Horizontal oscillating hook · Horizontal rotary hook · Vertical oscillating hook · Vertical rotary hook · Zig-zag mechanics · Zig-zag timing

Zig-zag hooks and their strengths and weaknesses

All zig-zag machines that I’ve got, are round bobbin. I know of one transverse shuttle zig-zag machine: it is by Seidel & Neumann, see www.naehmaschine-antik.de.

Round bobbin hooks can be classified like this: rotary versus oscillating, vertical versus horizontal, side-facing versus front-facing. Rotary and oscillating hooks have very similar performance characteristics when the other two parameters are the same. Rotary hooks are generally faster and may be smoother, but overall quality of the machine can change all that.

Forward-facing hooks offer the easiest solution to zig-zag: the needle oscillates within the width that the hook can use to pick up the thread. Side-facing hooks have to move the entire bobbin assembly left and right synchronously with the needle moving left and right. The entire mechanism is much more complex than that of forward-facing machines.

Direction of hook movement

For both rotary and oscillating hooks, the critical direction of movement is the direction in which the hook catches the loop: left to right or right to left for front-facing hooks and front to back or back to front for side-facing hooks. All the possibilities have been tried, but although some look better than others in theory, in practice there isn’t much difference in performance.

Needle movement

To achieve zig-zag, the needle needs to move left to right and right to left. This can be done in two ways: either the needle bar can be fixed near the top and swing left and right like a pendulum, or the needle bar can be fixed to an arm that swings horizontally, moving the needle bar along an arch.

The pendulum mechanism is the simpler of the two. It is sometimes called “swing needle” and is used with vertical forward-facing hooks (or transverse shuttles! Do let us know of any other hooks, but I haven’t seen any others). The needle hole in the needle plate is straight.

The needle is vertical in the central position but in all other positions it enters the material at an angle. This means that the eye comes up higher on the left or on the right than in the centre. Admittedly, this is only a very small difference in height, but combined with the hook curvature, it doubles. So if you are having skipped stitches with zig-zag, it is important to check needle height in the left and right position and not only in the centre.

The other type of needle movement comes in two variants: with the needle bar oscillating around a point behind it, or in front of it.

In both cases the needle hole is curved to match the arch along which the needle bar is moving. The corresponding zig-zag foot should have a matching needle hole! Or at least an oversized needle hole to fit all different configurations.

The needle remains strictly vertical in all positions, and also drops down to exactly the same distance. However, depending on the type of the hook, the distance between the hook and the needle eye may differ.

Horizontal hooks

The horizontal oscillating hook was introduced by Singer in model 66 (see my previous post for details). It was a side-facing straight stitch only hook. Later Singer added zig-zag and made the hook forward-facing, such as for example on model 328K (this is not the first model of this kind, but that’s the one I’ve got among my machines). Singer also developed this hook into a rotary version: side-facing straight stitch in model 201 later evolving into a forward-facing rotary version such as in Slant-O-Matic machines like model 401 (again, I am not sure that this is the first occurence of this hook).

In all these cases the hook is found on a ring of the same diameter that either oscillates or rotates around its centre. The side-facing oscillating hook picks up the thread with a cut-out on its inner side, so the needle enters into the hook and is threaded left to right. The hook moves front to back when it picks up the loop.

However, the side-facing rotary hook picks up the thread with a hook on its outer side, so the needle stays on the outside of the hook and is threaded right to left. The hook moves back to front when it picks up the thread.

When Singer had to turn these hooks to face forward, the oscillating hook presented no problems as the needle was now forward-facing and threaded from front to back. However, the rotary hook needed to be modified because having your needle threaded from back to front is really not practical. So rotary horizontal forward-facing hooks are very similar to oscillating ones with the hook having a cut-out on its inner side and the needle entering inside the hook. In both forward-facing hooks the hook moves from right to left when picking up the loop.

On Singer 328K, the needle bar oscillates around a point in front of it, moving along the arch matching the curvature of the hook (here is that photo again).

This is perfect because the needle is always strictly vertical and the needle to hook distance remains the same in all respects at every point.

The only hindrance that remains, is the rotation of the hook – it can’t be in every position simultaneously. ๐Ÿ˜‰ The hook moves from right to left, so it meets the needle in the right position earlier than in the left position, which means that it meets the needle at a different height during zig-zag. On a 4mm wide zig-zag of Singer 328K, this difference in needle height is about 1mm – not that much to be problematic, yet too significant to ignore, considering that the needle scarf (and the thread loop) is only about 2.5mm long.

