SCAT Electronic News 29 April 2001 issue 574
- Details
- Category: Archive 2001
- Hits: 1178
SCAT Electronic News 29 April 2001 issue 574
Table of Contents
=================
Rolled Tailbooms - Blackam
digital servo Van Wallene
Finally arrived in the US! - Limberger
Muncie paving FFAMS Meet - Campbell
True Type Font - Hensley
Bill Watson's Glider for Sale - Ivers
Balsa booms - Joyner
Rube Goldgerg Wing twist solutions - Skykieng
Rolled Tailbooms
=================
Author : This email address is being protected from spambots. You need JavaScript enabled to view it.
I built a lot of rolled balsa booms over the years up to 1993. After that
I used Al/CF/Al (either my own or commercial booms), wich are more
durable. I reckon I got balsa booms to a good stage by the time I stopped.
I used a re-tapered short pool cue as a form. Didn't like the original
taper (too large a diameter at the little end) so I used a (hand) plane
(not an air-plane!) to re-taper it. A good plane is a very accurate
wood-working instrument. I fine-sanded the boom by chucking the big end
into a lathe and running at lowish speed while applying 2-300 grit paper.
My desired taper for the form was about 29mm dia tapering to about 8mm
dia over 800mm length. In the finished boom it saves a lot of weight and
is stronger if you keep the small end of the boom to 9 or 10mm outside
dia or less.
The balsa blank was cut from 5 lb/cu ft, A grain (back-cut) 1/16 sheet.
Timber this light needs to have a defined grain or it will be too easily
broken (cheesy) across the grain. The blank was cut 100mm wide tapering
to 32mm wide over 770mm length, this blank typically weighs 6.5g. In
order to get a straight finished boom it's useful to cut the taper
equally on both edges of the blank (I learned this in many years of
building and flying F1D). The rear half of the blank is tapered in
thickness to 0.6mm. Use a flat sanding block with 100-120 grit aluminium
oxide paper, sand essentially across the grain and don't use much
pressure. After sanding the blank typically weighs 5.2g.
The blank is soaked in hot water (DON'T use ammonia, it damages the cell
structure and reduces the timber strength) and wrapped onto the form with
a crepe bandage. The blank edges will overlap slightly, try to keep the
overlap straight along the form. The boom form is put into an oven (no,
it doesn't completely fit but it works anyway) at about 120 deg Celsius
for 20-30 minutes. This baking process is important; it sets the blank
really well and more importantly reduces the weight by reducing the
moisture content in the timber quite dramatically. After baking the
bandage is unwrapped and a sharp blade run through the two overlapping
layers of balsa. The boom is removed from the form. Weight at this stage
is typically 4.9g.
Next, a release film is wrapped onto the boom form (I used lightweight
clear mylar film as used for tail covering material). This is most easily
spiral wrapped. After this a coating of wax release agent can be applied
(it helps!).
Two pieces of 3/4oz glass cloth are cut about the same size as the
original balsa blank or perhaps a little wider. Grain is +/- 90 degrees.
One piece is laid out on a sheet of heavy plastic or mylar and wetted out
with your favorite laminating epoxy (I use Ciba LY5052). It's essential
to wet the glass well but it's even more essential to remove most of the
epoxy because light glass cloth needs very little epoxy for the bond and
epoxy is HEAVY. Use paper towels to soak up most of the epoxy.
Lay the boom form down along one edge of the wetted glass and roll the
glass onto the form.
Replace the balsa blank, making sure the seam is straight and strap it in
place with mylar tape (preferably) trying not to crush the timber. I
usually 'cook' this in the oven (100 deg C for 20 minutes) to speed up
the epoxy hardening, but it's not essential.
After the epoxy has set, unwrap the boom and lightly sand any
imperfections away. Slightly crushed areas can be dampened with water to
swell the timber back out, then dried with a hairdryer.
On my later booms I applied some longitudinal CF for stiffness. These
were approx 3k tows (1/2 of a standard 6k tow). I applied 4 to the boom,
each (logically) at 90 degree points around the boom. It's really
essential that these are applied with the same tension as each other
(otherwise they'll cause a bend in the boom) and to achieve this I did
the following:
Cut each 1/2 tow to perhaps 100mm longer than the boom. Mark the start
and finish points for each tow at the two ends of the boom. Attach one
tow (with thin CA) at one end of the boom (easiest to use the small end).
Attach a weight of, say, 50-150 g (not critical) to the other end of the
tow. Hold the boom form vertically (small end up) and arrange so the tow
is hanging and laying against the form and aligned with the correct mark
on the large end. Spot glue the tow to the large end with CA.
Repeat for the other 3 CF tows. Finally, apply laminating epoxy to the 4
tows (dab it on), wetting them thoroughly and dabbing up any excess with
paper towels.
Now, wet out the 2nd layer of glass cloth as before and roll it onto the
boom form. Wrap over it with mylar tape and leave to harden (or cook it
if you like). Once the epoxy is properly hardened (and, if cooked,
completely cooled) the boom can be unwrapped and removed from the form
(it will probably take the underneath mylar with it, which can be dug out
fairly easily).
