Searching \ for '[EE]: Orienter w electromagnet ViBRATOR' in subject line. ()
Make payments with PayPal - it's fast, free and secure! Help us get a faster server
FAQ page: www.piclist.com/techref/index.htm?key=orienter+electromagnet
Search entire site for: 'Orienter w electromagnet ViBRATOR'.

Exact match. Not showing close matches.
PICList Thread
'[EE]: Orienter w electromagnet ViBRATOR'
2002\10\13@214500 by Patrick J

flavicon
face
I am building a dirtcheap mini orienter unit, something like this one:
http://www.ecomat.se/sweden/conveys.htm
(look at the drawings/pictures to get an idea what I am talking abt)


I have made the springsystem, and it works (transports objects forward)
when I use my engraverpen to create the vibrations.
From what I can tell the pen seems to fly out to a middle position
when the power is switched on. Then it oscilliates around that position to
create the vibs. Freq is 50 Hz. The bigger the amplitude is, the faster
the objects move forward (say from 5mm/s to 20 mm/s)


The problem is that my electromagnet doesnt accomplish the same
sort of vibration. Could be too slow or too weak, not sure.
So I wonder if it is possible to use a standard electrmagnet
(with a rod inside that pull on the springs) to create the right sort of
vibrations... or I must get a special magnet to do it ?

(simple cheap suggestions to create the magic vibrations are most welcome
:-)

TIA!

--
http://www.piclist.com hint: To leave the PICList
spam_OUTpiclist-unsubscribe-requestTakeThisOuTspammitvma.mit.edu


2002\10\13@222557 by Jim

flavicon
face
Seems like a standard soleniod with ac driving it would work.

Regards
Jim M



At 03:43 AM 10/14/02 +0200, you wrote:
{Quote hidden}

--
http://www.piclist.com hint: To leave the PICList
piclist-unsubscribe-requestspamKILLspammitvma.mit.edu


2002\10\14@022839 by Peter L. Peres

picon face
Not magic but gutting a cheap aquarium pump should yield what you need.

Peter

--
http://www.piclist.com hint: PICList Posts must start with ONE topic:
[PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads


2002\10\14@043611 by Patrick J

flavicon
face
Thats what I tried; 24vAC to the solenoid (electromagnet) but
I could hardly detect any movments (vibrations) at all.
If I use 24vDC it moves it but only 1-2 mm and it takes way too
much time to do it.
I'll ease up on the springs and see if that helps. Since the objetcs
only is 6 gram, current springs might be a bit stiff.

/me have gone back to the workshop


----- Original Message -----
> Seems like a standard soleniod with ac driving it would work.
>
> Regards
> Jim M

> >I am building a dirtcheap mini orienter unit, something like this one:
> >http://www.ecomat.se/sweden/conveys.htm
> >(look at the drawings/pictures to get an idea what I am talking abt)

--
http://www.piclist.com hint: PICList Posts must start with ONE topic:
[PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads


2002\10\14@043612 by Patrick J

flavicon
face
Interesting suggestion, what can be found in its belly ?
/Patrick

----- Original Message -----
> Not magic but gutting a cheap aquarium pump should yield what you need.
> Peter

--
http://www.piclist.com hint: PICList Posts must start with ONE topic:
[PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads


2002\10\14@064830 by Russell McMahon

face
flavicon
face
> > Not magic but gutting a cheap aquarium pump should yield what you need.

> Interesting suggestion, what can be found in its belly ?

Usually a solenoid with a magnetic circuit that has three sides fixed and
the 4th "side" is a moving arm that is sprung loaded slightly open. Mains
cycles attract the arm and it vibrates a diaphragm in a very crude pump with
flap valves. Pressure and flow obtained is small but adequate for fish tank
use. An arrangement almost identical principle is used in some electric hair
trimmers.

The air pumps often have an adjustment screw which allows the moving arm to
be adjusted until it just doesn't close the gap fully at closest approach.
If allowed to touch it makes a LOUD buzzing noise at twice mains frequency
(but probably does no great harm short term). Minimum airgap consistent with
quiet operation would result in lowest operating current.

A hair trimmer may be even better suited to providing a vibration source
than an airpump as it already drives a mechanical device with some force.


       RM

--
http://www.piclist.com hint: PICList Posts must start with ONE topic:
[PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads


2002\10\14@080357 by Olin Lathrop

face picon face
> Thats what I tried; 24vAC to the solenoid (electromagnet) but
> I could hardly detect any movments (vibrations) at all.
> If I use 24vDC it moves it but only 1-2 mm and it takes way too
> much time to do it.

A normal solenoid uses an unpolarized plunger.  You would get better
vibration and at half the frequency if the plunger were polarized.  This is
how loudspeakers work, except that the coil moves and the plunger is fixed.
Perhaps you could hack up an old loudspeaker.  A motor with an off-axis
weight should provide vibration too.