With increasing zig-zag width, the difference in needle height also increases. Here is a rotary hook on Janome’s Memory Craft 6000 with 7mm zig-zag:

The difference in needle height here is about 2mm.

The timing should be set such that there is a sufficiently large loop to catch in the right position, so the needle must be already past its lowest point and going up; but also the needle must not be gone too far up in the left position so that the hook could catch the loop.

Older machines of this design with 4-5mm wide zig-zag, have a very stable zig-zag stitch, but the left needle position remains the weakest point. As the maximum zig-zag width increases, stability of stitch formation decreases, although still not presenting any problems for quality built machines. The central position is the most stable and is best used for straight stitch. The right position might sometimes have problems with loop formation – some twisted threads have difficulties forming a loop, so the central position gives them a little more time.

I don’t know of any horizontal side-facing zig-zag hooks and I can’t quite imagine how it would work, so again, if you know of one, do tell!

Vertical hooks

Vertical hooks come in a much greater variety as different designs were introduced by different makers.

I don’t know of any side-facing vertical rotary hooks that can do zig-zag, but why not – they probably exist, so do tell!

Forward-facing rotary hooks have had quite a few very different bobbin case designs.

These hooks also exist as clockwise and anticlockwise (but not in the same hook!), and there does not appear to be any particular preference in terms of quality of stitching. These are excellent hooks, and with respect to timing for zig-zag, they follow the same guidelines as the forward-facing vertical oscillating hook.

Vertical oscillating hook is based on Singer 15 which was originally a side-facing straight stitch hook moving back to front when picking up the loop.

Singer later swapped the direction of movement in models 15-89, 15-90 and 15-91, but then returned to the original design as the benefit seemed to mainly be theoretical.

This hook also exists in a forward-facing version, which is perhaps the most common hook in domestic sewing machines.

Zig-zag with the side-facing vertical oscillating hook

This is the only side-facing zig-zag hook that I know of. To achieve zig-zag, the whole bobbin assembly has to move left-right synchronously with the needle. The needle must remain strictly vertical, so the needle bar can’t just swing as a pendulum but has to be moved as a unit. All of this makes for a more complex mechanism, so it is not surprising that simpler forward-facing hooks eventually took over. Yet the side-facing hook gives a much better stitch stability for zig-zag, at least in theory. I think that quality of stitching mostly depends on the quality of the machine rather than on the type of hook, but it is still an interesting hook to consider, especially in a top quality machine. ๐Ÿ™‚

I have only seen such machines with the needle bar oscillating horizontally around a point behind it, so the needle moves along an arch curved up.

Because the bobbin assembly moves left and right with the needle, the hook to needle distance remains always the same horizontally, at least if your machine is well-made. ๐Ÿ˜‰

However, vertically there is a small difference: in the left and right positions the hook meets the needle a little lower than in the centre due to the curvature of the hook, plus there is the timing issue as the hook moves from back to front.

In the left and right positions, the needle meets the hook earliest, so the timing must be sufficiently advanced so that the needle would be past its lowest point and going up. In the central position, the needle is a little higher, but the hook is now in its highest point, so the hook and the needle meet almost at the same height as in the left and right positions.

On Janome’s New Home 580 the difference in needle height at the point where it meets the hook is about 0.2mm – so small, I can’t measure it more precisely.

This is the most stable zig-zag design that I’ve seen, with every needle position being almost equally reliable. The central position is the weakest, but only with the smallest of differences.

And now it also becomes clear why a large number of such machines designate the left needle position as the position for straight stitch, not centre. Left (or right) is just ever so slightly better. ๐Ÿ™‚

Forward-facing vertical hook

This configuration gives the greatest variety of designs. Not only can you have an oscillating or a rotary hook moving in either direction, but also the needle motion can be either swinging or oscillating along an arch.

Zig-zag with the forward-facing vertical oscillating hook and swing needle

Designs with a swinging needle are simpler, but as noted above, the distance between the hook and the needle’s eye depends on the needle position.

In the left most position, the needle is slanted, so its lowest point is heigher than in the centre; it meets the hook at the earliest because the hook is moving from left to right; and the hook is at its lowest position. If the hook meets the needle too early, that is, when the needle is in its lowest point but has not yet started going up, there would not be any loop yet for the hook to catch.

