The infamous Seiko 5606 day-date corrector wheel rocker - sorted, I made a new wheel

[mikepilk]

I bought a Seiko LordMatic cal 5606A and noticed the quick change day-date wasn't working.  
A bit of Googling found out that they rarely do on these movements, due to the plastic wheel on the day-date corrector wheel rocker breaking (see pic). 

As it's such a common problem, I thought I was lucky to find a NOS one on ebay.
Unfortunately the plastic wheel on that was already broken - I've since read that this is common on NOS items, the plastic ages and breaks.

The one I have at the moment is spare time, so I found a bit of brass from an old pocket watch, got my finest file (#6), drilled a 1 mm hole,  and started filing ...

 

 

It took a while ....

But got something close :

 

I think it's OK as a first effort at making a part. 

It's not perfect, far from it, but doesn't need to be as it only needs to push the day-date wheels over.

What I found was :

 I couldn't have done it without my stereo microscope

Although a #6 file seems very fine, it's way too course to finish the part. I believe you can get #8 and #10 cuts. 
I sharpened the end of an old screwdriver to use like a chisel. I finished with some 3000 grit paper, but it's not easy to fix a small enough piece to use. 

I haven't stripped the movement yet, so don't know if the part works - can't see why it shouldn't

Mike 

 

 

 

Edited March 30, 2020 by mikepilk

[rogart63]

Looks good. I have used VTA star a few times to save old 5606  date mechanism.   

[mikepilk]

5 minutes ago, rogart63 said:

Looks good. I have used VTA star a few times to save old 5606  date mechanism.   

Hi @rogart63, who /what is VTA? 

[rogart63]

Just now, mikepilk said:

Hi @rogart63, who /what is VTA? 

It's a company in Australia that sells parts for Seiko. His name is Adrian Sellick and is a watchmaker from down under. VTA stands for Vintage Time AU. Don't know if he sells on Ebay . But did before under the name halfoeight. Have the nice forum Wristsushi also. 

[aac58]

Yes I also bought the star from VTA, the advantage is that he make it metalic so it's a permanent fix. I think they are out of stock though.

If your star works then it is a perfect fix, and also permanent I think. Very good job Mike.

Here's a video by Adrian showing how to install it:

 

 

[rogart63]

4 hours ago, aac58 said:

Yes I also bought the star from VTA, the advantage is that he make it metalic so it's a permanent fix. I think they are out of stock though.

If your star works then it is a perfect fix, and also permanent I think. Very good job Mike.

Here's a video by Adrian showing how to install it:

 

 

 

It could be made a little better but works very good . heard of some that has problems when the movement have a day disc. But has worked fine for me . Still have one of them left. But some small tabs up would make it perfect.

Edited March 30, 2020 by rogart63

[RemovedUser10118]

On 3/30/2020 at 5:31 PM, rogart63 said:

It could be made a little better but works very good . heard of some that has problems when the movement have a day disc. But has worked fine for me . Still have one of them left. But some small tabs up would make it perfect.

I made one, thanks to finding Mike’s post (thanks Mike!).

I made it from 0.3mm brass (an old dial) doing exactly as Mike did above, only I carved it out with X-acto knives instead of files. It went very quickly as brass carves like wood at that scale, but I couldn’t have done it without my stereo scope.

Mine would not engage the day disk, so I carefully bent the teeth up slightly under the scope while clamping the gear in a set of flat needle nose pliers so it would not deform.

The carving, including test fitting and  finishing touches, took about an hour. The teeth engage both day and date portions correctly and the star wheel slips properly when resistance is met. Very easy job.  
 

 

Edited August 2, 2021 by PinLeverPete

[mikepilk]

I used a slightly thicker piece of brass as I noticed the teeth where higher than the central circular part - hence you needing to bend them up. It wasn't easy reducing the depth of the central circular part. I used an old screwdriver as a scraper.

Initially I made the teeth too wide, which caused jamming, but once filed down worked fine.

I found marking the part out difficult - all tools seem so big and clumsy on that scale.

[steve1811uk]

After seeing this I was inspired to make myself a 5606 star wheel. Made from 0.7mm brass plate, sanded down to 0.5mm with 600 grit paper as last step. Marked out with a sharp sewing needle in a pin vice. Shaped with a Dremel cut off type abrasive disc (was approx 0.8mm thick). Hole was 1mm and then carefully broached to be a friction fit on the gear wheel. After getting the flat shape correct I transferred the star to a cocktail stick and then brought the abrasive disc down vertically inside of the 4 corners (to make the raised teeth and lower central part). Finished the central lower part using a sharpened screwdriver as a chisel. Finished with fine abrasive paper 2000 grit. Took me three attempts to get something I was happy with (lost one when the Dremel spun it off the cocktail stick never to be seen again). Star wheel works really well.