*****************************************************************
Embed Inc, embedded system specialists in Littleton Massachusetts
(978) 742-9014, http://www.embedinc.com

--
http://www.piclist.com hint: PICList Posts must start with ONE topic:
[PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads


2002\10\14@125043 by Roman Black

flavicon
face
Patrick J wrote:
>
> I am building a dirtcheap mini orienter unit, something like this one:
> http://www.ecomat.se/sweden/conveys.htm

> I have made the springsystem, and it works (transports objects forward)
> when I use my engraverpen to create the vibrations.

> The problem is that my electromagnet doesnt accomplish the same
> sort of vibration. Could be too slow or too weak, not sure.


Definitely try a loudspeaker, these are quite
energy efficient and you can glue pretty much
anything to the cone and drive it with any cheap
audio amp chip like a LM386. You can also get
"chest thumpers" from games, which are a speaker
driver you wear to give body impact and I have
seen very cheap as the sales were a bit of a flop.

You may also find driving it with a sawtooth wave
gives more inertia in one direction, and should give
very good results with your one way vibration
transporter, and it *might* be possible to transport
in EITHER DIRECTION based on the wave shape you
use.

One of the coolest things I ever built was an
inertia/friction engine, a sealed box about the size
of your hand that would move and even push/carry
things across a table. It would carry almost any
weight, only getting slower with heavier weights.
I was able to stand on it and move across a hard
floor, although only at a 1mm/sec.

As long as you have some reasonably reliable friction
you can build some really nice inertia transport
stuff. If you get it up to ultrasonic you can get
things flying across the table with no (external)
moving parts. :o)
-Roman

--
http://www.piclist.com hint: PICList Posts must start with ONE topic:
[PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads


2002\10\14@152449 by Peter L. Peres

picon face
On Mon, 14 Oct 2002, Olin Lathrop wrote:

*>> Thats what I tried; 24vAC to the solenoid (electromagnet) but
*>> I could hardly detect any movments (vibrations) at all.
*>> If I use 24vDC it moves it but only 1-2 mm and it takes way too
*>> much time to do it.
*>
*>A normal solenoid uses an unpolarized plunger.  You would get better
*>vibration and at half the frequency if the plunger were polarized.  This is
*>how loudspeakers work, except that the coil moves and the plunger is fixed.
*>Perhaps you could hack up an old loudspeaker.  A motor with an off-axis
*>weight should provide vibration too.

The aquarium pump vibrators I have dissected all used a ferrite magnet on
the moving armature.

Imho Russell M. was confusing the tuning of the aquarium pump gap with the
tuning of the Oudin coil gap <grin>.

Peter

--
http://www.piclist.com hint: PICList Posts must start with ONE topic:
[PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads


2002\10\14@152452 by Peter L. Peres

picon face
On Mon, 14 Oct 2002, Patrick J wrote:

*>Interesting suggestion, what can be found in its belly ?
*>/Patrick

An electromagnet that runs on 220V ac and is tuned to vibrate at 50Hz or
100Hz. It has all the contraptions required to mechanically tune it. If
you couple this mechanically to your table it should do the right thing.

I hope you know how to compute the self resonant frequency of a spring +
mass system. Weighing the mass (table) and testing k of the spring (load
table and measure displacement) would tell you all you need to know and
which way you are off. Then to decrease frequency add mass or weaken
spring, and vice versa.

I have once played with a prototype orienter/mover that used a bass
speaker and two model servos that inclined a triangular table coupled to
the speaker with a glued stick. The table had a shallow fold down the
middle that did the orienting. The servos held the other two corners of
the table. The speaker ran directly off a halogen light transformer (mains
50Hz).  What are you using it for ?

hope this helps,

Peter

*>----- Original Message -----
*>> Not magic but gutting a cheap aquarium pump should yield what you need.
*>> Peter
*>
*>--
*>http://www.piclist.com hint: PICList Posts must start with ONE topic:
*>[PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads
*>
*>
*>

--
http://www.piclist.com hint: PICList Posts must start with ONE topic:
[PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads


2002\10\14@155315 by hard Prosser

flavicon
face
How about the guts from an electric toothbrush?
Richard P





Patrick J wrote:
>
> I am building a dirtcheap mini orienter unit, something like this one:
> http://www.ecomat.se/sweden/conveys.htm

> I have made the springsystem, and it works (transports objects forward)
> when I use my engraverpen to create the vibrations.

> The problem is that my electromagnet doesnt accomplish the same
> sort of vibration. Could be too slow or too weak, not sure.

--
http://www.piclist.com hint: PICList Posts must start with ONE topic:
[PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads


2002\10\14@220235 by Patrick J

flavicon
face
Well today I made some real nice progress !
Changed the two springs to weaker and now the solenoid has
enought strengh to to its job. I placed a 3 kg iron block on the
vibrator and it moved the same speed as the 6 gram plastic part.
I was quite impressed by that.
And the best part is: ITS 99,9% QUIET !


I plan to drive this with a PIC that generates an adjustable
freq. and pulsefactor PWM 24vDC squarewave.
Something like 10-50 Hz, 5-90% PWM

The solenoid is supposed to 'let go' as fast as possible when
power is switched off. I wonder what effect the anti-transient diode
in parallell with the coil has on this... On a motor it would work
as a breake, but what will it do to the solenoid ???