In the right most position, the needle is slanted symmetrically to the left most position, and the hook drops equally low, but the meeting takes place slightly later, so that there is always a loop to catch. The danger is however that if the timing is set slightly too late, the needle would have gone too high and the hook would miss the loop for that reason (notice the difference in needle eye height in the above photos).

The right needle position is the weakest in this design, and the central position is best for straight stitch because this is where the needle is strictly vertical. Also, there is always a loop to catch as even to most difficult threads will have developed a loop by now. ๐Ÿ™‚

On my Haid & Neu Primatic with 4.5mm wide zig-zag the difference in needle height between left and right positions is about 1.5mm. This machine is a vertical oscillator manufactured to very tight specifications, and it works absolutely beautifully with this design. But what if zig-zag was wider?

This is Pfaff Creative 7510 with 9mm wide zig-zag and a counter-clockwise rotary hook. I marked where the hook was meeting the needle in the left and right positions – the difference in needle height is just over 3mm! It is longer than the average scarf (2.5mm), and the top meeting point is in fact well beyond the scarf. To improve stitch formation, the hook comes very close to the needle, in fact it is touching the needle.

This hook moves in the opposite direction to the oscillator, so the weakest point here is in the left most position. Again, the central position is best for straight stitch because the needle is strictly vertical here, but stitch formation is more secure on the right. However, Pfaff have been building these hooks since 1937, and they have a few tricks up their sleeves to make it work reliably left, right and centre. ๐Ÿ™‚

Zig-zag with the forward-facing vertical rotary hook and oscillating needle

In a different design used by Lada T132, a counter-clockwise vertical rotary hook is combined with a horizontally oscillating needle bar. The hook and the needle move in different planes – at a right angle to each other. The needle comes in front of the hook, and the hook moves from right to left.

In the right position, the hook is closest to the needle in the horizontal plane, it meets the needle at the earliest time and it is in its lowest position. So the timing has to be set such that the needle is already past its lowest point and is going up so that there is a loop to catch.

In the central position, the needle is a little higher but the hook is also in its top position, although it is furthest away from the needle in the horizontal plane. However, the thread loop is already well developed at this point and is easy to catch.

In the left position, the needle is again a little higher while the hook is a little lower, but it is also closest to the needle vertically which helps to catch the loop. The timing should be set in such a way that the needle is not too high in this position.

The difference in height between right and central position is about 0.5mm, and the difference in horizontal distance to the hook is about 0.3mm, giving an extra distance of about 0.6mm from the triangle. The difference in height between right and left position is about 1mm, and the horizontal distance to the hook is the same (practically zero). So this makes the left position the weakest one for stitch formation.

And the winner is…

First of all, let me re-iterate that the stability of zig-zag mostly depends on the quality of the machine, namely precision of manufacture and assembly, as well as correctly set timing according to the design.

However, comparing the designs with their different distances between the hook and the needle, we can rank them as follows:

  1. Side-facing vertical hook with oscillating needle.
  2. Forward-facing vertical hook with oscillating needle.
  3. Forward-facing horizontal hook with oscillating needle.
  4. Forward-facing vertical hook with swing needle.

I must admit I was quite surprised to find the old-fashioned side-facing vertical oscillator win on all counts when it came to zig-zag. But there you have it! Some of the best machines have this design, including Necchi Supernova with its embroidery that works on almost everything. The side-facing hook can also be used with a twin needle (with a special needle plate and foot), so that plain zig-zag turns into cross-stitch. I feel there are even more miracles lurking underneath – well worth investigating!

Buy me a coffeeBuy me a coffee ๐Ÿ™‚

19 thoughts on “Zig-zag hooks and their strengths and weaknesses

  1. Elena, this is fascinating! I must admit it will probably take a few reads to really digest. And I did not realize the swing needle design was actually ‘fixed’ at the top and pivoted from right to left.
    I have a Necchi Supernova freearm on my sewing table right now (needs a little tension work, I was fiddling and put it back together wrong) so after I set that right will be a perfect opportunity to take a look at how the zigs and zags actually work.


    1. It took me a long while to figure out! So confusing… but I’ve been noticing for a long time that different machines respond differently to difficult fabrics – some tend to skip stitches on the left, others on the right, and at first I thought I was imagining things, but then started looking closely. Not imagining! The mechanisms are different!

      Liked by 1 person

  2. Elena,
    Very well done. Explaining all that is hard – I’m sure this took an age to produce. I learnt an awful lot – and I thought I knew these machines!

