HMB Endeavour tiller and steering question

[shipaholic]

Robin

They are both dated July 1768, the writing is slightly different, one is annotated a copy. They must have made more than one at the same time. Didn't have printers and photocopiers back then and I suppose they would have needed more than one plan, perhaps one to use and one to keep as a record.There may have been several copies made

[dashi]

Something to consider if I've got this right?

 

The steering gear is a combination helm and tiller above decks using block and tackle wound around a drum attached to the steering wheel. The rope would need to remain under a constant tension for this to work right? There is no where apart from the lanyard lashings of the rope to the tillers goose neck to take up the tension from what I can see. You cannot have any blocks on the tiller because of the chimney which I assume is the reason for that gooseneck on the end of the tiller. Now I see a problem with a rigid tiller to rudder post attachment with this configuration because at 21 ft the tiller will follow a large arc governed by the angle of the stern post. This arc of the tiller would affect the rope tension differently as it moves through the arc during steering. When the tiller is mid way between port and starboard then its at its highest and would put the most tension on the drum rope. But at port or starboard the rope would be slack and the steering sloppy or not possible. The rope would be constantly slackening and tightening with every movement of the wheel which is not desirable I'd imagine. I don't know how they get around this problem on the replica unless the lanyard lashing to the gooseneck is a bungy?

 

A sweep with pivoting tiller would be a solution because it can be built to allow the tiller to move relative to the deck and tackle thereby maintaining an even tension throughout steering. When the rope slackens over time then it can be taken up by the lanyard at the gooseneck.

 

Does this make sense or have I got it terribly wrong?

[jbshan]

Dashicat, that's part of the reason for the quadrant, where it is used.  The tiller rope runs along the outside arc of the quadrant keeping the tension more constant as the rudder and tiller swing.  There is also a tackle to take the slack out.

[druxey]

Joel has this right. There was a brilliant analysis of the slack/tight line tension and methods of overcoming this by John Harland some years ago in Mariners Mirror. The sweep was the first of these methods.

[dafi]

It was always my understanding, that the sweep was to keep up the tension of the tiller rope and not in first place to support the tiller - somebody here in the forum put it once in a nice way: How is the tiller suppose to stand the violence of the sea if it can´t support it´s own weight :-)

 

But this is what is irritating me. To correct the difference tension of the tiller rope, the sweep has to be at the curve of the tiller rope, in this case the outer end of the goose neck. So why is it at 2/3 of the tillers length?

 

XXXDAn

Edited March 28, 2016 by dafi

[jbshan]

The below-deck tiller and sweep has the sweep closer to the end of the tiller and small rollers on the rim of the quadrant to reduce friction on the tiller rope.

This would be for larger ships, although Endeavour has headroom enough below, and if there is horizontal room for the tiller to reach it's functional limits on deck, there will be room enough in the cabin.

I think some of the drawings may be for a refit, if not for the voyage and the replica got it wrong, then for afterward.  That gooseneck bothers me a bit, also.  It's only a foot or so long; you wouldn't gain much advantage from the extra length.  Why not simply do away with it and put the ropes on the wooden part as was done on many, even fairly large, vessels, but perhaps not ocean-going.  It doesn't make sense, and if it doesn't make sense, maybe there's something not right going on.

[jbshan]

Joel has this right. There was a brilliant analysis of the slack/tight line tension and methods of overcoming this by John Harland some years ago in Mariners Mirror. The sweep was the first of these methods.

And Harland expanded on this in his 'Capstans and Winches' of about 2004.

[dashi]

The below-deck tiller and sweep has the sweep closer to the end of the tiller and small rollers on the rim of the quadrant to reduce friction on the tiller rope.

This would be for larger ships, although Endeavour has headroom enough below, and if there is horizontal room for the tiller to reach it's functional limits on deck, there will be room enough in the cabin.

I think some of the drawings may be for a refit, if not for the voyage and the replica got it wrong, then for afterward. That gooseneck bothers me a bit, also. It's only a foot or so long; you wouldn't gain much advantage from the extra length. Why not simply do away with it and put the ropes on the wooden part as was done on many, even fairly large, vessels, but perhaps not ocean-going. It doesn't make sense, and if it doesn't make sense, maybe there's something not right going on.

