Conference vmszoo::rc

Anybody out there have plans for an electronic speed control for an
    RC car. I'm not into cars but a friend of mine is and he's looking
    to build one.
    
    Thanks,
    Dan Eaton
    

    O.K. what features are you interested in?
    I would like to see the following:
    
    1)   It should be able get speed information directly
         from the receiver (eliminate the throttle servo).
    
    2)   It should also have a power tap for the receiver.
         (eliminate the battery pack).  
    
    3)   If item 2 is implemented it should shut down the
         throttle when the batteries get low.
    
    4)   It should be light enough to be used in electric
         planes (my interest) or cars.
    
    5)   Since you can buy one of these things off the shelf
         It should be cheap and easy to build. (Don't know
         if this ones possible)
    
    Anybody got any other suggestions?
    
    Wayne...

    
    One more,
    
    If possible, I would like to have at least one reverse speed, to
    get out of a state of stuckness. (RC offroad)
    
    							Lorrin Lee

    
    
    What about six or seven 25 amp FETS in parallel, this would minimize
    voltage drop.  One 25 amp fet for brake.
    
    No reverse by now... Sorry Lorrin thats coming in the second revision.     
    
    The reverse can easily be implemented by a relay to switch polarity.
    
    We need to switch on and off the FETS to achieve the variable speed
    effect.
    
    How do we drive the FETS and decode the receiver commands?
    
    I propose a modular design in which the power switch section (FETS)
    lies (ellectrically - not phisically) separated from the command 
    decoding function. In that way it's easier to modify the design
    to fit most popular receivers (let's say FUTABA, AIRTRONICS, etc).      
                                                                 
    Let's go on...
    
    Orlando.
    

In my musings of such a design, I thought of taking the 1-2
millisecond decoded pulse from the RX and using it to key a frequency
to voltage converter.  The output of the VFC would charge a
capacitor/resistor network to an isolating OPAMP, which in turn drives
the FETs.  Howzatt?

    
    We need to provide to the FETS a square waveform of varying duty
    cycle to achieve the variable speed effect. 
    
    How do we get this from the VFC output?
    
    Orlando.
    

    
    	Put the VFC output through any digital gate, or perhaps if 0.8v
    to 4.0v are not the levels there are lots of level converting types
    of circuits you can use.
    
    					Mike
    

Fingers going faster than the brain...what I meant was FVC;
frequency-to-voltage converter.  For instance, the National LM2907 or
LM2917 Frequency to Voltage Converter (Linear Databook, 1982,
National Semiconductor, pg 9-135).  The voltage output of the
converter would be buffered by the opamp to directly drive the FETs. 

    RE: .9
    
    You don't really want a FVC.  The frequency of the pulse train coming
    out of the receiver is constant (well, approximately).
    What you really want is a pulse width to voltage converter.
    The simplest approach is to charge a cap from a constant current source
    for the duration of the pulse.
    The longer the pulse, the higher the cap's voltage. When the pulse
    is over, save the voltage in a sample and hold ckt of some sort.
    Also, directly driving the FETs to control the motor voltage wastes
    the battery since there is a lot of dissipation in the FETs.  Instead
    drive the FETs with a pulse width modulatior (PWM).  A 555 or 556
    could probably be rigged up to work. An interisting possability
    would be to use an ordinary servo control IC as the PWM generator
    but use tachometer feedback (maybe a FVC using a one-shot) rather than
    position feedback. This would allow accurate motor speed control based 
    on the transmitter's throttle position, and handle the interface to 
    the receiver.

RE: -1
Yeah, after I wrote that, I realized the problem with the
FVN approach...the demodulated channel signal is a nominal
1.5 millisecond pulse with an approximately 18 millisecond
interval.  Probably would not work with the FVN.  

Your approach is a little more complex, but, you're right on
the FET current drain.  The big hitch that I see for a start
is that the servo driver IC's from Signetics are hard to
get.  I think that ACE and DigiKey have them, but that's
about all.  

