Raymarine St4000 with Trim Tab Problem

[Rustic Charm]

Originally Posted by atoll

no reason why it should not work,as long as the ram is in the middle stroke length,and port/stb sense is correct.

probably want the AP on its lowest response setting,and trim tab at 15 cm to start with

Will do.

You do have personal knowledge of this working though don't you? I mean the picture you drew and the mounting of the auto pilot onto the tiller?

[atoll]

Originally Posted by Rustic Charm

Will do.

You do have personal knowledge of this working though don't you? I mean the picture you drew and the mounting of the auto pilot onto the tiller?

I have a similar set up for my 63 footer that worked well when tested.

we never really used it as the trim tab was a problem when reversing onto a dock,and our hydraulic steering needed disengaging when in use which made the whole system cumbersome as the hydraulic shutoff was in the back of the engine room.

so we ended up using a auto helm 3000 wheel pilot on the helm in the pilot house
.
we did find it useful briefly when our helm pump packed up and had to rely on the pulley driven emergency steering located at the back of the boat,where the tiller pilot was located for the trim tab.

bear in mind our helm is about 30ft away from the rudder,and the hydraulic shutoff is down a set of stairs,in the back of our walk- in engine room.so it proved to be not really practical without a steering position at the aft end of the boat due to the time it took from going from hand steering to trimtab steering.

[Rustic Charm]

Originally Posted by atoll

I have a similar set up for my 63 footer that worked well when tested.

we never really used it as the trim tab was a problem when reversing onto a dock,and our hydraulic steering needed disengaging when in use which made the whole system cumbersome as the hydraulic shutoff was in the back of the engine room.

so we ended up using a auto helm 3000 wheel pilot on the helm in the pilot house
.
we did find it useful briefly when our helm pump packed up and had to rely on the pulley driven emergency steering located at the back of the boat,where the tiller pilot was located for the trim tab.

bear in mind our helm is about 30ft away from the rudder,and the hydraulic shutoff is down a set of stairs,in the back of our walk- in engine room.so it proved to be not really practical without a steering position at the aft end of the boat due to the time it took from going from hand steering to trimtab steering.

gee, your helm is almost the lenght of my boat.

Well, hopefully my welder guy will do the modification quickly and I can check it out on the weekend. The only thing left I Will need to do is to move the 12v plug for the ram, which i'll move to the centre of the cockpit.

[atoll]

Originally Posted by Rustic Charm

gee, your helm is almost the lenght of my boat.

Well, hopefully my welder guy will do the modification quickly and I can check it out on the weekend. The only thing left I Will need to do is to move the 12v plug for the ram, which i'll move to the centre of the cockpit.

[Snowpetrel]

Can't you just make up a triangular plywood plate to bolt onto the existing trim tab tiller to trial thenew setup. Then temporarily extend the cord and test it all. I wouldnt want to pay someone to weld up anything until I was sure it was going to work well.

[Rustic Charm]

Originally Posted by Snowpetrel

Can't you just make up a triangular plywood plate to bolt onto the existing trim tab tiller to trial thenew setup. Then temporarily extend the cord and test it all. I wouldnt want to pay someone to weld up anything until I was sure it was going to work well.

Ok, I can certainly do that. I just need to organise some 'U' bolts to bolt it on without damaging the trim tab tiller.

Your starting to worry me Ben, do you not have confidence that this 'redesign' would work?

[Snowpetrel]

I am picking will work, but without any damping it still might hunt a bit. Thats why you prototype these kinds of things before you build them all fancy.

Im certainly should work a whole heap better than the old setup, which by the sounds of it didnt work at all. But it may be that a system with some positive feedback will work better, giving a smoother more proportional response. Hard to know for sure without playing with a few options first, or copying exactly from a proven design.

This is a simple way to get it functioning. Hopefully it will work well enough, then you can go ahead and make all the permenant modifications, but until you know for sure id make it as cheaply as I could.

Ideally a welder shouldnt charge too much to weld a bit of flat bar onto another peice of flatbar, all of an inch or so of weld, but then he will want to grind it and polish it, and round off all the edges etc if hes a yachtie type of welder...

Im picking he is likely to charge about $100 or more, because there is some sort of rule in australia that says you arent allowed to charge less than $500 for any job on a boat, no matter how small it is. And so he will think he is doing you a big favour by charging such a small amount for a 10 minute job.

[Rustic Charm]

Try again. What do you think?

[Snowpetrel]

Very nice. I'd add a couple more holes in the new trim tab tiller bit, running further out to give you more options to reduce the movement if needed.

Also if you have some strong bungy cord handy that can be helpful to tie off the main tiller to add some extra damping if needed in big seas.

I do have one criticism though, they look like shiny stainless steel u bolts, far too yachtie. You should have used a couple of old rusty ones you salvaged from a rubbish bin someplace for full marks. oh, and also I see evidence that you used a ruler to mark out the holes instead of crude guess work... shame on you!

[Snowpetrel]

Also, I assume the old bit of stainless steel trim tab tiller is not going to foul the ram when the trim tab turns?

[Rustic Charm]

Originally Posted by Snowpetrel

Very nice. I'd add a couple more holes in the new trim tab tiller bit, running further out to give you more options to reduce the movement if needed.

Also if you have some strong bungy cord handy that can be helpful to tie off the main tiller to add some extra damping if needed in big seas.

I do have one criticism though, they look like shiny stainless steel u bolts, far too yachtie. You should have used a couple of old rusty ones you salvaged from a rubbish bin someplace for full marks. oh, and also I see evidence that you used a ruler to mark out the holes instead of crude guess work... shame on you!

Blimy your hard to please and then 'u' bolts are cheap mild steel so they will rust pretty quick.

