Raketa 2609.HA Rebanking

[VWatchie]

Apologies for the unclear title—I struggled to summarize the issue. If you have any suggestions for a better title, please let me know.

I recently serviced this Raketa 2609.HA, and overall, I’m very happy with the results. The timing machine shows a straight graph, and the amplitude is strong in all positions—both fully wound and after 24 hours.

However, I’ve noticed an unusual issue. When the mainspring is wound to about 2/3 or more, the movement starts behaving erratically in the dial-up and dial-down positions—but only when the amplitude reaches 295 degrees and above. In contrast:

• In all four vertical positions, it runs perfectly, regardless of the power reserve.
• In horizontal positions, it also runs fine as long as the amplitude is below 295 degrees (i.e., when it’s not fully wound).

I encountered a similar issue with another Raketa 2609.HA, where too much shellac on the entry pallet stone collided with the centre wheel when fully wound. However, that doesn’t seem to be the problem here.

At first, I suspected re-banking, but after listening carefully, I didn’t hear the characteristic galloping sound. The only visible anomaly is the hairspring—the coils appear to be slightly bundled on one side. Could this be causing coil-to-coil contact and how bad is that?

Additional Observations

In this video, the mainspring is only 2/3 wound, and the timing graph looks perfect.

In this video, the mainspring is fully wound (I really can't see any difference from the previous video, which is a bit concerning).

If the hairspring is the issue, I’d appreciate any hands-on guidance. I know you experienced watch repairers often adjust it in situ, whereas enthusiasts like me tend to overthink and remove it before making adjustments. I’m open to either approach—whatever it takes!

(One unpleasant discovery: While troubleshooting, I found a 20mm-long thin white fibre (possibly dust or skin residue) wrapped around the roller. I initially thought this might be the cause but removing it didn’t make any difference. You can see part of this fibre in the 5th picture below.)

 

[nevenbekriev]

Press a little here

 

[Knebo]

2 hours ago, VWatchie said:

Apologies for the unclear title—I struggled to summarize the issue. If you have any suggestions for a better title, please let me know.

I recently serviced this Raketa 2609.HA, and overall, I’m very happy with the results. The timing machine shows a straight graph, and the amplitude is strong in all positions—both fully wound and after 24 hours.

However, I’ve noticed an unusual issue. When the mainspring is wound to about 2/3 or more, the movement starts behaving erratically in the dial-up and dial-down positions—but only when the amplitude reaches 295 degrees and above. In contrast:

• In all four vertical positions, it runs perfectly, regardless of the power reserve.
• In horizontal positions, it also runs fine as long as the amplitude is below 295 degrees (i.e., when it’s not fully wound).

I encountered a similar issue with another Raketa 2609.HA, where too much shellac on the entry pallet stone collided with the centre wheel when fully wound. However, that doesn’t seem to be the problem here.

At first, I suspected re-banking, but after listening carefully, I didn’t hear the characteristic galloping sound. The only visible anomaly is the hairspring—the coils appear to be slightly bundled on one side. Could this be causing coil-to-coil contact and how bad is that?

Additional Observations

In this video, the mainspring is only 2/3 wound, and the timing graph looks perfect.

In this video, the mainspring is fully wound (I really can't see any difference from the previous video, which is a bit concerning).

If the hairspring is the issue, I’d appreciate any hands-on guidance. I know you experienced watch repairers often adjust it in situ, whereas enthusiasts like me tend to overthink and remove it before making adjustments. I’m open to either approach—whatever it takes!

(One unpleasant discovery: While troubleshooting, I found a 20mm-long thin white fibre (possibly dust or skin residue) wrapped around the roller. I initially thought this might be the cause but removing it didn’t make any difference. You can see part of this fibre in the 5th picture below.)

 

Catchy thread title, Håkan 

While I am by no means a hairspring master like Neven, I don't see anything wrong with your hairspring. The terminal coil does seem to be closer (or even touching) the outer regulating pin, though. Hence, Neven's suggestion, no ? But I would not expect it to be the source of the behaviour you show on the TG. 

Before bending anything, maybe check two things:

1. Verify amplitude visually with super slow-mo video when the strange behaviour is occurring. Maybe the lift angle is somehow wrong (I can see that you put the usual 42° that it should be, but maybe it has been altered be a previous watch destroyer). I would still see rebanking as the No1 suspect. Everything seems to point there. 