Gonna extract your note and kick it around a little. 

        An additional feature...
        
        EMF braking. Something like 1/2 the "reverse" range gives braking,
        more reverse actually gives reverse power. EMF braking turns the
        motor into a generator, actually charging the battery from the
        cars kinetic energy. 

    For those of you still looking for a schematic for a speed control.
    The April 87 issue of Model Builder has two. The first is a digital
    proportional  control with reverse for 6-24 VDC. The second is a
    ON-OFF motor control circuit. You will find them in the electronics
    corner on page 32. 
    
    - Bob

        Mind filling me in? I don't get RCM. Can you give more details on
        the article, and what the control is for, cars or planes? Does it
        have reverse or braking? What does it cost? Do I have to etch a
        board? Etc, etc? 

    The speed control is for planes. The module measures 1" X 2.6" X
    .75" and weighs about an ounce. The following are available from
    the author:
    
    		PC board	$5.00
    		MOSFETS (4)    $12.00
    		all parts      $25.00
            assembled unit     $39.95
                 
    If you're interested, I can send you a copy of the plans. The article
    includes the schematic and PC artwork.
    
    					- Ron

       Hi, I have just brought my son a Hornet kit for his Christmas
       present.
       
       I have noticed that there has been a discussion on electronic
       speed controllers (note 8). The conclusion was that a schematic 
       was published in the April '87 issue of Model Builder. As 
       locating this publication here in the U.K. could be a bit of a 
       problem...
       
       My question is this. Does anyone have a copy of this article, and
       would they be willing to copy/post it to me?
       
       My mail address is;
                           Howard Easton @REO G/3-2
       
       Regards, Howard.

    Hi Howard!
    
    I had read you note on 6 Nov. and was wondering if anyone had gotten
    a copy of the the April issue to you.  I could probably get you
    one.  But I did notice another set of plans for Electronic speed
    controller in the December 1987 issue of Radio Control Modeler.
    On Pg. 184, plans, description and ordering info are all provided.
     The controller referenced here is good for boats and cars.  One
    point though, it isn't full speed reversing.  The author suggests
    that full speed reversing strips gears in cars and floods boats.
    Anyway, if you're interested and can't get a copy over there, let
    me know and I be happy to send you one.
    
    Dave Teerlink.
    

    If still interested see note 8.19.
    
    Dave T.
    

        The May '87 issue of RCM and the April '87 issue of Model
        Builder had articles on building an electronic speed control.
        I'll try to give a summary of the 'features' of the circuits in
        both articles with some comments from a car enthusiast's point
        of view. 
        
        
        RCM - May '87 (page 176)
        
        The speed controller described in this article uses mosfets in
        the motor drive circuit. The controller provides proportional
        speed control with a 'fail-safe' circuit that shuts down the
        motor if the radio signal is lost.  Features missing are reverse
        and braking.  Circuit schematics, parts list, and foil patterns
        for the pc board are included in the article. A brief overview
        of circuit design and operation is given as well. IRFZ30 mosfets
        are used in the design resulting in a .12 volt drop at 10 amps.
        The author indicates that using IRFZ40 mosfets the drop can be
        cut in half. This is still far from being as efficient as the
        better mosfet controllers you can buy. Parts are available
        postpaid from the articles author. 
        
        	Howard W. Cano
        	7057 Vivian Ct.
        	Arvada, Colorado 80004
        
        	PC Board			$ 5.00
        	IRFZ30 Mosfets (set of 4)	$12.00
        	Complete Kit			$25.00
        	Assembled and Tested Unit	$39.95
        
        
        Model Builder - April '87 (page 32)
        
        This article actually has two 'motor' controllers. One is a
        simple on/off controller using a relay as the switch mechanism.
        Clearly this is of no interest to anyone wanting a 'speed'
        controller. The other circuit is implemented using bipolar
        transistors. The bipolar transistors have a 1 volt drop at 15
        amps so this makes a poor efficiency unit that results in very
        hot transistors. The transistors have a 25 amp continuous rating
        and 40 amp surge. If your interested in building a decent speed
        controller this design may be a good example of what not to
        build. 

