PWM project for tiny DC motor turning painting

I have all the parts in hand for my project including
a solderless board for proof of concept. It has 63
by 10 connection points plus 4 by 63 connection
points 2 on each long side of the board marked +
and -.

What are the conventions for connecting components?

What gauge wire is needed for connections between
components on a breadboard? The guy at RadioShack
may know but I thought I should ask beforehand. I
assume it comes in a little spool.

Thanks in advance for your answers.

Michael Eisenstadt

 

I misspoke myself when describing the solderless board.
The outer lines of connection points, the ones marked
+ or -, start at 3 and end at 61, skipping connection
points at 8, 14, 20 and so on.

Michael Eisenstadt

 

It might be handy for the breadboarding stage to leave the leads of the
components uncut and use them instead of wires. If you need more wire you
can buy flexible wire on a spool, just explain the RadioShack guy you want
to use it for prototyping on a solderless breadboard. You could also buy a
few meters of different colors, quite handy when debugging your work.
Before you can use freshly cut wires you'll need to solder the ends, in
order to make them a bit stiff and keep the 'hairs' together and avoid
shorts. For prototyping, work without a powersource and check everything
carefully before you connect it.

Does your project includes ic's? They are sensitive to static charges, so
keep them in a safe container until you've finished everything and don't
put them in place until you've finished the other parts of the project.

I don't know much about conventions, but when making a 'final' version,
try to keep the number of crossings low, and if you point the directions
of the colorbands on the resistors all in the same direction (i.e. the
gold band always left/up for example) it's much easier reading the values.
This might help finding mistakes.

I assume you're project isn't about some high power/current/voltage
device?

have fun, bye, Olaf

 

Thanks, Olaf, for your prompt reply.

I don't think I made my questions sufficiently clear.

http://www.austincc.edu/umlauf/PWMproject.htm shows
the schematic, the parts list and a photo of the board
I bought.

You will see from the holes that I haven't attached the
ground post or the in and out posts yet.

I don't understand the conventions of the board. Are
connection points a through j which are perpendicular
to the numbered line internally connected or are the
numbered lines internally connected? Or neither? How do
the outer lines of connection points differ from the
10 inner lines of connection points? What are the +
and - marks meant to indicate? I am guessing that a
connection to any of the - points from the grounding
post is internally connected to the other - connection
points. Or maybe not. Then what do the + connection
points connect to? The voltage in post or the voltage
out post?

The NTE-2382 which gets hot and is sensitive to static
electricity is put into the board last, I got that point.
Its 3 legs seem longer than the depth of the board but I
guess they don't have to be bottomed. The other end of
it is a tab with a hole in it to which presumably the
heatsink is attached.

What part do I start with and where do I put it?

More generally how do I organize the placement of the
components on the board?

Thanks in advance for your help.

Michael Eisenstadt

 

Hi Michael,

I got the point. First the layout of the board. This is important to know,
otherwise it won't be possible to do anything with it. Do you own a
multimeter? In that case it is quite simple to test which holes are
connected and which arent. If you don't have one, consider buying one.
They are quite useful (essential) and a simple one costs a few euro's,
dollars, ...

But to make things a little easier for now, I have the same breadboard,
except for the black backside. So: The two outer rows on both sides (top
and bottom) are connected from left to right. In total four separate rows
of connections. Usually you use these for positive and negative voltage
from your powersource. This is the meaning of the blue - and red +. I
suspect these lines are connected to the in and out on the post. But I
don't know for sure! Use a multimeter to check, or a small lightbulb and
some wires, or a led and a resistor, or examine the backside of the post
or...

The other holes are connected by pairs of 5: 1a-1b-1c-1d-1e are connected,
so are 13f-13g-13h-13i-13j. You can use these pairs of five holes to
connect components, usually every thick dot in your circuit will be one
row of five holes.

If you want to use ic's (such as the CMOS 4011, which is also quite
sensitive to statics ;-) you need to place them over the central 'valley',
i'm sorry, don't know the correct English word. This is the only way to
put in an ic in without shorting the pins.

The leads of your components should be put firmly into the holes: they
should be clamped by the metal in order to make good contact, but it's not
necessary that they reach the bottom of the holes.

