Arm Winch

[Nigel]

I have been wondering about the arm winch offered by Kbits. It is based on the Jumbo sized Hitec digital servo that puts out 19kg of torque.
The question is, is so much torque actually required for an IOM and has anybody ever done some detailed research on the sheetingloads on our boats

Cheers

[awallin]

I have been wondering about the arm winch offered by Kbits. It is based on the Jumbo sized Hitec digital servo that puts out 19kg of torque.
The question is, is so much torque actually required for an IOM and has anybody ever done some detailed research on the sheetingloads on our boats
Cheers

A note on terminology / physics:

The servomanufacturers habit of expressing torque in "kg/cm" is as far as I can see unfortunate (some would say horrible) because the real unit of a torque is ofcourse a force (which can be expressed as the weight of a mass if you really want to) multiplied by a distance...

This means that the actual force pulling on your sheet is the torque of your winch divided by the length of the arm on an arm-winch and divided by the radius of the drum on a drum winch.
This means that in principle you could increase the force available by shortening the arm. However I think with an arm-winch this will lead you nowhere as you then also have move the sheeting points closer to the boom pivots and this has the opposite effect of requiring more sheet force... with a drum winch this works OK because you can increase number of turns the winch travels. (you will lose speed ofcourse...)

To further complicate matters you have to take into account that what is really driving the arm or drum is an electric motor controlled by a servo algorithm. Now an electric motor gives out maximum power and torque when spinning around at some rpm. So in fact I think that the holding torque that is available in this kind of system is less than the maximum output torque of the motor transformed into arm torque via the gears.

Now the real reason I think people are using this winch is that it is very very fast. It is clearly faster than an RMG.
In addition to the Binks brothers Lester had one of these winches in Vancouver and I'm sure he'll respond to this post

Another reason that I heard of in Vancouver was that an arm winch makes it possible to fix the mainsheet and jibsheet at different positions on the arm and thus get a different amount of travel for the main and jib. Wether or not this is class legal I do not know, there was something about this in the latest intrepretation.

Anders

[Nigel]

Well my knowledge of physics is only very basic, as this was one of the subjects I was very glad to give up in school as early as possible but I of course I am aware of the influence of the arm length.
The point I am trying to get to is, is a servo of this size necessary or would something in the range of 12 Kg have enough power at the same given arm length??

The point is, that the Servo the Binks brothers use ways in at >140g while a digital Hitec Servo with 12kg of "torque" ways only 60g. and is even faster!

[Chairman]

would something in the range of 12 Kg have enough power

Hi Nigel

The torque of 21 kg.cm from the 1/4 scale Hitec digital servo is only enough to sheet in on my Italiko when the wind speed is under 4.5 m/sec. If a gust comes in, I have trouble. The smaller servo with 12 kg.cm IMHO won't handle it at all on an IOM at the top of No.1 rig...

[Chairman]

has anybody ever done some detailed research on the sheetingloads on our boats

<CENTER></CENTER>
Photo ©2002 Ken Binks

Here are some rough calculations. Let's have an IOM on the run, right at the top of no.1 rig. Wind speed about 5 m/sec, boat speed about 1.2 m/sec, apparent wind on sails about 3.8 m/sec therefore. (Don't like m/sec, prefer knots? Conversion factor is actually 1.944, but use 2 and that's good enough. 2 knots = 1 m/sec will do fine. Wind is about 10 knots, boat is sailing at around 2.5 knots...)

IOM sail area of no.1 rig is about 0.6 sq.m, the coefficient of drag C<SUB>d</SUB> is around 1.8. (Rho is the density of air, around 1.2 kg/cu.m.) So:
<BLOCKQUOTE><CODE>Drag force = C<SUB>d</SUB> × 0.5 × rho × speed<SUP>2</SUP> × area (N)
= 1.8 × 0.5 × 1.2 × 3.8<SUP>2</SUP> × 0.6 (N)
= 9.35 N</CODE></BLOCKQUOTE>So the force on the sails is about 9 N, about 0.9 kg, about 2.1 lb. Let's have the winch pulling this in. The sheet attachment radius is 80 mm, doubled because we are using a pulley to give a "double" purchase on the sheet, so our torque of 21 kg.cm from the winch yields around 1.3 kg force on the sheet, and we haven't yet allowed for friction losses. So this winch is going to struggle to haul the sails in under these conditions, and my experience is that it will in fact not be able to do it as soon as the wind speed gusts just a bit, up to, say, 6 m/sec.

[Nigel]

Thanks Lester, that's really what I was looking for.

