NH36 Hairspring not central(?)

[TimD]

I'm new to this and have one NH35 watch and one NH36 watch that I recently assembled myself (not the movements!).  I adjusted the NH35 using Watch Accuracy Meter and it now runs to approx. +/- 2 secs per day.  Good enough.  On the NH36 I was adjusting it and all was going well when suddenly it went way out of wack and I can't get it back.  The WAM is telling me numbers greater than 400secs/day and 4.0ms beat error.  What have I done to suddenly cause this issue?  The movement is from a reputable supplier.  When I compare the two balance wheels running, the NH36 is not as central as the NH35 with the coils looking dangerously close to each other.  Is this the problem?  The arc of the wheel movement also seems way less and the motion of the hairspring is not so smooth as the NH35.  I can see no permanent deformation of the hairspring.  It seems to be free running in the regulator pin.  The stud seems to have a firm hold of the end of the hairspring.  I have not accidentally turned the regulator pin as it seems to be quite firm in its arms.   The plane of the hairspring looks fine.  I have demagnetized the watch.  If I very gently poke the hairspring it moves fine and springs back into place - just the wrong place!  Any thoughts on this would be greatly appreciated. 

[nevenbekriev]

The red arrow shows where the hairspring touches the kock. I can not be so sure from this poin of wiew, but what I see  it seems like the studd holder and regulator have popped up thus causing the problem. The blue arrow shows the direction the carrier should be deiven back in it's place.

Edited February 19 by nevenbekriev

[Nucejoe]

The coil should be level and not rub on any other part, outermost  might be hitting where neven showed or hit the stud as the coil breaths.

I wouldn't touch the regulator arm, shown in blue. Do the correction at the bend ( expand the wide angle at the terminal curve), futhure might need a wee bit of  tweak at the stud. 

Edited February 19 by Nucejoe

[CYCLOPS]

magnetism or oil on hairspring?

[Waggy]

Maybe my eyes are playing tricks on me, but it looks like the coil is slightly concave shaped with the outer coils slightly higher (closer to the cock) than the inner ones creating a dish shape? If this is the case then I would look to see if the coil is hung up in the regulator pins, maybe just needs freeing so the coil can sit lower and level out?

Edited February 19 by Waggy

[TimD]

Wow!  Sharp eyes Neven.  Yes I think it was touching the underside of the cock.  I very gently poked about with the spring a bit more and the hairspring cleared the cock and the rate  went down to around 70s/day and then suddenly if went out of wack again back up to the 600s/day.  So something is not quite right/stable in there.  The amplitude looks very, very low regardless of the rate.  The following two photos were taken after I stopped the movement with the crown and then gently rotated the balance wheel clockwise until it could not be moved.  This 'unwound' the hairspring so I can see better.  You can see on the first photo the hairspring looks to be still touching the underside of the cock.  In the second photo it looks to be clear.  These two photos looks like it does when running.  How do I move/tip the hairspring to get clearance on both sides?   The hairspring could be slightly concave as Waggy noticed but it does not look to be hung up on the stud or Regulator pin.  I just don't see how/where to bend the hairspring to straighten it out.  Also a related question, when I stop the movement with the crown, does the hairspring/balance wheel always stop in the same place?   Thanks again.

[Bert]

Hi Tim, I am new to this forum and also I have been working on Seiko and NH36 Hairsprings lately. I can relate to your dilemma and recently thought an accidental slip whilst adjusting my SKX mod may have caused permanent damage. But, it is possible to fix depending on where your skill level, experience and knowledge have developed. Touching a hairspring that is the most delicate part in the watch is risky and the right tools and special care are needed. So progress slowly with this. Please let me know if you resolved the issue or need more help. As I said, I can relate as I successfully fixed my SKX mod that also has a NH36 movement. After I bent the hairspring I used a timegrapher to see how the movement behaved. And I can share that Beat Error went unstable and way off the desired between 0.0 and 1.0 and Amplitude also was erratic and dropped from 250 degrees to 200. It was also erratic. This amounted to a similar rediculous timekeeping of +- 100s of seconds variance in accuracty. But with the correct approach this can be rescued.

[nevenbekriev]

Sorry for the delay. The toppic has sunk and I have forgotten it.

One of possible reasons is magnetisation of the studd carier/regulator, so they can atract the hairspring by magnetic force when it gets closer up. Try to demagnetise the movement to see what happens. If no improvement, then take out the balance with the cock and put it upside down (balance up), make some pictures and show them.