And now for the fun part, the layout. Well, that's your puzzle ;-) For
your circuit I'ld start by choosing a position for the 4011 somewhere in
the middle. (Don't put it in, just choose). Use the upper + and - line for
power. You have to be a bit creative, for example C1 connects the
powerlines, so you can put it in on top, connecting the blue and the red
row. Use short wires or pieces or flexible thin copperwire to connect pin
2 and 3 of the 4011. In general, the layout doesn't matter, as long as the
connections are the same. So pin 2 and 3 of the 4011 and one side of R1
should be connected to one each other somehow, not more, not less. And so
on. It takes some experience to put the components in cleverly, but this
won't be a problem with such a small circuit. You'll just end up using
more space and extra wires than necessary. But who cares ;-)

Do you know which lead is which on the NTE-2382?

Will this help? Anyway, just ask when in doubt, bye, Olaf

 

The proto board has lines of holes connected in two ways. There are
long busses that parallel the long edges of the board (labeled + and -
in this case) so those entire rows are one long connection node. The
rest of the board consists of groups of 5 holes connected in the other
direction (from the center groove out to the busses).

So you plug the chip straddling the central groove and you get 4 extra
connection points for each pin. 24 or 22 gauge wire is a good fit for
the connectors (the leads on a 1/4 watt resistor is just about right)
and the pins oh power devices like your big MOSFET may over stress
the spring contacts so they will not make reliable contact with wires
afterward. I often solder a half inch of clipped off resistor lead to
such devices to make the proto board last longer. This also raises
the device up to get its heat sink further from the board. Radio
Shack may sell a little kit of precut and pre stripped bits of wire in
color coded lengths for use with these boards. If things get crowded,
it is handy to have some insulation stripped off the next heavier
gauge wire to slip over long component leads to allow then to run past
each other without shorting.

By the way, this circuit needs a diode from the +6 volt supply to the
drain of the mosfet to prevent over voltage damage when it switches
off into an inductive load like a motor. The cathode goes to the +6
supply to reverse the diode when the fet is on.

 

Olaf,

Thanks to your hand holding I am making some progress.

A quick question about the pin configuration of the
14 pin 4011 CMOS NAND Gate. Pin 1 is identified on
the case but I dont know whether the number of the
pin opposite pin 1 is pin 8 or pin 14.

Thanks in advance.

Michael Eisenstadt

 

14. Conventionally the pins are numbered anti-clockwise. The datasheet
will usually have a diagram too.


Tim

 

The pin numbers circle the chip, so the highest pin is opposite pin 1.

 

good you're moving on! You can find answers on similar questions by
googling for a datasheet. Try 'datasheet 4011' in google and you'll get a
number of links where you can download it. For common ic's, transistors,
etc, this usually works fine.

bye, Olaf

 

I'm not up to that point of the project yet but I don't want lose sight
of your tip. Could you possibly specify a part number for a diode that
would be suitable (schematic at http://charlesumlauf.com/motorcon.gif)?

The motor is nominally rated at 12VDC @ 0.3a.

Thanks.

Michael Eisenstadt

 

Almost any fast diode rated for a quarter amp or so will work. 1N4148
(100v 200 ma switching diode) will work but may overheat on motor
stall:
http://www.fairchildsemi.com/ds/1N/1N4148.pdf

1N5817/8/9 1 amp schottky diode is just about right:
http://www.onsemi.com/pub/Collateral/1N5817-D.PDF

There are lots of other 1 amp schottky diodes in the catalogs that are
fine, also.

 

You should use #22 or 24 solid wire on these solderless breadboards.

I believe Radio Shack may sell a kit of pre-cut and stripped wires,
but I'd buy a 100 ft reel of #24 and cut off pieces as needed. (some
people have suggested using scrap telephone cable, if you can find
it.)





--
Peter Bennett, VE7CEI
peterbb (at) interchange.ubc.ca
new newsgroup users info : http://vancouver-webpages.com/nnq
GPS and NMEA info: http://vancouver-webpages.com/peter
Vancouver Power Squadron: http://vancouver.powersquadron.ca

 

I use scrap CAT-5 cable. It's everywhere, easy to strip, and the twisted
pairs sometimes come in handy for things like power, switches, speakers,
etc.

 

I use scrap CAT-5 cable. It's everywhere, easy to strip, and the twisted

Thanks for the suggestions about using telephone cable.