Thanxs everyone for your input so far :-)


{Original Message removed}

2002\10\15@011709 by Russell McMahon

face
flavicon
face
> The solenoid is supposed to 'let go' as fast as possible when
> power is switched off. I wonder what effect the anti-transient diode
> in parallell with the coil has on this... On a motor it would work
> as a breake, but what will it do to the solenoid ???

Slow the release time.

Solenoid current will circulate while decaying exponentially.
Initial energy loss will be diode loss plus resistive losses - 0.6I + I^2R.
You can work out time for energy to dissipate but it will be easier to just
try it and see :-)

   RM

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics


2002\10\15@035219 by Alan B. Pearce

face picon face
>If you get it up to ultrasonic you can get
>things flying across the table with no (external)
>moving parts. :o)
>-Roman


Careful Roman, you will be making piezo motors next :)))

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics


2002\10\15@043651 by Alan B. Pearce

face picon face
> The solenoid is supposed to 'let go' as fast as possible when
> power is switched off. I wonder what effect the anti-transient diode
> in parallell with the coil has on this... On a motor it would work
> as a breake, but what will it do to the solenoid ???

If you need fast release time then you will probably be better off with an
RC snubber network than a diode. This may also require you to use a
transistor with a higher voltage rating than you may have otherwise used.

Check out the various chapters here
http://www.leachintl2.com/english/english2/vol6/index.html

Right down the bottom is a page with useful formula, but chapter 1.3 may be
useful to you as well.

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics


2002\10\15@053920 by Roman Black

flavicon
face
Patrick J wrote:

> The solenoid is supposed to 'let go' as fast as possible when
> power is switched off. I wonder what effect the anti-transient diode
> in parallell with the coil has on this... On a motor it would work
> as a breake, but what will it do to the solenoid ???

The diode will give you slow-decay of the magnetic
field which is the opposite of what you need!
The typical way of fixing this is to put a zener in
(reverse) series with your diode, so that the coil
back-emf is clipped at (say) 30v. This fully protects
your driver semi provided it is rated at higher than
30v.

I have seen this used in commercial high current
relay drivers where it is important to make the
contacts open as fast as possible to reduce arcing.
For small/cheap designs you can just put a resistor
in series with the diode, providing you check back-
emf on the cro to be within safe levels.
:o)
-Roman

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics


2002\10\15@054544 by Roman Black

flavicon
face
Alan B. Pearce wrote:
>
> >If you get it up to ultrasonic you can get
> >things flying across the table with no (external)
> >moving parts. :o)
> >-Roman
>
> Careful Roman, you will be making piezo motors next :)))


I was making them in the mid '80s, at least
related to the inertia/friction engines. I have
an urge to build a sumo class robot using two
inertia engines... Having a solid metal box that
will push just about anything sounds like a great
3kg class sumo design. Sigh, put it on my list.
:o)
-Roman

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics


2002\10\15@081640 by Olin Lathrop

face picon face
> Solenoid current will circulate while decaying exponentially.

Accross a resistor, yes.  It will decay pretty much linearly accross a diode
because of the "constant" voltage.


*****************************************************************
Embed Inc, embedded system specialists in Littleton Massachusetts
(978) 742-9014, http://www.embedinc.com

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics


2002\10\15@143541 by Peter L. Peres

picon face
Will you post a photo of it ?

Peter

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics


2002\10\15@143542 by Peter L. Peres

picon face
On Tue, 15 Oct 2002, Patrick J wrote:

*>The solenoid is supposed to 'let go' as fast as possible when
*>power is switched off. I wonder what effect the anti-transient diode
*>in parallell with the coil has on this... On a motor it would work
*>as a breake, but what will it do to the solenoid ???

Brake ... put a zener specced to as high as you dare in series with it
(Vzener < Vceo - Vsupply) .

Peter

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics


2002\10\15@232805 by Patrick J

flavicon
face
From: "Peter L. Peres"
>>The solenoid is supposed to 'let go' as fast as possible when
>>power is switched off.

> Brake ... put a zener specced to as high as you dare in series with it
> (Vzener < Vceo - Vsupply) .
> Peter

Thanxs, that word looked funny but I couldn't get it right.
Ok, Zener seems to be the way to go, I'll give that a try once
the PWM-program works :-)


> Will you post a photo of it ?
> Peter
Sorry, I don't have a digital camera yet :-/

/PJ

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics


2002\10\18@003956 by Patrick J

flavicon
face
From: "Roman Black"
> The diode will give you slow-decay of the magnetic
> field which is the opposite of what you need!
> The typical way of fixing this is to put a zener in
> (reverse) series with your diode, so that the coil
> back-emf is clipped at (say) 30v. This fully protects
> your driver semi provided it is rated at higher than 30v.

Any idea how much current can be expected, rule-of-thumb ?
I know u dont have the specs for the solenoid (nor do I), but
say would it be more or less than it need to operate ?

