"Toy" DC Motor Control
Christopher Fairbairn
Hi,
I am a 15 year old and tired of going around in circles!
I am building a robot which is microprocessor controled (ATMEL AVRs and
PICs) and I keep running into a brick wall when trying to design the DC
motor control circuitry.
What I have is a gear box with two small toy DC motors which provides tank
style steering on my robot. The motors require 3 - 5V at about 1/2 an Amp.
At first I wanted to use an H-Bridge chip like the LM293N but thats
difficult for me to get in New Zealand (where I live) and if I buy one from
America it will cost me an arm and a leg in postage (Its unbelievable how
much different companies want to charge me).
I then thought about using a LM18293 which National Semiconductor says is a
direct replacement for the LM293 but it dosn't contain inbuilt diodes so I
would have to include 8 diodes on my PCB which I don't like for two reasons:
Reason 1: I am not that good at producing PCB layouts so I want a simple
design.
Reason 2: The space that the motor controller PCB neededs to go into is
small and there isn't enough room for the eight didoes.
I then thought of using 2 DPDT 5V relays with two TIP31 transistors to
enable PWM speed control (simply make the transistor switch on and off the
3V going to the relay's contacts). But I've been told by so many different
people that I would have a problem with the relays.
Problem : A 5V DPDT DIP Relay with 100VDC 2 AMP contacts won't be able to
handle my 3V 0.5Amp motors because of the High Rate of Switching of the
signals going to the contacts ie: the PWM signals (which I think is strange
since the voltage of the motors is well within the specs of the relays and
PWM dosn't produce that higher frequencies considering the relays I want
to use can handle AC fine).
So what do you think I should do? I would like to follow the last idea
beause in the case of a failure I will be able to get parts eaisly in New
Zealand. Will the fact that the signals going to the contacts of the relays
are PWM signals damage these DIP relays?
Thanks,
Christopher Fairbairn,
lgc...@es.co.nz
or if that fails try,
chris...@computer4u.com
r...@netgate.net
>What I have is a gear box with two small toy DC motors which provides tank
>style steering on my robot. The motors require 3 - 5V at about 1/2 an Amp.
>
>At first I wanted to use an H-Bridge chip like the LM293N but thats
>difficult for me to get in New Zealand (where I live) and if I buy one from
>America it will cost me an arm and a leg in postage (Its unbelievable how
>much different companies want to charge me).
First, you want the L293D, not L293N: if I recall correctly, the L293N is also missing the protective diodes.
Do a search on Dejanews for postings with "L293D" in the subject line, and "@aol.com" in the From: field. There's a guy who organizes group buys on hard-to-get parts, and he had some L293D's left over last time. This was several months ago, so they might be gone now, but it's worth a try: he'll probably be sympathetic to a fellow hobbyist, and not try to make a profit on shipping ;-)
>I then thought of using 2 DPDT 5V relays with two TIP31 transistors to
>enable PWM speed control (simply make the transistor switch on and off the
>3V going to the relay's contacts). But I've been told by so many different
>people that I would have a problem with the relays.
This is very strange: the only thing I can think of is that they thought you were going to try to PWM the relay coils, rather than the DC going through the contacts. The approach you're suggesting has been used a lot, and should work fine, as far as I can see (btw, don't forget to put diodes across the relays: they can blow up transistors, too).
The TIP31 may not have enough gain to do what you want, though: if something like a TIP120 won't work for you, you might need to set up a transistor or op-amp buffer to give it a little more base current. The PIC can *probably* handle it, but I dunno about the AVR.
Ran
Dan Danknick
r...@netgate.net wrote:
> >At first I wanted to use an H-Bridge chip like the LM293N but thats
> >difficult for me to get in New Zealand (where I live) and if I buy one from
> >America it will cost me an arm and a leg in postage (Its unbelievable how
> >much different companies want to charge me).
>
> First, you want the L293D, not L293N: if I recall correctly, the L293N is also missing the protective diodes.
>
> Do a search on Dejanews for postings with "L293D" in the subject line, and "@aol.com" in the From: field. There's a guy who organizes group buys on hard-to-get parts, and he had some L293D's left over last time. This was several months ago, so they might be gone now, but it's worth a try: he'll probably be sympathetic to a fellow hobbyist, and not try to make a profit on shipping ;-)
It's Richard Hughes and his website is at:
http://members.aol.com/rich924/index.html
Top notch individual from my experience.
Dan
Jason T. Harris
Christopher Fairbairn wrote:
>
> Hi,
>
> I am a 15 year old and tired of going around in circles!