HMB Endeavour tiller measurements from the AOTS book:

 

Full length from projection aft of rudder post including gooseneck: 21 feet

Aft projection to forward side of rudder post: 2 feet

From rudder post to end of wood (gooseneck connection): 16 feet

Iron gooseneck: 3 feet

Rudder post to end of gooseneck: 19 feet

 

Postion of the quadrant or sweep on the 1771 draft from rudder post: approx 11 to 12 feet or 2/3rds along the wooden part of the tiller.

 

To summarize thus far: I think we are still in need of more information because:

 

We can assume that the steering gear is part of the 1668 refit as it is drawn in red on the pre-refit draft. This profile draft dosn't show the sweep, but as Robin points out this is also the case in at least one other ship. On this pre-refit draft what might be a brace drawn in red can just be seen connected to the top of the rudder post to forward to the tiller, but is not seen on the post refit draft of 1668 unless someone has a better copy and can see it?

 

The sweep and tiller are not drawn on the plan drafts of 1668. This means the absence of the sweep in these drawings dosn't mean it wasn't there. Other ship drafts of this era don't show the tiller either so their absence on the drafts is not uncommon.

 

Cooks log indicates that the tiller braces were repaired twice in October of 1769 and January of 1770 when he was navigating New Zealand.

 

The 1771 Woolwich draft has the tiller and sweep but they are not drawn as additions which means the sweep could have been existing as we know the tiller hasn't changed from the 1668 drafts.

 

The even rope tension needed for this steering configeration is provided by use of a sweep. But we don't know how they keep an even rope tension on the replica?

 

Have I missed anything?

Edited June 4, 2016 by dashicat

[dafi]

Interesting discussion and thanx for the hint to Arming and Fitting Robin!

 

After the introduction of the wheel it took another 60 years where the slack tiller lines were tolerated until the Pollard system was introduced using the sweep to compensate the different traveling ways of the tiller ropes. And this was only in 1771, the year of the drawing.

 

Also AaF shows several sweeps in exact the same position 2/3 the way, only reason could have been imho to support the tiller. And I do not believe, that the Endeavour would have been one of the first vessels to be equipped with this newest device ;-)

 

XXXDAn

[shipaholic]

Maybe it's not a coincidence Dafi that the Pollard system was introduced in 1771 the year of the Endeavour refit, they might have used the Endeavour as a test case to trial the new system.

I agree that this has been a very interesting discussion, I wish there was more like this especially on the other disputed details about the Endeavour

Cheers

Steve

Edited March 29, 2016 by shipaholic

[dashi]

Wow this is great. Thanks Robin and Joel for mentioning those books. From what I can glean, the sweep was in use since the whipstaff and carried on with the use of the wheel as a tiller support whether above or below the quarterdeck and Pollard refined this idea in 1771 by inventing a system where the rope followed the sweep thereby solving the problem of slack in the steering ropes. Pollard did not invent the sweep.

 

According to 'The arming and Fitting by Lavery' as mentioned by Robin a sweep following the arc of the tiller was commonly installed on the quarterdecks of smaller vessels that employed a wheel (or or maybe whipstaff) where there wasn't room for a heavy tiller and sweep to fit under the quarterdeck well prior to 1771. In this situation it was fitted 2/3 rds of the tiller length forward of the rudder post primarily as a tiller support where a heavy tiller (which I assume due to the size of the vessel couldn't be operated directly) was rigged to a wheel as in the case of the Bark Endeavour. The sweep on the 1771 drawing is for this later purpose and I don't think related to Pollards design.

 

Not only that but I think the physics of having a sweep provided a minor solution to the rope tension problem by removing the arc produced by the angle of the stern post imposed on a rigid tiller. Which brings me to the purpose of the tiller brace which I now don't think was there to form a rigid connection to the rudder post but instead to help brace a pivoting connection needed for a heavy tiller to ride on the sweep as intended. How exactly the tiller brace did this as yet I haven't figured that out but am working on it and have some ideas and hopefully more research will reveal something.