[Rustic Charm]

Originally Posted by Snowpetrel

Also, I assume the old bit of stainless steel trim tab tiller is not going to foul the ram when the trim tab turns?

Oh damn it it will do too.

It needs to be on a different level doesnt it.

[GILow]

So I am coming in a little late on this, but I would like to contribute something because I have been down this path recently.

What is being proposed should work, but as acknowledged, it may not have the right kind of feedback to prevent hunting.

So I would like to step back a bit, and remark on something about the tiller pilot software. From my own personal experience of reinventing that particular wheel, I discovered that the TP software must work on the assumption that the rudder position has moved with the movement of the tiller pilot. This seems obvious, but consider this, the corollary is that the TP does not think the rudder has moved BACK to the centre until the TP moves it back. With this basic assumption, a whole lot of the parameters of the TP software are defined.

So for the TP software to work optimally, you really want the activation of the trim tab to move the tiller to a point and HOLD IT THERE, until the TP reverses the stroke. Any other system is going to defeat whatever self learning systems are programmed into the TP software, and probably cause it to hunt much more than is good for it.

To do this you need a very strong imperative in the design that will make the rudder want to slavishly track the position of the TP arm.

Jim hinted at it in post #19, and I assume he discovered this principle when designing and building his own trim tab powered auxiliary rudder windvane. You need a system that brings the rudder over and holds it in that position until the "signal" from the vane is cancelled. The simplest way I found of doing this with a trim tab rudder is to place the pivot point of a SECOND longer control arm slightly FORWARD of the pivot point of the rudder, with a more of the second control arm length forward of the pivot point than aft of it. This has two effects. It will reduce the movement of the trim tab in relation to the movement of the TP ram, and it will cause a self-centring of the trim tab as the rudder swings across, because the forward arm of the second lever is quickly moved to a "neutral" position relative to the rudder by the movement of the rudder arm. Also, if the rudder "overshoots", the same mechanism will reverse the trim tab to bring the rudder back into alignment.

A little fiddling with my favourite prototyping tool (Lego) found a nice apparent balance with the pivot point of the second arm approximately half the distance forward of the rudder pivot point as the trim tab pivot point is aft. After that, a 1:4 length ratio of the longer arm about the pivot point also seemed to "feel" about right, but this is clearly one of those cases where you drill half a dozen holes in the second pivot arm and move the attachment point of the TP ram until the whole system responds as you want it to.

What is really clear from the prototype is that this system will utterly and totally self correct. If you have some Lego I suggest you give it a go and see for yourself.

This will allow the TP software to work exactly as though it were connected to the rudder, and make full use of whatever A.I. was programmed into it to improve its response and minimise actuator time.

Matt

[Rustic Charm]

Originally Posted by Rustic Charm

Oh damn it it will do too.

It needs to be on a different level doesnt it.

Ok, to temp fix this, ill remake the trim tab ply frame with the tiller pointing out the back. Then it will be out of the way.

[Rustic Charm]

Originally Posted by GILow

So I am coming in a little late on this, but I would like to contribute something because I have been down this path recently.

What is being proposed should work, but as acknowledged, it may not have the right kind of feedback to prevent hunting.

So I would like to step back a bit, and remark on something about the tiller pilot software. From my own personal experience of reinventing that particular wheel, I discovered that the TP software must work on the assumption that the rudder position has moved with the movement of the tiller pilot. This seems obvious, but consider this, the corollary is that the TP does not think the rudder has moved BACK to the centre until the TP moves it back. With this basic assumption, a whole lot of the parameters of the TP software are defined.

So for the TP software to work optimally, you really want the activation of the trim tab to move the tiller to a point and HOLD IT THERE, until the TP reverses the stroke. Any other system is going to defeat whatever self learning systems are programmed into the TP software, and probably cause it to hunt much more than is good for it.

To do this you need a very strong imperative in the design that will make the rudder want to slavishly track the position of the TP arm.

Jim hinted at it in post #19, and I assume he discovered this principle when designing and building his own trim tab powered auxiliary rudder windvane. You need a system that brings the rudder over and holds it in that position until the "signal" from the vane is cancelled. The simplest way I found of doing this with a trim tab rudder is to place the pivot point of a SECOND longer control arm slightly FORWARD of the pivot point of the rudder, with a more of the second control arm length forward of the pivot point than aft of it. This has two effects. It will reduce the movement of the trim tab in relation to the movement of the TP ram, and it will cause a self-centring of the trim tab as the rudder swings across, because the forward arm of the second lever is quickly moved to a "neutral" position relative to the rudder by the movement of the rudder arm. Also, if the rudder "overshoots", the same mechanism will reverse the trim tab to bring the rudder back into alignment.

A little fiddling with my favourite prototyping tool (Lego) found a nice apparent balance with the pivot point of the second arm approximately half the distance forward of the rudder pivot point as the trim tab pivot point is aft. After that, a 1:4 length ratio of the longer arm about the pivot point also seemed to "feel" about right, but this is clearly one of those cases where you drill half a dozen holes in the second pivot arm and move the attachment point of the TP ram until the whole system responds as you want it to.

What is really clear from the prototype is that this system will utterly and totally self correct. If you have some Lego I suggest you give it a go and see for yourself.

This will allow the TP software to work exactly as though it were connected to the rudder, and make full use of whatever A.I. was programmed into it to improve its response and minimise actuator time.

Matt

I see what you have drawn, and this 'counter' would mean the auto pilot will operate in the same direction as the main tiller. So no need to reverse the control head wires.

But i dont understand how all of this counters it's desire to over reaact is it is now?

Are you suggesting Atoll's will not fix the problem ive been having?