2. How did you clean the hairspring? Any chance that the coils are sticking a bit when they touch at high amplitude? I've seen the coils come extremely close in my recent Vostok, but it didn't impact the rate/timegrapher readings. I've also had cases (particularly with Russian watches) where old/contaminated degreaser led to sticky coils. 

[mikepilk]

At high amplitude the coils are clearly touching near the end of the terminal curve, which would explain your timegrapher plot. You can see the much narrower gap where I've put the arrows compared to the bottom of the spring. 

Pressing where Nev suggests should open the gap. If not, just tweak it until the gap looks even.

[VWatchie]

Thanks for your input guys! Much appreciated!

I have some additional questions and concerns but I'll have to get back about it as I'm currently busy working.

[VWatchie]

On 2/24/2025 at 8:52 PM, nevenbekriev said:

Press a little here

 

Does it make a difference if I push where the red arrow points? I'm asking because I was reviewing an old thread about another movement with bunched-up coils, where @nickelsilver suggested applying pressure to where the red arrow points.

 

[nevenbekriev]

It will have the same effect. First apply some pressure and see what happens. This will be elastic and will return back if the force is small enough. Thus You will be able to see what the effect will be if pressing more to achieve non elastic deforming.

[nickelsilver]

Double check that it's centered in the regulating pins when you're happy with the overall centering. This can be adjusted at the stud without affecting the overall centering.

[oldhippy]

Does this movement suffer from a “psychotic disorder”?

I do.

[RichardHarris123]

1 minute ago, oldhippy said:

Does this movement suffer from a “psychotic disorder”?

I do.

I think most of us do OH. 

[VWatchie]

After the adjustment, the spring looks a lot better. A special thanks to @nickelsilver and @nevenbekriev for the guidance!

However, the problem remains. As soon as the amplitude reaches near 325 degrees, the "disorder" on the timing machine reappears, especially in the dial-up position where the amplitude gets slightly larger. Perhaps I should mention that after a while, the amplitude drops a bit, and the graph becomes normal, but as soon as the amplitude increases again, the disorder returns. BTW, the Before/Afer image is in HD if you wish to zoom in.

I certainly didn't hear a galloping sound, so I didn't suspect re-banking. Anyway, to make sure, I decided to make a mark on the balance rim at 180 degrees opposite its resting position to manually determine the lift angle, and it would seem it is closer to 46 than 42 degrees. BTW, apologies for the background noise in the videos!

After this, I made another mark 180 degrees opposite the first mark (balance in its resting position). I then wound the mainspring fully and recorded the video. I don't think it is re-banking. Then again, the fork horns are quite wide. I find it very difficult to tell. What do you think? Remember, I can't hear a galloping sound 

Does the tweaking of the hairspring look good enough now, or does it need more tweaking? I'm sure I'm using the original mainspring (T-shaped bridle). Perhaps move the pallets out a bit? Then why should I if it's not re-banking?

I'm out of ideas of where to go from here.

 

 

Edited February 28 by VWatchie

[Knebo]

6 hours ago, VWatchie said:

After the adjustment, the spring looks a lot better. A special thanks to @nickelsilver and @nevenbekriev for the guidance!

However, the problem remains. As soon as the amplitude reaches near 325 degrees, the "disorder" on the timing machine reappears, especially in the dial-up position where the amplitude gets slightly larger. Perhaps I should mention that after a while, the amplitude drops a bit, and the graph becomes normal, but as soon as the amplitude increases again, the disorder returns. BTW, the Before/Afer image is in HD if you wish to zoom in.

I certainly didn't hear a galloping sound, so I didn't suspect re-banking. Anyway, to make sure, I decided to make a mark on the balance rim at 180 degrees opposite its resting position to manually determine the lift angle, and it would seem it is closer to 46 than 42 degrees. BTW, apologies for the background noise in the videos!

After this, I made another mark 180 degrees opposite the first mark (balance in its resting position). I then wound the mainspring fully and recorded the video. I don't think it is re-banking. Then again, the fork horns are quite wide. I find it very difficult to tell. What do you think? Remember, I can't hear a galloping sound 

Does the tweaking of the hairspring look good enough now, or does it need more tweaking? I'm sure I'm using the original mainspring (T-shaped bridle). Perhaps move the pallets out a bit? Then why should I if it's not re-banking?

