Seiko 7A38 - by the numbers

Ok potentially dumb question, but I don't know the answer and don't have a 7A28 to test on.

So, in the interest of knowledge and providing a reference point I ask the question:

Are 7A28 and 7A38 modules interchangeable?

Part of the reason I thought this was that for some mysterious reason, it;'s almost as if there is a committed collecting community and forum for one and less so the other; 7A38s seem to hold much better prices than the non-ornate 7A28s and there would be the potential for some mix and match modding amongst the bargain hunters amongst us (ie. me).

So cheapo 7A28 with an ok case + destroyed but internally superb 7A38 = modding heaven?

AND i'm certainly not endorsing the cobbling together of such watches for resale (Really! actually being totally serious here for once )

Funny you should have asked the question James. 

One of the articles I had planned to write, in the near future, was '7A38 into 7A28 won't go'.  

Why do you think I bought a 7A28-7040 last year ? Not just because I thought I ought to have one.

It was because I wanted to try it for myself - including taking all the relevant internal measurements. 

The reverse - 7A28 into a 7A34 / 7A38 / 7A48 case is entirely feasible, because the 7A28A movement, without any kind of day / date or moon-phase complication, is approx 1.0mm thinner. However, you might need to be a bit 'creative' with dial ring spacers to ensure proper crown and pusher alignment. As a general sweeping statement all 7A34, 7A38 and 7A48 modules are (easily) interchangeable.

You may remember this 7Axx Franken I built last year, from a 7A34-7019 sample case with a 7A38-6109 609L dial and hands:

The question has been asked a few times before, James ....

Here's the (second) earliest reference I've found so far - on the old Network54 SCWF (Mirror site), dated 20th Aprli 2006:

http://seiko-divers.info/scwf/index.php?mod=103&action=1&id=1145586286

The reply by Hung Pham ('Time2Fly') reads:

From what I've read, the pushers and stem will line up, but the 7A38 movement is thicker than the 7A28, due to the day/date wheel.

I have never actually tried a swap. You might want to try swapping them, then examine the sideview very carefully to see if the 7A38 movement protrudes above the 7A28 case.

In actual fact, the crown and pushers don't line up. See my photo in a subsequent post.  

Edit: I found an earlier thread, from the same source:

http://www.network54.com/Forum/78440/thread/1139944086/1139967435/Are+the+7A28+and+7A38+movements+the+same+dimensions-

Again, the quote from Hung Pham ('Time2Fly') reads:

Most 7A models have inner tachy ring that mesh with cut-outs on the dials. If you know how to remove the tachy ring from the case (I haven't figured out yet) to go with the movement, that's one less issue. Otherwise, you'll have to make one movement/dial match the other's tachy ring feet.

Again, a slightly incorrect assumption / statement, on Hung's part - as it isn't (always) the Tachymeter ring which ultimately determines the precise height at which the dial / movement sits within the watch case. Or, at least not in every case, anyway. Of more anon. 

I'm still googling. 

Another thread from the old Network54 SCWF from 2007: http://www.network54.com/Forum/78440/thread/1180935483/7A28+v+7A38

An assertive statement by Kurt reads:

No chance ! The casings are different in height ! Due to the additional day/date wheel the 7A38 movement won't fit in the 7A28 casing. The 7A28 movement won't fit in a 7A38 casing.

Regards Kurt.

Pity he got the second part wrong. 

Other noted 7A experts, and one gentleman in particular, have also made categorical statements that 7A38 into 7A28 won't go. 

I'm still looking for another more recent thread on the old SCWF, and when I do find it (and trust me, I will) I'll quote him accordingly. 

Myself, I've always remained open to the possibility - at least until fairly recently anyway. Some of this stems from my old slot-car racing days, and shoe-horning much larger motors (Mabuchi 36D ring any bells with anybody ?) intended for 1:24 scale into 1:32 scale cars.