/PJ

--
http://www.piclist.com hint: To leave the PICList
.....piclist-unsubscribe-requestKILLspamspam.....mitvma.mit.edu


2002\10\18@005937 by Russell McMahon

face
flavicon
face
> > The typical way of fixing this is to put a zener in
> > (reverse) series with your diode, so that the coil
> > back-emf is clipped at (say) 30v. This fully protects
> > your driver semi provided it is rated at higher than 30v.

> Any idea how much current can be expected, rule-of-thumb ?
> I know u dont have the specs for the solenoid (nor do I), but
> say would it be more or less than it need to operate ?

The initial current is the coil current at turn off time and it decays from
there due to energy dissipation considerations.


       RM

--
http://www.piclist.com hint: To leave the PICList
EraseMEpiclist-unsubscribe-requestspam_OUTspamTakeThisOuTmitvma.mit.edu


2002\10\18@041307 by Roman Black

flavicon
face
Russell McMahon wrote:
>
> > > The typical way of fixing this is to put a zener in
> > > (reverse) series with your diode, so that the coil
> > > back-emf is clipped at (say) 30v. This fully protects
> > > your driver semi provided it is rated at higher than 30v.
>
> > Any idea how much current can be expected, rule-of-thumb ?
> > I know u dont have the specs for the solenoid (nor do I), but
> > say would it be more or less than it need to operate ?
>
> The initial current is the coil current at turn off time and it decays from
> there due to energy dissipation considerations.


This is not exactly true when using the zener
in series with the diode, as only the very peak of
the back emf is clipped by the zener+diode.

In the spirit of answering the question as best as
possible, he is using a 24v 800mA solenoid, and
if clipped by a 30v Zener in series with the diode
there will only be a few percent of the total
back emf energy passed through the diode, and even
a small wattage zener will run fairly cool.

For instance a 12v zener might run warm as it will
clip a much higher percentage of the back emf energy.
A 50v zener will run stone cold. Remember only the
PEAK is clipped, the rest of the energy decays as it
would in an open circuit situation, which in effect
it is because the zener is off.

The main requirement is to clip the back emf voltage
before it reaches the safe voltage limit of the
switching transistor.

For a 24v 800mA solenoid being PWM'ed I would start
with a fast 1A diode like a 1N5819, switched with
a 60v transistor, and a 1W 32v zener in series with
the diode. For mega reliability upgrade to a 1N5822
and 3W zener, a cheap 115v transistor like TIP41C,
and maybe add a small snubber 33nF and 22 ohms.

Also I try to avoid zeners over 36v, they seem to be
a lot less reliable than the lower voltage parts.
I think it is something to do with the physical
construction.
-Roman

--
http://www.piclist.com hint: To leave the PICList
piclist-unsubscribe-requestspamspam_OUTmitvma.mit.edu


2002\10\18@084150 by Olin Lathrop

face picon face
> Any idea how much current can be expected, rule-of-thumb ?
> I know u dont have the specs for the solenoid (nor do I), but
> say would it be more or less than it need to operate ?

The current thru the discharge diode when an inductor is shut off is always
the same as the current right before it was shut off.  That's what an
inductor does.  So if the solenoid is driven with 200mA, then the discharge
circuit needs to handle 200mA.


*****************************************************************
Embed Inc, embedded system specialists in Littleton Massachusetts
(978) 742-9014, http://www.embedinc.com

--
http://www.piclist.com hint: To leave the PICList
@spam@piclist-unsubscribe-requestKILLspamspammitvma.mit.edu


2002\10\18@084204 by Olin Lathrop

face picon face
> This is not exactly true when using the zener
> in series with the diode, as only the very peak of
> the back emf is clipped by the zener+diode.
>
> In the spirit of answering the question as best as
> possible, he is using a 24v 800mA solenoid, and
> if clipped by a 30v Zener in series with the diode
> there will only be a few percent of the total
> back emf energy passed through the diode, and even
> a small wattage zener will run fairly cool.
>
> For instance a 12v zener might run warm as it will
> clip a much higher percentage of the back emf energy.
> A 50v zener will run stone cold. Remember only the
> PEAK is clipped, the rest of the energy decays as it
> would in an open circuit situation, which in effect
> it is because the zener is off.

The inductor is storing a fixed energy when it is shut off.

 E = I**2 * L / 2

For sake of example, let's say the inductance is 10mH.

 E = 800mA**2 * 10mH / 2 = .64 * .01 / 2 = 3.2mJ

This energy has to go somewhere.  If I understand your circuit correctly,
then the only place it can go is into the zener.  It doesn't matter at what
voltage that takes place.  For a perfect inductor and perfect zener, the
time of the back pulse will be inversely proportional to the voltage of the
back pulse.  Note that this comes out to the same energy at all
combinations.  This principal is the basis for single-inductor boost voltage
converters.

Why not just use a resistor in series with diode instead of a zener?  The
current is known, so the resistor can be easily selected for the maximum
initial back voltage that the switching element can tolerate.  Let's say
that's 35V, and to be conservative that the diode drops 500mV.