Edited March 29, 2016 by dashicat

[jud]

Perhaps rigged like this will minimize the slack problem down to a manageable level.

 

Looks like I may need to start using a white background instead of black like have for years.

TILLER1_Image.pdf

TILLER2_Image.pdf

Edited March 30, 2016 by jud

[dashi]

Perhaps rigged like this will minimize the slack problem down to a manageable level.

 

Looks like I may need to start using a white background instead of black like have for years.

 

lst983 copy1_Image.pdf

Jud I can see how that might work as it balances out the math because of running the helm ropes through blocks on the bulwark and then blocks on the tiller head and then aft to the tensioning blocks at the bulwarks again (at least that's what I think but not sure). I've just done the math for the drum and instead of 5 turns on the helm drum if the rope were lashed to the tiller head, you would need 9 turns because you have doubled the robe needed for the full movement of the tiller. This in turn would affect the steering by making it twice as light and half as responsive, meaning the helmsperson would need to turn the wheel twice as much to get the same angle of turn at the rudder. I don't know if this would be a good or a bad thing?

 

Another thought I got from looking at old pictures of helm drums last night was that some of them are concave. It has just occurred to me now this would help take up slack that might be produced on the side the tiller is moving to as the rope from that side is wound on the drum. The side that is wound on would run up the concave taking more rope per turn compared to the side that is winding off as it runs down the concave realising less rope per turn. But you would need to know how much slack to expect from a parallel drum in that situation and then you could work out how much to concave the drum. At least that's the theory that just popped into my head, so who knows if it's right?

Edited March 30, 2016 by dashicat

[jud]

robin b; Changed the sketch in my post, primarily to see how line widths are effected when printing cad to the Cute PDF software and then bringing that to this site. Refined the original and added a Full Left Rudder configuration. Agree that rigging this way will effect the wheel movement and tiller response, don't consider that much of a problem but if it is, doubling the circumference of the wheel drum would negate the effect. This rigging setup allows plenty of room for the arc but the smoke stack would be in harms way. Are you sure about the additional turns on the wheel drum would need to be increased rigging this way?

jud

Edited March 30, 2016 by jud

[jud]

Looking at the build log of the HMB Endeavor by Kiwiron/Occre kit, post 1 and 245 it would appear that I was reinventing the wheel. Worth a look.

jud

[dashi]

Thanks Robin, we could do with those plans if anyone has them.

 

Jud, I've taken a closer look at the plans and see it has an 18" drum so with doubling the blocks I think you could get away with 7 turns. However I've been drawing lots of lines all over the deck trying different things and don't think a double block configuration will take out the slack, but it would half the torque at the wheel maker for a much lighter helm. As for that chimney I can't find it or the fire place marked on any of the 1768 or 1771 plans and have only seen it on the replica and AOTS plans so it's presences needs verification.

 

From drawing and rubbing out lines all over my tracing of the quarterdeck what I found was that the position of the blocks on the bulwark is critical to reducing the helm slack. However this is a compromise compared to Pollard's method. So to find the position of the blocks draw these lines; one at right angles to the tiller head when at top dead centre to the bulwark; one from the rudder head at 25 deg to the centre to the bulwark; the arc of the tiller head from TDC to the bulwark. Now measure the distance between where the tiller arc line and the first right angle line cross the 25 deg line. Half this distance and take that mark to the bulwark which should give you the anchor point for the block. To check mark several positions along the tiller arc and measure the distances from the tiller head to the SB block and tiller head to the Port block and add these. If right then they should all be equal. This needs verification.