[watchweasol]

Well done, Its a worker.

[mikepilk]

Nice one!  It's very satisfying when you can make a part that works, without having all the necessary tools.

Using a needle as a marker is a good idea.

[Michael1962]

Kudos to all of your effort gents. Well done.

Would the wheel measure about 3mm across the points?

Small piece to make. 

Edited March 22, 2022 by Michael1962

[steve1811uk]

9 hours ago, Michael1962 said:

Kudos to all of your effort gents. Well done.

Would the wheel measure about 3mm across the points?

Small piece to make. 

Original is 2.5mm across the diagonal, 1.8mm square, although mine did end up a touch bigger, 2.7mm diagonal. Any bigger than that and it will bind and not rotate.

[suomaf]

Hey guys, 

I am new here and new to fiddling with watch parts. I tried to 3D print the part out. Did not get anywhere. My Photon S just does not have the resolution to make parts that small. I then thought I have this other material. Bought this for another project a few years back. What do you think? Another question, Is there anyway for me to test out if this work without buying a staking set? The lordmatic was listed as junk and $40 so I was wondering if there is anyway to muddle my way through and get it working?

[suomaf]

To answer my own question, this is a no go. At the thickness, it is too soft.

[Nucejoe]

 Quick date set mech on 5605 and most other watches is flawed . You have  a free wheel  ie;  not in sync with teeth on  date plate so there is no gurantee they engage pefectly every moment you activate the quick set device, therefor you have the choice of either teeth to brake ,

Choice one , the teeth on the corector ,

Choice two , the teeth on date wheel.

To make both out of metals simply means one is unaware he is making  teeth on both vounerable to breakage.  Its not perfect as is called in above posts.

Good design would be to have the corrector and date disc in constant contact  " in sync "  and just activate-deactivate it somewhere in the set mech.

When a piece breaks in a watch, you replace it with new. 

Thats not the case with airborn delivery vehicles, where if a part misfunctions up there, the designer team on the ground is unlikely to live to tell the story of the explosion to others. 

My advice would be to make a plastic corrector like Seiko designers did, so to save the date disc. 

 

 

 

 

 

[f355]

In case somebody stumbles on this old thread, I've managed to 3D-print a replacement. Here's the video of it in action: 

 

[Nucejoe]

5 minutes ago, f355 said:

In case somebody stumbles on this old thread, I've managed to 3D-print a replacement. Here's the video of it in action: 

 

Can we get some pictures too, the Vid doesn't open in some regions. 

TIN

[f355]

52 minutes ago, Nucejoe said:

Can we get some pictures too, the Vid doesn't open in some regions. 

TIN

Here's one. There are also renders of the model at Thingiverse (see the link above).

[mikepilk]

If it proves robust enough, maybe you could make a batch and sell them on ebay?

[arkobugg]

On 3/30/2020 at 5:12 PM, mikepilk said:

I bought a Seiko LordMatic cal 5606A and noticed the quick change day-date wasn't working.  
A bit of Googling found out that they rarely do on these movements, due to the plastic wheel on the day-date corrector wheel rocker breaking (see pic). 

As it's such a common problem, I thought I was lucky to find a NOS one on ebay.
Unfortunately the plastic wheel on that was already broken - I've since read that this is common on NOS items, the plastic ages and breaks.

The one I have at the moment is spare time, so I found a bit of brass from an old pocket watch, got my finest file (#6), drilled a 1 mm hole,  and started filing ...

 

 

It took a while ....

But got something close :

 

I think it's OK as a first effort at making a part. 

It's not perfect, far from it, but doesn't need to be as it only needs to push the day-date wheels over.

What I found was :

 I couldn't have done it without my stereo microscope

Although a #6 file seems very fine, it's way too course to finish the part. I believe you can get #8 and #10 cuts. 
I sharpened the end of an old screwdriver to use like a chisel. I finished with some 3000 grit paper, but it's not easy to fix a small enough piece to use. 

I haven't stripped the movement yet, so don't know if the part works - can't see why it shouldn't

Mike 

 

 

 

How do you separate the axle from the bridge?

[mikepilk]

2 hours ago, arkobugg said:

How do you separate the axle from the bridge?

 

I used a punch in the staking set and it just pushes out.

[arkobugg]

2 hours ago, mikepilk said:

I used a punch in the staking set and it just pushes out.

That is what I now did this evening!