 R = V / I = (35V - .5V - 24V) / 800mA = 13 ohms

The initial back voltage will be the same as with a zener, but it will drop
exponentially instead of staying flat until the inductor energy goes to 0.
This will cause the magnetic field collapse to be slightly slower, but still
much faster than with just a diode.  Given the better reliability of a
resistor compared to a zener, I would use this method unless I really needed
the little extra speed.

Now for the power dissipation.  For the 10mH example, 3.2mJ will be
dissipated every time the solenoid is switched off.  If this happens once
per second, the average dissipation is only 3.2mW.  If it happens 10 times
per second, then it is 32mW.  It's unlikely he is trying to drive a
mechanical solenoid faster than that, so power dissipation is not an issue
given the 10mH and 10Hz chosen for this example.


*****************************************************************
Embed Inc, embedded system specialists in Littleton Massachusetts
(978) 742-9014, http://www.embedinc.com

--
http://www.piclist.com hint: To leave the PICList
KILLspampiclist-unsubscribe-requestKILLspamspammitvma.mit.edu


2002\10\18@111438 by Peter L. Peres

picon face
On Fri, 18 Oct 2002, Olin Lathrop wrote:

*>Now for the power dissipation.  For the 10mH example, 3.2mJ will be
*>dissipated every time the solenoid is switched off.  If this happens once
*>per second, the average dissipation is only 3.2mW.  If it happens 10 times
*>per second, then it is 32mW.  It's unlikely he is trying to drive a
*>mechanical solenoid faster than that, so power dissipation is not an issue
*>given the 10mH and 10Hz chosen for this example.

This is very correct except most solenoids I've seen are 1H-ish (okay,
0.5H I've seen too). You need a lot of turns to make mechanical force. The
power dissipated in the switch recovery element will be significant imho.

Peter

--
http://www.piclist.com hint: To leave the PICList
RemoveMEpiclist-unsubscribe-requestTakeThisOuTspammitvma.mit.edu


2002\10\18@123249 by Roman Black

flavicon
face
Olin Lathrop wrote:
{Quote hidden}

With a perfect inductor your argument is true. However,
real circuits are far from perfect. :o)

Consider the case where the solenoid is switched and
there is NO diode. When the switch opens there can be
NO current as the circuit is open. The field collapses
and tries to produce infinite voltage, which can't happen
due to the inherent resistance of the coil and also
circuit capacitances which cause a crude "snubbing"
effect.

This back-emf is a fast high voltage spike which then
decays exponentially as can be seen on a cro. The
zener solution was chosen (it's not my invention!)
because it clips the PEAK of the spike only, and the
rest of the energy is absorbed into the real world
innefficiencies of the open circuit as it would have
been if there was no diode.

Rule 1: if a magnetic field collapses and there is a
complete circuit, then current will be induced.

Rule 2; if a magnetic field collapses and there is
NO circuit, NO current will flow! (rotate an open
circuit DC generator if you don't believe me!)

> Why not just use a resistor in series with diode instead of a zener?  The
> current is known, so the resistor can be easily selected for the maximum
> initial back voltage that the switching element can tolerate.

In the case of a resistor in series with the back-emf
diode, the resistor DOES provide a circuit for the
back-emf and the entire energy of the collapsing field
will be dissipated by the resistor apart from a small
amount by the diode. This is ok with small coils. :o)

The zener system is superior as only the PEAK is
conducted, the bulk of the energy (the area under the
curve as it decays exponentially) is at lower voltages.
When tuning a zener flyback system on the cro (as
I have done) you can tune *what percentage* of the total
energy is clipped.

Using a zener to clip SOME of the back-emf energy
from a coil is common in many stepper driver circuits,
and is described by "Jones on steppers" as an excellent
way to give fast coil performance without generating
excessive back-emf voltages. I have also seen this
in commercial large solenoid drivers when I was an
industrial technician as it gives fast contact opening
to reduce arcing.

An RC series "snubber" has a similar effect, the capacitor
allows the peak to be clipped but the bulk of the energy
again is dissipated in an open-circuit situation.

> This energy has to go somewhere.
> This principal is the basis for single-inductor boost voltage
> converters.

Yes, but single inductor boost converters are very
inefficient, and is the reason that generating high
voltages (like TV flyback) is always done by T:T
transformer rather than a simple boost inductor.

If you monitor the open-circuit back-emf on a cro
you can actually see the "clipping" occur, and see
what *percentage* of the energy you clip with your zener
or gas arrestor etc. :o)
-Roman

--
http://www.piclist.com hint: To leave the PICList
spamBeGonepiclist-unsubscribe-requestspamBeGonespammitvma.mit.edu


2002\10\18@214415 by Patrick J

flavicon
face
From: "Thomas Lehmann"
> Long discussion over this topic,
Yes, most interesting to watch the gurus fight it out the finer details ;-)
Very helpful info for me tho.


> 1. What do you have is a spring-mass-system with its natural resonant
> frequency and damping.
yes


> 2. A directed movement of a part for transportation purposes could only be
> achived with a unsymetrical waveshape.
It might work good with a 50% pwm squarewave.
I should know for sure sometime next week.


