Thread tension and twisting

[Constabulary]

It basically does not matter HOW you apply tension to the thread (or should I say friction between thread and what ever...) but it matters THAT you apply tension and the amount of tension (top and bottom) depends on the kind and the thickness of material you want to sew and the type and thickens needle and thread.

There are several different type of "tension devices" (meaning different systems) and their purpose always is to apply the necessary amount of tension to the thread that you need to sew your chosen material with the chosen thread am needle. The way they (the tension devices) do this (apply tension) is not really important. The point it that they to have apply a certain and constant amount of tension during the sewing process. You can achieve this by just tension discs with different strength springs or numbers of wraps around a post or a pulley or combinations of them (as on early 1900´s patchers or saddle maker / harness maker machines). There are even disc tensioners with a pulley between the tension discs (F.i. on different CLAES machines). The BUSMC #6 even has a totally different tensioner system. Sewing machine engineers have developed different tension systems over the decades - some have survived some not. Nowadays - because of simplicity and effectivity - you mainly have tension discs + springs in different strength (depending on your application).

I run a 45K which has the same tension system as the Adler 4, 5, 104, 105 (and so forth) and many many similar more modern type machines of this class and I DO NOT wrap the thread 1 + 1/2 time around the pulley, I just apply tension with top tension discs. I have the "tension" on the lower unit just as low / high that the pulley can spin. The Singer 136K / 133K (and their modern clones) have a similar system (incl. upper tension unit with discs). But there is not really a pulley (with V or U grove) it rather it a plain roll (more or less) and a massive one sided tension disc. On the other side of this tension disc is the roller in combination with a massive disc that is spinning while the thread is running between it - kinda hard to explain ). Basically the same system abut technically a bit different. Pretty sure I somewhere have posted a pictures of a disassembled 133K / 132K lower tesnion unit)

So again - it´s not important HOW tension is applied - import is THAT a certain amount of tension is applied and it does not not matter how this is achieved.

EDIT

Addition:

The CLAES 213 / 214 is the same class machine (same hook, same needle, same feeding system for same applications and optically and technically quite identical...) as the Singer 45K or above mentioned Adler models and so forth. They are using all the same type of tensioner (top and bottom) with the difference that you cannot apply any tension to the lower tension unit on the CLAES machines but their lower "tension unit" has the pulley which is just held by a screw (no tension spring) and the top tension unit has a small pulley between the discs. You know what I mean? It´s not important HOW - important is THAT...

Edited January 8, 2020 by Constabulary

[Gymnast]

1 hour ago, Constabulary said:

The Singer 136K / 133K (and their modern clones) have a similar system (incl. upper tension unit with discs). But there is not really a pulley (with V or U grove) it rather it a plain roll (more or less) and a massive one sided tension disc. On the other side of this tension disc is the roller in combination with a massive disc that is spinning while the thread is running between it - kinda hard to explain ). Basically the same system abut technically a bit different. Pretty sure I somewhere have posted a pictures of a disassembled 133K / 132K lower tesnion unit)

Thank you, Constabulary. I got a link here to seller with photo of a tension unit to a Singer 132k. You are able to magnify the Photo there:

https://www.trojansewing.co.uk/product/tension-complete-97800/

@Constabulary is this the kind of tension unit you describe?

[Constabulary]

Yes - the bottom part.

2 hours ago, Constabulary said:

Addition:

The CLAES 213 / 214 is the same class machine (same hook, same needle, same feeding system for same applications and optically and technically quite identical...) as the Singer 45K or above mentioned Adler models and so forth. They are using all the same type of tensioner (top and bottom) with the difference that you cannot apply any tension to the lower tension unit on the CLAES machines but their lower "tension unit" has the pulley which is just held by a screw (no tension spring) and the top tension unit has a small pulley between the discs. You know what I mean? It´s not important HOW - important is THAT...

And this is the CLAES tension "setup" w/o a spring that applies tension to the pulley on the bottom part - if that is of any interest. The top tension has a small pulley between the discs. So no matter what kind of grove U or V the pulleys have (or what ever) the main thread tension usually is applied by the tension discs (with or without pulley) on the upper tension unit - in my opinion. Others may have different opinions but this is mine

[Bert51]

Of all the machine I have seen with this type of tension system, they all have "V" and the thread wraps around it 1 2/2 times.

Some time back I looked at a flat bed singer 45 and the centre hole was over 3/8 in size (quiet out of round as well) so it had been used with high tensions.

Bert.

[Gymnast]

17 hours ago, Constabulary said:

It basically does not matter HOW you apply tension to the thread (or should I say friction between thread and what ever...) but it matters THAT you apply tension and the amount of tension (top and bottom) depends on the kind and the thickness of material you want to sew and the type and thickens needle and thread.

I agree with this statement, if you have got no thread twisting problems. But some people have reported thread twisting problems, causing kinks being formed somewhere in the thread path or other inregularities that cause variations in needle thread tension. So that is why I adress this question on how the disc tensioner and roller tensioner share how much needle tension they provide. You can deal with thread twisting problems in many ways too as have been pointed out already. So how the two tensioners for needle thread tension is sharing the load is in my oppinion just one way more to deal with the twisting problem.

Ryan Neel have made two very similar videos on how to adjust thread tension on the Cowboy CB4500 and CB3200 machines. And I think quite many machines are like that. Here is a link to one of them:
https://youtu.be/FdViQ_tv9Rc

Ryan adwise, that when you need some more needle thread tension you may increase the primary roller tensioner by one turn (right) and half a turn on the disc pre-tensioner. So you always turn half as much on the pre-tensioner as on the primary tensioner. It is not exsactly the same as written in the manuals. But Ryans adwise makes better sense to me, because when you increase the tension of the main roller tensioner, then you like to have some more tension on the pre tensioner as well so you assure the grip on roller.

But I have not yet found adwise from the suppliers or manufacturers of these machines on how the two tensioners should share the added tension they provide together. The only adwise is in the naming "main tensioner or primary tensioner" and "secondary tensioner or pre-tensioner". From this you may get a hint, that the primary tensioner should provide most of the tension.

I have tried to measure a bit on the grip or the friction on a roller like this with 1½ turn of some different polyester sewing threads. Based on that I should recommend as a rule of thumb, that the primary roller tensioner provide 2/3 of the total thread tension, at the point where the thread leaves primary tensioner. And then the pre-tensioner should be adjusted, so 1/3 of the tension is on the thread before the roller tensioner. You like to have less tension on the disc tensioner to avoid too much twisting. On the other hand you want to avoid that the thread starts to slip on the roller, and therefore it can be risky to go much above 2/3 of tension from the roller tensioner. But with some more sticky thread, you can go above 2/3 of tension. I have to stress, that this adwise is very much as a rule of thumb, and of cause other factors influence this. And I make this adwise because nobody else have done it. Perhaps somebody else will do that.

[Bert51]

Since I read "gottoknow's"write up on twisting thread and how to set it up your thread by raising the height from the spool to the height of the thread stand, I have not had any further problems with my thread twisting.

I can put as much pressure on my thread as I need with out it twisting. I also applied the same to my Singer 132K6, which has two tension systems and it has no problems with twisting.

I have pushed tension further then I needed to and when I stopped the long run the thread melted from the heat of the needle.

Bert.

[Gymnast]

The Smart Anti Kink Device

For people having issues with kinks being formed near the spool, I suggest this “Smart Anti Kink Device” to be placed over the spool and perhaps mounted on the thread stand.

It consists of a bended piece of steel thread (this is Ø1.5 mm), and a small block of expanded polyurethane foam, as has been suggested by @mikesc and @RockyAussie. On the horizontal steel part, you can twist the thread around to adjust the direction and the amount of thread twisting being delivered back to the spool. You are in this way able to absorb most of the twist caused by the thread coming off the spool end, and then kinks are not formed.

This is not the first prototype. The way the steel thread turns to guide the thread are done carefully to avoid a birds nest or any other change of function. But of cause you can adapt.

When I do the clips test shown in first video, I get this twisting depending on how the thread turns around the steel:

When you have a new spool of serafil 20, you can get about 4 turns of twist each meter with the thread coming off the spool, and then you would select 0,5 turn on the steel thread. Later on when the thread is almost empty, you get about 10 turns of twist each meter and then you set 1,5 turns on the steel thread.

The foam block will absorb most of the twist from the sewing machine, but not all. So you may have to compensate a little for that too in the number of turns on the steel.

I have only tried this with two kinds of thread, but I suppose you will get a similar function with other kinds of sewing machine threads.