How does this work in practice? Do change the whinch in your boat according to the weather you are expecting??
The Binks brothers both used the arm whinch in Vancouver as far as I know and that event so a fair bit of wind at some times didn't it. How did this work for them?

Cheers

[JohnB]

Thanks Lester, that is a good explaination, but can that formula be used for the forces required for close haul sailing ?

[Chairman]

Do change the winch in your boat according to the weather you are expecting??

Hi Nigel

Er, nope! I run the 1/4 scale winch at all times, not too concerned with its weight since this is very similar to a RMG, which is the winch I used previously and would return to if the arm arrangement disappoints.

The Binks brothers both used the arm whinch in Vancouver as far as I know and that event so a fair bit of wind at some times didn't it. How did this work for them?

Ken and Trevor are way better sailors than I am, and their installation isn't quite the same as mine... But you need to get this from Ken. E-mail him at K.Bits, his site URL is on my "Links" page.

[Chairman]

can that formula be used for the forces required for close haul sailing ?

Hi JohnB

The formula is good for obtaining the amount of force exerted by anything with some area A moving at a given speed S through some medium of a specified density rho. Close-hauled, looking to calculate the lift force of the sails, the coefficient of lift C<SUB>l</SUB> might be, say, 1.0:
<BLOCKQUOTE><CODE><PRE>Lift force = C<SUB>l</SUB> × 0.5 × rho × speed<SUP>2</SUP> × area (N)<BR> = 1.0 × 0.5 × 1.2 × 3.8<SUP>2</SUP> × 0.6 (N)<BR> = 5.2 N</PRE><CODE></BLOCKQUOTE>and then the amount of lift would be around 5.2 N, assuming an apparent wind of 3.8 m/s.

[ReyNewman]

Since I use arm winches in both my IOM and Soling 50, I calculated mainsheet load and jib sheet load for the IOM A rig for various wind speeds. The Hitec 5735MG digital servo theoretically does not run out of steam till ~18knots wind speed.

I used the following formulas for calculating approximate mainsheet and jibsheet loads(you can cut and paste them into an Excel spreadsheet and substitute the variables for cells containing the proper parameters):

Main sheet (courtesy of Sail magazine):
load(lbs)=((e^2)*(p^2)x0,00431*(v^2))/((p^2+e^2)^.5)*(e-x)
Where
e: foot length (in feet)
p: luff length (in feet)
v: wind speed (in knots)
x: distance of sheet attachment point from end of boom

Jib sheet (courtesy of Harken, http://www.harken.com)
load(lbs)=((i*j)/2)*v^2*0.00431
Where
i= jib leech length (in feet)
j= jib foot length (in feet)
v= wind speed (in knots)

Rey

[ReyNewman]

PS: I have used the HS5735 with the A rig without any problems up to 13kts (didn't have a B rig.) The holding power of the 5735 is phenomenal - better than the 815BB, but it is more power hungry.

If you plan on using a 5735, use the programmer to slow it down - it is way too fast.

Rey

[Chairman]

If you plan on using a 5735, use the programmer to slow it down - it is way too fast.

Hi Rey

It certainly is fast. Ken Binks tells me that is one of its very strong selling points. It allows you to execute a "flick gybe" without touching the rudder. I've won more than one club regatta recently by being able to gybe faster than the guy in front...

[Chairman]

Was sailing with Graham Bantock and Ken Binks at the Brentwood club's "Icicle" event on Sunday (they finished in that order, Peter Stollery third, yours truly fourth). Graham said that the new SAILSetc catalogue and Web site (in about 4 weeks) will feature an arm for this winch with adjustable attachment points etc, and will also feature a purpose-designed deck moulding for mounting the winch (and rudder servo, pot holder, etc) just aft of the fin box.

Ken explained to me that the digital technology may need a different approach to working the sheets. In particular, he was emphatic that I really did need to take care if I wanted to use the winch to "sheet vang" -- that is, pull the main boom down at close-hauled -- and he could not recommend it. The reason is that the digital winch is controlled by a micro-processor, and this little computer is programmed to increase power to the motor if it finds the arm isn't moving to its commanded position. And it increases power and increases power until something gives. The older analogue winches can't and don't do this, they are set up to cut power as the commanded position is approached. So if for any reason your sheeting doesn't allow the winch to take up the commanded position (a sheet snag, say, or "sheet vanging" with an incorrectly-set end-point), an analogue winch will take its time about burning out, while the digital one will do so in somewhat of a hurry...

