Trying to source mainsprings and nothing comes up.

[Birbdad]

So i'm officially done trying to use a mainspring winder on thin seiko mainsprings, it's just a crapshoot whether it will destroy them it seems, something i'd heard about but only just now experienced when my winder ate two of them. It's fine, they were old and probably should have been replaced anyways. THese are the measurements my calipers came up with. I'm rounding slightly up or down to standard sizes and nothing comes up on cousins because of course it didn't. NOS ones are crazy expensive. I'm trying to be more independent at this but every time i try to source generic parts like this i come up empty handed.

Specifically what i want is:
7s26 mainsprings.
4006 mainspring.
4006 alarm mainspring.

If ANYBODY has cousins ref numbers or general resortes ref numbers for these that confirm work i'd really appreciate it.

[Waggy]

I feel your pain, I pretty much always hand wind Seiko springs as the odds of getting it back in the barrel are much greater. For new springs here I what I use:

 

7009, 7S26/36 and NH35/36 I use the below from Ali Express, I was a little skeptical at first, but I an now very happy with them.

For the 6309 I use an ETA 2892a2 (see pdf attached for CousinsUK reference material), there other equivalents I have tried but they have too large a center coil and the arbor cannot hook it, but these work well (as long as you install them the right way around).

 

Equivalent Mainspring ETA 2892a2.pdf

[Birbdad]

4 minutes ago, Waggy said:

I feel your pain, I pretty much always hand wind Seiko springs as the odds of getting it back in the barrel are much greater. For new springs here I what I use:

 

7009, 7S26/36 and NH35/36 I use the below from Ali Express, I was a little skeptical at first, but I an now very happy with them.

 

For the 6309 I use an ETA 2892a2 (see pdf attached for CousinsUK reference material), there other equivalents I have tried but they have too large a center coil and the arbor cannot hook it, but these work well (as long as you install them the right way around).

 

Equivalent Mainspring ETA 2892a2.pdf 223.82 kB · 0 downloads

Thanks for that. I might give em a shot. What i'm really trying to do is to figure out how to source generic mainsprings in general. I have to make another cousins order anyways and if i coudl find what i need and get it all in one place that would be ideal, i hear GR mainsprings are fantastic and i'd love to try them. Professional watchers can just measure a barrel and arbor and spring and find high quality generics...i want to know how the heck you do this.

So googling other places all i could find was this https://www.cousinsuk.com/product/under-100mm-height?code=GR2378X It was listed in a forum as the correct mainspring for a 7s26, yet when you click on it it only lists it as correct for 4006. So this might be correct for both as they are fairly similar barrels, the arbors are different sizes but I know how to close an inner coil on an arbor to fit a smaller one. It's also a lot of time to wait and money to spend to just find out if something works -_-. 

So that brings me to the alarm mainspring for the 4006. Given how expensive these are and how few seem to be out there you would think if a generic worked for it somebody would have found it. THe only hit i've gotten on it is ancient and the one they bought didn't work.

[Waggy]

Try the files below, I had to trick the system into thinking they were pdfs to upload it so just delete the .pdf and leave it ending .xlsx

The first lets you find the GR reference from the watch movement information and the second lets you calculate the spring dimensions from the barrel information, based on the youtube from Watch Repair Tutorials

hopefully they are useful

 

Mainspring GR Data.xlsx.pdf New Spring Sizer.xlsx.pdf

[Birbdad]

9 minutes ago, Waggy said:

Try the files below, I had to trick the system into thinking they were pdfs to upload it so just delete the .pdf and leave it ending .xlsx

The first lets you find the GR reference from the watch movement information and the second lets you calculate the spring dimensions from the barrel information, based on the youtube from Watch Repair Tutorials

hopefully they are useful

 

Mainspring GR Data.xlsx.pdf 775.78 kB · 0 downloads New Spring Sizer.xlsx.pdf 5.93 kB · 0 downloads

I'll check it out! Also great channel with a great video i forgot he made on the subject!

So arbor dimensions i see in other places are a common issue people run into with mainsprings and generics. Mainly like what you said, the inner coil is too large. He also has a video on how to reshape them without breaking the thing.

Edited February 14 by Birbdad

[Waggy]

Here is the link to the specific video I make my spreadsheet from

 

Edited February 14 by Waggy

[Birbdad]

2 minutes ago, Waggy said:

Here is the link to the specific video I make my spreadsheet from

 

Hah, ok, we watched the same vid! Taking a look. Looks like there's no entry for 7s26 and the 4006 alarm spring. Maybe if i find something that works you can update it. 