Edited March 31, 2016 by dashicat

[jud]

Dashicat, How I did it was to draw a line from the block anchor point on the tiller to the end of the arc that I had drawn representing the travel line of the end  of the tiller. At the point that that arc was tangent, 'the part of the arc furthest from the first line I drew', to the first line, I made a mark and drew a perpendicular from the first line thru the mark on the arc I had just made. Found the mid point along that perpendicular between the first line and the ark, drew a line from the beginning point of that first line through the midpoint to the rail, then used equal distances from that point of intersection with the rail forward and another equal length towards the stern for block positions, repeated until I had two points with a pleasing spread, looking like it would work. That way should keep the tension as even as you could without leading the tiller rope along the arc of the tiller as the wheel was turned. A mirror image used for the other side.

Need to remember that this is just a method I might try if the problem was mine to take care of, there are some attempting to root out some facts, they may not find any, that would be a shame, the drawings we have seen do not look like a working solution to some of us. The Laws of physics are the same today as they were then, so we do have a method of judging any method shown in documents of the past for any deficiency's and conflicts with reality.

Turns on a drum, would not need to increase the turns unless there was slippage. The nature of a drum in that it lets out and equal length it takes in regardless of the number of turns as long as the number of turns is within reason.

I don't think I can add anything more that would useful to this discussion, I do find it interesting but probably need to let it go and watch for more facts.

jud

Edited March 31, 2016 by jud

[shipaholic]

Dashicat

 

In answer to your query about the great cabin stove, it is clearly visible in red in this draught (3814a) - you can see the shape of the stove and the chimney protruding above deck

Steve

Edited March 31, 2016 by shipaholic

[shipaholic]

Its also interesting that on plan 3819 (the 1771 one) the drum is on the aft side of the wheel, whereas all other profile plans show the drum forward of the wheel. Another thing, 3819 doesnt show a fireplace its labelled as a pantry instead

[dashi]

Jud thank you so much for your time and input to this discussion. Just had confirmation of the aft chimney so the steering ropes have to clear that.

 

Shipaholic thanks for this but you posted a similar plan to this earlier and it doesn't have the chimney or fire which I was also referring too. Could you post this recent one again but with a wider view from rudder post to wheel? The chimney is back in so I can't rig it per the Caldercraft rig as the steering rope will foul the chimney so looks like it will need to be made off as in the replica - via lanyard lashing to the tiller head.

 

Robin I hope your house restoration goes well. I guess at the end of the day as Jud says, the physics don't change so eventually I'm going to have to weigh up the evidence to hand and make a decision as to that.

 

lavery's book corroborates the common use of a sweep as a tiller support for steering gear below ships decks and above decks on smaller vessels that are of a size to require a long heavy tiller with wheel such as barks etc. The sweep has been in use since the time of the whipstaff from the 1600's. The only Bark Endeavour deck drawing to show the tiller also shows a sweep but with no evidence as to whether it is part of the 1771 refit or an existing part of the tiller gear from earlier. All other deck plans to my knowledge don't show the tiller so why would they show the sweep? The physics supports the use of a sweep for removing the horizontal arc of the tiller and it's contribution to the slack problem. The position of the blocks is critical to reducing the slack problem but didn't alleviate it until Pollard 1771 ran the rope around the sweep because according to him, there had been fatalities from the slack allowing I assume the rudder to whip back and take up the slack with such force to the wheel it had thrown the helmsman overboard.

 

I definitely can't say I am certain about a sweep being used on Cook's voyage, but there appears to be corroborating evidence in favour of there being a need, and common practice for the installation and use of said sweep on vessels of the size of HM Bark Endeavour.

 

That's all I have go on for now until I or someone comes across some more factual evidence which might help clear things a bit more. So thanks again everyone for a great discussion which I for one have learned a great deal from.

Edited April 1, 2016 by dashicat

[druxey]

Jud: your illustrations appear to show the aft tiller line crossing over the chimney area and so would get cooked.

[jud]

druxey; Believe I made a note about that on one of those drawings. They are not finished plans, only a quick illustration of an idea. Also the end of the wood in that tiller is shown with a hidden line. ' dashed ', at the point the Iron begins. I did not reduce the width there because the photos of the rebuild of the Endeavour's tiller in post 31 do not show a narrowing from the wood section of the tiller to  the iron end, Shows clearly in DFX but not so well in PDF, one reason for the drawing was to find these things out.

jud

Edited April 1, 2016 by jud

[dashi]

I've done some calculations as to the possible weight of the 18 foot wooden part of the tiller if made from oak. I am not an engineer but I've got a figure of around 280 pounds - 320 pounds 130 - 150 Kgs.