If someone said it couldn't be done, I'd give it a try ! Especially when you consider certain similar 7Axx case designs - the 7A28-702x and 7A38-706x for example - and how much deeper the Tachymeter dial ring spacer is on both those, compared to other models. 

Besides, it's been done once already - albeit not entirely successfully.

Have a read of this old thread on RLT while I'm uploading the photos: http://www.thewatchforum.co.uk/index.php?showtopic=53256

This 7Axx Franken was listed on eBay, by a seller in Turkey with the user ID 'led-sw', at the end of March, 2010.

The eBay item # definitely was 280485498778, but it seems to have disappeared off eBay history altogether. It was listed simply as:

VINTAGE SEIKO EARLY QUARTZ CHRONOGRAPH WATCH USED (no mention of 7A28 nor 7A38 in title or description)

If you've read through that thread, you'll see that I'd already flagged it up in the 'Heads Up On A Rare 7A38 Diver On Ebay' thread.

In this other thread linked above, DaveS failed to notice it was a franken at first - just stating that it was on a non-original bracelet.

I had eschewed 7A38 'Divers' for the first 18 months of my collecting, and didn't actually buy my first 7A38-7070 until August 2010,

and so wasn't that familiar with the case shapes - hence why I'd initially mis-identified it as a 7A38-706A dial in a 7A38-7070 case.

Later I realized my mistake - that the case was actually from a 7A28-7040 (or 7A28-7049), rather than a 7A38-7070. 

Which of course made it a complete and utter franken, of the first degree. 

Dave's reactions (in post # 13 of that thread) included: 

Well I'm blowed! How did we miss that mate? 

Actually, I did wonder why the case wasn't quite the same shape as mine. I thought it was just because the pics were taken from a different angle. Never occured to me that he'd shoe horned a 7A38 into a 7A28 case.

Wonder what size hammer he used to put the back on?  

As you will also have seen, from post #14 of that thread, the buyer, who had paid $183.50 for it, wasn't entirely happy with it.  

Here's where he subsequently re-sold it, back in January 2011: 

http://www.ebay.co.uk/itm/110642966485

Vintage Seiko Quartz Chronograph with day date

Here is a nice Seiko Chronograph from the 80's. It is sold as is since the chronograph function doesn't work properly. The watch keeps excellent time and the day/date function work as they should, but the chronograph function is not reliable. One day when I was working in the yard I looked down and saw that the chronograph function had already started. Until this day I can't get it to work right. It will set back to zero by pulling the crown out to the first click, but will start up again at any given time. I have enjoyed this watch and hope someone can fix it. Sold as is no return. Movement measures 41 mm across excluding crowns. Please see my other auctions.

You'll note that it sold for a mere US $78.77 (approx. £49.72) that time around - due, I'd say to this seller's more honest description.

Still, it intrigued me. 

Hindsight, as the saying goes, is a wonderful thing. 

Particularly as I'd had the the photo below on file since as far back as May 2009 - though I can't remember now, where I saved it from.

I'd captioned it simply as: '7A38-7070-Divers-Stainless+Black-With7A28'. Shows you how little I cared for 'Divers' or 7A28's at the time.

It's what appears to be a very nice example of the 7A38-7070, with the head (only) of a 7A28-7040 (or possibly a 7A28-7049).

In the photo, you can clearly see the quite different shapes of the lug ends of the two cases.

Later on, by early 2011, I had actually come to quite like the 7A38 'Divers', and went back and looked long and hard at this photo.

It occurred to me that not only did the all-black dial of the 7A28-7040 look nicer than the 7A38-7070, but so did it's case and bezel.

So I decided to build myself a 7A38 Franken Diver, with an all-black dial, and try to make one that looked more like a 7A28-7040,

but obviously with the all-important day-date feature. This became the watch that, by now, most of you will already be familiar with:

Although I was (and, indeed still am) pleased with the end result, it was still mostly comprised of 7A38 parts (the all-black 709L dial came out of a 7A38-7080), with very little in the way of actual 7A28 content - only the (NOS) hour and minute hands of a 7A28-7040.

As it happened, I'd bought a NOS 7A28-7040 bezel, from Cousins (back in the days when they still had stock). Being a bit naive, and already knowing that the the 7A28-7040 and 7A38-7070 not only shared the same crystal part number (310W64GN00) and used the same rotating bezel O-ring retaining gasket (0Z3404B02), but also that the insert used in both bezels was the same, I thought I'd try fitting the 7A28 bezel onto my 7A38-7070/7080 Franken ....

It wouldn't fit. The 7A28 bezel is only 3.3mm deep, compared to 3.5mm of the 7A38 - the grooves for the O-ring machined accordingly.

I probably should have taken that as a warning, and left it there. But No. 

Instead, in August last year, I went and bought myself quite a nice 7A28-7040, not for it's own sake; with only one purpose in mind. 

Inspired by that Turkish eBay 7A28-7040 / 7A38-706A Franken, I planned to take my predominantly 7A38 Franken one stage further.

It wasn't until a month later, the penultimate weekend in September, last year, that I finally got around to trying my little experiment.

As it happened, the natural available daylight wasn't the best, so I ended up having to take the majority of the photos by flash

but I trust they'll be clear enough to demonstrate the various points I'm trying to explain. If not, you can always try it for yourself. 

Here is the standard 7A28 704L dial / movement (left) with my 'Franken' (7A38-7080) 709L dial / movement from my 7A38-7070.

As they now shared the same hand set, I was pleased how similar they looked side-by-side (apart from the day / date windows):

What is probably less apparent in these two photos is the slightly different sandwich construction of the two dials. They both have recessed sub-dials. The way this was achieved was by using a 0.3mm thick lower dial plate (which includes the sub-dial printing), on top of which is laid a 0.1mm thick upper plate, which carries the remainder of the main dial printing. The two plates are fixed together by microscopic rivets on the back of the batons and the SEIKO badge logo. Where these two dials differ slightly (the same variable applies across the 7A38 range), is that the upper plate on the 7A28 704L extends the full diameter of the dial, whereas on the 7A38 709L, it extends just beyond the circumference of the batons, so that the Tachymeter dial ring spacer sits on the lower dial plate.

Here's the empty 7A28-7040 case, showing some of the machined 'steps' in case, in the area under the under the crystal:

Again, with its own plastic Tachymeter dial ring spacer dropped into place: 

And with the (painted metal) Upper Dial Ring spacer (p/n 84304363, which is common to the 7A38-7070 and 7A38-7080) in place:

Here's an underside view of the 7A28-7040 case, still with it's original scratched crystal and both dial rings fitted:

In close-up, just underneath the end of the pusher and crown tube, you can see a machined step - effectively a dial locating flange.

Just for comparison, here are a couple of shots of the 7A38-7070 case (still fitted with both dial rings and crystal gasket):

Here's the 7A28-7040 dial and movement in its own case. Note how any further upward movement is limited by the case's dial flange.

Here's my Franken 7A38 709L dial / movement in the 7A28-7040 case. The upper dial plate is 0.1mm nearer the top than the 7A28's ! 

Here's a shot from underneath, albeit not a particularly clear one, of the 7A38 dial / movement in the 7A28 case:

In close-up, you can clearly see the mis-alignment of the pushers and the lever actuators of the movement:

Here are the dimensions I took at the time, though the first group were academic / irrelevant to the exercise: 

7A28-7040 Vs 7A38-7070 Comparative Dimensions (millimetres)

Lug end to lug end 42.75 / 43.0

Case Ø (w/o buttons) = 40.5 / 40.25

Bezel knurled rim Ø = 39.5 / 40.25

Overall thickness = 10.5 / 11.5

Bracelets B1075S and B1241S both taper from 24mm to 16mm, but link thickness 2.4 vs 2.7

