Bessler's drawing techniques

[KAS]

Having studied all of the available MT drawings, I couldn't help noticing the inconsistency in Bessler's drawing techniques.

I refer to the originals only.

Some are quite complex and drawn really well whilst others are crude and (dare I say it) childish to say the least.

For instance, I notice that MT101 has been drawn with a 2 dimensional base whilst the upper part of the workings is drawn in 3 dimensions.

In another, (MT 143) the weights appear to swing horizontally which, in turn will see saw the beam but the swing paths for the weights, which should also have been drawn horizontally, has been drawn vertically.

This is all very confusing and only adds to the problem when trying to visualize the animation.

It is almost as if there were 2 draughtsman involved with 2 completely different styles.

Does anyone else think this?

Kas

[keith sinclair]

hi

Id also figured that there were two artists involved .But i came to the conclusion that a lot of the pendulems and outside stuff is merely window dressing . for inertia we tried and found successful using a large weight in the middle of the machine ,

"there was an inner wheel or millstone "

And the addition of weight in the centre in the form of an inner heel attached to the outer does give the same effect as was described " slow at first then gathered speed till it reached its full working speed"

keith

[Stewart]

Hi KAS

I just noticed this topic as I was about to create a new topic about the Merseburg wheel drawings, and so I will post here instead.

You raised an issue that is actually an important thing to understand when studying images created by Bessler. He had some artistic skill, but was by no means a master of copperplate engraving or woodcutting. One difficulty he had was clearly representing things in the third dimension. He frequently mixes orthographic and perspective views for the same object, and this leads to a lot of confusion for some, but as long as you are aware of this then actually on the whole the images are clear and well presented. The second problem is he doesn't always break the correct object when two objects overlap. It's easy to understand how he can make the overlap errors when you imagine what is involved in the copperplate engraving and woodcutting process (working in reverse with cutting tools) and how difficult it is to correct any mistakes in those mediums. I often make the same mistakes when sketching designs with overlapping parts, but at least I can rub out the lines that shouldn't be there! Anyway, bearing these two quirks in mind when looking at Bessler images makes them easier to understand.

There isn't enough evidence to say that anyone other than Bessler worked on the MT woodcuts, but it certainly isn't out of the question; he could have given a sketch of a design to someone else to reproduce in wood.

All the best
Stewart

[Stewart]

The Merseburg wheel images that are seen in GB and DT have always caused confusion for the very reasons mentioned above, but are in fact quite an accurate representation of the exterior setup. Most of the setup is pretty straight forward with only the pendulums causing any real confusion. The GB image is a copperplate engraving, and is more accurate than the DT woodcut of the same image. It would seem Bessler was more adept at copperplate engraving than woodcutting, and I can imagine that being an easier medium to work with. Bessler mentions in AP that he learned how to do copperplate engravings. The GB drawing also has "Fecit Orffyre" at the bottom, so I think it's safe to conclude that he made that one. The woodcut image in DT is not so accurate, but I think this is only due to Bessler struggling in that medium, and does not necessarily mean it was done by someone else, although again this isn't out of the question. There are a few slight changes and additions, which I think would be unlikely to have been added by anyone else who would probably only have had a print of the copperplate engraving to work with.

So, taking the GB engraving as the most accurate, with the DT Merseburg and Kassel wheel woodcuts as a guide also, plus bearing the drawing issues previously mentioned in mind, it is quite easy to reconstruct the drawings and specifically the pendulums. Ed and I will have a complete animated 3D model of the Merseburg (and Kassel) wheel on our website when it opens, but for now I'll show you some pictures of the pendulums that I've coloured to make them easier to understand (see attached images). I've attempted to describe the images at the end of this post.

I hope this helps, although ultimately it doesn't get us any nearer to a working wheel. However, Bessler tells us that these pendulums were only added to even the motion, and in doing so slowed the wheel a bit, but could be removed. My current theory is that these pendulums where only needed to help smooth the motion of the bi-directional wheels, as I haven't found any mention of them in relation to the uni-directional wheels. We know the bi-directional wheels were initially balanced and needed a push to start them, so maybe at certain times in its revolution this balanced state re-occurs. When running without any load this temporary loss of torque would probably be imperceptible, but under load the pendulums could help to power it through those low torque moments.

Stewart


Although the three depictions of the pendulums in the Merseburg wheel image appear to be conflicting, they actually aren't, but the best one of the three to look at when trying to understand them is the middle one....