Finished boom weight was normally 9-11g (dependant on the amount of
epoxy). Add a fin at 1.2g, joiner at 2g, tail mount and fittings of 1g
and the all-up weight can be as low as 13.2g which compares pretty well
with a modern Al/CF/Al boom.
Incidentally, I tried many combinations and permutations to arrive at
this structure, including various tissue/dope/kevlar combinations. I
won't go into them all, but the epoxy assembled glass/balsa/glass far
outweighs any other system (at least those based around a balsa core) for
strength/stiffness/lightness and durability. Primarily it's because you
can use a lighter balsa core with the glass/epoxy skins.
Richard Blackam
digital servo
=============
From:This email address is being protected from spambots. You need JavaScript enabled to view it.
Hi Roger,
The whole story is rather complex, in fact it is enough to write an article
about. Yesterday I measured the regulating characteristic of the new
Graupner servo. I will post the graph on SCAT a.s.a.p.
Now to your questions:
> What is digital about the digital servo ? is it just the technology that is
> used in servo ? or does it include the communication between the R/C Rx and
> the servo ?
The digital servo is supplied with pulses from a standard remote control (or
timer), so no change there. So what's different??
The incoming pulse width (time) is measured by a microcontroller and converted
to a digital value that represents the 'desired' servo position. Lets name
this value A.
The actual position of the servo disc is measured by AD converting the
potmeter voltage. Lets name this value B.
So inside the microcontroller, the error value A-B is used to supply the servo
motor with pulses. This pulse frequency is a lot higher (about 400 Hz) compared
to old type servos (50Hz).
The duty cycle of the pulses determines the power applied.
The Graupner also measures the current going through the motor to give an
indication of the load applied to the servo. The higher the load, the more
power is applied to the motor. This gives the high 'holding torque'.
All in all the regulating characteristics are a lot better than analog
servos.
But there are more pro's:
-No more twitching when switching servo on
-Servo operates constantly independent of temperature, analog servos slow
down when it gets below 0 degrees C.
-No temperature drifting
-Dead angle is less than 0.4 degrees compared to 1.5 degrees for analog
servos
-Higher holding torque
-More accurate and powerfull for small movements, so a smaller servo can be
used.
Some con's
Slower
More power consumption
Not yet available with ball bearing (Graupner)
A lot more expensive (factor 3 to 5)
All in all, this servo will be THE solution to currently known servo
problems.
> I was considering making a board that would replace the one in the servo
> and give didgital communication but this would be non standard . Is this
> what Torleif Jensen did ?
Yes and no. The servo is stripped from its electronics (except potmeter).
All regulating (AD conversion, error feedback, control) is done by the timer
microcontroller. Left and right steering is done by a classical H-bridge (2 N
channel and 2 P channel FETS) which are on the timer circuit board. I was
planning to put these inside the servo. Taken from the Torleif manual, there
are 5 leads coming from the servo (plus, GND, left, right, potmeter) so no
measuring of servo load which would require a 6th lead.
I will try to establish a value of the servo load by taking the derivative of
the potmeter voltage vs. time. The speed of the servo disc gives an indication
of the loads applied. For instance, a blocked servo has
zero speed (max load).
At zero speed there is no risk of overshoot (instability) so more power can be
applied to the motor. If there is no load, the speed of the servo is maximum
and the power applied to the motor should be less (lower duty cycle) to avoid
overshooting the desired position (instability).
Maybe you can post your question and my answer on SCAT. It might trigger
Torleif to reply as well.
I will send the curve I made of the Graupner servo regulating characteristic
tonight (error signal versus power supplied to motor) in PDF format.
Regards, Allard
Finally arrived in the US!
==========================
Sender : This email address is being protected from spambots. You need JavaScript enabled to view it.
Hi!
Well, it seems i finally made it to the US. After endless months of
struggling with immigration i finally arrived in the US. My wife and I
no live in the Los Angeles area and i hope i will soon be able to give
my gliders some fresh "US Air". I am also looking forward to continue my
work on the ABAS software system and hope that Allard and I will have
the next generation system "ready for prime time in october" (depending
on my free time...).
In case someone would like to contact me about ABAS or whatever, feel
free to use either my work numbers (at the bottom) or my home phone:
310-234-8933.
ps: please update my SCAT subscription e-mail from
to . thanks!
--
Rene Limberger
Sony Pictures Imageworks
phone: 310.815.4673
fax: 310.840.8100
pager: 310.960.6099
Author : This email address is being protected from spambots. You need JavaScript enabled to view it.
Subject : Digital servo
Hi,
Yesterday the digital servo arrived, and I performed some tests on it.
Servo: Graupner DS361, plastic gears, no ball bearing, torque 31 Ncm,
holding force 78 Ncm, speed 0.15 sec/40 degrees, weight 18
gram
dimensions 28 x 13 x 30 mm
So torque is high and holding force (force required to move the disk
away from ist position) is very high. Speed is slow.
Temperature:
This servo is extremely temperature stable. In fact, cooling it down
more than 20 degrees, it didn't move a bit.
Dead angle is very small. I had a hard time measuring it, so it must be
a few tenths of degrees only.
Small movement behaviour is very good. So when the servo receives a
command for a small movement it does move to the new position, and with
lots of power too. Ideal for accurate small angle steering.
Max. current consumption is about 500 mA with a fully blocked motor. The
maximum power to the servo motor is applied if the error (wanted
position minus actual position) is 10 degrees or more.
No instable switching on/off behaviour (no twitching or erratic disc
movements).
This servo is a bit on the bulky side, but Graupner will release a
smaller one by the end of this month: DS281. Specs: 22x21x11 mm, 9
grams, torque 13 Ncm, holding torque 50 Ncm, speed 0,16 sec/40 degrees,
plastic gears, no ball bearing.
So holding torque is high compared to conventional servos of similar
size, and it should be suitable for stab control. However, it is very
slow!
As with all digital servos, idle current consumption is more than double
of that of a conventional servo. So switching it off after a steering
command is essential for a battery charge to last the whole day.
regards, Allard
[Allard did supply a PDF graph that we will send separately]
[What is the good in having strong holding torque if you have to switch
the servo off ?
Does it go faster on 6 volts ?]
Muncie paving FFAMS Meet
===========================
Author : This email address is being protected from spambots. You need JavaScript enabled to view it.
The paving is complete on the perimiter road on the south end of the
Muncie site. The field is in excellent shape.
The first meet of the year is the FFMAS meet on 05/12-13/01.
Information is available at
This is an Americas Cup meet for the small classes.
Lee Campbell This email address is being protected from spambots. You need JavaScript enabled to view it. 1-765-289-7753
True Type Font
==============
Author : This email address is being protected from spambots. You need JavaScript enabled to view it.
I am trying to find a true type font for "Registration Numbers", or USAF
numbers for model airplanes. Saw an old message from
George A. Bredehoft
Volare Products
FAC model airplane plans
http://www.battlecreek.net/volare
Any help would be appreciated.
Robert Hensley
Engravers Network
Arlington, Texas
This email address is being protected from spambots. You need JavaScript enabled to view it.
www.engraversnetwork.com
Bill Watson's Glider for Sale
==============================
Author : This email address is being protected from spambots. You need JavaScript enabled to view it.
What is Bill's email address? If that glider he has for sale is the small
(A1) class, then I am interested.
Dick Ivers
Balsa booms
===========
Author : This email address is being protected from spambots. You need JavaScript enabled to view it.
Roger:
In the "good old days" before the advent of ready-made carbon-aluminum tail
booms, I tried all manner of things to stiffen up the rolled balsa booms.
Tissue and carbon strips on the inside seemed to work fairly well. However,
the best thing I came up with was to run all the lines down the center of
the boom. (Running the lines along the inside near the top would result in
the boom bending upward when the lines were hooked up.) All that was
required was the addition of a balsa former near the big end of the boom and
thin plastic tubing to lead the lines from the top of the boom to the center
of the former. With the lines running along the boom centerline, bending was
eliminated. (Going to lighter lines also helped.)
I have long felt that the development of the lightweight carbon-aluminum
boom is as much responsible for the "modern" F1B as all the other gadgets
combined. Getting the tail lighter has always been known to be a good thing.
We just didn't know how good a thing until Peter King's December 1995
AeroModeller article quantified it.
My lightest rolled 1/32 inch balsa tailboom still weighed more than a
complete carbon-aluminum tail boom, rudder, fittings, AND STAB!
Why would anybody want to make a balsa boom now?
Louis Joyner
(new address)
183 Civitas St.
Mt. Pleasant, SC
29464
Rube Goldgerg Wing twist solutions
==================================
Author : This email address is being protected from spambots. You need JavaScript enabled to view it.
Wing Twist
> ======================
> Author : This email address is being protected from spambots. You need JavaScript enabled to view it.
(SNIP)
>
> All cambered airfoils have this nose
> down tendency, roughly related to the slope of the mean camber line at the
> trailing edge.
(SNIP)
>
> F1C ships show this characteristic during the climb.
if there were an aileron like affair near each tip,
> it possibly could counter the twisting tendency. Of course the aileron would
> have to be supported structurally and actuated by electronics or mechanics.
(GOOD GRIEF! SNIP)
>
> There is possibly another solution. Simply put, sweep the wings forward.
> This puts the center of lift of each wing panel farther forward relative to
> the root chord thus producing a nose-up tendency to counter the camber effect
.
>
> Bill Bogart
It is crystal clearly embodied within the above text that there is an
absolutely elegant solution to this vexing wing problem that Bill
Bogart has for some reason ignored in his listing of "fixes."
simply put: USE A SYMMETRICAL AIRFOIL
Feeling slighted -- whether deservedly or not -- Skyfolder
>
...................................
Roger Morrell
