From the Lab January 1997 [Archive]

skranish

Jan 01, 1997, 01:00 AM

Why 'From the Lab'? 

Well, maybe the 'Lab' isn't the right term. Perhaps it should be 'From the Technical
Side'.  This column covers the technical side of  electric flight in a
not-too-technical manner. The reason is simple: even if you don't intend to build your own
electronic systems, or design your own plane, you really should understand how and why
things work. If you do not understand the basics, you will never be able to figure out
what went wrong when things don't work. 

Have you ever wondered how medieval engineers designed and built the enormous
cathedrals found in the major European cities? Almost 1000 years ago they designed and
built huge buildings with unobstructed interiors hundreds of feet high, and relatively few
interior columns. These buildings have stood, largely intact, for centuries. They were
build without the aid of any sort of computational equipment, or even any engineering
texts. How did they do it? They looked at other buildings, based their work on prior
experience, and frankly, they guessed a lot. If something failed, they strengthened it,
and made it better on the next one. 

How is this significant to us? Well, frankly, most model airplanes are designed an
built a lot like medieval cathedrals: a mix of prior work and some guesses for good
measure. How do you decide what size spar to use? Or whether or not something needs to be
reinforced? Probably you base your design decisions on prior experience, and if that
is not enough, well, you probably guess. 

I have been going through exactly this process trying to select a wing joiner for an
RCHLG. I looked at the wing joiner that I built into my Airtronics Square Soar, which I
know was based on the joiner in my Graupner Hi-Fly. This is simply a 6" long,
3/16" music wire rod, set into a brass tube which is placed between the spars. The
space between the spars is filled with balsa and a mix of microballons and epoxy, and
then a shear web holds the whole thing together. The RCHLG is smaller than either of these
gliders (it is a little smaller than a speed 400 class sailplane), and will not be
launched on a full size high start, so it does not need to be as strong. Will 1/8"
music wire do? How about 5/32"? How do I decide? Well, I will probably guess,
and go for the 5/32", because it will be easier to replace with a carbon fiber rod if
I want to save a little weight. 

It doesn't really need to be like this. There is actually a lot of good technical
information available, either specifically for model airplanes, or at least applicable. I
would like to put together a web page with a bibliography of technical information for
model builders. The information may be books, magazine articles, or academic papers. If
you know of any appropriate articles or material, please email me with the particulars. 

B2 Streamlines sells a book on 'Structural Dimensioning for RadioGuided
Models' which may provide an answer for my wing joiner question. There was also
a series on model structures in the early issues of QFI. 

<hr>

My Editorial Policy 

Somehow it seems like destiny that I would wind up writing my own webzine column. My
father is an old-time newspaperman and long time science writer who wrote about things
like the first Vanguard satellite 'launching' (the Navy said it rose to a height of 87
feet, without noting that 84 of those feet were the height of the rocket) and the
government experiments that exposed a lot of people to unnecessary radiation without
bothering to actually TELL them. My younger brother is a political correspondent for
a major newspaper. He has personally interviewed several presidents and a lot of
congressmen, and  seems to be impressed by none of them. So now I am a journalist,
too. With that job comes the issue of journalistic ethics. I have established for myself
the following policies for both this column and any product reviews that I write.
(note that these policies apply to MY writing, and everyone else is on their own):

<ul>
<li>Nothing in this column will be a paid advertisement in disguise. Unless otherwise noted,
all products described here are things I have purchased with my very own money and
actually used. If something is given to me free of charge for evaluation or review,
that fact will be clearly noted. It will not affect my approach to the review, but
the reader should know about it. </li>
<li>I will not reprint press releases and promotional material. If I write about a product
(or book, or magazine, or whatever) my writing will be based on actual, personal
experience. If I am simply mentioning something I am aware of, but have not actually used,
I will mention that fact. </li>
<li>I will only review things that I like. I really see no point in doing a completely
negative review, unless a product is dangerous, in which case the purpose of a review is
to protect potential purchasers from injury. This does not mean that I will not make any
negative comments about products - if you have read any of my reviews, you should
realize that I am openly critical of things I do not like. But it does mean that I
will not write any flaming negative reviews. There are too many worthwhile things to write
about, and frankly, if something is that bad, I am not likely to do enough with it to
justify a review. </li>
<li>All opinions expressed here are my own. Period. If you want to disagree, write your own
column. If you have a problem with something I have written, please feel free to contact
me, and If you actually have something worthwhile to say, I may include dissenting
opinions in the next column. If you really have a problem with something I have written,
please contact Jim Bourke, publisher of this WebZine. </li>
</ul>