    I have completed a speed controller recently.  It was intended to
    be as much a learning exercise (about the Radio Control part of RC)
    as an applied gizmo.  Since there are several high performance
    speed controllers on the market (eg Novak), I did not expect to
    break new ground.
    
    The unit is based on my own functional analysis and a published
    aircraft speed controller which used a CMOS HEX Inverter for the control
    function.    It was designed initially to provide either Forward-
    Neutral-Brake or Forward-Neutral-Reverse which was switch selectable.
    It is also fully proportional and operates in a PWM mode. It
    automatically shuts down if the receiver losses contact with the
    transmitter. 
    
    I used 3 CMOS IC's, the HEX Inverter, a quad XOR, and a quad AND.
    The output was thru an H-bridge amp consisting of paralleled IRFZ40's
    for forward branches and single IRFZ40's for the Brake/Reverse
    function.
    
    The unique, and limiting part of this design (other than having reverse
    as an option) was the stacking of the control and receiver logic
    power on top of the motor power pack voltage.  There are 2 advantages
    to this.  1) I can get well over 10 volts (12-14) driving the gates
    of the output HEXFETS which allows the HEXFET ON resistance to approach
    the rated 25 milliohms each, thus increasing energy transfer between
    battery and motor, 2) The radio/servo battery prevents loss of or
    unpredictable control when the motor battery begins to reach the
    end of its discharge cycle.  The limiting part is obviously the need
    for the receiver/servo battery (although this can be replaced with a 
    DC-DC converter, you would lose advantage 2).
    
    After bench testing the design and being satisfied, we went to the
    field. There we found something was causing the etch which provided
    current during Reverse to overheat and lift.  Since we wanted to go
    forward with the unit and not spend too much time chasing it down,
    I reconnected the 8 HEXFETS in a more conventional source follower
    forward, short the motor windings brake mode.  In this configuration
    the controller has a less than .004V/A drop.  The unit also remains
    relatively cool to touch after a full charge run on the track using
    a stock Reedy motor.
    
    I constructed the unit in a 1.5 x 1.5 x .5 package which was potted
    in polyester foam and coated with plastic film.  Including wires
    and connectors it weighs 2 oz.
    
    I spent about $35 all together on new materials (I had everything
    except the HEXFETS already). It would cost about $45 in material to 
    duplicate what I have. The equivilant unit in the marketplace 
    (ie Tower Hobbies) sells for about $140-$170.
    
    I am willing to share the schematic and layout to anyone brave or
    fool enough to want to do one yourself.  If you are not good
    technically though dont even consider it.
    
    Walt

>    I constructed the unit in a 1.5 x 1.5 x .5 package which was potted
>    in polyester foam and coated with plastic film.  Including wires
>    and connectors it weighs 2 oz.

Walt - I was also playing with a soldering iron lately.  How did you go
about the "polyester foam and plastic film coating"?  I have a simple
open circuit board that I would like to pack and protect.  

Please describe the materials and procedure.

Bye          --+--
Kay R. Fisher  |
---------------O---------------
================================================================================

    I have a piece on that in ELECTRO_HOBBY. Note 607 I think.
    
    Basically I created a teflon coated mold form. I set the module
    into that and injected polyester foam from a can of that stuff
    that Grossman's sells to insulate your inside electrical outlets from
    drafts.  It takes about 18 hours to cure out, and I had to add more
    foam and reinsert the unit into the mold twice to fill the voids
    completely.
    
