approximately with the same
force needed to press a key.  It ended up being somewhere between 1 1/2
to 2 1/2 lbs of initial force. I think anything over 3 might be
damaging.  The travel needed is .75". Although if I grid it out I can
split the distance between the solenoids (I believe).

Thanks.

Lonlaz

The internals of an IBM Selectric boil down to six or seven bars that are
pulled in various combinations, then a 'print' bar is pulled and the
character types.  Oddly enough, the bars have a 1:1 correspondence with
the bits in EBCDIC - I can't _imagine_ how that could be.

So you can modify a Selectric into a printer with a few solenoids
underneath on a plate, and an interface box, and still use it as a
typewriter.

--
Tim Wescott
Control systems and communications consultinghttp://www.wescottdesign.com

Need to learn how to apply control theory in your embedded system?
"Applied Control Theory for Embedded Systems" by Tim Wescott
Elsevier/Newnes,http://www.wescottdesign.com/actfes/actfes.html- Hide quoted text -

- Show quoted text -

*gag* I am a COBOL programmer by profession, I do electronics to
get *away* from EBCDIC.

Unfortunately it won't be that easy,
I have a typewriter that I have in mind, a LC Smith Super Speed
purchased from a Thrift store for $15. Still mulling over a
design. I am going to wind up a solenoid, just to find out how
boring an ineffective it is and convince myself to just buy some.

Lonlaz

 

[Rich Grise]

On Mon, 07 Apr 2008 14:42:42 -0700, Tim Wescott wrote:

Jonathan Montoya wrote:
....
I thougt I'd make a kind of a 'manual' word processor. You could type
a
page, it would store the text info, and you could print it back. I
wonder
if I can't adapt an X Y solution under the keyboard if the solenoids
are
too cumbersome.

I think it has 45 keys, I'll have to count again tonight.

So put the button-pusher in a box that can register its position to the
typewriter, so you can easily lift it away.

I still vote for robot hands that look like hands, probably because I'm
not the one who has to make it work...

I envision this, and the idea of a pair of disembodied hands (and maybe
forearms) seems like it would look a little creepy, like some of the
"mechanisms" in "Animusic": http://www.animusic.com/

But it would be fun, if you have a proper machine shop and stuff.

I'm thinking solenoids like from an old "ding-dong" doorbell, or from
a pinball machine. I was once going to buy 88 "drop target reset"
solenoids to electrify my piano, but the approx. $3.00 a pop price tag
kind of slowed me down on that.

With a jig and some proper cores you could make them yourselves, but
there must be solenoids on the market somewhere.

And yes, I think it's a cool idea, also using it for input, albeit
then you'd need to figure out how to mount all those contacts.

Cheers!
Rich

 

[Rich Grise]

On Mon, 07 Apr 2008 10:56:43 -0700, Jonathan Montoya wrote:

On Apr 7, 11:20 am, Mike Silva <snarflem...@yahoo.com> wrote:
On Apr 7, 11:37 am, Jonathan Montoya <lonlaza...@gmail.com> wrote:

Hi, I've done a couple microcontroller projects, but otherwise I'm
pretty new to electronics.  I recently bought an antique typewriter
that I'd like to automate for fun.  My idea is that I could mount
solenoids under each key(which are essentially levers that push up on
some rods) and pull down with the same force that it would take press
key.  I haven't measured how much force is necessary yet, and the
stroke is probably about an inch.

Before I started doing some expirements, I wanted to ask the experts
to see if this is at all feasable.  Would solenoids be the best way to
do this? I picked them because they seemed like they'd better be able
to simulate the fast sharp action needed to get the antique key to
impress the ink on the paper.  Will I be able to jam 50 or so of these
parallel and under the keys (given that they'll be mounted under a
desk or something like it), or will the size for the work needed
require more creative mounting?  And, am I assuming correctly that I'd
have to make these myself, since they probably have speifications
different from what most solenoids are made for?

I appreciate any guidance I can get.

Wow, deja vu!  I wanted to do this in about 1980, and I think I even
bought a load of surplus solenoids to do it, but never went beyond
that.  Instead I bought a dot-matrix printing mechanism, built a
driver board and made a printer driven by my Z-80 CP/M system.  Hint
on that one - buy lots of extra printhead pins to replace the ones you
break while debugging the software!

Theoretically it is certainly possible.  If you use stiff wires or
rods between the keys and the solenoids that should make your physical
construction a lot easier.  And make sure your power supply, drivers
and circuitry can all handle the current spikes (and resultant
noise).  Use a short-timeout watchdog or gate the firing pulses with
an external signal (555?) to assure that you don't hold a solenoid
down too long by mistake.  Make sure all your solenoid drivers power
up in the OFF state (imagine all 50 trying to fire on power-up!).

And be sure and post a video to youtube when you've got it
working. - Hide quoted text -

Heh, look for it on You Tube in about 15 years.

Nah. If you've got the parts and access to a decent shop, you could
slap something like this together in just a few months, assuming you
already have the micro and know how to program it.

Cheers!
Rich