<hr>

Beginner's Corner: AmpereHours, Watts, and WattHours 

In the last edition of Beginner's Corner, we covered Amperes and Volts. If you have not
read the previous edition, or need to review it, please do so before attempting these more
complex subjects. 

AmpereHours 

If you remember from last time, Current, measured in Amperes,
is the quantity of ping pong balls (well, electrons, really) that pass a given
point in a second. More specifically, an Ampere is a flow of 6.2422E+18 ping
pong balls per second. An AmpereHour is the number of ping pong
balls that pass a given point in a hour. If the current is constant, this is actually
the number that pass the point in a second, times the number of seconds in an hour,
which is usually 3600. So One AmpereHour, which is one Ampere supplied continuously for
one hour, is (6.2422E+18 * 3.6E3) = 2.2472E+22 ping pong balls, which is an awful lot of
ping pong balls. 

An AmpereHour is a measure of the ability to supply current over time, so it is
a convenient way to rate the capacity of a battery. Actually, we usually measure battery
capacity in milliAmpereHours, which is 1 milli, (that's a metric milli, not a Milli
Vanilli) or 1/1000 th of an AmpereHour. If we say that a battery has a capacity of
1000milliAmpereHours (normally abbreviated 1000mAH), it has the ability to deliver a
current of 1000mA (or 1A) for 1 hour before becoming fully discharged. (We will discuss
the fuzzy issue of exactly what 'fully discharged' means in a later column). 

It is important to understand that the capacity rating for a battery is usually
specified at a particular current, such as 20% of total capacity. In other words, the
rated capacity is measured at a current that will take 5 hours to fully discharge the
cell. The capacity of the cell will be reduced at higher currents - a 1000mAH cell can
provide 200mA for 5 hours, but will not provide 10A for 6 minutes. There are a lot of
factors that affect cell capacity at the current level we use to run motors, and cell
impedance (discussed below) is the probably the most important one. 

Typical battery capacities for R/C applications range from small 225mAH packs for RC
HLGs to typical 600mAH receiver packs to 1700mAH 1700SCRC flight packs. Much smaller and
much, much larger batteries are available for other applications. Sanyo makes cells from
45mAH to 20AH! 

If you look at a typical multimeter, it has an input and a display scale to
measure current from milliAmperes to Amperes, but no scale for measuring AmpereHours. The
reason for this is that the meter is actually measuring the number of ping pong balls
passing a specific point in a fraction of a second, and then scaling up the number to
be the number of ping pong balls that would pass that point in an entire second.
Amperes are really an instantaneous measurement, while AmpereHours is a time measurement. 

To measure AmpereHours, we need to measure current (count ping pong balls) and count
time. We would really like to know how many seconds (or what fraction of an hour) the
current can be supplied for. If we want to do it accurately, we have to measure the
current, say, every second, because it may vary over the time measured. We also need to
measure the voltage, to determine when the cell(s) are fully discharged and the
measurement is complete. A typical battery cycler applies a constant current load to
the battery, and then counts time until the battery is fully discharged. The cycler I
designed for Royal Electronics many years ago, and many similar units, simply display
the amount of time that the load current can be supplied for. More recent cyclers
actually display battery capacity in mAH. To convert from time to milliAmpHours, you need
to know the load current. Capacity = Hours * Current. If the load current varies
during the discharge of the battery, the calculation is far more complicated, because
you need to integrate current over time (gasp!). So it is actually much easier to just use
a constant (fixed) current load. 

Watts 

Our electric flight systems are designed to do the work of turning a
propeller. Turning a propeller 12000 times in a minute (or 200 times in a second) is
the same amount of work as turning it 12000 times in an hour, but the rate of work
is much higher - 60 times higher to be exact - because the work is done much faster. 

Power, measured in Watts (that's watts, not whats!) is the
instantaneous rate of doing work, which in our case is turning a propeller. The output of
the system is mechanical (turning the prop), so it may make sense here to consider the
mechanical definition of a Watt: a watt is 1 Joule per second, where a Joule is
the work done to move something 1 meter against an opposing force of 1 Newton, and a
newton is the push that will accelerate 1Kg at 1 meter per second per second. Whew! 

The input to the system is electrical, so in electrical terms, a Watt is a current
of 1 Ampere at a Potential of 1 Volt. It is really that simple: Power = Volts * Amps.


A Speed 400 motor may take in about 80 Watts, and a Speed 600 motor may take in 150
Watts. Larger cobalt type motors may take in hundreds of watts. As a point of reference,
one horsepower = 746 Watts. 

There are instruments available that will measure Watts directly, but it is really more
practical for our purposes to simply measure both voltage and current, and multiply the
numbers. 

Watt-Hours 

In the above description of Watts, you may have noticed that Power is an
instantaneous measurement - a snapshot in time. A power measurement tells you what is
happening at one particular moment, but most of us strive for flight times that greatly
exceed a single moment. While the instantaneous measurement of power at the motor may tell
us something about the motor itself, it does not tell us much about our power source, the
battery. The battery output power may be measured by measuring its output voltage and
current at any instant, but we may also want to know how much energy - power over
time - is provided by the battery. 

Energy, measured in Joules, is a total amount of work.
You may also want to consider it to be a Watt-Second, a power of 1 Watt supplied for
1 second. You may further want to consider it as a current of 1 Ampere at 1 Volt
supplied for 1 second. A somewhat more convenient measurement is the Watt-Hour, which is
1 Ampere at 1 Volt supplied for 1 hour. An this is really the measurement we are
interested in. 

A typical battery pack for a Speed 400 sailplane is a 7 cell KR600AE pack. It is
nominally rated at 1.2V per cell (8.4V for the entire pack) and 600mAH capacity. This
means, in theory, that the battery pack can supply 600mA at 8.4V for 1
hour, 5.04 Watt-Hours. That is actually not a lot of energy - remember that your
friendly local electric utility bills you by the kilowatt-hour, or thousands of
watt-hours. 

But the reality is somewhat different. I have measured a Speed 400 6V motor with a
Robbe 6-3.5 prop and a freshly charged KR600AE pack at 11.6 Amperes with a voltage of 7.6
Volts. This is a power of 88.1 Watts. At the end of discharge, just before the throttle
cuts off the motor so that the radio still has some battery left to power it, the voltage
will be 6.0V, and the current will be similarly reduced, to about 9.1 Amps or less. This
is a power of only 54 Watts. As a result, measuring Watt-Hours is not simple, because
we need to measure both the current and voltage many times. Nonetheless, Watt-Hours are
actually a good measurement for comparing battery packs. 

Instruments are available to measure Watt-Hours - there is probably one on the outside
of your house, placed there by your friendly local electrical utility. There are no
similar inexpensive units available for our purposes - YET. 

Additional Reading 

Want to learn electronics in on single volume? The absolutely best book I have ever
seen on electronic circuits (and I own plenty) is 'The Art Of Electronics', by Horowitz
and Hill, published by the Cambridge (UK) University Press. About $60 and worth every
penny. Circuits, signals, device, and applications, all with a purely practical
orientation. There is a little bit of math, but mostly practical, useful information. 

Next Time 

Is there something that you don't understand that you would like to see explained?
Please email me at <a href="#skranish(at)world.std.com">skranish(at)world.std.com</a>


<hr>

Cell Impedance: The Spec You Need to Know, But No One Tells You


A Nickel-Cadmium (Nicad) cell is not a perfect voltage source. No battery or power
supply is. Most textbooks will describe the electrical model of a battery or power supply
as an ideal voltage source in series with a resistor. The resistance models the output
impedance of the source. The impedance of a nicad cell is actually very
complicated, and cannot be accurately modeled over the entire discharge range with such a
simple model. But the model persists, so let's take a look at some published
specifications, and what they do and do not mean. 

The output impedance causes the output voltage of the cell to drop as current is drawn
from the cell. As more current is drawn, the voltage drops more. If enough current is
drawn to cause the cell to heat up, the cell impedance will increase. The cell impedance
also increases as the cell discharges, and this poorly documented characteristic is what
make some cells better than others. 

I have a set of Sanyo Cadnica catalogs from 1992 and 1994. They are not exactly up to
date, but they do contain some interesting information. Sanyo specifies the cell impedance
at 50% discharged, and a 1000Hz frequency. This means they are measuring the cell
impedance with something that resembles a pulse width modulated signal at 1000Hz, when the
cell is about half discharged. In addition to the single number, they also provide
a graph of cell voltage over time for high and low discharge currents. The highest
current they show on the graph is always 8 times the cell capacity, which is low for our
purposes (for a 600mAH cell this is only 4.8 Amps) but it is nonetheless useful
information. For some cell types, the discharge voltage curve is very flat (constant
voltage) for most of the discharge time, but for some cells it is more of a gradual
decline in voltage as the cell discharges. 

The following table lists some common cell types, the published cell impedance (in
milliOhms), and a description of the discharge voltage curve. Note that for smaller cells,
the tested current is lower, so the description of the voltage curve may be deceptive. 

Sanyo Cell Impedances (as published)

<ul>
<li>Cell                        
Impedance               Discharge Curve </li>
<li>N-150N                   27mOhm
                 slightly declining at
about 1.18V </li>
<li>N-225AE                 20mOhm   
              almost flat at 1.15V </li>
<li>N-270AA                15mOhm    
             slightly declining at about 1.15V </li>
<li>N-500AR                 9.0mOhm  
             flat at 1.2 V </li>
<li>N-600AA                12.0mOhm  
            slightly declining at about 1.15V </li>
<li>KR-600AE               8.5mOhm    
           sharply declining </li>
<li>N-650SC                  6.0mOhm  
             slightly declining at about 1.15V </li>
<li>N-1000SCR             4.5mOhm       
        almost flat at 1.18V </li>
<li>KR-1200AE             7.6mOhm      
         fairly sharply declining </li>
<li>N-1400SCR              4.0mOhm      
        slightly declining at about 1.15V </li>
<li>N-1700SCR              4.0mOhm      
        slightly declining at about 1.18V </li>
</ul>

In selecting a cell for a particular application, you need to consider not
just the single number for cell impedance, but also how the cell voltage changes
as it discharges. 

Perhaps the best example from the above list is to compare the N-500AR to the KR-600AE.
Both cells are the same size, and both weigh 19 grams. In theory, the KR-600AE has 20%
more capacity, which is probably fine if you are going to run primarily at low
throttle. The voltage curve for the N-500AR is much flatter, so the
available voltage stays almost constant until the battery is almost discharged. This
should be much more suitable for pylon racing, where the motor is run at fully throttle
for the entire flight. Even if there is less total capacity, there is probably more usable
capacity for a full throttle application. 

I will try to provide a detailed comparison in a future column. Measuring cell
impedance is not simple, although there is some equipment made for the RC car market that
claims to make the measurement. 

By the way, SR Batteries does not publish data of this sort, so I cannot comment
on their cells until I get some instrumentation built up and measure it myself. 

<hr>

Visibility of Model Airplanes 

This past summer I did a lot of flying with my NSP Neon, a Speed 400 class electric
sailplane. I quickly discovered three things: the Neon can go up very quickly in
a thermal, it can be very hard to see at a distance, and my eyesight (which was never
great) is getting worse. (Actually I discovered a fourth thing, a little secret:
Speed 400 sailplanes have an outrageous fun-to-hassle ratio). 

Although I always plan my color schemes with visibility in mind, I have made some
adjustments to the Neon since I finished building it. The basic color scheme is
transparent yellow all over, with a 6" wide, 45 degree dark blue stripe on the bottom
of each wing, and fluorescent orange top tips. The tail also has fluorescent orange top
tips, and dark blue stripes on the bottom of the stab. 

The usual recommendation for sailplane color schemes it to make the top of the wing
light, and the bottom dark. The wide stripe on the bottom provides enough dark area, while
also providing some orientation information. At a distance, everything turns grey, anyway.
I did find, much to my surprise, that the plane was harder to see on clear days than on
cloudy days. Contrast against a clear blue sky was much worse than against clouds. 

The fluorescent tips are VERY visible on cloudy days, but only at moderate distances.
The covering is responding to ultraviolet light (remember 'black' lights? Same thing!) so
it is much brighter on cloudy days when it is not drowned out by visible light. I highly
recommend the use of fluorescent colors for trim, although they are generally too heavy to
cover an entire plane. I have been using Oracover/Ultracote, and I am very pleased with
the result. I have heard comments that the fluorescent versions of MonoKote and some other
materials may not stand up well in the sun. My typical flying session is only an hour or
so, and I live in New England - so someone in Arizona, who has the time to spend all
day at the field, may have different results. 

When the plane was high in a thermal, and I had to spiral it down to keep it in sight,
I found that on sunny days the sun would reflect off the shiny (Oracover) covering, with
the result being flashes of light as the plane spiraled down. Sometimes I think those
flashes were ALL I could see. This is a really good reason NOT to use tissue as a
covering, and perhaps an argument against Micafilm, too, because the surface is not
adequately reflective. 

Anyway, I wanted to add something to improve visibility, especially on sunny days. Phil
Pearce of Tempe, AZ sells two types of reflective metallic tape to improve visibility of
models. I ordered a packet of 'SkyShine', which is a multifaceted metallic film. Phil
suggests putting it on vertical surfaces and wing tops, so I put a stripe on the vertical
stab (fin + rudder), and stripes on each wing tip, just inboard of the fluorescent
trim. The stripe on the fin is perhaps 4 square inches total, and the stripes on the
wing tips are about 6 square inches total. I was only occasionally able to get a 'flash'
from the fin, and have almost never seen a flash from the wing tips. I only got a flash
when I was between the model and the sun, and the SkyShine was oriented almost vertically.
I was rather disappointed, so I called Phil, and he told me that SkyShine should be used
in pieces of at least 9 square inches, and 12 square inches is better. Also, the best
place to use it is on vertical areas such as fuselage sides. The Neon does not have a lot
of fuselage side area, so this material may be better suited to old timers with large
fuselage sides. 

I then ordered some 'SkySheen', a holographic type tape which comes in 1" and
2" wide strips. This is one of those materials that has no obvious color, because the
color changes as the light angle changes. It is recommended for leading edges, and that is
where I put it. I wrapped 1" wide strips about 12" long around the inner
panel leading edges. On sunny days, there is a distinct, multicolored flash every time the
leading edge faces the sun. The exact angle does not seem to matter. I have not had the
Neon back into the cloud base since I installed the SkySheen, but I think it will be
a help for keeping the model in sight on sunny days, which is when I seem to need to
most help. I installed the tape with about 50% of the width on top and 50% on the bottom,
and I think that most of the tape on the top is wasted. It should be wrapped around the
leading edge, but I will put more of it on the bottom in the future. 

Recommended if you need to keep your plane in sight! 

