Wow! That may be my most risque title yet.  The twins are 2 model 99k machines I picked up over a month ago now.  I got both of them a day apart.

So, I sat down at 10pm, and figured I’d try to figure out what was going on with the 99s.

Using the one bobbin case between the two, one machine insisted on the bobbin tension being fully tightened in, and the other one only worked with 0 tension.  You could actually see the spring wasn’t touching anything on the non-screw side.  You could stick a dime in between the spring and the body of the bobbin case.  Both gave almost, but not entirely acceptable tension balance results. Additionally, the tension seemed to fluctuate on Jellybean, and possibly the other machine as well (testing was more extensive on Jellybean).  Sometimes it would seem that the needle tension was too loose, but a few inches later, with no adjustments from me, it would change, sometimes perfect, sometimes too tight.

The machines were cleaned a couple of days ago, including complete dis-assembly of the upper tensioner, and removal of the bobbin case, and the entire bobbin area cleaned.

Same bobbin case, extremely different results. It seems to point toward the needle tensioner being out of whack, doesn’t it?

Not in this case.  I even swapped the upper tensioners into the opposite machines, and got the same results.   Yuck.  I was hoping for a quick fix, and got a mystery.

The singer manual says the following for adjusting bobbin tension:

To Adjust the Bobbin Thread Tension
First adjust the needle thread tension, as instructed above. Then, using No. 50 mercerised thread in both the needle and the bobbin, and using two thicknesses of thin material under the presser foot, turn the numbered dial by means of the thumb nut, to bring the numeral “4” opposite the center line. A few stitches should now be made in the material and then examined to see if the stitch is properly locked in the material. If the bobbin thread shows on top, the tension on the bobbin thread should be increased. If the needle thread shows on the bottom, the tension on the bobbin thread should be decreased.

 

This is all well and good on a properly adjusted machine with no “issues” needing to be addressed first. No setting on the upper dial would give good consistent tension.

I also noticed in the manual that it says:

Under no circumstances must the screw EE be loosened. The loosening of this screw will change the clearance for the thread between the bobbin case and the bobbin position bracket.

Screw EE being the screw that holds the bobbin positioning finger (bracket) in place.

What the heck, I’m an adventurous person, I threw caution to the wind and decided to disregard Singer’s warning.  Besides, the screw on both machines was far from unmolested.  Both showed evidence of having been tampered with by a non-mechanically inclined person.  Additionally, they both were set a little differently from each other.  In fact, besides the bobbin spring tension, it was the only thing I could find that looked different.

I made careful note of the position of the positioning bracket on each machine so I could put it back if I needed to, and proceeded to disassemble.

The first thing I noticed is that the position of the positioning finger did affect tension, but in the completely opposite way I had guessed that it would.   Moving it further away from the bobbin case caused the upper tension to end up way loose. If the tension was almost balanced at a 4, it now needed a 7 or greater to not ball up under the fabric.

I played with it for a bit longer, on both machines, and didn’t really get the results I was looking for.  By now it’s after midnight, and I’m starting to get frustrated.  I knew I had to be missing something simple.

I hit google and bing and struck out.  I read a bunch of tension tutorials hoping to find something I was missing.  Nope, it didn’t seem like it.   I tried googling for model 99 or 66 specific tension issues and decided that google’s polluted.   The results weren’t useful at all.

Finally, I recalled a set of pdfs, I’d downloaded a few months ago and went looking for them again. This is where I’d originally found them.
http://www.tfsr.org/publications/technical_information/sewing_machine_manual/ – 2016-10-21 The manuals from TFSR are offline at this time.  They can still be found at: https://web.archive.org/web/20160901183756/http:/www.tfsr.org:80/publications/technical_information/sewing_machine_manual/

In the end, this is what saved my sanity.

I basically stepped through this tutorial, and in the end had a working pair of machines, even though the issue I appear to have been suffering wasn’t specifically addressed in the tutorial, it was a comment about what would cause the bobbin tension to not have enough adjustment that lead me to the solution.

Check that there is some resistance when you pull the end of the thread. To be precise, this should be equivalent to 1 oz (28 grams).

You can test this by hanging a 1 oz weight or equivalent (3 one pound coins) from the thread and turning the bobbin holder nearly vertical. The thread should just about leave the bobbin.

In my case, I got a ziplock baggie, and filled it with 1oz of Kamut berries, tied a piece of the thread from the bobbin to it and tested that way.

Adjust the tension as required. If you can’t get enough tension, it is probably one of two reasons:

1. The spring has been damaged. Fit a different one if you have a spare, or note for attention on the checklist.
2. Fluff has collected under the spring. Remove the spring, clean and refit.

When the thread just unwound regardless of the tension, I knew where the problem was, and promptly disassembled the bobbin case spring from the case.

There was oil and debris under the spring, but nothing I would expect to hold it out and make it unadjustable.  Still, I reassembled, and tested again.  I still couldn’t adjust the spring enough. Damn, it looks like the spring must be damaged.

I took a really close look at it (and unfortunately forgot to take a picture) and found that there seemed to be the tiniest little kink in the spring, just past where the thread enters the bobbin case, which meant that it didn’t lay flat against the case, like it does in all of the other tension springs I’ve ever looked at.

Figuring that I had nothing left to lose with this spring, I gently reformed it against the bobbin case, so that it lays flat.  I reassembled and tested again.  The bobbin case held the Kamut off the desk.   Success!

In the long term, I will likely order a new spring (Singer’s part number for this is 32567 and it’s available at some of the online shops.)

Looking a little further in the tutorial, it shows how to properly adjust the bobbin position finger to keep the mechanism working, and somewhat quieter:

The screw hole in the plate is larger than the screw, so that although the plate is held firmly in the notch, the bracket underneath can be moved from side to side. It is essential that, when the screw is finally tightened, there is a small gap between the back of the bobbin holder and the position bracket to allow the top thread to slip through unhindered when the stitch is made.

If a large gap is left, the machine will work, but can be noisy.

 

I learned:

• tension problems can be anywhere, and often not where you think they are
• dirt and oil can get in everywhere! Remove those tension springs and clean in between periodically.  If your bobbin tension is good prior to this, tighten the spring completely, counting the number of turns it takes to get to full tight (write this down), then remove the spring. This will save you the effort of having to “weigh” the tension like I did above.
• Nearly everything is fixable if you approach it logically and with your “eyes open”.  You should never be afraid of disassembling something to take care of it for fear of messing it up.  Be careful, be engaged in what you’re doing, and enjoy it.  This isn’t something that has to be stressful.
• not all parts need to be replaced, sometimes you can fix them, if only as proof of concept
• Sometimes you have to disregard warnings and “Do NOTs” in the original owner’s manuals.  As long as you proceed carefully, and document well so you can put it back together, it’s often OK to do this. This is especially true when working on something that you’re not the original owner of.
• Google’s polluted.
• statistically, you’re never going to be the first person to experience a problem. Also statistically, chances are, your solution is already documented somewhere, you just have to find it.
• pack-ratting is good.  It helps you “find” your solution faster if you already have it saved and stored somewhere close.  😛

Save