First lets look at how they pivot: the horizontal bar that the pendulums attach to actually pivot with them; they don't pivot around the horizontal bar. That horizontal bar has two pins which fit into two bearings, one on the main wheel support and the other on the door frame (or for the left pendulum it goes into a wall/post mounted bracket - see the woodcut version - also look at the pendulum setup in the Kassel wheel image which has the water screw in it). The semi-circular bearing is what can be seen on the frame in the front view of the wheel. The pendulum itself is comprised of a vertical rod (blue) with a U shaped piece (pink) at the top which is attached to the pivoted horizontal bar. You don't see this in the front view because it goes into the page. The purpose of this U piece in my opinion is to provide support and help prevent any movement/wobble in that plane. You'll see why this is particularly important in a minute. The rest of the pendulum construction is fairly obvious with the horizontal part (yellow) and its two weights mounted over the pivot rod and supported by the two angled brackets (green). The final part to understand is how it is connected to the crank at the wheel's axle. This is also the part where the mix of perspectives causes confusion. It looks like one of the angled brackets extends above the yellow rod, but there is actually either a separate small rod, or the bracket is bent at ninety degrees. The crank connecting rod attaches to this. I've deliberately set the camera angle in the 'Perspective View 01' image to show how the bracket and connection piece can look like they are one, but in fact be at different angles. You should also be able to see why the U shaped pieces are needed for support to stop the pendulum from twisting when the connecting rod pulls on the jutting out piece.

As for the left pendulum it is suffering badly from the overlap problem, but you should be able to figure out how it is set up with the connection piece on the right rather than the left.

[Stewart]

In another, (MT 143) the weights appear to swing horizontally which, in turn will see saw the beam but the swing paths for the weights, which should also have been drawn horizontally, has been drawn vertically.

What you're referring to as MT143 is not a woodcut print like most of the images in MT, but is in fact a hand drawn picture on the back of 136 and isn't itself numbered. The levers and weights are geared together, and the gears are mounted one on top of the other so that when you move one weight the other one follows. They open vertically, so the dashes are correct.

The drawing is not meant to be a possible PM device but is a demonstration of a device that has become known as a 'Roberval Balance' since before Bessler's time. The parallelogram with two fixed pivots means that you can balance two equal weights attached to the vertical bars on either side, no matter where their centres of gravity are in relation to their attachment point. There are many examples of this type of demonstration apparatus, but Bessler's is by far the best I've ever seen, as by using the geared weight-levers you can adjust the centre of gravity for each pair on a sliding scale from far right to far left, and thus prove very easily and conveniently that any combination of centre of gravity positions on either side results in balance. It's a useful demonstration for explaining why certain PM designs look overbalanced but are actually balanced and won't turn by themselves. MT50 is the perfect example.

I built a quick and simple model of Bessler's Roberval Balance demonstration some time ago and I've attached images of it below.

Stewart

[scott]

Wow Stewart, thank you for the *very* informative posts. It is so good to have you back.

Fascinating stuff about the pendulum. And that is a fantastic model of MT143.

-Scott

[jim_mich]

When I look at a Roberval Balance I see a "pair of pairs" of weights that are inter-connected using gears.

I think that many Roberval Balances are designed such that they can be rotated or spun around their vertical axis to demonstrate how CF acts on the weights.

[KAS]

Thanks Stewart.


I can see that you have given this some study time. Your explanation is very descriptive and informing. I found this it great help in understanding the draughting techniques of the early 18th century.

Thanks again.


Kas

[Stewart]

Thanks Scott, it's great to be back, I've missed this place!

When I look at a Roberval Balance I see a "pair of pairs" of weights that are inter-connected using gears.

It's the parallelogram that is the Roberval Balance, Bessler has added his geared paired weights to it rather than have fixed arms with adjustable weights like Desaguliers' apparatus (see attached image). I've not seen this type of apparatus used to demonstrate centrifugal force, there are other types of apparatus for that purpose, but I think you are probably talking only about the geared weights, in which case I can see how they are similar to other CF demonstration apparatus.

I should just point out that in my model I've used a piece of cord in a figure of 8 arrangement around two pulleys instead of gears (the cord is nailed to the pulleys where the rods attach); this allows the device to function exactly the same as with gears except it's much quicker and easier to build.

