Guidelines for repair: Hermle floating balance

[doc_fields]

If you ever scrap out a Hermle movement with a floating balance in it, SAVE the balance! It is probably the best balance made for the Hermle movements, and is no longer being made. The new one isn't as frictionless as this one is, and not as well-made. Here are a few guidelines and pointers on adjustments after rebuilding one of the Hermle floating balance movements.
As I have said before, certain things normally do not change when repairing a movement. I have seen some posts about this unit that talk about punching out the stop pins for either increasing or decreasing the time adjustments on the floating balance. You have no reason to do that, and I will show you why. If you do those things, you're not really attacking the problem the right way, and you are fudging it, like many shade-tree mechanics do the cars they work on.
If you find you cannot increase your time adjustment or decrease it enough to have the clock run correctly and on time, and reached its limits, DON'T punch the pins out. Look at picture #1 below.

My tweezers are pointing to a stop tab on the left side where the fork comes to rest. Now look at the next photo:

My tweezers are pointing to the right side of this also. What's my point? Adjustment of these two pins, either in, or out increases the beat amplitude, making it run faster (closer together) or further apart, run slower. These two tabs should only be adjusted with smooth-jaw pliers, to avoid scratch marks on the unit. This should be a last resort repair, if previous + or - adjustments do not work. And, it only takes a little bit of movement to adjust these tabs, don't get crazy. Lastly, my tweezers are pointing to a brass tab at the top of the coil, which, when adjusted left or right, puts the balance into beat.


Using your Microset clock timer, you can put this unit and movement into beat on the "Beat error" setting of your timer by rotating this tab either left or right.
I hope this helps..............doc

 

[doc_fields]

If you ever scrap out a Hermle movement with a floating balance in it, SAVE the balance! It is probably the best balance made for the Hermle movements, and is no longer being made. The new one isn't as frictionless as this one is, and not as well-made. Here are a few guidelines and pointers on adjustments after rebuilding one of the Hermle floating balance movements.
As I have said before, certain things normally do not change when repairing a movement. I have seen some posts about this unit that talk about punching out the stop pins for either increasing or decreasing the time adjustments on the floating balance. You have no reason to do that, and I will show you why. If you do those things, you're not really attacking the problem the right way, and you are fudging it, like many shade-tree mechanics do the cars they work on.
If you find you cannot increase your time adjustment or decrease it enough to have the clock run correctly and on time, and reached its limits, DON'T punch the pins out. Look at picture #1 below.

My tweezers are pointing to a stop tab on the left side where the fork comes to rest. Now look at the next photo:

My tweezers are pointing to the right side of this also. What's my point? Adjustment of these two pins, either in, or out increases the beat amplitude, making it run faster (closer together) or further apart, run slower. These two tabs should only be adjusted with smooth-jaw pliers, to avoid scratch marks on the unit. This should be a last resort repair, if previous + or - adjustments do not work. And, it only takes a little bit of movement to adjust these tabs, don't get crazy. Lastly, my tweezers are pointing to a brass tab at the top of the coil, which, when adjusted left or right, puts the balance into beat.


Using your Microset clock timer, you can put this unit and movement into beat on the "Beat error" setting of your timer by rotating this tab either left or right.
I hope this helps..............doc

 

[harold bain]

Doc, very good information. I repair a lot of clocks with these balances, and find them to be quite reliable. Very rare to find a problem with the balance itself. Just to add to your post, never oil these, and don't clean them in your ultrasonic. I use camping fuel (naptha gas), which evaporates quickly off the unit, after a short time soaking.
Harold

 

[bangster]

Hey Doc...
In the spirit of a gadfly, let me explain, in my newbie way, why I'm not satisfied with the method you describe. You and others will point out where I go wrong.

With balance escapements, the received way of adjusting the rate is by adjusting the rotation speed of the balance wheel. Rotates slower, clock runs slower, and so on. There are two ways of doing this. One is to shift mass toward, or away from, the perimeter of the balance wheel. As mass moves outward from the center, wheel rotates slower. (Analogous to moving the pendulum bob outward from the suspension point.) As it move inward, wheel rotates faster.

