Seiko 7A38 - by the numbers

Subtitle

Forums

Post Reply
Forum Home > Parts Info, Tech Tips and Tinkering > Sticky: How to set the Day / Date on a Seiko 7A38 - and when NOT to do it

Seiko7A38
Site Owner
Posts: 14020

A pretty basic question really - and one I'd expected might have been asked in this thread in the main 7Axx General Discussion area:

seiko7a38.com/apps/forums/topics/show/8474319-everything-you-always-wanted-to-know-about-seiko-7axx-s-but-were-afraid-to-ask-


But if you don't have a copy of the Seiko 7A38 instruction manual and haven't managed to find an online .pdf version ....

http://www.Seiko7A38.com/Seiko7A38-UserManual-Scan.pdf (a link is included on our 'Other Links' page) .... You might well ask. 


This morning, with my 'Admin' hat on, I ran my usual daily check on incoming google searches - and noticed this one in particular:

how to set day date on seiko 7a38 Try it yourself: Google search (Belated edit: this topic is now the first result returned).

The visitor from Larbert in Scotland certainly had a good look around the forum, but I'm not convinced he found the answer he sought.


It is of course on page 4 of the Seiko 7A38 instruction manual (and hopefully common knowledge to everyone on this forum):


January 24, 2013 at 10:13 AM Flag Quote & Reply

Seiko7A38
Site Owner
Posts: 14020

However, the 'Note' at the bottom of the page should probably include a stronger warning:

Rather than merely stating: 'the calendar will not change to the next date' ....

Perhaps it ought to read: the auto day change-over might never work again ! 


You can of course change the day / date (as on any watch) by advancing the hands through 24 hours (or more) by turning the crown.

But the Seiko 7A38 has a 'quickset' day / date feature - as shown on page 4 of the instruction manual. It's one of the more modern features that those of us who love these watches appreciate. It is also the watch's Achilles Heel. If you read on the 'Net elsewhere about Seiko 7Axx's, you'll sometimes see the phrase 'All Metal Gear Train' touted. I've used it myself in eBay listing descriptions. 

It's not strictly true. There is one plastic gear in a 7A38 - the Day / Date driving wheel (Seiko part  # 867.725). It is the weakest link.


January 24, 2013 at 11:26 AM Flag Quote & Reply

Seiko7A38
Site Owner
Posts: 14020

Here's a few of Simon's photos, showing the day / date wheel, which I've borrowed from another thread. I'm sure he won't mind. ;)



January 24, 2013 at 11:56 AM Flag Quote & Reply

Seiko7A38
Site Owner
Posts: 14020

It's not just the fact that it's plastic (it can be moulded in either white or black), and that it is perilously thin in a couple of places ....

But also it's position relative to the battery (on the other side of the movement) which makes it potentially even more vulnerable. 

Here's one of my own photos showing the teeth of a white plastic day / date driving gear underneath the main-plate battery recess:




Here's another of Simon's photos (borrowed from another thread):




So any prolonged leakage of either fluid or fumes from an old run-down battery could also easily weaken it further, by making it brittle.


January 24, 2013 at 12:30 PM Flag Quote & Reply

Seiko7A38
Site Owner
Posts: 14020

The Seiko manual states: If the setting of the calendar is made when the hands are pointing between 9 P.M. and 3 A.M. ....

Whenever I've observed them, the day / date changeover sequence of my 7A38's goes as follows:

11:00 PM Date wheel starts to move and is usually completed by 00:30 AM (the following day).

01:00 AM (previous) Day starts to move. By 01:30 AM changed to 'alternate' language; by 02:30 AM the Day changeover is complete.


I'm not entirely certain how it works (it's obviously not possible to see with the day and date display wheels in place), but those two thin lever arms on the plastic day / date driving wheel engage the teeth on their respective wheels during this change-over period. Thus common sense would tell you not to mess wiith the day / date settings, using the crown's quickset mechanism between these times.