Maybe you need to turn on the other motor.
> What I have is a gear box with two small toy DC motors which provides tank
> style steering on my robot. The motors require 3 - 5V at about 1/2 an Amp.
>
> At first I wanted to use an H-Bridge chip like the LM293N but thats
> difficult for me to get in New Zealand (where I live) and if I buy one from
> America it will cost me an arm and a leg in postage (Its unbelievable how
> much different companies want to charge me).
Have you tried Ben Wirz? I bought some of these chips from Ben a
couple of months ago and the service was just fine. Sending them to
NZ by airmail wouldn't cost very much because they are small and
weigh next to nothing. I'm not sure how much you were wanting to
spend but $30 US should put a couple of them in your mailbox.
> I then thought about using a LM18293 which National Semiconductor says is a
> direct replacement for the LM293 but it dosn't contain inbuilt diodes so I
> would have to include 8 diodes on my PCB which I don't like for two reasons:
I would say get a free sample of a LM18200 from NSC, but as I recall
you need a 12V or higher motor to use this driver.
>
> I then thought of using 2 DPDT 5V relays with two TIP31 transistors to
> enable PWM speed control (simply make the transistor switch on and off the
> 3V going to the relay's contacts). But I've been told by so many different
> people that I would have a problem with the relays.
>
> Problem : A 5V DPDT DIP Relay with 100VDC 2 AMP contacts won't be able to
> handle my 3V 0.5Amp motors because of the High Rate of Switching of the
> signals going to the contacts ie: the PWM signals (which I think is strange
> since the voltage of the motors is well within the specs of the relays and
> PWM dosn't produce that higher frequencies considering the relays I want
> to use can handle AC fine).
I tried something like this once. ie: Use relays to switch directions,
use power FETs for the PWM speed control. Yes it works: but..
1) The end result is bulky
2) There is certainly more labor involved than using a driver chip
3) You will still need protection diodes
4) You need power FETs to drive the relay coils
5) It is slow to switch
6) It is noisy (electrically an audibly)
7) It pulls a lot of current
And some special Jason touches..
8) I got some weird coupling between the two motors (???)
9) I ended up blowing up a power FET
When the last thing happened I got religion and decided to use
motor driver chips designed by professionals, even if it was a
pain in the arse to get them.
> So what do you think I should do?
I think you should use an LM293
--
Jason H.
To send me email replace nospam with jasonh.
Murray McCulloch
Christopher Fairbairn wrote:
> I am building a robot which is microprocessor controled (ATMEL AVRs
> and
> PICs) and I keep running into a brick wall when trying to design the
> DC
> motor control circuitry.
[...]
> I then thought of using 2 DPDT 5V relays with two TIP31 transistors to
>
> enable PWM speed control (simply make the transistor switch on and off
> the
> 3V going to the relay's contacts). But I've been told by so many
> different
> people that I would have a problem with the relays.
[...]
> So what do you think I should do? I would like to follow the last idea
>
> beause in the case of a failure I will be able to get parts eaisly in
> New
> Zealand. Will the fact that the signals going to the contacts of the
> relays
> are PWM signals damage these DIP relays?
I don't see why you want to use relays (big, mechanical, noisy), when
there are surely electronic alternatives (what about thyristors?). The
18 July 1997 issue of ETI has an article on DC motor control which shows
a few circuit examples of varying complexity. They may be more complex
than you need or want, but you will probably learn a lot by having a
look.
Murray
phracker
This sounds exactly like what I need for one of my projects - thanks for the
information! I'll stop by the www.national.com website and check out the specs.
he...@hpapcfla.nsr.hp.com wrote:
> My robot uses National LMD18200. Theses parts are complete H-Bridge drivers
> with a microprocessor interface. They can drive up to 55V at 3 amps continuous
> and 6 amps peak. Protection diodes are internal. The only downside is that
> this part costs about $18 in the USA. Check it out anyway.
>
> Henry
he...@hpapcfla.nsr.hp.com
John Atwood
>> >At first I wanted to use an H-Bridge chip like the LM293N but thats
>> >difficult for me to get in New Zealand (where I live) and if I buy one from
>> >America it will cost me an arm and a leg in postage (Its unbelievable how
>> >much different companies want to charge me).
>http://members.aol.com/rich924/index.html
>Top notch individual from my experience.
That looks good, I may get a handfull myself. Another approach might
be to salvage some chips off floppy disk drives; my local shop sells
'em for $1us.
See the section, stepping motor sites, at:
http://www.doe.carleton.ca/~jknight/RobotBookmarks.html
and this page has specs on many of the chips you're likely to find:
http://www.mrc.uidaho.edu/mrc/people/jff/giicm/driver.html
--
--Office phone: 541-737-4145 (Dearborn 311) home: 757-8772
Office mail: 303 Dearborn Hall, OSU, Corvallis, OR 97331
Greg Blackwell
In article <633q52$f...@Chaos.es.co.nz>, lgc...@es.co.nz (Christopher
Fairbairn) wrote:
> At first I wanted to use an H-Bridge chip like the LM293N but thats
> difficult for me to get in New Zealand (where I live) and if I buy one from
> America it will cost me an arm and a leg in postage (Its unbelievable how
> much different companies want to charge me).
>
Chris,
I've bought LM293B's and LM293D's from Veltek, here in Australia.
(http:\\www.veltek.com.au)
They don't always have them in stock and I had to wait until someone
else put in a bigger order but the postage to NZ should be lower.
(Their web page now has stock listing pages. Check out the SGS Thomson stock.)
LM293D: 0.6A drive capacity with internal clamp diodes
LM293B: 1.0A drive capacity but no internal clamp diodes
'The 6.270 Robot Builder's Guide' shows a LM293D soldered piggyback onto
a LM293B to give 1.6A drive with internal clamp diodes.
Greg Blackwell
gbla...@labyrinth.net.au
Edwin Wise
gbla...@labyrinth.net.au (Greg Blackwell) wrote:
>In article <633q52$f...@Chaos.es.co.nz>, lgc...@es.co.nz (Christopher
>Fairbairn) wrote:
>> At first I wanted to use an H-Bridge chip like the LM293N but thats
>I've bought LM293B's and LM293D's from Veltek, here in Australia.
>LM293D: 0.6A drive capacity with internal clamp diodes
>LM293B: 1.0A drive capacity but no internal clamp diodes
I keep seeing these messages about buying motor controller chips...
but they seem overly expensive.
I was wondering, why not use a digital-input-level MOSFET in a simple
H-bridge, and a bit of external logic to prevent cross-firing and
turning it into a smoke generator?
I mean, I assume I am missing something. The MOSFET I have played
with is the RFD14N05L from Harris, and it only cost me a couple of
dollars (if I remember correctly). It takes direct TTL input, and can
drive 14 Amps.
Edwin
Sarath R. Sirimanne
Hosfelt Electronics (800-524-6464) also sells L293D now for $2.95.
Shipping costs about $7.00. It is not cost effective if you are buying
only that.
Sarath
Dan Danknick wrote:
>
> r...@netgate.net wrote:
> > >At first I wanted to use an H-Bridge chip like the LM293N but thats
> > >difficult for me to get in New Zealand (where I live) and if I buy one from
> > >America it will cost me an arm and a leg in postage (Its unbelievable how
> > >much different companies want to charge me).
> >
> > First, you want the L293D, not L293N: if I recall correctly, the L293N is also missing the protective diodes.
> >
> > Do a search on Dejanews for postings with "L293D" in the subject line, and "@aol.com" in the From: field. There's a guy who organizes group buys on hard-to-get parts, and he had some L293D's left over last time. This was several months ago, so they might be gone now, but it's worth a try: he'll probably be sympathetic to a fellow hobbyist, and not try to make a profit on shipping ;-)
>
> It's Richard Hughes and his website is at:
> http://members.aol.com/rich924/index.html
>
> Top notch individual from my experience.
>
> Dan
Todd Schifferdecker
You are correct sir! Give that man a cigar... or do you roll your own
also? :)
Why spend big bucks for a wimpy little 1 Amp controller, when building
one is so easy?
Check out:
http://www.cadvision.com/blanchas/h-bridge.htm
There is a good circuit, and a great explanation of MOSFETs. The circuit
described uses N-Channel and P-Channel MOSFETs. Logic-Level MOSFETs are
not needed for this circuit, but can be used.
Mouser (www.mouser.com) has, for example:
RFP12N10L : Logic-Level N-Channel, 100V, 12A for $1.75 US each
RFP12P10 : P-Channel, 100V, 12A for $3.95 US each
A 1 amp version would be even less expensive.
P-Channel MOSFETs are harder to find, and more expensive, but make the
drive requirement much simpler. To use all N-Channel, you need to pull the
gate of the upper transistor at least 5 volts (10 is better) above the high
side of your motor supply.
See the excellent discussion of P-Channel only version at:
http://www.cadvision.com/blanchas/nch-brdg.htm
Or go to AltaVista Advanced Search and enter:
h-bridge or "h bridge" and mosfet and schematic
for lots of great links.
Like Edwin mentions, add logic to not allow the state marked "Not
Allowed" on the schematic. Don't forget heat sinks.
The faster you switch, the better. A MOSFET dissipates essentially
nothing when it is off, and very little (Rds_On times Current) when it is
fully on, but will use power while it is switching.
The 1N4000 diodes shown in the schematic are probably OK (especially if
you are not switching at PWM type frequencies), but a High efficiency
rectifier (HER103, etc), or a Schottky (1N5819 thru 1N5822) is probably
better.
HTH
--
Todd Schifferdecker
Send direct replies to:
jo...@i1.net
#include <disclaimer.std>
Edwin Wise <ew...@simreal.com> wrote in article
<345dd411...@news.continet.com>...
The2X4
In article <01bcea70$2e3abf50$7c3ce6cf@todd_pc>, "Todd Schifferdecker"
<jo...@NOSPAM.i1.net> writes:
>
> P-Channel MOSFETs are harder to find, and more expensive, but make the
>drive requirement much simpler. To use all N-Channel, you need to pull the
>gate of the upper transistor at least 5 volts (10 is better) above the high
>side of your motor supply.
Discrete bipolar tranisitors don't need this 'Super voltage' which a lot
of times is not
available on robotic systems. Or at least not on beginners systems. If the
5 or 10 volt excess is available I would think the design is complex. Or so
that the beginner might do well to work on a more simple logic level drive
to bipolar devices. Such a design is at http://efjdevices.com
zwge...@gmail.com
Founded in April 1997, Shenzhen ZhaoWei Machinery & Electronics Company, is a national high-tech enterprise which engages in research and development, production of gear transmission mechanism, providing all-round services for customers with the transmission scheme design, parts mould design and manufacturing, component production and integration assembly.
Our company has passed ISO9001, ISO14001 and automotive industry TS16949 system certification, won the title of national high-tech enterprises in 2013. Our company attaches great importance to scientific and technological strength and independent research and development. We have achieved 40 patents, including 17 invention patents. We take active participation in a number of scientific and technological projects around nation and Shenzhen city, and obtained remarkable achievements.
Zhao Wei is committed to the information and communication and new energy smart transmission mechanism, independently developed 4G base station antenna power transfer system, smart home, intelligent robot core driver module, in the international advanced level. And established long-term strategic partnership with HUAWEI, Germany BOSCH, Panasonic and other world top 500 enterprises, business growth rapidly and development prospects brightly. ZhaoWei is manufacturing enterprises which focus on precision molding parts injection and precision driving system. And started the Huizhou Machinery & Electronics Company project, as a sign to the world stage, and made a great contribution for economic development and social progress.
Products are mainly related to automotive components, industrial equipment, smart home, medical equipment and standardized gear box, providing services and solutions for the world's intelligent driving.
With 20 years professional design, development and production, provide you with one-stop service from design simulation to the production. ZhaoWei Machinery & Electronics, you are worthy of trust!
Company: Shenzhen ZHAOWEI Machinery & Electronics Co., Ltd
URL: http://www.zwgearbox.com/
Contact: Anny Liu
Tel：+86-755-27322652
Fax：+86-755-27323949
E-mail:sa...@zwgearbox.com
Add: Blk. 18, Longwangmiao Industry Park, Fuyong Tn., Bao’an Dist., Shenzhen 518103, Guangdong, China
cadc...@gmail.com
> Reason 2: The space that the motor controller PCB neededs to go into is
> small and there isn't enough room for the eight didoes.
>
If it is going inside a toy car, consider cutting off the top of the car, or even the entire upper body. Besides, it is good to show off that the item is not just a stock toy, but has been substantially modified. If you, or others, know 3D modeling, and you have access to a 3D printer at school, you might take advantage of it, and create a custom enclosure for your electrics.
-Joe
Jürgen Exner
comp.robotics.misc:
the Month.
>> Reason 2: The space that the motor controller PCB neededs to go into is
>> small and there isn't enough room for the eight didoes.
>>
>
>Don't get too caught up trying to keep the size small.
>At your age, you are undertaking a lot already, so you will run into plenty of other frustrations.
1997, I am pretty sure today he won't take kindly to being told he is
taking on a lot for his age.
jue