SAFE Winch Launching

At 15:53 13 July 2009, bildan wrote:

Look, I didn't start this thread. You don't have a perfect winch and
neither does anybody else. There's a LOT of room for improvements in
every aspect of winch design. Keep an open mind.

If you think that automatic gear changes don't matter, put a tension
logger on yours for a few launches. The results will astonish you.

What gets my dander up is "emotional" opposition to all attempts at
innovation based on the idea that your domestic winch is 'perfect'.

Sorry to ask you this again Bill, but could you please quantify what
problems your alleged 'tension spikes' actually cause? We just don't
seem to get weak link or cable breaks associated with gearchanges. Even if
they do exist, there is enough elasticity in a 1km plus winch cable to
absorb them. As you have been told repeatedly by everyone with a decent
quality automatic gearbox winch, the gear changes are totally smooth and
imperceptible at the glider end.

And can you prove that your mega expensive superwinches will do any better
in rough conditions with a wobbly student pilot at the controls of the
glider?

Derek Copeland

On Jul 13, 10:30*am, Del C wrote:
At 15:53 13 July 2009, bildan wrote:

Look, I didn't start this thread. *You don't have a perfect winch and
neither does anybody else. *There's a LOT of room for improvements in
every aspect of winch design. *Keep an open mind.

If you think that automatic gear changes don't matter, put a tension
logger on yours for a few launches. *The results will astonish you.

What gets my dander up is "emotional" opposition to all attempts at
innovation based on the idea that your domestic winch is 'perfect'.

Sorry to ask you this again Bill, but could you please quantify what
problems your alleged 'tension spikes' actually cause? We just don't
seem to get weak link or cable breaks associated with gearchanges. Even if
they do exist, there is enough elasticity in a 1km plus winch cable to
absorb them. As you have been told repeatedly by everyone with a decent
quality automatic gearbox winch, the gear changes are totally smooth and
imperceptible at the glider end.

And can you prove that your mega expensive superwinches will do any better
in rough conditions with a wobbly student pilot at the controls of the
glider?

Derek Copeland

This is just more rabble rousing. You know the answer, you've seen
the data from winches in your own country. Seeing tension oscillate
by 80% of the tension limits is very scary and it does lead to
problems including wire and weak link breaks. You actually begged me
once not to publish the tension charts or even mention the problem.
(I saved the e-mail in case you deny it.) I have a LOT more tension
charts - really ugly ones. You don't want me to publish them either.

All the current winches are priced within a few thousand dollars of
each other so there are no "mega expensive winches" - except they're
unfortunately all too expensive.

My agenda is simple; help build the best winches in the world without
constant and distinctly unwelcome interference from people with
nothing constructive to add. In all the years you've been posting
about winch construction, you've never added a single new idea. But
you've campaigned loudly and often against any every offered by anyone
else.

I've read every BGA publication on winch launch that's ever been
published. I've read every training BGA syllabus - and I've read ALL
your accident reports. Your accident record is far from the best.
It's in a three ring binder about 4" thick on my bookshelf. Your
stuff is OK but no better than other sources.

So, here's an offer. Stop attacking everything we do and I won't
comment about the Skylaunch. We are going to build, publicly test and
write detailed articles about our winches. I expect no derogatory
commentary from you.

We are not British Subjects and certainly not your students.

At 01:07 14 July 2009, bildan wrote:

This is just more rabble rousing. You know the answer, you've seen
the data from winches in your own country. Seeing tension oscillate
by 80% of the tension limits is very scary and it does lead to
problems including wire and weak link breaks. You actually begged me
once not to publish the tension charts or even mention the problem.
(I saved the e-mail in case you deny it.) I have a LOT more tension
charts - really ugly ones. You don't want me to publish them either.

All the current winches are priced within a few thousand dollars of
each other so there are no "mega expensive winches" - except they're
unfortunately all too expensive.

My agenda is simple; help build the best winches in the world without
constant and distinctly unwelcome interference from people with
nothing constructive to add. In all the years you've been posting
about winch construction, you've never added a single new idea. But
you've campaigned loudly and often against any every offered by anyone
else.

I've read every BGA publication on winch launch that's ever been
published. I've read every training BGA syllabus - and I've read ALL
your accident reports. Your accident record is far from the best.
It's in a three ring binder about 4" thick on my bookshelf. Your
stuff is OK but no better than other sources.

So, here's an offer. Stop attacking everything we do and I won't
comment about the Skylaunch. We are going to build, publicly test and
write detailed articles about our winches. I expect no derogatory
commentary from you.

We are not British Subjects and certainly not your students.

I am struggling to understand what cable tension has to do with anything.
I have been winch launched 10000 times and have driven a winch for many
more launches than that over 50 years and the only thing that concerns me
is the speed of the cable retrieval at the delivery end. Yes I use the bow
in the cable to keep the speed constant but this is a relative thing, not
an absolute. I accelerate quicker than most drivers to keep the ground run
as short as possible and to give the pilot aerodynamic control as quickly
as possible. I also reduce the speed at the top of the launch often
causing the glider to back release.
It really does not matter how it is done but the important thing is that
the control applied by the winch driver is reflected in the response of
the winch to his input at the business end.
I can think of few, if any accidents that have been caused by "tension
spikes" whatever they are, and a lot caused by varying amounts of a lack
of skill.
It is a real shame that people are being denied an economic alternative
launch procedure by ignorance. It does not have to be complicated, in fact
given the use to which it is put the simpler the better.
The winches we have in the UK work, and they work well. Reading some of
the postings about tension control has been interesting in an amusing sort
of way, well actually it's been a hoot.
Learn from the mistakes of the Munster Van Gelder winch, an excellent
winch with many advanced features, most of which are switched off in
normal operation. The servicing overhead is massive and the only people
able to run them, with one exception is the RAF. Yes it gives an
excellent launch, probably the best I have ever had but at a huge cost in
maintenance overheads. Why, because it is too complicated with too many
advanced features even when driven by what are really proffesional winch
drivers.

I am also struggling to understand why so much notice is being taken of
someone who obviously knows so little about the practical operation of
glider winches. As Wayne has said, almost no winching is carried out in
the USA so I would have thought it sensible to consult those who do
operate winches rather than some crackpot with wild ideas and zero
knowledge.

Just exacly where are you making these tensions measurements????

Are you making them at the weak link??

Are you making these measurements at the winch end of things?

I'd really like to know...

At 22:34 14 July 2009, Gary Emerson wrote:
Just exacly where are you making these tensions measurements????

Are you making them at the weak link??

Are you making these measurements at the winch end of things?

I'd really like to know...


That is the question that is puzzling me too. The tension sensed by the
winch will, and should be constantly changing during the launch, measuring
that would be, how can I put it, useless.
The tension at the glider release hook will be relatively constant but
measuring that introduces the problem of the addional weight of the device
that does it and a reliable transmission of data to the ground.

At 00:30 15 July 2009, Don Johnstone wrote:
At 22:34 14 July 2009, Gary Emerson wrote:
Just exacly where are you making these tensions measurements????

Are you making them at the weak link??

Are you making these measurements at the winch end of things?

I'd really like to know...


That is the question that is puzzling me too. The tension sensed by the
winch will, and should be constantly changing during the launch,
measuring
that would be, how can I put it, useless.
The tension at the glider release hook will be relatively constant but
measuring that introduces the problem of the addional weight of the
device
that does it and a reliable transmission of data to the ground.

That is the bit that the proponents of constant tension winch launching
seem to gloss over. How do you measure tension in a long and thrashing
cable that is being wound into a winch at up to 70 knots as the glider
lifts off? There are gadgets called running line tensiometers that are
used in underground cable laying, but they don't work at the sort of
speeds required.

The other approach is to mount a load cell at the glider end and transmit
the data back to the winch in some way, such as a radio link.

Even if you can directly measure tension in the cable, you have to bear in
mind that the glider has to pull against that tension and that changes in
its pitch angle due to pilot inputs or gusts will temporarily increase or
reduce the tension. You would therefore have to damp the system or it
would tend to hunt. I know as a glider pilot and winch driver myself, that
the less you can change the pitch angle of the glider and the throttle
setting of the winch, the steadier the launch becomes. If the pilot is
trying to control the airspeed with the stick at the same time that the
winch is trying to sense and maintain a constant tension, there is a risk
that they will end up chasing each other, especially when separated by a
long, sagging (even Dyneema sags slightly) and slightly elastic cable.

In practice the Skylaunch approach of limiting the power setting as
appropriate to the glider type, headwind component and weak link strength
seems to work well. If the pilot pulls back harder, or the glider runs
into a thermal, it slows the winch engine down by just the right amount.
Easing forward or hitting sink allows the engine to speed up. Since we
replaced Tost winches with Skylaunches, broken weak links have almost
become a thing of the past and the vast majority of the launches are
correctly speeded given correct pilot inputs. The pilot can control the
airspeed by changing the pitch angle.

Derek Copeland

People trying to design tension controlled winches are re-inventing
the wheel. Conventional winches already do that. IF the power setting
is retained constant in conventional winches with a torque converter,
the tension will be constant excluding the inertia factor. Every winch
will have inertia and delay and none will be able to respond
instantaneously to changes in wire load.

A tension meter can be installed in any type of winch, electric,
hydraulic or conventional. There is no difference between the
different types there and power can be controlled by one if desired.
The issue is, does it make sense? Is the approximation calculated by
the British design good enough? Or do we need a real time tension
control winch? Would the inertia i the system even make this possible
or desirable?