The other way is to adjust the effective length of the balance spring. Longer spring, slower rotation; shorter spring faster rotation. This is the rating method used for standard hairspring balance escapements. But not the only possible way. A regular balance wheel COULD be adjusted by adding weights near its edge, leaving the hairspring alone.

The spring of a floating balance leaves no way to have its effective length altered. Instead, it's rotation speed is adjusted by turning the rating fingers, which move weights in or out in the balance wheel. It is designed to have its rate adjusted by adding or removing weight to its rim. It's designed with movable weights for just that purpose.

But there's another way altogether to change the timing of a balance-operated clock. Instead of adjusting the speed of the balance wheel, leave that constant and modify the amount the fork moves, between oscillations of the balance wheel. This is doc's approach. It takes longer for the fork to traverse a wider arc than a narrower one. So, the wider the arc, the longer between each capture and release down at the scapewheel end of the lever. He adjusts the arc by bending the banking pins closer together or further apart. He says that's the right way to do it, and to adjust rate by manipulating weights is a shadetree mechanic's bodge. (I wish he hadn't said that.)

If this approach were correct for a floating balance, it also ought to be correct for an ordinary balance-wheel balance (indeed, for any lever escapement). Instead of adjusting the effective length of the hairspring, we ought to leave the hairspring alone and bend the banking pins, to modify the arc of he fork. If it's the only right way for one, it's the only right way for all.

But I don't think that's the right way to adjust the rate of a standard balance escapement. And so I don't think it's the right way to adjust a floating balance escapement. Instead of bending parts not designed to be bent, rather than adding or removing weights designed to be added or removed, strikes me as just wrong.

My reasoning is no doubt screwed up somewhere, so I'll hunker down and let y'all beat me up. I probly deserve it.

bangster
uppity newbie

 

[rocky]

I've used both of the above methods at times. I've also found that the depthing of the fork can cause problems. A double ended key works well when trying to adjust the the tabs doc refers to.

 

[Berry Greene]

Old posts but contain all the questions if not all the answers! I have seen these floating balances with both a beat and balance spring rating adjustment. Yet the WUBA has only a beat adjustment and the rate adjustment seems to be exactly as described above by Bangster. Namely weights on the balance that can be moved in & out - if you're lucky and there's any adjustment left in the direction that you want to go. If there isn't - what to do? Adjust the banking pins? Coarse & crude maybe but sometimes necessary. To make a living. To get a result. to save a headache. To get it done. Not though the first thought surely? I read somewhere here on the forums how long the floating balance should rotate with the lever removed before it stops. A measure of its friction. I think that's a pretty good place to start if it won't rate for lack of adjustment. The power from the lever is another issue before we think too much about its throw - isn't it? How much time have we got for this? How much will we get paid for? Integrity; if you can afford it - 'tis a noble thing indeed.
It doesn't take much extra weight on the balance rim to slow it down. Change its natural beat/rate/resonance. Tippex/Snowpake will change it. How do I know? I just do!!
Thank you all for this. It helps to gel the thoughts
BerryG

 

[Berry Greene]

With the benefit of more experience I can add this.

DO NOT OIL – Clean with Naphtha, camping gas, or lighter fuel.
NOTE: There are 3 types ALL are highly inflammable and very etherial.
1) Petroleum Naphtha,
2) Natural gas Naphtha,
3) Distillation of coal tar & peat is true Naphthalene (C10H8) and is the main constituent of mothballs.It is white in colour and solid at room temps.

Lighter fuel is essentially NAPHTHA. Known as Zippo in US or Ronosol in UK and is OK to use even though it may also contains some very light oils. It almost evaporates. What some are maybe thinking of though is a gas that evaporates like Ether. In my view this could crack the jewels via violent cooling.

BALANCE TEST
Disengaged rotation from a 270 deg displacement should be 3 -5 mins.
The wire is 0.009” & a guitar string works as a replacement. A 0.010” fits but is too tight.
{BTW – the wire anchors are tricky – 2 off – but the wire doesn’t meet itself or tie together. Wind around anchors having stretched it straight to thread through the tube}.

