Wonky anchor? Koma standard lacks power.

AndyDWA

Registered User
Dec 26, 2013
621
6
18
Western Australia
Country
Region
I have a Koma standard which is a gift for a relative. I believe it is a variation of Plate 1683.

I have been right through this clock finding and fixing issues as I go, including a bent centre arbor, removing a spring tension washer that didn't belong and adding a dome tension washer that should have been there. It even had a piece of steel-nylon fishing trace as a suspension wire!

The fork was bent and twisted in unusual ways so I've straightened that and ground the inner surface smooth with a sharpening stone (it still might be a problem, but I don't have a spare fork and it look good to my eyes).

I've cleaned everything and removed and serviced the mainspring (which is a whole 'nother story!). It now winds without clunking.

I have fitted a new .0035" suspension wire.

But... I can get no more than 180deg rotation with not much over-swing. For a standard, this seems grossly under-powered.

The anchor concerns me and I need some advice. I straightened the anchor pin but, given all the other things I've discovered on this clock, it's entirely possible the pallets have been messed with too.

In the attached photo you can see that one pallet touches the floor while the other is held above the floor. They also seem very short compared to pallets on Kundo models. I assume they should be the same relative length but that leaves the question - which one do I adjust?

It's difficult to see lock and drop with the small viewing space - and my ageing eyes.

Any thoughts appreciated.

anchor.jpg
 

Tinker Dwight

Registered User
Oct 11, 2010
13,666
94
0
Calif. USA
It is hard to tell. Even though it is a little gap,
the lever arm is relatively straight up.
One can't tell much when it is not in place.
Tinker Dwight
 

AndyDWA

Registered User
Dec 26, 2013
621
6
18
Western Australia
Country
Region
That's what bugged me. The arm is vertical but the pallet isn't resting on the floor.

In place, when I push the anchor with a toothpick, it seems to jump on centre or slightly past centre when pushing toward that side (toward the right from the back view). But it jumps just before centre when going back to the left. That's what made me take a look at the geometry.

I'll try and get a better view of what's happening with lock and drop and I'll re-read the chart in the book. I do know the exit drop seems very small - the pallet looks like it could almost touch the rear of the escape tooth as it travels downward.

Thanks Tinker

EDIT:

I managed to set up a magnifying lamp to get a closer look.

Both locks look good but exit drop is tiny and less than entrance drop.
 
Last edited:

AndyDWA

Registered User
Dec 26, 2013
621
6
18
Western Australia
Country
Region
Update:

I took some video of the escape action and provide screenshots below. Note that the fork is set quite low as it refuses to run if I move it higher.

The entry lock seems small and the tooth may even land right on the corner of the impulse face. I don't know if the low fork is contributing to some sloppiness here.

The exit/outside drop is tiny but the entry/inside drop is pretty good.

The anchor pin seems to favour the right side when running but the pin is vertical when sitting on the bench, though the pallets are at different heights as as noted earlier. I suspect the different pallet heights are causing the pin to lean to the right in the clock, but adjusting the right pallet will affect lock on both pallets so I'm not sure what to do except, perhaps, bend the anchor pin toward the right so it would be perpendicular to the pallets.

Drops are supposedly adjusted by raising the anchor. This clock has a removable anchor bracket but it has no freedom so is unable to raise/lower the anchor.

The front plate houses an eccentric adjuster (shown) but it is not one with a slot so I don't know how it would be adjusted without damage.

There is no obvious sign that either the pallets or the eccentric have ever been tampered with (ie: no damage)


backplate-1683.jpg eccentric-pivot.jpg lock-drop.jpg
 

Tinker Dwight

Registered User
Oct 11, 2010
13,666
94
0
Calif. USA
You have excess inside drop. To fix this requires
using the eccentric.
It is possible that the anchor is not original to this
clock and was a replacement.
Raising will reduce the locks more. That would require
adjusting the pallets.
Poor lock on the entrance is an indication of friction
on the exit pallet.
Either surface or bent escapement teeth.
Bent teeth drag more because of the curve of the
pallet.
I could also be off centered pallets as you noted
with your table test.
Tinker Dwight
 

AndyDWA

Registered User
Dec 26, 2013
621
6
18
Western Australia
Country
Region
Thanks Tinker.

I have re-polished the pallet faces as best I can. They are as shiny now as any others I've seen. I haven't checked the EW but I did give it my usual attention when I did the first clean and didn't notice anything unusual.

I have also dismantled and re-finished the fork. It was in pretty bad shape when I started and it was still a bit bent after I "repaired" it. I noticed it was leaning to one side when suspended, even with the pendulum on. So I have straightened it out as well as I can and re-ground the inside faces to get it as smooth and flat as possible (I've made the inside faces slightly convex along their length to minimise friction). I'm hoping this will help fix the entry-lock issue since it was binding a little before.

I haven't adjusted the eccentric, partly because I'm not sure how to do it with this type and partly because I've never adjusted one before.

For completeness, I should have mentioned earlier that the clock easily passes the three-click test with the anchor out.
 

AndyDWA

Registered User
Dec 26, 2013
621
6
18
Western Australia
Country
Region
UPDATE:

I tweaked the front eccentric up a tiny bit and the drops now look even.

I lowered the exit pallet a tiny bit and the entry lock is now good.

When I do the toothpick test, pushing the anchor pin back and forth under load, the anchor does not jump until the pin is vertical - in both directions. I believe it is jumping too late.

Does this suggest my lock is now too deep? There's no fouling of teeth and pallets and the clock will "run", but it still lacks power, with barely 180deg rotation and very little over-swing.

Is it worth raising either pallet a tiny bit to reduce lock a fraction?

I also notice some "recoil" when the entry pallet locks but it is most definitely not hitting the impulse face.

I hung a different suspension and pendulum on it (Kundo standard), to see if a different fork might make a difference but it still required the fork to be low and still offered barely half a turn of rotation. I realise that different weights and suspension thickness may render this test meaningless.

I have removed and re-cleaned the EW and will test again to see if anything has changed but the toothpick test is bugging me.
 

AndyDWA

Registered User
Dec 26, 2013
621
6
18
Western Australia
Country
Region
I guess the jump happens when a tooth moves from lock to impulse and pushes the anchor (with no pendulum/fork attached).

I've been reading the Schatz thread in which you've gone into great detail and I suspect I am at the stage where I could apply what you've written about having the anchor approx 1/3 on impulse when the lever is vertical, and adjust the pallets accordingly.

I assume to keep the adjustment even, I would raise both pallets a tiny bit.

Thanks as always Tinker.
 

AndyDWA

Registered User
Dec 26, 2013
621
6
18
Western Australia
Country
Region
UPDATE:

If I do the "upright lever" test Tinker Laid out in the Schatz 49 pallets thread, I get something like the following (photos were unclear so I illustrated instead).

It is very difficult to judge the actual position of the tooth because it can only be viewed from an angle and is slightly obscured by the pallet when viewing through the window. But the exit pallet is clearly still locked with the anchor lever vertical.

I guess I need to raise that exit pallet again :( but that would only get me back to where the clock wasn't running anyway.


impulse-locks.jpg

EDIT: Subsequent tests, using the suspension wire as a visual guide, show the exit tooth is much closer to impulse (less lock) than shown. It would be fair to say it is right on the corner. The entry tooth is also probably closer to the lock face than shown (not as far into impulse) but still resting on the impulse face.
 
Last edited:

Tinker Dwight

Registered User
Oct 11, 2010
13,666
94
0
Calif. USA
One tooth will always be on the lock surface while the other
is on the impulse face. One has to escape the wheel by rocking
the lever to see the location on the other pallet.
The trick is to get the lever to exactly the same location
for each half check.
Do also remember, no pallet is made perfectly right. You may
have to increase the amount of lock by moving up the impulse face
a little.
Friction is the biggest issue. The exit lock is facing up and is most
likely to have dust on it and slight corrosion. ( polishing of the lock surface )
Also, the escapement wheel tooth shape is more critical on the exit pallet.
If warn down, the tooth will make contact in two places
on the exit pallet. Once at the tip and once at the knee corner between
lock and impulse.
Both of these will cause more drag, resulting in more backward kick
on drop. This makes it more likely to miss the entrance lock.
One last thing is the balance of the anchor and the centering
of drops. If the centering is way off, the lever will extend more
away from the suspension spring on one half swing than the other.
This is why I try to center the drops at the same time I'm
adjusting the 1/3 trick for the lock adjustment.
Since I know you have working clocks, you can look at this
same location on a working clock. Most will be at 1/3 or a little
higher( 1/4 ).
Making it higher increases the locks, allowing the fork to be moved
lower for more power transfer before flutter.
I've not got that far on the other thread.
Tinker Dwight
 
Last edited:

AndyDWA

Registered User
Dec 26, 2013
621
6
18
Western Australia
Country
Region
Thanks Tinker.

The drawing showed two separate situations (one for each lock) but we can ignore that now because 1: Further testing showed the drawing to be wrong, and 2: I've gone ahead and adjusted pallets since then.

A comedy of errors ensued as I scribed the pallet I was moving, moved it too much, then discovered it hadn't scribed at all. So I took a guess, put it back in the clock and locked up the escapement. Oops!

I also discovered that I moved the wrong pallet yesterday as I forgot to note that I had flipped the anchor over to get to the screws, so the exit pallet was now on the left but I adjusted the one on the right. But that pallet did scribe, so I could get it back home.

Luckily, I could refer to the photo I posted earlier and reset the unscribed pallet using tarnish markings on the side of the pallet. So I did that then made small adjustments to account for the lever test.

It's going again, but so far still less than 180deg rotation with very little over-swing. I will redo the lever test tomorrow when I'm awake.

Thanks for your patience. Your instructions on the other thread have proved very useful too.
 

Tinker Dwight

Registered User
Oct 11, 2010
13,666
94
0
Calif. USA
If you go a little higher, it should be able to let you lower the
fork more. That should get more swing.
If not you may have to take the anchor out and inspect the
pallet faces, and locks.
The escapment wheel may also be the issue, as noted before.
Tinker Dwight
 

AndyDWA

Registered User
Dec 26, 2013
621
6
18
Western Australia
Country
Region
It's run all night but rotation remains at just over 180deg total with about 15-20deg over-swing which is better than before. I adjusted the fork upward a little before reading your reply, so we'll see what happens next.

The pallets are as clean and polished as any I've seen before and the wheel looks good. I've checked full rotation of the EW and see no short or bent teeth. I've also cleaned every tooth under a magnifyer using a flattened toothpick and polish.

If things don't improve after the last fork adjustment, then I feel I have two choices: 1: accept that this is one of those clocks that only gets 180deg rotation or 2: Buy a new mainspring since it's possible this one is set (it did not expand significantly when I removed it but I don't really know what a set spring looks like or if there is a "minimum" diameter this spring should expand to when it's free). I was more expanded after stretching during the cleaning process.

It wasn't coned at all, so I'm certain it's straight enough.

It may even be the wrong mainspring if someone's replaced it before. But I'm not sure how I would tell.

@Tinker: Do you mean move the pallets upward then set the fork higher for more rotation or lower for more over-swing? I consider this fork to be set quite low compared to the book and to other standard suspensions I have (not Komas).
 

AndyDWA

Registered User
Dec 26, 2013
621
6
18
Western Australia
Country
Region
UPDATE:

Things are looking good with a higher fork and a bit more winding (I've been testing on around a half-wind). I now have more rotation and some over-swing.

Thanks Tinker. I've learned some new tricks.
 

AndyDWA

Registered User
Dec 26, 2013
621
6
18
Western Australia
Country
Region
UPDATE:

Getting 270deg rotation now with comfortable over-swing.

Now to put the motion works, dial and hands back on - and hope it keeps working.
 

Tinker Dwight

Registered User
Oct 11, 2010
13,666
94
0
Calif. USA
Fork height is an interesting thing, when raising it, you get more swing but
actually from the pallet adjustment, lowering it delivers more power.
Using more lock allows one the increase the power, by lowering the
fork to a optimal running point before flutter.
It is one of those adjustments that have opposite effects from where
the adjustment is made.
Being an older clock of unknown spring, I'd say live with 270 degrees.
Tinker Dwight
 

Forum statistics

Threads
177,526
Messages
1,555,869
Members
53,605
Latest member
Morbier
Encyclopedia Pages
909
Total wiki contributions
3,054
Last edit
Illinois Watch Company by Kent