> 3. This technique for moving parts is available since several decades.
yes, I don't claim to have invented it :-)
I am just making a cheap & quiet one. The commersial ones I've seen
was way too expensive and noisy for me.


> 4. You need no microcontroller for it ;-).
No, but admit it has a higher impress-the-chics-factor ;-)
And I can remotely control it by serial or IR or <whatever> and/or have it
adjust itself. The list is endless.

The primary features:
- adjustable freq. (5-50 Hz)
- adjustable PWM pulsefactor (5-90%)
- 2 independent channels (if that is messy ill haveto use 2 pics)
- show the freq & pwm settings on a LED display (nice, but will work w/o)

So far I have the PWM going 0-100%. The tricky part is to change
the freq without affecting the PWM or the other way around.
Actually that works too.. tho it doesnt change linear, but in a X^2 fashion.
(Freq = 2952,9 * delay ^-0,9829   that equation wont be fun to use in asm)

This non-linear freq. update behavior have the effect that when the user
presses the freq-down, or up,-button  it will change rapidly around 50Hz
and slow around 5Hz.


The real problems begin when the 2:nd chan is to be done... Independent
PWM factor is easy, but independent freq.. hrm


> 5. Enclosed an old fashioned schematic for mains applications and some
> pictures to give you an idea how to do it.
The mechanical part of it is already done, and works fine !
(and it is a simular in design to the one in your picture)
I work on the PIC program and electronics to drive it as much as I
have time to.

> Hope this helps.
> Thomas
Always interesting to learn :-)
Thanxs.

--
http://www.piclist.com hint: To leave the PICList
TakeThisOuTpiclist-unsubscribe-requestEraseMEspamspam_OUTmitvma.mit.edu


2002\10\19@051940 by Roman Black

flavicon
face
> From: "Thomas Lehmann"
> > 2. A directed movement of a part for transportation purposes could only be
> > achived with a unsymetrical waveshape.

I don't think this is true. There are *many*
mechanisms where transportation will work well
with an equal waveshape;

* when the back/forth movement has a vertical
component, as shown by the 20 degree angle of
the device in your picture.

* when the friction surface has a uneven friction
system, like the angled plastic "carpet" fibres
shown on the web page for that commercial system.
Horizontal vibration is all that is required.
-Roman

--
http://www.piclist.com hint: PICList Posts must start with ONE topic:
[PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads


2002\10\19@141050 by s Lehmann

flavicon
face
> > 2. A directed movement of a part for transportation purposes could only
be
> > achived with a unsymetrical waveshape.

>I don't think this is true. There are *many*
>mechanisms where transportation will work well
>with an equal waveshape;

>* when the back/forth movement has a vertical
>component, as shown by the 20 degree angle of
>the device in your picture.

>* when the friction surface has a uneven friction
>system, like the angled plastic "carpet" fibres
>shown on the web page for that commercial system.
>Horizontal vibration is all that is required.
>-Roman

--
Hello Roman,

of course you are right from the electrical point of view.
The coil of an electromagnetic vibrator accepts a
lot of different waveforms.
When you are looking at the mechanical waveshape of the move-
ment of the part, it is unsymetrical.
The part to be moved, is accelerated from the vibrator surface against
the force of gravity, so the part leaves the surface and will do a parabolic
movement.
During the "flighttime" the vibrator surface moved back a little and
the part falls down to the surface and is accelerated again.

If the acceleration of the parts is low (ex.: heavy mass) they did not leave
the surface and
movement of the parts could be achived through the difference between
slipping and sticking friction.

So in both cases you can get a movement of the parts in a predefined
direction.

In a transportation system you must be able to change the speed and the
movement direction
of the moved parts (forward and backward like a conveyor) and this is IMHO
only possible with
unsymetrical electrical waveshapes. Of course you can additionaly change the
frequency.

By the way I have build your current limited SMPS on my workbench with great
success; more publications
like this please :-).

-- Thomas

--
http://www.piclist.com hint: PICList Posts must start with ONE topic:
[PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads


2002\10\19@153445 by Peter L. Peres

picon face
On Sat, 19 Oct 2002, Roman Black wrote:

*>* when the friction surface has a uneven friction
*>system, like the angled plastic "carpet" fibres
*>shown on the web page for that commercial system.
*>Horizontal vibration is all that is required.

More exactly one of the vibrations, or the contact friction must be
non-linear, resp. asymmetrical.

There was a model 'robot' demo that used two toothbrush heads glued under
a vibrator unit, both inclined at about 15 degrees in the same direction.
I built it once, it works. If you couple two such vibrators loosely with
an axle then you could have the equivalent of a two-wheel desktop robot.
No gears, moves slowly, high mass is not a problem, noise is, won't go
backwards (I have ideas on this using two vibrators, one x one y, and
control the phase between them, with straight brushes). Travel could
be measured by a free coasting wheel and encoder, or better, build an
optical mouse into the axle between the drive units, and slightly behind
them so it tends to run straight. Maybe this is the future of desktop
robots. Sunday I'll build one to see (note to self:  buy 4 hard bristled
toothbrushes).