Phil Pearce, 111 East Geneva, Tempe AZ 85282 USA   (602)-966-6384  
 102165,2325(at)Compuserve.com 

Also available from New Creations R/C , PO Box 496, Willis TX 77378 USA
 (409)-856-4630 

<hr>

Measuring and Keeping Track of Battery Capacity 

I am trying to keep track of the capacity of my flight packs, so I made up a chart
to track charge and discharge capacity. Using my handy dandy wordprocessor, I made up a
table with the following columns:

<ul>
<li>Date </li>
<li>Pack Identifier </li>
<li>Charge Current </li>
<li>Charge Peak Voltage </li>
<li>Pack Usage (flight) </li>
<li>Delay before discharge </li>
<li>Discharge current  (for complete discharge) </li>
<li>Discharge Cutoff Voltage </li>
<li>Discharge Current </li>
</ul>

I have a Robbe Infinity charger, so I can program charge and discharge current, along
with the discharge cutoff voltage. After some flights, I fully discharge the pack so that
I have some idea about the residual capacity after cutoff. I occasionally cycle packs so
that I have a full measurement of charge and discharge capacity. I do not fill in the
table for every cycle of the pack, but even occasional use allows me to track the pack
capacity over time. It is also useful to have this information to compare packs that
should have identical characteristics. 

<hr>

The Reader's Survey 

I received very few responses to the December column. Please feel free to email me and
let me know if the material is interesting to you. 

<hr>

Tool of the Month 

Several people suggested the Dremel 'Mini Mite Rotary Tool Model 750'. This is a small,
nicad powered grinding tool and drill that uses the standard Dremel bits. It is a
relatively low torque tool, so it will not work well on metal or plywood, but it is great
for grinding away soft material such as foam or balsa. Actually, it is better on soft
surfaces than the AC powered tool, because it works slower and is easier to control. About
$35 from Tower. 

If you do not currently have an AC Dremel tool or equivalent, this may be all you need.
If you already have an AC tool, this may be useful for serious field surgery, but
otherwise offers no real advantage over a variable speed AC tool. I won mine in a raffle. 

If you would like to recommend a 'Tool of the Month', please feel free to contact me at
<a href="http://rcgroups.com/shared/nospam.php?u=skranish&d=world.std.com">skranish(at)world.std.com</a> 

<hr>

<h2>Source of the Month </h2>

Rather than a 'Source', I am going to write about two print magazines that you should
be aware of. 

'Sailplane & Electric Modeler' is the latest incarnation of  'Sailplane
Modeler', with the addition of more material about electric planes. It includes a column
by Aveox's Matt Orme, and a LOT of reviews. Not a lot of color, but very nicely done, if
you are willing to ignore the name misspellings. The ads are very well focused on the
intended audience, and it is interesting to see who is and is not there.
There is a lot of coverage of contests and fly-ins. See
http://sailplanemodeler.com, call (509)-627-0456, or write Sailplane & Electric
Modeler, PO Box 4267, W. Richland WA 99353 USA. 

'Aviation Modeler International' is a fairly new magazine from the UK,
covering all aspects of model airplanes: free flight, control line, indoor, outdoor,
electric, slimer, sailplane, and RC. There is a lot of beautiful color, and each issue
comes with several free plans, which are printed on newsprint and folded. A one year
subscription will give you a lifetime of projects to do! There is coverage of
a variety of international events. In the UK, contact Model Activity Press Ltd, 22
Star Street, Ware Hertfordshire, SG12 7AA. In the US, contact Wise Owl Worldwide
Publications, 4314 West 238th Street, Torrance CA 90505-4509 USA. I have been buying my
copies at Barnes & Noble. 

I recommend both magazines, but then I go through a large pile of magazines every
month, so maybe I just like to read model magazines! 

If you would like to recommend a 'Source of the Month', please feel free to contact me
at <a href="http://rcgroups.com/shared/nospam.php?u=skranish&d=world.std.com">skranish(at)world.std.com</a> 

<hr>

<h2>COPYRIGHT </h2>

This document is copyrighted (c) 1997 by Steven Kranish, and may not be copied or used
in other forms of publication (electronic or paper) without written permission from the
author. I will probably grant permission, but I would like to know about it, so go ahead
and ask. 

<hr>

<h2>CONTACTS </h2>

If you have any questions, please feel free to contact me at <a href="#skranish(at)world.std.com">skranish(at)world.std.com</a>