I'm out of ideas of where to go from here.

 

 

I'm still convinced you're dealing with re-banking. 

The 325° amplitude you mentioned, is that with the lift angle set to 42 or 46° ? The difference in setting should easily add 25° of amplitude. 

I'm not sure how pronounced the galloping sound is if the rebanking is marginal (i.e. just "blocking" only the last hypothetical 5° of an oscillation) 

[GPrideaux]

I'm no expert, and the numbers do suggest re-banking, but it also looks to me that coils of the spring might still be touching when the amplitude is large. In the video, compare the coil spacing where it goes under the cock (at about 5 o'clock in the pic) with where they go under the stud carrier 180 degrees away.

 

Edited March 1 by GPrideaux

[Neverenoughwatches]

On 2/26/2025 at 1:58 PM, VWatchie said:

Does it make a difference if I push where the red arrow points? I'm asking because I was reviewing an old thread about another movement with bunched-up coils, where @nickelsilver suggested applying pressure to where the red arrow points.

 

I wouldn't think it makes that much difference between the two pushing points. The aim being looks to put pressure at that first bend where the terminal curve starts, as opposed to where the spring exits the stud. I'd have thought with the regulator being close to the stud...the spring will lean against the regulator pin for leverage and push the coils further out at around one oclock where the terminal curve finishes.

21 hours ago, VWatchie said:

After the adjustment, the spring looks a lot better. A special thanks to @nickelsilver and @nevenbekriev for the guidance!

However, the problem remains. As soon as the amplitude reaches near 325 degrees, the "disorder" on the timing machine reappears, especially in the dial-up position where the amplitude gets slightly larger. Perhaps I should mention that after a while, the amplitude drops a bit, and the graph becomes normal, but as soon as the amplitude increases again, the disorder returns. BTW, the Before/Afer image is in HD if you wish to zoom in.

I certainly didn't hear a galloping sound, so I didn't suspect re-banking. Anyway, to make sure, I decided to make a mark on the balance rim at 180 degrees opposite its resting position to manually determine the lift angle, and it would seem it is closer to 46 than 42 degrees. BTW, apologies for the background noise in the videos!

After this, I made another mark 180 degrees opposite the first mark (balance in its resting position). I then wound the mainspring fully and recorded the video. I don't think it is re-banking. Then again, the fork horns are quite wide. I find it very difficult to tell. What do you think? Remember, I can't hear a galloping sound 

Does the tweaking of the hairspring look good enough now, or does it need more tweaking? I'm sure I'm using the original mainspring (T-shaped bridle). Perhaps move the pallets out a bit? Then why should I if it's not re-banking?

I'm out of ideas of where to go from here.

 

 

I'm with the rebanking idea. If it were the coils touching I'd have thought that the effect would remain, for a while at least. If its rebanking then that is an impact, a more severe action that takes place and is quickly delt with until the amplitude catch back up.

Interesting  to discover which symptom the watch has, but whatever, the root cause appears to be too much amplitude. The cure would be to reduce it...oil the lever pivots,  change the mainspring for a weaker one, flatten off the balance pivots which would better average out the positional amplitudes. 

A good habit to get into is measuring the mainspring for correctness when servicing it. Someone might have put a stronger spring in for some reason. Possibly a reason that you've fixed during a service or an adjustment that you've made. Handy to know whats in the barrel before diagnosing an issue. Similarly a doctor should want to know information about a patient's heart,  have they got a pacemaker fitted ? before making comment on their blood pressure. 

Edited March 1 by Neverenoughwatches

[nevenbekriev]

 The amplitude looks to be about 320 degr.  If You need to know if it is rebanking, then force it by adding torque winding by the crown and holding after the full wind, if the amplitude can rise more than what it is on the video, then still not rebanking.

Edited March 2 by nevenbekriev

[VWatchie]

On 2/28/2025 at 7:40 PM, Knebo said:

I'm still convinced you're dealing with re-banking.

I'm starting to lean toward that conclusion as well.

On 2/28/2025 at 7:40 PM, Knebo said:

The 325° amplitude you mentioned, is that with the lift angle set to 42 or 46° ?