I have here a Bentima fitted with a 340-020 Hermle chiming movement that has a floating balance. Fortunately I have two such balances:
a) I repaired the broken wire with 0.009" guitar string. These are hardened and cruel!
b) I cleaned with lighter fuel, no oil, except possibly the residues.

Both now pass the 3 minute test. I have got a result in the clock with my newly wired balance but the rotation isn’t much over +/-90 deg. I think it should be more like +/-135 deg but I cannot verify that. It's hard to know if the power is sufficient because it is certainly very slight. The beat is centred and even, The balance starts very willingly.
The movement has been dismantled, cleaned, pegged, polished and bushed as seemed necessary. Should I let down the power or wait to see how long it will run? So far it is close to the correct rate without adjustment.
As a bonus the striking & chimes are correct too!
BerryG

 

[Kevin W.]

Berry you may have a lot of slop around pivots, regular wearing. In which case clock needs a overhaul, to improve balance amplitude.

 

[Berry Greene]

Thank you Kevin. Much appreciated. Yes I think you are right. Especially now having read so much more about these Hermle 340-020 movements. They seem a bit finicky - built down to a price - they lack true longevity - and the pro's who need a "proper" lasting job, seem more likely to replace the whole movement. That tells me a lot.
However, this is a project clock not far into my embracing chiming clocks in my collection. Although of interest, not least the true floating balance in this one, it is useful to practice the re-assembly procedure and chime/strike train alignment and lever adjustments.
I will avoid Hermle in future. This is a wholly poor clock from the 1970's. It came from a charity shop and its a good thing I didn't pay much for it. Ideal to learn from. It has now run 4 days and I can settle for that if everything else works as indeed it seems to. I haven't yet seen it specifically stated just what rotation the balance should be making. I have watched it in particular as it approaches the hour where the lift of the levers is highest, and therefore the load on the time train at its greatest, and the balance doesn't seem to lose any rotation remaining at a shade over +/-90 deg. The point is that it seems a whole lot better since I swapped the balance. Perhaps its the geometry of the lever and impeller post and not so much the balance pivots? I can't tell. It must be a very small difference but maybe that's all part of the close tolerances withing the general design?
It would be a lot better if the 8 day clocks in my collection would actually run for 7 but I can wind up twice a week if necessary. I have another (Carriage clock) that cannot quite make 7 days that I wind twice each week. I'm guessing that its mainspring has been shortened as it winds right out but just cannot make it to the seventh day.
Regards & thanks
BerryG

 

[doc_fields]

........ I'm guessing that its mainspring has been shortened as it winds right out but just cannot make it to the seventh day.
Regards & thanks
BerryG

Hi Berry;
Usually if an eight day clock will not run it's allotted time of eight days, the more probable cause would be a "set" mainspring, i.e., it has been coiled up so long that the metallic grain of the spring has weakened for want of a better word, and lost it's strength or tension. New replacement springs can be had for these, and/or new barrels with mainspring included.

For my experience of 17 years in this craft, I have never seen but one or two examples of a shortened mainspring in a barrel, and that never really tastefully done but crudely.

Another factor can have a bit of a play in this also, such as misalignment of wheels due to wear with the pivots cutting into the plates. I have been rebuilding these, some with success, others not so, just for the challenge of it. I keep a dozen or so in a special box to tinker with or replace in clocks I sell. I am also thinking about trying a jeweled time train from above the minute wheel arbor to the escape wheel. The mainspring barrel, first wheel and minute wheel arbor will be done with Butterbearings©. Trying to keep the friction down!.............................gary

 

[Berry Greene]

Thanks Gary
Very interesting. Jewelled bearings eh! That would up the quality when done would it not? Set mainsprings. Yes I have heard of that but not knowingly encountered one. That carriage clock could be I suppose....
My Hermle 340-020 is still going. Just 12 more hours to 7 days now. Quite a surprise. I will pay closer attention to the lever pivots and surfaces before I re-install the movement.
Thank you so much for your post and your interest. