It doesn’t have anything to do with keeping the gondola stable, or flat though, having only one fastening to the tube /bearing doesn’t change the centre of gravity of the gondola, and doesn’t make any difference to the stability. It takes a bit of a leap to think about it, but the shape of the arm that links the tube to the cord doesn’t matter at all, it’s the position of the attachment point with regards to the tube that makes the difference. So the cords are still pulling in line, on the same point.

The third bearing is redundant as a bearing, yes, you’re right, you’re not missing anything. It’s there as a weight, and that’s really all. The cord attaches on an axis that runs down the centre of the middle bearing, so to hang straight, there needs to be equal weight on both sides, and keeping a bearing there was the simplest way to do it.

In practice there is almost always more weight on the back side (the end facing out) because the tubing is slightly longer, and even though the pen is light, it sticks out a long way. So I often stick an extra bit of weight (like a bag of pennies) on the stabiliser if it’s important to me that I get sharp corners.

You’re very welcome – happy to help 🙂
sn

A query about the gondola though – I notice you’ve got three bearings in the gondola – and I can see in your original gondola pictured at the top of this thread what the purpose of these 3 was – with the right hand cord linked to the centre bearing, and the left hand cord connected to the top and bottom bearings, they’re both pulling at the same point on the pen, which means the pen should stay straight. against the paper when its traversing the page.

In your current kit though, you’ve still got the 3 bearings, but watching your video, only 2 of these have arms connected to them.

I’m assuming you’re using 2 arms rather than three as the addition of the round plate now means that the pen is kept flat against the page, and connecting a third arm in the same configuration as your original is now redundant. Is that the case? Did you find that having the cords pulling at slightly different points on the pen doesn’t really make a differences?

And secondly, if you’ve got 2 arms instead of 3, isn’t the third bearing now redundant also? Watching your gondola assembly video, I couldn’t make out what the bottom bearing was doing, other than acting as a spacer – unless I’m missing something.

I totally agree with your observation on the “friction-beset” gondola. Even if you had the exact same friction on each pivot of the arm, I was thinking if you had a long horizontal line where you could get the speed up, you might also see the gondola swinging as it accelerates/decelerates along the vector.

Thanks again for your content,
Andrew

I think the issue with any offset arm gondola is that the orientation of the gondola itself becomes important, and that is never going to be predictable because it’s controlled by friction of the pen against the paper, the pivot joints, even the stiffness of the cords. The reason I use expensive ball bearings on the polargraph gondola is not for weight, it’s to remove any impediment to the arms turning. In principle they should always be in-line with the cords.

In practice that’s not the case, there are still occasions where I see the angle of the cords change, the and the angle of the arms on the gondola lags behind because the force on them is not yet great enough to overcome the initial resistance. This causes the pen tip to be dragging behind where the true tip of the triangle is, by a tiny bit.

I don’t genuinely think it makes any difference in a way that is a problem. I’m joking in this post because my awesome discovery is clearly something that everybody else already knew (and had probably already patiently explained to me). It’s a commentary on my style of working or thinking, and is funny because it reveals how reluctant I am sometimes to actively rock the boat when it means challenging ideas that I like.

These machines (for me) are about reveling in the inaccuracies and the foibles, the peculiar little physical tics that differentiate one machine from another. They’re about seeing the process in the result – seeing “the hand of the creator” (the machine) in the shape of the lines. If it were perfect every time, well, just use a printer.

Sure, with the pen lower you can draw that amount lower, but the gondola is all wibbly-wobbly down there. Who needs that?

I got a bad grade in Geometry (a cute girl sat next to me – long story), so I’ll trust the rest of you to work out that aspect of the argument.