46°! The movement is specified to a 42° lift angle, but I have to adjust my Weishi 1900 to 46° for it to display 180° when the balance wheel is actually swinging 180°. I find it strange and have no theory as to why, but I still believe that 46° is more accurate, at least for my particular example of the movement.

On 2/28/2025 at 7:40 PM, Knebo said:

I'm not sure how pronounced the galloping sound is if the rebanking is marginal (i.e. just "blocking" only the last hypothetical 5° of an oscillation)

This is also strange, and I don't have enough experience to know whether the galloping sound always accompanies a movement that is re-banking. Perhaps we are dealing with re-banking, but to such a minimal extent that it isn’t audible. I’ve recorded the sound and increased the volume but haven’t heard the slightest hint of a galloping noise. Puzzling!

On 3/1/2025 at 3:33 AM, GPrideaux said:

I'm no expert, and the numbers do suggest re-banking

Yes, it seems like that is probably the explanation, but it would have been nice to be able to verify it audibly.

On 3/1/2025 at 3:33 AM, GPrideaux said:

but it also looks to me that coils of the spring might still be touching when the amplitude is large. In the video, compare the coil spacing where it goes under the cock (at about 5 o'clock in the pic) with where they go under the stud carrier 180 degrees away.

You’re right, there is definitely a difference, but if you look at the picture of the adjusted hairspring when the balance wheel isn't moving (the AFTER picture), the difference is nonexistent. I’ve never thought about it, but maybe that’s how it always looks? I’m not an expert either, so it would be extremely interesting to get an assessment, or maybe even a qualified guess, from @nickelsilver and/or @nevenbekriev!?

On 3/1/2025 at 8:38 AM, Neverenoughwatches said:

I'm with the rebanking idea. If it were the coils touching I'd have thought that the effect would remain

This could be a strong argument for the theory of re-banking being the correct one. Again, I lack enough knowledge and experience to know how a movement behaves when the coils of the hairspring collide with each other. I actually asked this question in my first post: 'Could this be causing coil-to-coil contact and how bad is that?' So if anyone knows, it would also be very interesting to gain more knowledge about it.

On 3/1/2025 at 8:38 AM, Neverenoughwatches said:

Interesting  to discover which symptom the watch has, but whatever, the root cause appears to be too much amplitude.

I agree! The only thing we know for sure is that the symptoms occur at high amplitude, and since we don't want the amplitude to be too high, we could focus on trying to reduce the amplitude and worry less about what it exactly causes.

On 3/1/2025 at 8:38 AM, Neverenoughwatches said:

oil the lever pivots,  change the mainspring for a weaker one, flatten off the balance pivots which would better average out the positional amplitudes.

Thank you for the suggestions, but unfortunately, I don’t find any of them particularly attractive.

Oiling the pallet fork staff pivots is a temporary, albeit very tempting, solution. The problem will most likely resurface at the next service, regardless of whether you, I, or anyone else performs it. Granted, I’m an amateur, but I still have the ambition to try to work as seriously as I can. Perhaps it is still the most rational solution. Hmm...

I don't have the specifications for the original mainspring, but long story short, I find it relatively unlikely that someone has replaced the original mainspring. The movement was in good condition but had a broken winding stem, which was probably the previous owner's attempt to counteract dried-up oils. And when I look at the screws, they don't appear completely untouched, but they do seem to be in very good condition. I believe I’m the first to try repairing the movement, but of course, you can never be completely sure.

It's funny that you suggest flatter balance staff pivots because I was surprised by how unusually flat they were when I looked at them while disassembling the movement. So no, I don't think there's any room to flatten them more than they already are.

The action I’m leaning towards, since the we're dealing with fixed banking, is the one @nickelsilver recommends in this post. What do you and the others think about it?

33 minutes ago, nevenbekriev said:

The amplitude looks to be about 320 degr.  If You need to know if it is rebanking, then force it by adding torque winding by the crown and holding after the full wind, if the amplitude can rise more than what it is on the video, then still not rebanking.

Great tip, I'll definitely try that! Thanks! 

[mbwatch]

On 2/28/2025 at 6:17 AM, VWatchie said:

apologies for the background noise in the videos

I'm listening for rebanking but what I belive I am hearing is George Harrison stuck in there. He was the smallest of the 4 after all.

