Not satisfied with Windlass operation

[Muaddib1116]

Originally Posted by markxengineerin

Is your chain stopper acting on the chain, or on the gypsy wheel?
On my chain one, the chain can twist or enter the stopper area at a bad angle, preventing it from engaging very easily even when I'm helping it. And it's not very heavy to the point where gravity is a dominant force vs. a rushing chain.
My gypsy one, being horizontal, needs a spring, which I will add as a temporary improvement.

My chain stopper is a Lewmar stopper. It is bigger than yours, partly since my chain is 10mm, which is a larger chain size. It seems to do ok with grabbing the chain, even if there's twist, and the "gate" is heavy enough to work by gravity. The thing is actually hard to keep open when lowering the anchor. If the chain jumps too much, it can flip the gate back over and lock. I'm planning to add a bungee to keep it open when getting ready to anchor.

Originally Posted by markxengineerin

I don't think the motor needs to be very strong at all- just the same power as most are using currently, or less, as in theory it only needs enough torque to raise the anchor and chain against the force of gravity. Any extra breaking free force would come from wave action or the boat's engine power + inertia, with the stopper engaged.

This was more a response entertaining your fantasy windlass, which could pull the boat forward against 40kt winds and swell and yank the anchor out of the bottom all by itself.

[markxengineerin]

I am thinking to replace the cone clutch with something like this, in the ideal windlass:
https://www.bostongear.com/products/...ard-cgc-series

Adjustable detent clutches provide rotational timing and torque control that is more accurate than friction type products.
Fully automatic reset
Hardened components rated for thousands of overloads

Controlling the coefficient of friction is difficult- using springs seems better, and sounds similar in principle to how a power drill clutch works. Those have always worked well for me.

Thinking more about how to add shock absorption, the similar application that comes to mind is the automotive clutch disk:

The way clutch plates use springs could be built into a gypsy wheel, but I'm not sure a high enough spring rate or enough travel to compete with the traditional nylon snubber line could be achieved within the space available. It might have to be a stack of discs rather than just 1. A sample calculation:
Nylon 3 strand, 1/2", breaking strength around 6,000 pounds (I saw a variety advertised, centered around here). "Nylon will stretch by about 20% at 50% of its breakload." So that means a 6' snubber from the below Practical Sailor test can stretch 1.2 feet at 3,000 pounds.

At the 1500 or so pounds expected, it's stretching ~0.6 feet- that's a lot of travel, and the bare minimum of what I'd want if going through the trouble to make something like this. The radius of the gypsy wheel is about 2 inches, so that's about 0.6 wheel rotations desired from zero load to peak load. The desired spring rate = 4.2 pounds (as applied to gypsy radius) per degree, or 0.7 foot pounds per degree, if I got the math right. I will compare this to available springs and the space available another time.

I didn't mean to suggest anywhere that I want to pull the boat against 40 knot winds- of course that's unrealistic. But it's not realistic to want to weigh anchor in those conditions. From Nigel Calder's book:

Although windlasses are not designed to handle snubbing loads, at times, it is inevitable that they will.

So, I think it's insane that they're not designed to handle them, when it isn't going to be much extra effort on a $4,000+ piece of hardware.

Destroying my windlass was the first thing I did as a new boat owner. I wasn't bothering with a snubber or to even engage the chain lock- all loads were going right to the worm gears. My goal is for that level of ignorance to be ok- for improved convenience and improved safety.

[markxengineerin]

It seems to do ok with grabbing the chain, even if there's twist, and the "gate" is heavy enough to work by gravity. The thing is actually hard to keep open when lowering the anchor.

Good to know, if the gypsy wheel spring idea has unforeseen issues, I will try to improve the other stopper by making the bow roller a better fit to the chain, and probably adding some weight to the pawl- or another spring.

[markxengineerin]

I would look at the look at the vertical alignment of the bow anchor lock. You may want to raise it a tad, a thicker spacer underneath.

Noted, I can see how that might help.

[Muaddib1116]

It's an interesting thought experiment to think of adding a shock absorbing spring to a gypsy to reduce shock loads on the gearbox. I think your clutch issues are probably more of an issue with your windlass. The cone clutches seem to work ok for most people. To be clear though, your rotational spring gypsy would still not be appropriate to act as the anchor snubber full time.

Taking the load off the windlass is one effect of an anchor snubber, but it is not the primary purpose. Anchor snubbers are to reduce shock loads on boat hardware as well as on the anchor, to prevent the anchor from jerking out of the bottom. An equivalent 0.6ft absorbtion is not enough for this. A 6ft nylon snubber is far too short for heavy conditions.