Peter

--
http://www.piclist.com hint: PICList Posts must start with ONE topic:
[PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads


2002\10\20@085427 by Roman Black

flavicon
face
Thomas Lehmann wrote:

> of course you are right from the electrical point of view.
> The coil of an electromagnetic vibrator accepts a
> lot of different waveforms.
> When you are looking at the mechanical waveshape of the move-
> ment of the part, it is unsymetrical.

Sorry Thomas, I thought you meant the electrical
waveform.


{Quote hidden}

Sure, you definitely need asymetrical mechanical
movement.

> In a transportation system you must be able to change the speed and
the
> movement direction
> of the moved parts (forward and backward like a conveyor) and this is IMHO
> only possible with
> unsymetrical electrical waveshapes.

Ok, but the picture of the device you posted only
works in one direction? Looking at the angle of
the vibratory movement it would only need to be
oscillated, and the moved parts would always travel
in the same direction. Are you saying it is suitable
for both directions?

> By the way I have build your current limited SMPS on my workbench with great
> success; more publications
> like this please :-).

Thanks! :o)
-Roman

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics


2002\10\20@091200 by Roman Black

flavicon
face
Peter L. Peres wrote:
>
> More exactly one of the vibrations, or the contact friction must be
> non-linear, resp. asymmetrical.
>
> There was a model 'robot' demo that used two toothbrush heads glued under
> a vibrator unit, both inclined at about 15 degrees in the same direction.
> I built it once, it works. If you couple two such vibrators loosely with
> an axle then you could have the equivalent of a two-wheel desktop robot.
> No gears, moves slowly, high mass is not a problem, noise is, won't go
> backwards (I have ideas on this using two vibrators, one x one y, and
> control the phase between them, with straight brushes). Travel could
> be measured by a free coasting wheel and encoder, or better, build an
> optical mouse into the axle between the drive units, and slightly behind
> them so it tends to run straight. Maybe this is the future of desktop
> robots. Sunday I'll build one to see (note to self:  buy 4 hard bristled
> toothbrushes).


I can suggest another design that I have built
in the past, just replace the vibrator with a
solenoid, and use asymetrical waveforms to maximise
the intertia/friction difference.

The idea is that you accelerate the solenoid plunger
at a rate slow enough NOT to break the friction of
the robot to the table, but when the plunger reaches
the end of it's travel and hits the hard end stop it
breaks the friction point and the robot moves forward.

Then you release the plunger, slowly enough to return
to the home position gently so the friction is preserved
and the robot stays at the new position.

If the solenoid has a hard end stop at *each end* you
can use two different waveforms, one moves it forward,
one moves it backward. Obviously a solenoid on each side
of the 'bot will give fill differential drive, forward,
back, turn in place etc. You don't need the toothbrushes,
simply the friction of the metal base against the table
worked better than many friction systems I tried, and
any hard robot base against a hard table worked well.

I had no problems getting 6 inches per second movement
of a 300g metal box on a hard table, it used about
500mA at 6v going from memory.

I've been itching to make one of these for a while
now as a 500g class sumo bot, all it needs is two
solenoids, battery and circuit in a metal box.
They push or carry just about anything. :o)
-Roman

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics


2002\10\21@062202 by s Lehmann

flavicon
face
>> 4. You need no microcontroller for it ;-).
>No, but admit it has a higher impress-the-chics-factor ;-)
>And I can remotely control it by serial or IR or <whatever> and/or have it
>adjust itself. The list is endless.

>The primary features:
>- adjustable freq. (5-50 Hz)
>- adjustable PWM pulsefactor (5-90%)
>- 2 independent channels (if that is messy ill haveto use 2 pics)
>- show the freq & pwm settings on a LED display (nice, but will work w/o)

>So far I have the PWM going 0-100%. The tricky part is to change
>the freq without affecting the PWM or the other way around.
>Actually that works too.. tho it doesnt change linear, but in a X^2
fashion.
>(Freq = 2952,9 * delay ^-0,9829   that equation wont be fun to use in asm)

>This non-linear freq. update behavior have the effect that when the user
>presses the freq-down, or up,-button  it will change rapidly around 50Hz
>and slow around 5Hz.


>The real problems begin when the 2:nd chan is to be done... Independent
>PWM factor is easy, but independent freq.. hrm

Hello Patrick,

not shure if you solved it already.
I had to solve same problem for independant
selection of pwm and frequency without affecting each other,
and used a PIC12C761 with two pots on two analog input pins.
The range is 1 - 100 Hz and  PWM of 1 - 99 %.
The application is written in CCS C and uses timer0 interrupt.

Thomas

--
http://www.piclist.com hint: The PICList is archived three different
ways.  See http://www.piclist.com/#archives for details.