[Tiga]

Hi Rey,

can you tell me: how long is the arm of your hitec-servo?

thanx
Rainer

[Rob Davis]

Ken explained to me that the digital technology may need a different approach to working the sheets. In particular, he was emphatic that I really did need to take care if I wanted to use the winch to "sheet vang" -- that is, pull the main boom down at close-hauled -- and he could not recommend it. The reason is that the digital winch is controlled by a micro-processor, and this little computer is programmed to increase power to the motor if it finds the arm isn't moving to its commanded position. And it increases power and increases power until something gives. The older analogue winches can't and don't do this, they are set up to cut power as the commanded position is approached. So if for any reason your sheeting doesn't allow the winch to take up the commanded position (a sheet snag, say, or "sheet vanging" with an incorrectly-set end-point), an analogue winch will take its time about burning out, while the digital one will do so in somewhat of a hurry...

This same learning should also be applied to most any digital servo application. If one is used for the rudder servo it's important to insure the linkage is free of any binding as this will also cause the servo to drain the battery.

Rob

[ReyNewman]

Hi Rey,

can you tell me: how long is the arm of your hitec-servo?

thanx
Rainer

The Hitec 5735 comes with an arm that is 1inch long, from center point to end attachment point. I built a 3.75inch long arm, from center point (servo screw) to sheet pivot point, which attaches to the Hitec arm. With a 2:1 purchase, and 90deg of servo travel, I get approx. 13 inches of line travel.

Rey

[Chairman]

Gyula Ferencz has told me about an old trick to increase arm winch torque to around 175% of rated amount! I'll be trying it in a couple of weeks, and here is the spreadsheet: http://www.onemetre.net/Download/Cam/Cam.htm

[awallin]

Gyula Ferencz has told me about an old trick to increase arm winch torque to around 175% of rated amount! I'll be trying it in a couple of weeks, and here is the spreadsheet: http://www.onemetre.net/Download/Cam/Cam.htm

hmm, interesting idea. A bit like the counterweight on a lift...

Lester, If possible, can you measure the current the winch draws when you test this ?
I would assume that with this kind of system the winch works against the spring tension whenever the winch is not at its close halued position and that should lead to a higher battery drain...

[Chairman]

... can you measure the current the winch draws when you test this?

Hi Anders

What I normally do is charge up a battery to a known capacity, go and sail and measure the total time spent with the radio system on, and then measure the remaining capacity of the battery afterwards by discharging it fully. Typically, in light winds, the 1/4 scale Hitec takes about 250 - 300 mA per hour, and in heavy airs up to 500 mA per hour. I expect this arrangement to take about 500 mA all the time, but will certainly test it on the bench first.

I would assume that with this kind of system the winch works against the spring tension whenever the winch is not at its close halued position and that should lead to a higher battery drain...

Yes, if anyone is thinking of trying it with a digital winch, please remember that a digital winch develops much more power than an analogue when at, or close to, its commanded position -- heat! This tensioning arrangement means that the winch must always hold its commanded position (even at close-hauled) by pulsing the motor continuously -- lots of heat! There is a very good reason why the Hitec 1/4 scale has an integral heat sink. For me, the increased current drain isn't too much a problem, but the possibility of overheating the winch would seriously spoil my day!

[Ken Binks]

I have read with interest the discussion relating to the use of an arm winch for IOM sail control. The points discussed do raise some issue's which I feel should be clarified. The application of the Hitec 1/4scale DIGITAL aircraft servo as an arm winch, was instigated by K Bits to provide IOM sail control which was superior in resolution, speed and by arranging the sheeting angles gave an exponential effect for even greater control.

Resolution:

The feedback loop is mechanically tight, because of the 'normal' type of servo mechanics.

The DIGITAL amplifier updates info up to 6 times faster than normal AND with max power.

Speed:

The speed is VERY useful for 'flick' gybing on run, de powering quickly AND re powering in gusty winds. AFTER ALL we have proportional radio control! and can move both winch and rudder very slowly.

Exponential:

By having winch lines almost going over centre (10 deg) when close hauled, the transmitter stick provides a very smooth and very precise response from close hauled to around 1/4 travel. Then the sheeting becomes more linear and response is faster towards the running position.
This effect is the same whether using a basic 2ch system or a computer Tx.

Points to watch:

Max radius 85mm to 2:1 pulley

To increase throw, INCREASE angular change not longer arm.
Typically 120-140deg

Ensure servo arm does not 'hit' any boat structure. Also do not use winch as vang when close hauled.

Have servo programmed for extra throw and switch 'on' failsafe.
Do not use manafacturers modification to reduce water ingress as it will significantly limit the heat sinks effect. (tape over blue HS)

Have it modified to reduce water ingress using silicon sealant.


In December 2002 K Bits catalogue stated this winch was 'world class'. Some prophesy! Trevor Binks used one to become IOM W. Champion in June 2003.