On another note, while your spring gypsy idea, might mitigate/prevent damage to the gearbox by reducing shock loads, that doesn't mean that's the only failure mode of windlasses being improperly loaded.

[markxengineerin]

The cone clutches seem to work ok for most people.

The cones will get their fair test, I'm only anticipating issues based on my previous work with them. The clutch is well greased and cones were replaced a couple years ago, all in good condition. I'd like to remove the uncertainty of "tightening by feel" in a re-design, but this is a small complaint relative to the lack of shock absorption.

The 0.6 feet was a starting point, have to start somewhere and it provided easy numbers to work with. The working load limit of the chain is 3,900 lbs, which means 0.6 feet is enough to prevent the chain from breaking up to 60 knots for the 40' example boat. I haven't analyzed the bow roller, the pulpit, the gypsy wheel, and the pawl, but my guess is the pawl (pawl pivot bolt threads, specifically) is the weakest link in the things that get shock loaded in this concept.

To be clear- the gearbox and anything boat-side of the gypsy wheel will see zero load from shock, as anything greater than what the clutch can transmit will be reacted fully by the pawl.

Even if the integrated snubber has some wind/wave limit, I still think it's a nice feature. Fewer times messing with a snubber = less chances for pinched body parts and more time to focus on more interesting tasks.

[rslifkin]

For setting the windlass clutch, I always test it with the anchor fully retrieved. The clutch should be set to slip before the motor stalls, but still tight enough that it doesn't slip excessively during normal anchor retrieval.

[Muaddib1116]

I just don't think you can build in enough shock absorption to get rid of the need for a proper snubber, especially since the scenarios that inspired all this are extreme wind and wave conditions, which you would absolutely need a good snubber for in order not to drag anchor. Enough absorption to protect the windlass may be possible though.

Also, the pawl on the windlass and the cone clutch aren't really designed for what you want to use them for, thus their deficiency in that area obviously. The pawl is mostly a lock to prevent the gypsy from spinning when you loosen the cone clutch. This is mostly useful for windlasses with a capstan drum. Most of those work by having a gypsy that is engaged to the driveshaft with the cone clutch, and the capstan drum is locked in to the shaft with a spline or key. So you lock the pawl, loosen the clutch, and then you have a powered capstan independent of the gypsy.

The cone clutch is really more of a binary thing than you are thinking. It's an elegant way to engage the gypsy without worrying about clocking to a spline or something, and then to quickly free the gypsy and allow it to freespin and drop the anchor and chain with gravity. While you can tune the clutch tension to allow some slip in some conditions, I don't believe it's real intention was to act as a safety to protect the motor/gearbox. That responsibility lies with the operator.

[markxengineerin]

What you're saying makes sense. I am considering this the temporary prototype and something will probably break, but I'll learn something. Usually something I didn't think of no matter how hard I tried.

I always test it with the anchor fully retrieved. The clutch should be set to slip before the motor stalls,

. Sounds like a good tip, but to stall the motor seems to take a LOT- I can see the whole windlass assembly moving (minutely) relative to the deck at that point. I have never needed approach that level of clutch tightness in my use.

[Muaddib1116]

I look at it more as the clutch prevents the windlass from ripping my deck up.

[rslifkin]

Originally Posted by markxengineerin

What you're saying makes sense. I am considering this the temporary prototype and something will probably break, but I'll learn something. Usually something I didn't think of no matter how hard I tried.

. Sounds like a good tip, but to stall the motor seems to take a LOT- I can see the whole windlass assembly moving (minutely) relative to the deck at that point. I have never needed approach that level of clutch tightness in my use.

That makes me question if your windlass mounting is strong enough. It's entirely possible that the section of deck it's mounted into isn't strong enough without additional reinforcement.

[markxengineerin]

Ha, I'm laughing out loud at that one- the idea of ripping the entire windlass out of the deck. I prefer any other failure over that one. I have a feeling the movement is because I didn't torque the through bolts excessively when I installed the windlass- it's a balsa core deck, unknown how much compression load that can take, and I wasn't trying to find out with the bolts. There is a good sized backing plate, so rip out force is >>>>> force to wiggle it slightly, I think.

[Muaddib1116]

I know my current windlass flexes the deck slightly if I'm not careful. I'm installing a new windlass soon, and I'm planning to heavily reinforce the deck in that area.

[markxengineerin]

Yes, probably normal, unless some builders have better deck cross sections in the windlass area. Pacific Seacraft says they used plywood in highly loaded areas with fittings, the balance being balsa, but I don't actually know if that applies to the windlass or not, or if the installation was original or a later addition.

[Muaddib1116]

If it's a factory installation, I would expect they put plywood under the windlass.