2002\10\21@073546 by Patrick J

flavicon
face
From: "Thomas Lehmann"
> not shure if you solved it already.
> I had to solve same problem for independant
> selection of pwm and frequency without affecting each other,
> and used a PIC12C761 with two pots on two analog input pins.
> The range is 1 - 100 Hz and  PWM of 1 - 99 %.
> The application is written in CCS C and uses timer0 interrupt.
> Thomas

Hi !
No, its not solved yet, atleast not the way i want it (linear and 2
independent
channels) I've been thinking about using IRQ on some timer but havent
got so far yet.

>written in CCS C
That would be interesting to take a look at if you dont mind sharing. One
day
I must start with C. (this could be a good first project)

Microchip has some C-compilator for free I belive (for 16F877 ?)
CCS C isnt that one, right ?


Best regards
/Patrick

--
http://www.piclist.com hint: The PICList is archived three different
ways.  See http://www.piclist.com/#archives for details.


2002\10\21@153225 by s Lehmann

flavicon
face
part 1 1207 bytes content-type:text/plain; (decoded 7bit)



-
{Quote hidden}

Hello Patrick,

I cannot share the whole project, because it was contract work,
but in the attachment you find the pwm and frequency part with a
small description which might be of interest for you (also for assembler
programming).

The example shows the values for 25% pwm and 25Hz.

Pins 2 and 3 are for the gate-driver of the mosfet.

The PIC12C671 has enough resources to do some additional rs232 communication
at 1200 baud.

Hope this helps

Best regards Thomas



part 2 2822 bytes content-type:text/plain;
(decoded quoted-printable)

/////////////////////////////////////////////////////////////////////////
////                                                                 ////
////  Thomas Lehmann Revision B vom 23.05.2002                       ////
////  12C671 pin    BOARD:                                           ////
////     1          VCC   +5V                                        ////
////     2          GP5   Frequenz dig. OUT                          ////
////     3          GP4   Frequenz dig. OUT (invertiert)             ////
////     4          GP3   dig. IN, Starttaste (active low)           ////
////     5          GP2   analog IN, Einschaltdauer                  ////
////     6          GP1   analog IN, Frequenzpoti                    ////
////     7          GP0   analog IN, PWM-poti                        ////
////     8          GND   gnd                                        ////
////                                                                 ////
////                                                                 ////
/////////////////////////////////////////////////////////////////////////

/* This is a part from MAIN */

while (TRUE)
  {
  pwm_value=read_adc();            //channel0 = pwm
  var_A=pwm_value;                 // Beispiel: bei 25% duty cycle ist dieser Wert 64
  set_adc_channel(1);              // channel1 = frequenz
  var_B=var_A * 200;               // Beispiel: 64 * 200 = 12.800
  var_A=var_B / 512;  //256 * 2    // Beispiel: 12800 / 512 = 25
  pr2_reg=var_A;                   // Pulszeit   Beispiel: = 25
  pr2_reg_complement=100-pr2_reg;  // Pausezeit  Beispiel: 100 - 25 = 75
  freq_value=read_adc();           // channel1 = frequenz
  var_C=freq_value;                // Beispiel: bei 25Hz ist dieser Wert 64
  var_D=var_C * 200;               // Beispiel: 64 * 200 = 12.800
  var_C=var_D / 512;               // Beispiel: 12800 / 512 = 25
  set_adc_channel(0);
  duty_faktor_pos = (100 * pr2_reg) / var_C;            // Beispiel: (100 * 25) / 25 = 100  [
  duty_faktor_neg = (100 * pr2_reg_complement) / var_C; // Beispiel: (100 * 75) / 25 = 300
  }

/* This is timer0 interrupt service routine */


// Timer0 ISR:
  #int_rtcc
  tim0_isr()
  {
  set_rtcc(223);                               tmr2_reg++;
     if( !bit_test(userflag,output_state) && (tmr2_reg >= duty_factor_pos) ) // Pulsausgabe
       {
        bit_set(userflag,output_state);
        tmr2_reg=0;
        output_high(PIN_A4);
        output_low(PIN_A5);
        }
     if( bit_test(userflag,output_state) && (tmr2_reg >= duty_factor_neg) )  // Pauseausgabe
        {
        bit_clear(userflag,output_state);
        tmr2_reg=0;
        output_low(PIN_A4);
        output_high(PIN_A5);
        }
   }

part 3 131 bytes
--
http://www.piclist.com hint: The PICList is archived three different
ways.  See http://www.piclist.com/#archives for details.


2002\10\21@210558 by Patrick J

flavicon
face
> but in the attachment you find the pwm and frequency part with a
> small description which might be of interest for you (also for assembler
> programming).
> The example shows the values for 25% pwm and 25Hz.
> Pins 2 and 3 are for the gate-driver of the mosfet.
> The PIC12C671 has enough resources to do some additional rs232
communication
> at 1200 baud.
> Hope this helps
> Best regards Thomas

Hi, Thomas,
Thanxs for the snippet :-)
I havent written anything in C yet. I'll go through it and see what I
understand
of it and can use.

/Patrick

--
http://www.piclist.com hint: The PICList is archived three different
ways.  See http://www.piclist.com/#archives for details.


More... (looser matching)
- Last day of these posts
- In 2002 , 2003 only
- Today
- New search...