Another quick question. In the schematic, the 100k pot's 3
leads connect one of them to resistor R1, another to capacitor
C2 and another to pins 3, 5 and 6 of the CMOS NAND Gate.

The 100k linear pot I bought at Radio Shack (still in bubble pack)
has no markings on the 3 leads. Which lead goes to which connection?

(Please refer to http://charlesumlauf.com/PWMproject.htm)

Also what kind of socket is needed for the CMOS NAND Gate (mentioned
in the parts list)? I assume that it, like the MOSFET with its
heatsink, needs to dissipate heat.

Again thanks to all of you who have been helping me with this project.

Michael Eisenstadt

 

Pots almost always have the leads coming out in some 1 fold symmetry,
with two of the terminals symmetrical about a fold line passing
through the third. The two symmetrical terminals are the ends of the
resistor (100k, in this case) while the one on the line of symmetry is
the sliding contact that can be made to measure from 0 to 100k to
either of the others, by turning the pot. If in doubt, check the
resistance with a meter. This circuit has the wiper connected to one
end of the element to make a variable resistor in stead of a variable
voltage divider.

CMOS chips draw very little power so the proto board is all the socket
you will need. You may have to lay the chip on its side on the table
and form the leads a bit to make them fit the board spacing and plug
in easily. They are made with the lead angled just a bit wide, so
that they can be squeezed by a gripping head on automatic insertion
equipment, to give a precise size, when held for insertion. The
outward spring helps to hold the chip in the board till soldered. But
this feature may make it hard to get the chip into the proto board
without an adjustment.

It brings back memories.

I don't know if anybody has warned you about static. Both the CMOS
logic chips and MOSFET power transistors have inputs that are
capacitors formed with a very thin coating of silicon oxide. Normal
body static can easily apply enough voltage to this layer to blast a
hole in it, destroying the part. This is especially problematical in
winter, when the humidity is low. The manufacturers recommend that
you work on a static dissapative table and ground yourself with a
ground lead and lots of other static precautions when handling these
devices. But you should have little trouble if you just keep in mind
how static might be built up and discharges through one of these
parts.

A wooden table or a table covered with a piece or cardboard is more
static dissapative than a formica surface or a metal surface (the
metal surface allows very large static currents to flow, while the
plastic surface can hold stored charge for a long time. Do not walk
across a room and reach out and touch one of these devices before you
touch a grounded surface, first. Do not carry one of these devices
and make contact with a grounded surface with the device. Enclose it
in your hand, and make contact with ground with your knuckle before
opening your hand. Never walk up to a circuit and touch one of the
pins on a chip first. Always touch something grounded and then the
battery. Well, you get the idea.

Oh, one other gotcha. Scotch tape produces prodigious static when it
is peeled. Be careful after having opened boxes that you had to peel
tape or labels off of.

I have never used a body grounding strap, but with such simple
precautions, I have never had a device fail because of handling. Just
get in the habit of thinking that there may be a static spark each
time you reach for something.

But it doesn't hurt to have a few spare parts for any project in case
you happen to zap something or connect the power backwards. We have
all done that.

 

Remember that the two leads tied to C2 and R1 are actually tied together as
well; by connecting the leads together, the pot is being used as a rheostat.
The center lead of the pot is the wiper (arrow) and the outer two leads are
either end of the resistor. It shouldn't matter in this application which of
the two outer leads goes where.

On a solderless breadboard, no socket should be required. On a solder board,
I wouldn't use a socket either; logic chips are usually cheaper than sockets
where I buy them. You can use a socket (on a solder board) if you are really
concerned that you won't hook the wiring up correctly the first time. Logic
chips dissipate practically nothing; you won't need a heat sink.

 

I asked the terminals on the 100k pot.

If I understand you aright, and having tested with the multimeter, the
center terminal should be connected to R1 resistor, one outer terminal
to C2 capacitor and the other outer terminal to pins 3, 5 & 6 of the
NAND Gate. Is that right?

http://charlesumlauf.com/motorcon.gif

As for the static dissapative environment of the project, I
will try to remain optimistic ;-).

TIA

Michael Eisenstadt

 

All those get connected together.

 

I'm not entirely following you here.

Have I wired the pot's terminals correctly? I am asking before
soldering the 3 connections.

Please refer to schematic and picture at
http://charlesumlauf.com/PWMproject.htm

Again thanks for your indispensable help.

Michael Eisenstadt