 

I'm not sure what the replica tiller is made of but someone mentioned in an earlier post that it might be made from a laminate or it could even be composite which have a much better weight to load ratio than hard wood and therefore wouldn't need the same level of bracing and support. I read that laminate can be as much as 50 to 200% stronger. So until someone can confirm the specification and material of the replica tiller then references to it may not be relevant to answering this question for the possible need of an above deck tiller support at 2/3 rds along it's length.

 

Another consideration is if the helm tackle were to break then with out some form of stops you would have a 300 pound metal tipped 18 foot pole crashing into the deck and bulwark possibly in rough seas.

 

As for the abaft fire flu according to Lawrey's book its possible it is a wooden cowl with a swivel top the same shape and size as on replica. But the helm tackle would still need to clear it.

Edited April 1, 2016 by dashicat

[Cabbie]

Good Morning dash

 

I was wondering if the tiller had to be as long as it is currently?

Is there a rule about tiller length and the size of the ship?

If it was shorter, the rigging could all be behind the chimney, there would be less sag problems.

It Wouldn't need a great chunk of metal on the end.

 

Aots has it as Jud has drawn.

 

 

I am just coming from a novices point of view, and just look at things as if I had to do them.

And wondering why they designed it to be like it is.

Cheers Chris

Edited April 2, 2016 by Cabbie

[dashi]

Good Morning dash

 

I was wondering if the tiller had to be as long as it is currently?

Is there a rule about tiller length and the size of the ship?

If it was shorter, the rigging could all be behind the chimney, there would be less sag problems.

It Wouldn't need a great chunk of metal on the end.

 

Aots has it as Jud has drawn.

 

Tiller20160403_07485256.jpg

 

I am just coming from a novices point of view, and just look at things as if I had to do them.

And wondering why they designed it to be like it is.

Cheers Chris

Hi Chris, good questions.

 

Previous post numbers 24, 27 and 64 have profile or sheer technical drawings from 1668 and all show the tiller as it is with the metal uplift on the end to clear the chimney cowl from the great cabin. It's length with metal uplift matches the what is given in the AOTS book and the Caldercraft Kit.  

 

It can all get a bit confusing as I've found, because the post 1668 refit plan has the fire and chimney with tiller and metal uplift, but I'm not sure about the tiller brace, whereas the pre refit drawing of 1668 doesn't have the fire with chimney but still has the tiller and uplift as is, with tiller brace at the rudder head.

 

The AOTS book has the fire and chimney but they have forgotten to include that on the drawings relating to the helm tackle which would slice through the chimney cowl if it were there?

 

I'm building the Caldercraft 1:64 scale Endeavour and it doesn't even have the great cabin chimney cowl and so follows the AOTS helm tackle rig which crosses over where the cowl should be when hard to port or starboard. To avoid the chimney cowl the helm rope needs to be forward to a position at the bulwarks in a line at right angles from the end of the metal tiller uplift. Using just one set of blocks and lashing the helm rope to the tiller head via lanyards I've worked out that this configuration introduces tension to the helm robes when the tiller is moving close to port or starboard. I'm still working on this but it might be correctable by adjusting the length of rope anchoring the blocks at the bulwarks.

 

I find it's easy to forget the scale of things when looking at drawings and building the kit so I've made a 1:64 scale ruler. I've marked at 1ft intervals which translates to 3/16 of an inch per interval. This helps me keep everything in perspective. 

 

There probably is a rule somewhere regarding the ratio of tiller length to rudder and hull size. 

 

I came across a topic regarding the attachment of tillers to rudder head (which I can't seem to find) and it mentions something about the way the french did this but no drawing. However the description seemed to match the drawing of our tiller brace. I think it might pass over the rudder head and down over the aft tiller protrusion where a pin through the protrusion holds it and the tiller in the head. But I'm not sure I've got that right and need to find more information. But having the this brace could make sense if the helm tackle is introducing a forward pull on the tiller. 