Depth of Bezel (irrelevant) 3.3 / 3.5

Crystal Thickness (same used on both) 2.0

Depth of Case-back (incl. Threads) 3.0 / 3.2

Depth - Top Surface Glass to Case-back 10.3 / 11.3

Overall depth of empty case 7.8 / 8.8

Depth of Flange in case from case-back 3.9 / 4.7

Depth of Flange from Bezel top track 3.3 (7A38 n/a)

Thickness Upper Dial Ring Spacer 0.8 (same p/n used)

Thickness Tachymeter Dial Ring 1.5 (7A38 not measured)

7A28-7040 Movement + Top Dial plate to Anti-Magnetic shield 4.0

7A38-7070 Movement + Lower Dial plate to Anti-Magnetic shield 5.0

Average overall thickness (most) Seiko 7Axx Dials is 0.4

Thickness of Lower Dial Plate only (where applicable) 0.3

To make this work, and I still believe it CAN be made to work, one would need to grind away the dial mounting flange inside the case. 

It is that flange, not the Tachymeter dial ring, nor the dial ring spacer, which the defines the positioning of the movement in the case.

Effectively, to achieve good alignment between case / movement / crown stem and pushers, using this particular combination of 7A28 / 7A38, one would need to move the dial face / hands / movement upwards, approximately 0.9mm closer to the underside of the crystal.

By machining away the case's dial mounting flange, allowing higher vertical positioning, and eliminating the upper dial ring spacer, and possibly also a slight reduction in the thickness of the Tachymeter dial ring spacer (which would then need to be glued to the dial) ....

But the problem was, I'd bought myself a pretty decent 7A28-7040, and I wasn't about to start butchering it to find out. 

So, after replacing the scratched crystal in the 7A28-7040, and fitting the NOS bezel I had in stock, I put them both back together. 

My 7A28-7040 SPR013J - still looking just as Seiko intended it (unlucky this time):

Later, towards the end of September 2011, an empty, and slightly tatty looking 7A28-7049 watch case came up on eBay in the States:

Looks like someone had held the ends of a watch-case in a serrated jawed vice, but it would have been perfect for experimenting with.

IIRC, the eBay seller was another of these narrow-minded 'Will ship CONUS ONLY' merchants, and so sadly, I had to pass on it. 

But what about the Turkish eBay 7A28-7040 / 7A38-706A Franken ?  How did he go about it ? How big was his hammer ? 

I can only assume he did what I was considering, and must have ground away the dial mounting flange from inside the case. 

That would have allowed the 7A38-706A dial / movement to sit higher in the case. He also used the 7A38-706A Tachymeter ring, which must have left him a bit tight for space, as it's quite a bit deeper than most 7A38 dial rings. Naturally, he would also have omitted the 7A28-7040's (now unwanted) 0.8mm thick upper dial ring spacer. I suspect, to give himself a little more room to play with, he may well also have swapped the original 2.0mm thick crystal for a thinner 1.5mm one. Whatever was fitted, it's certainly not the original.

So where did he go wrong ?

If what the buyer, and subsequent re-seller wrote: The watch keeps excellent time and the day/date function work as they should ....

is true, then we must assume the Turkish watch-botcher got the case / movement crown-stem alignment right, because any resultant binding can be very detrimental. The erratic chronograph operation ? Perhaps the PCB had an intermittent fault. But I can't help noticing how the pusher buttons, particulary in the original eBay listing photos, appear to stick out much further than normal. They're probably not the original 7A28-7040 case pushers, and certainly not the pushers from the 7A38-706A (movement) donor, because those would be shorter. Perhaps they were some other odd ones, maybe not even Seiko pushers, just  of the correct 3.5mm Ø, which the watch-botcher had lying around, but with a longer stems - which might have resulted in them being in constant contact with the movement's actuator levers - almost like a 'hair trigger'. Hence the chronograph 'will start up again at any given time'. Who knows ?

While not entirely successful, he did manage to shoe-horn a 7A38 movement into a 7A28 watch-case, using fair means or foul. 

He showed it CAN be done - with a little intuition, and maybe some light machining work needed on the inside of the watch case.

That, Dear Readers, is one of the reasons I still haven't re-sold my 7A28-7040. I still haven't quite given up on the idea yet. 

Glad to have provoked a comprehensive response!

Upon reflection, it all becomes clear.  I've looked at many a 7A28 or 7A38 auction and something about the average 7A28 just didn't compare exactly with the 7A38 ('Yes, dear leader, I have seen the light...'    )

What that is is exemplified by the picture used above.  Of course it's a matter of taste but the 7A28 has a flatter, thinner appearance whilst the 7A38 appears 'chunky' and more solid.

And now I know why.  Because that indeed is the case.

That photo you cite (the first image in my last-but-one post) gives a false impression, James.

The 7A28-7040 is laying flat on it's case-back, whereas the 7A38-7070 probably has the Z-fold clasp, or at least the thickness of the bracelet links (2.7mm thick on the B1241S) between it's case-back and the sponge surface. Overall, the 7A38-7070 is only 1.0mm thicker. I took just about every significant dimension off both of them. That 1.0mm difference crops up in a few places, but it matters.

Hey 1.0mm is a lot when it comes to watch thickness!

I stand by my comment.  Look at any 7A28 and there is a flatter appearance than 7A38s which look bulkier.

It's a subtle piece of aesthetics but it's there nonetheless...

With respect, James - you are talking complete and utter tosh ! 

Forgetting for the moment, some of the odder looking assymetric Guigiaro designs, there are some very aesthetically pleasing 7A28's to choose from. I have often been tempted by the 7A28-7010 / -701A and 7A28-7039 / -703B, and I think the cases of both those are more attractive than, say, the stainless 7A38-7000, with it's odd protrusions ('wings', as a certain eBay seller once described them) betwixt pushers and crown. Indeed, not only is the 7A28-7040 / -7049 a more attractive looking watch than the 7A38-7070 IMHO (probably the reason far more were sold), but it's nicer to wear, too. With it's slightly thinner bracelet links, it's a better weight.

By the way, the culprit responsible for that 1.0mm difference in thickess, between 7A28 and 7A38 movements, is probably the least aesthetically pleasing component used anywhere in any 7Axx - the day / date spacer ring / dial mounting plate p/n 0105726*.

*The 7A48A movement uses p/n 0105725.

I presume it is the same 1.0mm overall thickness as p/n 0105726, with different internal cut-outs to clear the moon-phase complication.

Not tosh.

LOOKS flatter.

It is strange how the addition of a day/date window alters the visual impact of the two so strikingly.

The recent prices seem to reflect a distinct lesser interest in the (IMO) less visually impressive standard 7A28 range than its 7A38 counterparts.

As some of you are aware, I have an add-on statistics module running on the website. Not only do I find it interesting to see which threads visitors read, but sometimes how they arrive at the site. If not direct, it's usually by way of google searches - sometimes vague, sometimes very specific. Last night, a Spanish visitor googled exactly this phrase: 7a28-7020 vs 7a38-7060 which led him to the site, but sadly not to this thread. Which is a pity, because of the greater similarity in the watch cases of  those two models, they probably would actually have been a better choice for this exercise. I know, for instance, that the 7A28-7020 bezel will fit the 7A38-7060 case.

I've got plenty of the 7A38-706x variants in my collection, but despite bidding on eBay on various 7A28-7020's, and particularly the 7A28-702A, which I quite fancy, I've never managed to get hold of one. Does anybody have both models, and fancies trying the same 'will a 7A38 movement fit into a 7A28 case?' exercise, themselves - including taking all the relevant dimensions ?

Or would anybody trust lending me their 7A28-7020 for the purpose of such an exercise ? I'd be ever so careful with it, honestly ! 