 

[Waggy]

I'm sure there is a solution out there, always more than one way to skin a cat

[Birbdad]

3 minutes ago, Waggy said:

I'm sure there is a solution out there, always more than one way to skin a cat

Well if we can figure it out it'll be here to help others. 

That said cousins actually lists that one i linked to earlier as the specific one for the 4006. GR2378X Might be worth it to update your fantastic little database!

Edited February 14 by Birbdad

[mikepilk]

The 4006 mainspring is GR2378X as shown by Cousins (which agrees with the GR database).

I downloaded all the GR database PDFs from Cousins, but the spreadsheet is much easier, or go online here 

https://watchguy.co.uk/cgi-bin/mainsprings

I can't find any reference to the alarm spring. I have two BellMatics and didn't have to change either alarm spring, so didn't measure them.  Just use your measurements (are you sure about your thickness, it doesn't agree with the GR database?).   As it's only ringing the alarm you don't need to be an exact match.

People have listed various springs to use for 7S26. My last one, I used GR2377X

[Birbdad]

11 minutes ago, mikepilk said:

The 4006 mainspring is GR2378X as shown by Cousins (which agrees with the GR database).

I downloaded all the GR database PDFs from Cousins, but the spreadsheet is much easier, or go online here 

https://watchguy.co.uk/cgi-bin/mainsprings

I can't find any reference to the alarm spring. I have two BellMatics and didn't have to change either alarm spring, so didn't measure them.  Just use your measurements (are you sure about your thickness, it doesn't agree with the GR database?).   As it's only ringing the alarm you don't need to be an exact match.

People have listed various springs to use for 7S26. My last one, I used GR2377X

Fantastic! And so instead of measuring the thickness with calipers i just calculated the strength from the measurements from the video above. 

It looks like for the bellmatic alarm it's gonna be GR3650 is a pretty close match. 

And that one you recommended for the 7s26 it lists the 7005 as an exact match. I don't know about the 7005 but i know from the 7006 to the nh36 and everything in between it's the same mainspring so that's likely an exact match. 

This has been illuminating and i thank you all for your help, i'll report back when all this stuff arrives so we can see if it all works, specifically that alarm spring. 

Edited February 14 by Birbdad

[spectre6000]

I recall a while back watching a horology lecture about sizing mainsprings. Probably on YouTube. One of those unflashy, classroom style, guy in front of a whiteboard sort of affairs. Went through the whole deal on the formula for how to size a mainspring to a barrel. I imagine what you're missing in your attempts at going solo sourcing these is knowing where and how to fudge. I.e., you can get away with a thicker spring, just has to be longer by X% per .001mm thickness provided you don't exceed Ymm^3 or something. Learning the hows and whys of mainspring sizing will probably open your options up considerably by opening up the specific size to a range of sizes.

Edited February 14 by spectre6000

[Birbdad]

12 hours ago, spectre6000 said:

I recall a while back watching a horology lecture about sizing mainsprings. Probably on YouTube. One of those unflashy, classroom style, guy in front of a whiteboard sort of affairs. Went through the whole deal on the formula for how to size a mainspring to a barrel. I imagine what you're missing in your attempts at going solo sourcing these is knowing where and how to fudge. I.e., you can get away with a thicker spring, just has to be longer by X% per .001mm thickness provided you don't exceed Ymm^3 or something. Learning the hows and whys of mainspring sizing will probably open your options up considerably by opening up the specific size to a range of sizes.

Yeah my mistake was entering too much info into the cousins mainspring search. I found near perfect matches once i just put in the height, calculated the length and strength using the formula from the video above and found lots of good options. It wasn't till i rewatched the video above that i remembered there is a lot of leeway on mainsprings.

Do you have a link to that video by chance? 

[mikepilk]

5 hours ago, Birbdad said:

 there is a lot of leeway on mainsprings.

No there isn't. 

The height has to be close to fit in the barrel correctly. 

A spring is like a beam bending, whose moment of inertia is (height*thickness³)/12.  i.e. it varies with thickness cubed. So if you go from a thickness of 0.11 to 0.12mm, the 'strength' increases by (0.12/0.11)³ = 30%. That's a lot.
So you need to be close on thickness.

The length is not as critical, as long as it's not too long to fit properly. If it's shorter than specified, you will just lose out on run time.