    Walt

    A great place for FETs is Active Electronics in Westboro, Ma.  They
    carry a very wide variety, including most of the BUZ component lines
    (these seem to be the most popular in ESCs, BUZ11 is what I've seen the
    most of).  Also, they carry books, one particular set would be of
    interest to you, I own it, it's a MOSFET application guide and a MOSFET
    handbook (can't remember the manufacturer right now).  Very complete,
    with Industry Cross Reference.

    One of the nice things about FETs is, provided they have the necessary
    current handling capability, you can get lower on resistance from a high
    resistance part simply by adding more of the same FETs in parallel.  As
    you shop around, you might keep this in mind.  Very low On-resistance
    FETs are expensive, but a decent intermediate range component, in
    sufficient quantity to give you the desired resistance (when hooked up
    in parallel) might well price out much cheaper.  You will of course
    sacrifice some weight, but the FET's only weigh in at a few grams a
    piece, and there's about 28 grams to an ounce, so you could have quite a
    few FETs before you need to worry about how much they're weighing down a
    plane.

This looked like a good reference from the usenet.

Bye          --+--
Kay R. Fisher  |
---------------O---------------
################################################################################

Article         4351
From: [email protected]
Newsgroups: rec.models.rc
Subject: ESC references
Date: 10 Apr 91 06:12:48 GMT
Organization: University of Canterbury, Christchurch, New Zealand
 
Sorry about taking so long to post this, but when I had the spare time the
system was always down!
 
Well, here are the references I have to Electronic Speed Controllers. They are
in the format:
 
MAGAZINE, DATE, TITLE, AUTHOR, PAGE
	my comments on the article
 
********************************
 
Model Builder, Dec. 1990, Electronic Speed Controls, Bob Boucher (Astro
Flight), p48
	Good introduction to ESC's. Compares the theoretical performance
	of resistor type controllers, low rate controllers, and high rate
	controllers. Does NOT provide any circuits.
 
Model Builder, Feb. 1989, Making an Economy Electronic Throttle, Howard W.
Cano, p31
	Construction article for a simple and very compact proportional ESC.
	Contains circuit diagram, PCB artwork and circuit discription. Uses
	3 MOSFET's, a 4069 CMOS hex invertor, two transistors, and a few
	diodes, resistors and capacitors. Handles up to 24 cells and works
	by stretching the output pulses from the receiver.
 
Model Builder, Apr. 1987, Electronics Corner, Eloy Marez, p32
	Contains two circuits, both using the NE544 chip. One is for a
	proportional ESC with reverse for 6-24 volts, while the other is for
	a simple on-off controller using a relay.
 
Model Builder, Jan. 1987, Electronics Corner?, Eloy Marez?
	I haven't seen this article, however it is refered to in the April
	1987 article (previous reference).
 
Radio Control Models & Electronics, Aug. 1983, Motor Switch, Roger Todd, p650
	Gives circuit diagram, description and PCB artwork for a on-off RC
	driven switch. It has adjustable turn on and turn off points, and was
	designed to switch the motor from the elevator channel in a two
	channel system (full up/full down to switch on/off). It also
	incorporates a low voltage cutoff and battery eliminator circuit.
	Designed to work with 5 to 10 cells.
 
********************************
 
In my Electra I use an on/off type controller built around a NE544 IC. I
designed it myself, however it is very similar to that given in the April 87
issue of Model Builder. The motor that came with my kit had a capacitor
connected across the terminals for noise suppression. I found that I still
suffered from glitches, however (in this model I use a Futaba Sky Attack 3
channel radio on 40 MHz). To overcome this I soldered a capacitor (47nF)
between each terminal and the motor case. Since doing this I have had no
further problems with glitching, even though I have a direct connection between
my motor battery and receiver battery grounds.
 
*********************************************************
Chris Kaiser
Postgrad - Elec Eng Dept
Canterbury University
Christchurch, NEW ZEALAND
 
E.MAIL: [email protected]
*********************************************************
	"When you're fresh out of lawyers
 	 You don't know how good it's gonna feel"
		- Al Stewart, 1988
*********************************************************