As I had a quiet day house bound and inbetween phone calls to work, I spent some time tinkering 

I had a 7A34-7010 (that whilst almost in NOS condition just didn't really do it for me), a 7A28-702A that I had robbed a minute hand from, one of my many 7A38-7190's in general dissassembly and a 7A48-5000 that needed a clean and the sub-dial hands centering.

So, it's picture time 

Here's the 7A34 out of the case:

Hands and Dial off reveals:

Hmmm, there's a lot of bits that a) look familiar and b) aren't being used 

Let's take a look at the 7A38 again:

and from another angle:

So, the difference between the two is the missing 7 toothed gear wheel on the 7A34. This drives the quickset day function - so, another chance to remove that pesky c-clip:

which actually isn't such a bad guy after all. Underneath the day wheel we see what that 7 toothed gear wheel drives:

But they both share the same quickset for the date:

Now, how about the spacer ring:

OK - so it's 0.9mm thick.

Now, let's compare it with the spacer rings from a 7A38 and a 7A48:

they look the same 

and the measurements:

not a great pic, but trust me, that's 2.7mm (which happens to be 3 x 09.mm). That's a common part then.

Here are all 4 movements side by side:

and in glorious close-up. First  the 7A34:

the 7A38:

the 7A48:

and finally the 7A28:

looking positively naked compared to its big brothers.

I think I might have strayed off the original thread topic a bit in my previous post 

Back on-topic with just the 7A28 and 7A38 module question, here are the two circuits, first the 7A28:

and the 7A38:

They look the same (well, slight colour differences apart).

I've played mix and match a bit lately with circuits from all 4 of the 7Axx range that I've got  - 7A28,7A34,7A38 and 7A48 - and all apear to be the same and drive the movements correctly.

I'm a great believer in re-using something when it's right. 

More rain, another day indoors so time for some more pics.

Back to our 7A28 with the hands and dial removed:

Now, how do we get to the above with a 7A38?

Well, we start here:

and remove the day wheel:

let's zoom into that black plastic wheel:

here you can see it engaging with the date wheel - the plastic wheel is driven from the main hour hand gear (hidden in the above pic) - this advances the date every 24hours, then shortly afterwards (as the black wheel continues to turn, it engages with the underside of the day wheel to advance the day:

Now, to remove the day/date complication requires us to undo the 3 screws:

We can then lift off the top plate, and free the date wheel:

and see what's underneath:

this is the quickset for date and day. In the above configuration, all three gear wheels are engaged so turning the crown will advance the day.

Turning the crown in the opposite direction moves these gears to this configuration:

Here the 12 toothed gear wheel is no longer engaging with the gear that drives the day wheel, but has moved down and out to drive the date wheel. Smart eh?

Lifting these tiny gears off:

shows us this:

and we can see the slot that the gear moves up and down in.

Now lets remove the next plate:

to reveal this:

Now we can see what is driving the black day / date complication.

Let's remove that final plate:

that's 3 plates in total:

to reveal this:

we're nearly there!!

A quick look at the black plastic wheel, underside:

and topside:

now let's take it off:

look familiar?

The only difference I can spot between this 7A38 with the day/date complication removed and the 7A28 is the different cutout for the crown clutch assembly, shown here in the three crown positions:

and the much higher posts that the spacer block sits on:

 Here is the day date complication in a box:

I think I might have strayed off the original thread topic a bit in my previous post. 

Well, I didn't like to say anything, Simon  but .... carry on with the great   work !

I've edited James' original topic title, by adding a more all-encompassing 'all 7Axx movements'. 

PS - I guess I've only got myself to blame - after all I did start off by posting a photo of my 7A34-7010 / 7A38-6109 Franken.

Back on-topic with just the 7A28 and 7A38 module question:

The circuits are interchangeable but I've found a little difference in the function.

With a 7A28 circuit it is possible to reset to zero while the chrono works.

With a 7A38 circuit you have to stop first the measurement to enable the reset possibility (@ 4 o clock pusher)

In my opinion that is a needfull question before buying a 7Axx