[nevenbekriev]

As I actually don't expect this to be real rebanking, will try to say what else it can be. The hairspring of 2609HA has small gaps between coils. This brings to danger when the spring is wound (shrank) to the maximum balance amplitude the coils to touch or get very close to touching. More, the spring can vibrate (ring) and this will cause coils to touch when they are close. Now, What we can do is to try to better center the spring but not in rest position - in the max shrank position. AND, decrease the ability (the main reason) for ringing  by closing the distance between regulator pins as much as possible so no free play to be at all and still the spring not to get stuck when moving regulator. On the video, there is hairspring ringing, this is well seen when the playback speed is reduced

[VWatchie]

14 hours ago, nevenbekriev said:

On the video, there is hairspring ringing, this is well seen when the playback speed is reduced

Thanks for your insight and advice! Much appreciated! 

I’m having trouble seeing the vibration (ringing), probably because I don’t know what to look for. In which video is the vibration most visible? Does the vibration occur sideways or vertically, or maybe both? In short, what should I be looking for?

14 hours ago, nevenbekriev said:

Now, What we can do is to try to better center the spring but not in rest position - in the max shrank position.

How do you do that in practice? I assume that by "max shrank position," you mean when the hairspring is at its most compressed, that is, when the coils are closest to each other, yes?

14 hours ago, nevenbekriev said:

AND, decrease the ability (the main reason) for ringing  by closing the distance between regulator pins as much as possible so no free play to be at all and still the spring not to get stuck when moving regulator.

That’s also something I find unclear in terms of how it should be done. As far as I understand it, the index pin should always be parallel to the boot.

Here is my attempt at illustrating how I interpret your description of what needs to be done. Is this correct?

If so, I’m also wondering how this is done in practice. Do you first bend the regulator pin at its attachment point under the arm and then bend it back slightly lower down, towards the anti-shock setting?

Practical advice on how to proceed would be greatly appreciated!

Edited March 3 by VWatchie
Replaced "regulator pin" with "index pin"

[Knebo]

6 hours ago, VWatchie said:

The action I’m leaning towards, since the we're dealing with fixed banking, is the one @nickelsilver recommends in this post. What do you and the others think about it?

I'd say that's the most elegant solution. 

[RichardHarris123]

On 2/26/2025 at 4:52 PM, oldhippy said:

Does this movement suffer from a “psychotic disorder”?

I do.

 What degrees of amplitude is required for rebanking?  Obviously it will depend on the distance between the banking pins, compared to a full circle, but there must be a rule of thumb.  

[Neverenoughwatches]

6 hours ago, RichardHarris123 said:

 What degrees of amplitude is required for rebanking?  Obviously it will depend on the distance between the banking pins, compared to a full circle, but there must be a rule of thumb.  

It can vary from movement to movement. Since the reason being is the impulse pin striking the outside of the fork . The lift angle, the fork horns and impulse pin size being the deciding factors.  I wondered about reducing the fork horn but that also upsets the lever poise.

[VWatchie]

2 hours ago, Neverenoughwatches said:

I wondered about reducing the fork horn but that also upsets the lever poise.

True, but I generally admire your talent for thinking outside the box!  I need more of that.

[ColdWind]

Call me crazy but I do like to think outside the box. Once in a while I would have to fix customer's Chinese watch. These watches are equiped with low-end cheap Chinese automatic movements, so you would imagine the quality is not so great, the bridle of the mainspring is a bit too long which result in a watch re-banking like hell. So what did I do? I trimmed the bridle down a bit so it would be shorter. Then the watch would never be re-banking again and customers are sastified. I mean they are cheap Chinese watch, replicas and fakes even, and I can't think what else I could do when part of the movement is not right in the first place. But you are dealing with a manual wind mainspring so my thinking outside of the box is that I would bust out my dremmel and work on thin down the spring by just a bit. I never done it before thought. Also what would you do in both situations when the watch, being automatic or manual, come with a too strong mainspring originally? I saw a guy on Youtube put a bunch of something looks like Molykote DX on the barrel wall of the automatic barrel but I wouldn't think it as a permanent and reliable fix.

Edited March 3 by ColdWind

[tomh207]

Box? There’s a box? Wish I knew, would make things much easier.

 

Tom