Edited April 3, 2016 by dashicat

[jud]

After looking at a bunch of photos taken aboard the The Replica Endeavour, I am starting to look at the Tiller a little differently. Haven't changed my mind that the Tiller needs support to ease stress on the Rudder Head clear down to the rudder. I believe that to ease passage through the water the ship uses it's sails not only for power, but using them to cause the ship to want to track in desired heading. If the ship were balanced, only small rudder adjustments would be needed and the chimney would not be in danger of being broken with the Tiller Ropes. The Tiller Ropes appear to have been looped back through the Tiller ring and lashed to it's self. That lashing looked like it was placed so it could be quickly and easily cast free of the tiller. Hard over rudder positions are not the norm and usually are anticipated in advance. The Tiller is longer than what we think as the norm when using a wheel, perhaps the transition period during this place in time while perfecting the use of the wheel created a need to be able to use both wheel and hands on tiller. Ain't speculation cool.

jud

Edited April 3, 2016 by jud

[dashi]

Thanks Jud.

 

Had another look at how the block position opposite the tiller head affects the helm rope tension. Lengthening the block ropes didn't solve the tension problem. I put my good glasses on and made some more precise measurements and it appears that from the centre to hard over gradually introduces 1 ft of slack to the helm rope. Back to the drawing board to try Jud's suggestion of running closer to the chimney cowl. 

[dashi]

Thanks Jud.

 

Had another look at how the block position opposite the tiller head affects the helm rope tension. Lengthening the block ropes didn't solve the tension problem. I put my good glasses on and made some more precise measurements and it appears that from the centre to hard over gradually introduces 1 ft of slack to the helm rope. Back to the drawing board to try Jud's suggestion of running closer to the chimney cowl.

Several hours later and I try the next position as suggested by Jud...

 

Thanks, following your idea has found a workable solution to the helm slack exactly like you prophesied. Here are my measurements of the helm rope between the starboard and port tiller blocks at varying degrees if I move them 2.5 mm or 2/16" @ 1:64 scale aft of (their previous position at the bulwarks) the line at right angles from the tiller head. This brings the helm ropes to almost touch the (what could have been wood chimney cowl) at full over. My calculations suppose the tiller arc being only along the horizontal plane or sweep and does not include a vertical arc created by the angle of the rudder post relative to the level of the deck.

 

Centre: 42=42=84mm

1/4: 32.5+51.5=83mm (added from my later post #79)

1/2: 61+22.5=83.5mm

3/4: 71+13=84mm

Full Over: 80.5+5=86mm

 

As you can see only 2mm or approx 6" at 1:1 of slack is created after 3/4 Full Over which I think is a very workable solution.

Could someone please check this incase I've made an error (which I'm prone to do)?

Edited April 3, 2016 by dashicat

[jud]

dashicat;  Went back to my cad drawing and snapped some distances to see how it effected the Tiller ropes when full rudder movement was used. I snapped the distance between the center of the sheave axles rather than compute the ark length around the sheaves, would still show the effect by doing it the easy way.

 

Tiller amidships:  total length between block centers at the Vs   27.4 ft.

Rudder half Left:  total length between block centers at the Vs   27.2 ft.

Rudder hard left:  total length between block centers at the Vs   27.0 ft.

                                              Total Difference     0.4 ft.

Measurements from the drawing I posted several days ago, other drawings will get different numbers but the same trend.

 

Would need to rig the Tiller ropes with the tiller amidships

with no slack, at hard over there would be some slack, decreasing

as the Tiller swung back to C/L which verifies the effects you

computed. Because of the stove chimney, I believe that the tiller

ropes were removed for expected extreme rudder shifts.

All speculation on my part.

jud

Edited April 3, 2016 by jud

[druxey]

Harland made the point that at intermediate angles, the slackness/tightening varies in a very non-linear fashion. Did you take this into account on your calculation?