I would say that setting the desired tension according to type and
conditions by approximation, ala Skylaunch, and not worrying about the
fluctuations is plenty good enough. The main reason is because
designing a winch to be truly controlled in real time by the actual
and current tension, micro second by microsecond, is probably very
expensive, unproven and would not solve the inertia delay issue of the
winch. Trying to chase the gusts so to speak - how would that work? A
conventional winch type already lets the drum slow down and/or pulls
the engine revs down if the cable tension increases due to a gust or
thermal - thus reducing the tension - this is already a mechanically
built in constant torque effect.

This is by no means perfect, but certainly good enough to give mostly
smooth launches. In gusty conditions, there is no winch in the world,
not even on paper, that can react fast enough to dampen away the
oscillations in cable tension and airspeed, due to the time delay in
reacting to the changes in cable tension.

We just need to figure out how to make a cheap reliable winch and find
places to use them at. At $10-$15 a launch, a glider club or operator
would quickly make a nice income, like European clubs do. The real
cost of a winch launch is something like $3-5 per launch including
maintenance, reserves and replacements. 30 launches a week x $10
profit = $1,200 extra a month. That's one very nice glider upgrade
every 5 years. This is one of the main reasons why European clubs have
such nice equipment.
Tom

At 04:19 15 July 2009, tommytoyz wrote:
We just need to figure out how to make a cheap reliable winch and find
places to use them at. At $10-$15 a launch, a glider club or operator
would quickly make a nice income, like European clubs do. The real
cost of a winch launch is something like $3-5 per launch including
maintenance, reserves and replacements. 30 launches a week x $10
profit = $1,200 extra a month. That's one very nice glider upgrade
every 5 years. This is one of the main reasons why European clubs have
such nice equipment.

There are winches already in operation in the US, there are groups who
want to operate them, and there are groups trying to design and
manufacture them. I think we need all of the above.

I'm part of a wanna-be operating group, with limited funds. That pretty
much means either finding a rebuildable Gehrlein, importing an inexpensive
used winch from Europe, or homebrewing a new winch (based on the UK spec).
I prefer the latter, as I think there is still plenty of room for
optimization, both in fabrication and operation. While I respect those
who are trying to push the envelope with advance power systems and launch
automation, a simple single drum winch based on automotive components will
be good enough for our local group get started.

If you aren't a member of the Yahoo Winch Design group, please consider
joining. While we have our share of spats, there is progress being made
towards coming up with component designs for a cheap reliable
non-automated winch...

Marc

On Tue, 14 Jul 2009 21:19:50 -0700 (PDT), tommytoyz
wrote:


The real
cost of a winch launch is something like $3-5 per launch including
maintenance, reserves and replacements. 30 launches a week x $10
profit = $1,200 extra a month. That's one very nice glider upgrade
every 5 years. This is one of the main reasons why European clubs have
such nice equipment.


Unfortunately this is not the reason...

There might be a few clubs in Europe that charge the prices you
mentioned, but all the clubs I know charge between 2 and 5 Euro,
sufficient only to cover the costs.

Cheers
Andreas

Don Johnstone wrote:
At 22:34 14 July 2009, Gary Emerson wrote:
Just exacly where are you making these tensions measurements????

Are you making them at the weak link??

Are you making these measurements at the winch end of things?

I'd really like to know...


That is the question that is puzzling me too. The tension sensed by the
winch will, and should be constantly changing during the launch, measuring
that would be, how can I put it, useless.
The tension at the glider release hook will be relatively constant but
measuring that introduces the problem of the addional weight of the device
that does it and a reliable transmission of data to the ground.

Which is why I suggested we think about metrics. If there were not big
changes in tension at the Winch end you would have a REAL problem. Go
check it - on our 2km run the glider has not moved before the winch is
1-1.5s into power up - until then there is a lot of tension going into
the cable at the winch and zero happening at the glider end.

We have had a ham fisted learner winch driver snap the cable neatly
(1880KN breaking strain cable) at the drum without even moving the chute
at the launch point.

So - The place it has to be reasonably gentle (smooth curve) is at the
glider hook.

Now we get into a whole new debate about the relative advantage of
UHMDPE (Spectra / Dyneema etc.) and steel cable.

Steel has much more elasticity and inertia to absorb those transient
loads. The spring steel wire acts like a spring storing energy and
smoothing loads. Inconveniently this means that when it breaks under
tension you get a backlash that can result in big overwinds and injuries.

The plastic rope is - counter-intutively given the name, far less
elastic and very light. Result is that winches so equipped may need to
have smoother delivery. And (I am not a good enough engineer here) it is
possible that the tension at the winch on these is similar to that
experienced by the aircraft hook. Would depend on hysteresis and all
sorts of other things that are probably best measured rather than
calculated / guessed.

Any one with facts?