But people do make that mistake. I've done it myself once - and only once. Never again. I've also bought two 7A38's where the seller had obviously broken the day / date driving wheel in this manner - most recently my NOS Orient J39908-70, the subject of this thread:

http://www.seiko7a38.com/apps/forums/topics/show/8404123-my-newest-incoming-orient-j39-nos-but-more-disappointment

Thankfully, one of the topics read by our erstwhile visitor from Larbert was that one - which includes the salutory warning:

Rule #1 when setting 7A38's - NEVER mess with the day / date quickchange between 11:00pm and 3:00am.

He read it 3 times - so hopefully he got the message, if not the instructions on how to actually go about it.


One other little tip: If you buy a 7A38 with a flat battery (often a gamble in itself), or if you've left one for a long time, in a collection box or drawer, and the battery has run out, particularly if the day / date windows are fully aligned (regardless of the language shown) and the time is shown between 9:00 and 3:00 - you don't actually know where it stopped. I've found that 7A38's batteries will often expire after midnight, as the watch draws slightly more current while it's doing the day / date change-over. Once you've fitted a new battery, pull the crown out to the second click, and rotate the hands through at least 12 or possibly 24 hours, so that you are absolutely sure what time it is displaying: say 1:30pm rather than 1:30am. If you 'over-shoot' - wind it on manually another 12 hours, just to be safe.


Once you're out of the 11:00pm to 03:00am 'danger zone', it takes only a matter of seconds to reset the day / date using the quickset.

It takes rather longer to remove the movement; pull the hands; remove the dial and day / date wheels to replace a broken drive gear.

You know it makes sense. ;)


January 24, 2013 at 1:06 PM Flag Quote & Reply

Seiko7A38
Site Owner
Posts: 14020

Funnily enough, at the weekend I had to refer to the Seiko 7A48A Technical manual - probably the first time I'd read it properly. 

Top of the last page is this warning:




Yet no such warning appears in the 7A38A Technical manual, where it's really needed.

What makes this warning seem more pointless and misplaced is that the date driving gears in a 7A48 are metal - not plastic. :roll:


January 28, 2013 at 1:39 PM Flag Quote & Reply

Sir Alan
Member
Posts: 458

Here's an example of a damaged day/date driving wheel:



(the damage is at about 1:30pm)


November 25, 2013 at 4:15 PM Flag Quote & Reply

Seiko7A38
Site Owner
Posts: 14020

This thread got a little plug on TZ-UK yesterday: http://forum.tz-uk.com/showthread.php?316126-Seiko-7A38

So I guess I should add this as a further cautionary footnote for all 7A38 owners, existing and prospective ....


There is one plastic gear in a 7A38 - the Day / Date driving wheel (Seiko part # 867.725). It is the weakest link.


Unfortunately the part is now officially 'discontinued' by Seiko and no longer available from Cousins UK:(



Another very good reason to exercise care when setting the Day / Date on a 7A38. ;)


November 11, 2014 at 5:30 PM Flag Quote & Reply

Rob Benham
Member
Posts: 280

Hello, Paul.  My most recent forey into fettling has me looking out boxes of part done jobs where life got in the way of honest fettling. 

I purchased a 7A38 SGP with no small hands and declared for parts.  It was bright and clean inside but my heart sank when I found the accumulator minutes stem was broken off at the shoulder.  'I'll rob the rather disappointing black watch which was another disappointment.'

It's strange when things are too good to rob.  But the black watch 7A38-7810 ebay 24th Mar 18263549507297 was purchased all too hastily and had a corroded circuit and a day fault. But the movement is very clean.  However, it would not change day even by cranking it round. 

I removed the face and to my surprise found no stipped teeth.  But my white version of your black wheel (thanks Sir Alan) I may well check again having seen your detailed photo. I think it's okay, but an even closer look now I've seen yours. 