 

[Jon]

I use a mainspring for the Seiko 7009 for the 7S26C and get 300 degrees amplitude. The size is 0.95 x 0.115 x 380 x 10.

Those that say Seiko's only get a low amplitude is nonsense. I've never serviced a Seiko without seeing an amplitude of at least 270 degrees. The trick is using a new mainspring, which seems pretty obvious to me, but hey!

I've recently serviced a Seiko 7019A, which is over 30 years old with an amplitude of 290 to 300. a 7S26C should definitely get a great amplitude! Every student that services the 7S26C, B and even A variant gets well over 270 degrees amplitude. The problem is most Seiko's never get serviced regularly because it isn't cost effective, so they get run into the ground and parts wear, especially the barrel bridge arbor hole, but if the parts are good, then a good amplitude will be had the majority of the time. Seiko's with low amplitude seems to be an old wives tale that someone else then repeats and if it is said long enough it seems to become a fact, just like an 'over-wound' watch. Don't ever oil the pallet arbor jewels, even though it is recommended in the Seiko service manual. That is a sure way of losing a lot of amplitude. If I had a penny for every time I heard someone say Seiko's have low amplitude......

Edited February 15 by Jon

[spectre6000]

11 hours ago, Birbdad said:

Do you have a link to that video by chance? 

No. Sorry. If I recall correctly, I watched it in a sleep deprived state while bottle feeding my daughter, who is now 4. It probably got in front of me by way of a link at the bottom of some other video to boot, so I wouldn't even know how to find it really short of throwing search terms into the bar.

[Birbdad]

50 minutes ago, Jon said:

I use a mainspring for the Seiko 7009 for the 7S26C and get 300 degrees amplitude. The size is 0.95 x 0.115 x 380 x 10.

Those that say Seiko's only get a low amplitude is nonsense. I've never serviced a Seiko without seeing an amplitude of at least 270 degrees. The trick is using a new mainspring, which seems pretty obvious to me, but hey!

I've recently serviced a Seiko 7019A, which is over 30 years old with an amplitude of 290 to 300. a 7S26C should definitely get a great amplitude! Every student that services the 7S26C, B and even A variant gets well over 270 degrees amplitude. The problem is most Seiko's never get serviced regularly because it isn't cost effective, so they get run into the ground and parts wear, especially the barrel bridge arbor hole, but if the parts are good, then a good amplitude will be had the majority of the time. Seiko's with low amplitude seems to be an old wives tale that someone else then repeats and if it is said long enough it seems to become a fact, just like an 'over-wound' watch. Don't ever oil the pallet arbor jewels, even though it is recommended in the Seiko service manual. That is a sure way of losing a lot of amplitude. If I had a penny for every time I heard someone say Seiko's have low amplitude......

I don't even always use a new mainspring and i get at a minimum 275 amp and amazing accuracy if i take the time to do a little work on the hairpsring.

Heck i serviced a 7006 that was the most trashed movement i've ever worked on, horribly abused. It gets 285 amp.

My last two services of 7s26's i got about 275 amplitude, on one i got a delta of 6, on the other a delta of 8. On the first one average rate deviation ofver a week was 1 seconds per day and on the next one about 1.5 seconds per day. 

I too constantly hear about what crap movements seiko makes, meanwhile i get consistently amazing results when i work on them.

And you don't need to tell me to ignore the seiko service sheets lol. They're a total trainwreck. I use modern oiling theory and principles.

[Birbdad]

On 2/14/2024 at 3:47 AM, mikepilk said:

The 4006 mainspring is GR2378X as shown by Cousins (which agrees with the GR database).

I downloaded all the GR database PDFs from Cousins, but the spreadsheet is much easier, or go online here 

https://watchguy.co.uk/cgi-bin/mainsprings

I can't find any reference to the alarm spring. I have two BellMatics and didn't have to change either alarm spring, so didn't measure them.  Just use your measurements (are you sure about your thickness, it doesn't agree with the GR database?).   As it's only ringing the alarm you don't need to be an exact match.

People have listed various springs to use for 7S26. My last one, I used GR2377X

So i just put in the 2377x into a 7s26 barrel. It looks like the inner coil is too damn big for the arbor. Have you had an issue with this?

 

On 2/15/2024 at 3:00 AM, mikepilk said:

No there isn't. 

The height has to be close to fit in the barrel correctly. 

A spring is like a beam bending, whose moment of inertia is (height*thickness³)/12.  i.e. it varies with thickness cubed. So if you go from a thickness of 0.11 to 0.12mm, the 'strength' increases by (0.12/0.11)³ = 30%. That's a lot.
So you need to be close on thickness.

The length is not as critical, as long as it's not too long to fit properly. If it's shorter than specified, you will just lose out on run time.

 

I'm basing what i said on this vid. THis guy generally gives great advice. Length not being critical would fall under the category of there being leeway. I"m just saying you don't have to have one identical to the factory one from what i understand.

 

[Neverenoughwatches]

1 hour ago, Birbdad said:

So i just put in the 2377x into a 7s26 barrel. It looks like the inner coil is too damn big for the arbor. Have you had an issue with this?

 

I'm basing what i said on this vid. THis guy generally gives great advice. Length not being critical would fall under the category of there being leeway. I"m just saying you don't have to have one identical to the factory one from what i understand.

 

Is that a new spring you've put in there col ?  It's uniformity isn't perfect on the last coil, bit of a bend, couldn't say if this would have any effect with the arbor hook not catching. Have you tried any inner coil tightening techniques ?. Length is considered as not critical unlike the height and thickness, that has to fit the barrel height correctly and provide the right amplitude. Length would give more runtime and I think you do need to be pretty close to the movements designed length, a few centimetres either way ? I still imagine more or less coils will act differently as regards to the amount of energy that is delivered.  Too many or too little coils and the spring might struggle to unwind or have too much tension ?

[Birbdad]

10 minutes ago, Neverenoughwatches said:

Is that a new spring you've put in there col ?  It's uniformity isn't perfect on the last coil, bit of a bend, couldn't say if this would have any effect with the arbor hook not catching. Have you tried any inner coil tightening techniques ?. Length is considered as not critical unlike the height and thickness, that has to fit the barrel height correctly and provide the right amplitude. Length would give more runtime and I think you do need to be pretty close to the movements designed length, a few centimetres either way ? I still imagine more or less coils will act differently as regards to the amount of energy that is delivered.  Too many or too little coils and the spring might struggle to unwind or have too much tension ?

It's the brand new spring mikepilk says he uses for this movement. And no i haven't tried any inner coil tightening technique yet, far as i know they all involve annealing the coil which would mean i would have to take it out of the barrel which means since it's a short seiko spring there's a 50% chance my winder would eat it trying to put it back in. 

Is there any way to tighten up that inner coil with it in the barrel?

[Neverenoughwatches]

38 minutes ago, Birbdad said:

Is there any way to tighten up that inner coil with it in the barrel?

I don't imagine so Col. You would need to have good access to the coil to be able to manipulate it into shape. Any attempt to do it in the barrel might risk damaging the soft brass arbor hole. Unless someone has some fancy gadget that i haven't seen yet they would take it out to work on in. I would look at getting the right tool to fit these springs safely .

[Birbdad]

1 minute ago, Neverenoughwatches said:

I don't imagine so Col. You would need to have good access to the coil to be able to manipulate it into shape. Any attempt to do it in the barrel might risk damaging the soft brass arbor hole. Unless someone has some fancy gadget that i haven't seen yet they would take it out to work on in. I would look at getting the right tool to fit these springs safely .

There is no mainspring winder made for short springs of this size. 

[Neverenoughwatches]

2 minutes ago, Birbdad said:

There is no mainspring winder made for short springs of this size. 

I can only suggest checking the arbor fits and hooks nicely before fitting the spring in future. 

[Birbdad]

Just now, Neverenoughwatches said:

I can only suggest checking the arbor fits and hooks nicely before fitting the spring in future. 

Yup it didn't even occur to me the inner coil would be so damn big on this thing. I've never tried sourcing a generic before but i had a feeling it would be a pain in the ass. Seems like every post i read about using generics the inner coil being either too small or too large is an issue.

[Neverenoughwatches]

5 minutes ago, Birbdad said:

Yup it didn't even occur to me the inner coil would be so damn big on this thing. I've never tried sourcing a generic before but i had a feeling it would be a pain in the ass. Seems like every post i read about using generics the inner coil being either too small or too large is an issue.

It might be that generic springs of the same size can be used among different brands. Sometimes they fit the arbor ok and sometimes they dont. The spring's inner coil to fit the arbor is something that isn't specified when ordering, so we are either lucky or we aren't. Get yourself some practice springs to play around with heating and bending then when the time comes that you need to modify you have experience under your belt to do it for real.