I can't figure out if the kink is supposed to be here or if it got bent when mom sent it through the washer and dryer.

The relationship of the final arc over the field of the main coils looks pretty good to me. It lays flat and true, seems to oscillate fine.

Just looks odd to me. In the bottom left where the curve begins then is curved back to finish the radius to the pin.

I think you have a slight inward bend starting about where the red arrow is, working counterclockwise.

Mike,

Sometimes it's so obvious... I was so focused on the odd lazy S shaped bend for the over coil that I never even saw the sistered coils below.

I'm having trouble with erratic beat error and generally inconsistent timing numbers.

Sometimes I have full action, great time and beat error locked in at or about 5ms... and then I get beat error that floats from 4 to 9 and action drops to about 50 percent, but stays pretty close to 18000 beats per hour... bouncing 50 under to 50 over...

I thought I might have a loose staff rivet, loose roller/jewel or even a loose hair spring collet... but I've checked and double checked all of them a half dozen times.

Alternately, I was having trouble with the timing being way fast on a test run, followed by way slow on the next test run...

Overlooking a coil that is locking up and unlocking would totally explain all of the above.

I'll update this tomorrow or Saturday when I get my next chance to spend a few minutes at my bench.

Thanks a mil,

Paul

OK, It was poker night and someone brought chocolate covered espresso beans to snack on.

Between that and Mike's probable solution to my problem... I had to give up the notion of sleeping.

It took 15 minutes to pull the balance assembly from this anemic watch. Pull the hair spring, verify all my rotational tightness checks for the roller and staff and then select tools to recurve the hair spring.

The gar fish shaped tweezers were perfect for the task, ran them around the outer coil to the area Mike marked up and began gently squeezing and walking around the arc to the other side.

It took about three passes to gently open the curve of the outer coil to match the rest of the coils.

Put her back together and now we are in business.

Beat error locked on 4.5ms with no float or deviation. BPH = 18000 plus or minus 1 or less in six positions.

In my line of work, we refer to this as "being too close to the problem or too close to the solution"

I really appreciate the review and recommendation.

R/Paul-T

Hi Paul, I'm happy to have been able to help.
Many times if you can find the exact place of the bend you can "anchor" it with one tweezer and stroke the remaining coils outward (in this case) gently with a needle or similar tool to remove the bend and have everything fall into place.
A good microscope helps.
Regards,
Mike

There is a simple tutorial video on youtube how to fix hairspring basic issues. Just search for watch repair channel - nice videos.

Watch Repair Channel

Rafal