I forgot to mention in my last post that although this demonstration serves as a good experiment for proving why certain PM designs can't work (such as MT50), there is probably another reason Bessler is showing it. He has handwritten the following at the top of the page:

Demonstration: Etwas in Æquilibrio zu heben

'zu heben' is normally translated as 'to lift'. In fact that's probably the only definition you'll find in modern dictionaries. But when you research into the word 'heben' in old dictionaries etc. you'll find that it can also be used in the sense of 'halten' meaning to keep/hold, so we get either:

Demonstration: to lift something in balance.

or

Demonstration: to keep something in balance.

The second meaning seems to make more sense as that is what the device does, i.e. no matter what the positions of the weights are (one pair could be pointing in, the other pair point out), the whole device is kept in balanced.

I'm sure some of you will have ideas about how you might use this in an wheel, but just bear in mind that this is a false balance and only works if the two pivot points of the parallelogram remain one above the other. It's an interesting drawing and for those that have studied MT you'll know that these connected paired weights appear elsewhere, and of course the parallelogram features on the toy page.

I would just like to say here that Ed and I work closely together on Bessler research and anything I post here Ed will most likely have had a hand in and so he deserves equal credit.

All the best
Stewart

NB: This image can be seen in a Royal Society Journal entry titled:
An Account of an Experiment Explaining a Mechanical Paradox, Viz. That Two Bodies of Equal Weight Suspended on a Certain Sort of Balance (As in Tab. Fig. 2.) Do Not Lose Their Aequilibrium, by Being Removed One Farther from, the Other Nearer to the Center. By the Rev. T. J. Desaguliers, L. L. D. & F. R. S.

[Gregory]

Wow Stewart!

This MT143 replica is simply amazing! Thank you for show it!
For me it's very good to hear about that handwritten note by Bessler. I have this same thought for some time, and I believe this is a part of the principle in a specific way.

Thank you, you really inspired me.

Greg

[jim_mich]

Stewart, in your model I saw that you used a cord as a gear. A few times I've thought about using cord in such a way.


Somehow, somewhere I got the idea that some Roberval Balances were constructed such that they could be spun around their central support pillar. I must have been mistaken. But if you where to spin a Roberval Balance around its central support pillar and if the weights were not symmetrically placed then you will see that it no longer balances due to CF. This is the message that I saw over a year ago when looking at MT-143.

I see a similarity between MT-143 and the toys page hammer guys. Both show four interconnected weights.

[LustInBlack]

Hmm.


Gives me idea.. If you move the left weights farther from the balance, it's not more heavy on that side, so you can then give these weights to another mechanism, so this balance could be used as a idle shifter for the weights, that would stay in equilibrium until you decide to connect or disconnect weights on it .


I like to connect and disconnect things. ;]

[ken_behrendt]

Stewart...

I like your model of MT143...it looks like something Bessler could have constructed himself!

However, earlier, with regard to the pendula attached to the Kassel wheel, you wrote:

When running without any load this temporary loss of torque would probably be imperceptible, but under load the pendulums could help to power it through those low torque moments.

But, those pendula were supposed to be "counterpoised". Thus, regardless of what portions of their arc of swinging their bob weights were in, they could not apply any torque to the axle.

I do not view them as providing assisting torque to the huge wheel to carry it through "dead spots" when under load, but, rather, solely there to adjust the wheels maximum terminal rotation rate and, thereby, help keep the CF acting on the weights from becoming disablingly excessive.


ken

[rlortie]

But, those pendula were supposed to be "counterpoised". Thus, regardless of what portions of their arc of swinging their bob weights were in, they could not apply any torque to the axle.

"Counter poised" Where did you draw that conclusion from? Who said they were not intricate?. Like a two piston engine, they meet in the middle of a stroke. Would you call that Counter poised?

Besides even if one is moving up while the other is moving down, you still have the combined inertial force of both of them! Now isn't that to say they are capable of applying torque

As a matter of fact if one crank is 180 degrees out from the other and they are on a two part shaft, with the correct timing, one turning clockwise would induce the other to turn counter clockwise. A simple method of employing counter rotating transference without the use of gears or figure 8 belts on pulleys.

Your imagination is your only limits. Open it up and let it out of the tunnel.

Ralph

[Gregory]

Gives me idea.. If you move the left weights farther from the balance, it's not more heavy on that side, so you can then give these weights to another mechanism, so this balance could be used as a idle shifter for the weights, that would stay in equilibrium until you decide to connect or disconnect weights on it .


I like to connect and disconnect things. ;]

Or just put another fulcrum, maybe an anvil beneath them to "rest" on it... For a few seconds with greater leverage and torque. ;)