This was the main findings to be used. In the following I go into more lengthy details about it.

I have measured (with a lot of testing) when kinks can be expected to form on the thread leaving a spool with Serafil 20 thread. I did that with a test set up shown here:

I got a disc tension devise to adjust the amount of twist that returns to spool. Then I pull of thread and observe if kinks are formed. Twist from both spool and tensioner will increase the level of Z-twisting of the thread. With a distance of 10 cm from top of the spool to thread guide, kinks can be formed when the total twisting exceeds 8 turns each meter. With a distance of 48 cm from top of spool to thread guide, the total twisting need to exceed 15 turns each meter for kinks to be formed. Below these levels the thread may show twisting behavior, but kinks are not formed.

So I can with these measurements confirm, what @Constabulary and @Bert51 already have written, that a higher thread stand better resist kinks being formed.

These threshold levels for kinks of 8 and 15 turns each meter will of cause be different for different threads and different heights of thread stands. The kink threshold level can be below the level of twist the spool makes by itself by the thread coming off the end. Therefore the foam block alone making zero twist may not be sufficient in all cases and you need to actively create twist in right direction by the use of the steel thread of this anti kink devise.

When I used the disc tensioner in this set up, I discovered that a very low tension of 0.2 N setting was able to create twisting up to 2 turns each meter in opposite direction. When tension was increased to 0.5 N the twisting in expected direction occurred. A tension of 1.8 N caused 24 turns each meter. The reason for this low tension behavior can be explained by the thread construction. This is a close up picture of the thread:

When the tension discs press hard on the thread, then they engage the strands of the whole thread, and therefore twist are generated according to the twist of the bigger strands. When you got a very low pressure on the thread surface, then only the fibers in the strands are engaged. The surface fibers direction are parallel to the thread direction, or perhaps a bit opposite to the direction of the strands.

I think the polystyrene foam block acts in the same way as the gentle discs at low pressure, and therefore they cause no twisting and to some extent prevent twisting passing through.

In the first video, I used an improvised pre-tensioner, and a close up Photo is this:

When used to provide 0.3 N tension, it do create a twist of 25 turns each meter. With 0.7 N tension it makes 50 turns each meter. The Anti Kink Devices can handle this too, but this kind of tensioner is not the best choice.

[mikesc]

You are Heath Robinson ;-) or a worthy ( "heritier" ) descendant / kindred spirit of..
, and I claim my €5.00, ;)

...which ( along with other multiples of Euros,....in anticipation of your approval of the means of their dispersion ;) ) have already been sent ( as "electronic money", which will do far more direct good to Australians ( Fire-fighters and Residents ) and the Animals there ) not as "things" that they cannot manage, and the sending of which, costs more than the things themselves ) to Australia ..

Thank you for your "curiosity"...which ..despite my ( sometimes ) thinking that you may have too much time on your hands..is , indeed, very useful :)

Edited January 13, 2020 by mikesc

[Gymnast]

Thread net on spool.

Some people have reported a succes in using Thread Nets on the spool. Other names used are Thread sock. You can buy them from more suppliers (use Google) or you may find them used for food like when you buy garlic or grape fruits. Some use ladies nylons. They prevent the thread to drop under the spool, and add some tension to the thread, so I guess it reduce the chance of kinks to be formed right after the unwinding of the spool. Some have recommended to fold the net double to prevent the thread to get caught on the net edge. This is one video about this thread net:

https://youtu.be/udl-UnLdG-4

You can buy this net as a 10 yards long (more expensive) or as 50 small pieces 16 cm long in length.

This solution has been suggested some times before in this forum to help solving problems with kinks and thread drop Down under spool. I found these previous threads:

 I guess, than if the net press too hard on the spool, the thread tension may vary due to the thread comming of the spool near the bottom or near the top.

Edited January 25, 2020 by Gymnast

[Bert51]

I have found with some spools with the thread nets, the nets can cause to much tension. I no longer fold my nets, as that can give the fuller spools to much tension.

I use the small nets (12 to 16 cm) on  thread up to 75 M (T40) and the larger net on on the heavier thread spools, specially the nylon spools that sit there and slowly unravel. I'm about to order another 10 mts of the later.

Please don't do what I did the first time I got a 10 meter length, sat down and started cutting the net nearly the same length of the spool. thankfully I was use a high spool as my guide.

Bert.