    My two side facing oscillating machines produce the best looking zig zag stitches, which agrees with the theory. My second best zigzagger, the Bernina 807 is also your number 2 mechanism. All designs are of course a compromise and the side facing machines, because of their complex mechanism that’s oscillating back and forth, are in my experience are heavy machines that aren’t the smoothest, although Necchi did a pretty good job I have to admit.

    Are there issues for straight stitch quality with the forward facing vertical, oscillating needle? The maximum distance from hook to needle is on straight stitch with these machines. My Lada T132 is not great on straight stitch – whereas all the zig zag Singers I’ve come across do all do a really nice straight stitch. Also, like you say, sometimes it’s difficult to get a fair comparison in practice because the quality of the machine in general can override theoretical performance.

    Dan H


    1. Thanks, Dan! ๐Ÿ™‚ My Lada T132 is brilliant on straight stitch, especially with the straight stitch cam. But yours has a different hook! May be the distances are different too?

      As for the side-facing oscillators being heavy-footed, so to speak – I agree! Quality of manufacture becomes increadibly important here. Both my 1920 Singer 15K and my Necchi Supernova are as light as can be, but most of the Far Eastern machines were pretty clumsy. The problem appeared to be in the precision of assembly: once I overhauled each of them by going through every single screw and re-adjusting, the machines became very light going, delicate and precise. So if you get one of these, it’s worth while assuming that they were assembled by monkeys. ๐Ÿ™‚


  3. I wouldn’t have thought the differences in hooks between Lada’s should make a difference. All my Singers do have a dedicated straight stitch needle plate though and my Lada does not – I think that’s at least part of the reason the Singers perform better.

    I was trying to refer to the weight of the side facing hook machines. Singer 328 in it’s case (front facing) is 28lb. Necchi Supernova Julia (side facing) is 46lb!


    1. Oh I see about the weight – you are right! Also Necchi has a large cam driving mechanism and in general the Singer 15 hook with is surround is a lot heavier than the horizontal hook.


  4. Good article, thanks Elena, I have been imagining and trying to understand zigzag stitching machine how it works, but you have just elaborated every bit of it to me. Thanks a lot.


  5. Lovely article! I also plan to make this article like this but school is in my way so I just give time for searching thoughts about sewing machine hooks.

    I had this sewing machine back in June 1 of 2017, and it is my very first sewing machine. It is a cast iron, side-loading, vertical oscillating hook sewing machine made in China (I know their products are quite low) but it works like a charm for me. I guess I made the right decision of choosing this sewing maching with this type of hook mechanism and orientation (Yay! For being the first place in your hook designs).

    You can check the YouTube channel โ€˜theseargentโ€™ for his sewing machine collections. Youโ€™ll find the review about the side-loading, vertical rotating hook sewing machine. Have a good day ๐Ÿ™‚


  6. I have a Singer 237M, made in Monza, Italy. It is a side-facing vertical oscillating hook. The needle bar doesn’t swing, it moves left and right in tandem with the hook, which also moves left and right. The stitch isn’t formed with a cam, rather it has a sort of sliding mechanism, like the Necchi’s. The result is a perfect zip-zag. It’s one of my favorite machines.


    1. Thank you for alerting me to this model! I’ve seen them around and never looked twice… I am looking with both eyes now. ๐Ÿ˜ƒ


      1. I did have one issue with it, which is likely why I got it so inexpensively. I couldn’t get the tension to hold steady. It turned out the spline on the front tension assembly rod was cracked.

        Internally the machine is all metal. Externally it has some plastic covers and knobs. I can live with that. But the tension rod is not a solid machined piece like on most machines. The rod is metal, but the spline is a separate piece and is plastic. I couldn’t believe it was designed like that! I carefully JB welded it back together and have had no more issues with it. It makes beautiful stitches and holds tension well.

        I tried to see if I could replace it with an all-metal rod/spline. I have yet to figure out if tension rods are different sizes and if so to try to get my hands on one that will fit. The search continues but meanwhile, all is well.


      2. A plastic rod in the tensioner is a bad idea. ๐Ÿ˜• They are not all the same, you can’t just swap them out blindly. Many are interchangeable, but you need to find the same type. I’ve never seen one with a plastic rod on a decent machine though.


  7. Well, the actual rod is steel. It’s a two-piece part with the spline being plastic. Even so, it is a terrible design as evidenced by the fact that it failed. I’m just really lucky I was able to fix it. Otherwise, it’s really well-engineered. It boggles my mind that the designers would even consider a plastic spline considering how well it’s engineered otherwise. One never knows what goes on in the minds of people.


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.