This brings me to how to datumise the whole setup.  Is there any literature that shows the positions of the teeth for a given time.  It occured to me that perhaps setting the hands to 23:00 might give a clue, but I have no idea where the correct 23:00 should be so that day and date will motor round to the exact spot. 

I found that brass ring came off surprisingly easily, and the day disk was worn a bit where it touched, but I cleaned and lubed everything and reassembled.  The days flopped over nicely via the crown. However, when I crank it round, it starts moving but does not cycle to the end of its travel. Nor does the date. 

Having remembered your warning from my early days, I suddently realised that with the hands off on a 38, the angle that they are reapplied to is critical.  What's more, turning the crown while fully out with the hands off is doing whatever mischief it does while blind to the time. 


May 2, 2020 at 9:10 PM Flag Quote & Reply

Seiko7A38
Site Owner
Posts: 14020

Rob. It's probably best if I ask Simon ('Sir Alan') to answer your question, because he's now done many more of these, than myself. He doesn't post very much on any watch forum nowadays; he's too busy repairing and servicing other people's 7Axx's - including mine. It just so happens, that yesterday morning I posted my 7A38-728B off to him, for a service. It has a few issues besides, but also a problematic day and date changeover which never quite manages to properly align itself in the window.



I don't believe it's a problem with the plastic gear wheel's teeth - more likely the day changeover arm is weak or damaged. I've enclosed a NOS changeovever wheel with my package (the last one I bought from Cousins, before they ran out of stock and flagged it as NLA). I'm keeping my fingers crossed that Simon doesn't need to use it. I'll post a link to his strip and re-build photos when they become available.


May 3, 2020 at 9:48 AM Flag Quote & Reply

Rob Benham
Member
Posts: 280

Thanks Paul.  Seems to be good timing. 

At 80 years old I'm astonished I've not got any significant tremor.  Not sure how long that will go on for, but this period has bought back how much I loved working on the 7A** series. 

I too became disenchanted with P-bucket and will need to set up a hosting system.  I'll ask the chaps on PPRuNe dot org / Jet Blast if one can host one's own photos.  There's some full-time IT bods there sometimes. 

Having said this, the challenge is teasing at me. I must just be careful I don't bust someting before having the secret laid before me (as has happened so many times on this forum). :)


May 3, 2020 at 12:15 PM Flag Quote & Reply

Sir Alan
Member
Posts: 458

Herewith a brief study of the 7A38 date/day changeover mechanism.

I will exclude the quickset date/date setting mechanism (as they are independent).




In the very centre of the movement is the centre wheel and pinion - this is what the Minute hand is fitted to.

On the outer edge of the indented circle (at approx. 11pm) is a gear (the minute wheel) - this is driven by the centre wheel and pinion, and in turn this is what drives the hour wheel, fitted in the next picture.




You can see where the hour wheel teeth mesh with the teeth of the minute wheel. You can also see the black plastic wheel (the day & date driving wheel) - this is what drives the date dial and the day star with dial disk.




The above picture shows the calendar plate fitted (to hold the day & date driving wheel in place), the date jumper and the date dial.

The date jumper has a sprung 'finger' that engages the teeth of the date dial (in the above picture on the 5th).




The purpose of the date jumper is to ensue that the date dial advances by exactly 1 division (1 day) at a time.

Now we fit the date dial guard, held in place by three screws.




Then we can fit the day star with dial disk (held in place by the snap ring).




OK - so how does it work?

Let's look again at the plastic day & date driving wheel (spot the difference!!)




Other than the outer teeth, the obvious detail is in the two 'fingers' - of different lengths, and though not entirely obvious at different heights.

The 'date finger' at the bottom (in the above picture) drives the teeth of the date dial. 

As the centre wheel and pinion rotates (as the movement runs), the hour wheel is driven. Thus the Minute hand moves, and so too the Hour hand - in sync with each other (as they are geared together).

As the hour wheel rotates, the day & date driving wheel rotates - in sync with each other (as they are geared together).

At a particular point, the 'finger' makes contact with the teeth of the date dial. This finger pushes the date dial around, advancing it from one tooth to the next, with the date jumper ensuring it aligns perfectly each time.

The 'day finger' opposite it is longer and is designed to pass over the teeth of the date dial and to engage with the day star, which is hidden on the underside of the dial disk,



 

As the hour wheel continues to rotate, the day & date driving wheel also rotates. The 'day finger' makes contact with the teeth of the day star. This finger pushes the dial disk around, advancing it from one tooth to the next, with the day jumper ensuring it aligns perfectly each time.

 

Hopefully the above makes sense. It's not helped by the name of the parts, which don't exactly roll off the tongue.

Remember that the Minute hand is press fitted onto the centre wheel pinion. And the Hour hand is press fitted onto the hour wheel.

A good watch maker will do his/her best to fit these hands so that they both align perfectly at 12pm, and where the movement has a date/day complication that the date changeover completes as close to 12pm as possible (it's quite fiddly so +/- 5 mins or so is quite acceptable).

A bad watchmaker or a careless tinkerer will often fit the hands so they don't align at 12pm, and the changeover can take place at any point. This is a very good (or bad?) sign that a watch has been badly serviced.


May 7, 2020 at 3:22 PM Flag Quote & Reply

Seiko7A38
Site Owner
Posts: 14020

Simon. Thank you for the comprehensive explanation. 


@ Rob. Per my earlier post, thinking it might provide a possible answer to your problem, I was hoping Simon would use the grotty 7A38-728B I'd sent him as an example. He finished servicing it yesterday. Turns out there was actually nothing wrong with the plastic day / date changeover wheel in my watch's movement.




The watch had suffered a minor battery leak and appears to have been given an amateurish enthusiastic over-oiling at some point. There was lots of sticky residue and battery sulphate deposits all over the movement, including the area of the day / date changeover mechanism, which was causing additional friction. For example:



May 8, 2020 at 12:07 AM Flag Quote & Reply

Rob Benham
Member
Posts: 280

Thank you both very much. Sir Alan, I wish you'd been an aircraft type-training lecturer.

 

My days are rather taken up with helping folk over this miserable period, but I was also diverted by an irrational desire to mend a coil. It became a battle. More on another thread.

 

I'm surprised I didn't spot the difference in the finger heights. It's in pieces again this afternoon. I have a very positive click into place with the crown in the quick set position. It's only when I crank it around in the third position that both stop too early. I put the hands on loosely to get some idea where the procedure will start, but of course, if my plastic wheel is out of phase with its driving information, I'm still in danger of harming the teeth. I'm turning the crown with the sensitivity of an eye surgeon.

 

For now, can I assume the pictures above show the basic phasing of the gears? It won't take long for a test run with new phasing, but I'm hesitant to trust a circuit to it, so the hand cranking with have to suffice.


May 12, 2020 at 1:52 PM Flag Quote & Reply

Rob Benham
Member
Posts: 280

Aaaah.  That Day locating spring on mine is wrong - inasmuch as it exceeds the lenth to the point it just abut touches the Day drive teeth.  It is also clearly bent.  The apex of surfaces (anticlockwise) number 2 and 3 should point straight up at the axises of the hands shafts.  Mine does not.  The length of the deadly-critical spring is clearly bent.  The Day wheel crushes on top of it if one is really not careful. 

I imagine it's a one-go trial to get it right.  Lucky it's only a practice/parts watch.  However, what is not so lucky is me coming in here last night tired.  The little brass bush pinged off and headed towards my left shoulder.  I'm up to about 2, maybe three hours 'floor time'. :roll:  I even left my clothes in my den last night.  I thought I was way past sending things into hyperspace, but it seems not.  While I was looking, I found a pusher C clip.  Small compensation. 

Now very motivated to get hosting. 

May 14, 2020 at 7:33 AM Flag Quote & Reply

Rob Benham
Member
Posts: 280

I found the brass C ring. Less said about that the better, except that it was now, and probably had been, an unreliable circumference. I looked at the steel tube it was to go on at some 90X magnification. The restraining contour is minimal. I found a drill bit that was near to diameter and crushed the ring onto it. It went onto the tube with a distinct click, but I’ve no way of knowing the design tension. A guess at best. Now was the time for the day dial to be fitted.

The cross section akin to a little house on the end of the long sprung steel strand HAS to have its ridge pointing at the centre shafts. Mine had been so bent that the days landed some 20% off the mark; a declared fault at point of sale. The little house has a hole in it which aids the most cautious bending with tweezers. The entire arm probably needs to be raised above the surface plate but I’m guessing.

Happily, the day dial has slots around its cog’s circumference. It allows one to see the metal popping in and out of the day registration dips. You know then that the dial is not lying on the spring which I imagine is what caused the original failure. (there were some clues it was a trodden path.)

I took some time to be sure the day dial was fully down and then covered the C ring in plastic before attempting the simple click down. It gave a positive click into place and I compared the height with a blown up picture from Sir Alan’s list. It looked the same.

 

I dabbed the hours shaft with ink and watched the rotations. For the first time the days registered in the window from 24 hour cranking. Phew. And thanks again.

The phasing of the plastic wheel, I seemed to be able to copy from the pictures. I was way out, and that in itself could have endangered it.

There is no doubt it is vital to observe the setting rule this thread is about.


May 16, 2020 at 6:07 PM Flag Quote & Reply

Chronos
Member
Posts: 10

@Seiko7A38, in my quest for a new day/date changeover gear, part number 867.725 which is NLA, I now wonder if I can use a changeover gear from another Seiko. For instance, do you know if all 7A38's had dual language day/date wheels? How about 7A28's? Do you know what other Seiko watches used the gear 867.725?

 

Thanks in advance. By the way, great detailed post about setting the day/date on the 7A38 and even greater details regarding the changeover gear. Really appreciate all the information.


June 8, 2020 at 12:23 PM Flag Quote & Reply

Seiko7A38
Site Owner
Posts: 14020

Firstly, thanks for moving your post here. ;) As I wrote in my PM, I believe your questions are more pertinent to this thread, than the 7A38-xxxx model specific thread where you originally posted it. To answer them, simply:


1) Not as such. Whereas I've seen various other similar-looking plastic date / date changeover wheels used in a number of other Seikos, both quartz and mechanical, I don't believe there is an alternative part.


2) All 7A38's (regardless of the model variant suffix) use dual language day / date wheels. The two main variables between them are the second language (besides English) printed on their surfaces, including French, German, Italian, Spanish and of course Kanji Japanese on the JDM variants, like the titanium 7A38-7030. The other variable is the colour of the background: White lettering on Black backgrounds (like yours) or Black lettering on Silvery-White backgrounds. They are all mechanically identical underneath the paint and are all driven by the same Seiko p/n 867.725 changeover wheel (be it moulded in black or white plastic).


3) Nope. 7A28's don't have a day / date function. The complete (approx 1.0mm thick) complication is omitted in that caliber. Although it doesn't have a day window (only date), the same p/n 867.725 changeover wheel is however also used in the Seiko 7A34. These can sometimes be picked up on eBay in the States fairly cheaply, especially the less desirable 7A34-7019 gold-tone version. Obviously the same day / date changeover wheel is also used in all the re-branded versions of the Seiko 7A38 movement: Shimauchi V906; Orient J3920, Cartier Ferrari Cal. 531, etc.


PS - I've sent you another PM. ;)


June 8, 2020 at 12:41 PM Flag Quote & Reply

Chronos
Member
Posts: 10

@Seiko7A38, thanks again for the detailed answer. I really appreciate it. With the gear being used in other movements, there's a much greater chance of finding one without having to pay a steep 7A38 price.

June 11, 2020 at 10:40 AM Flag Quote & Reply

You must login to post.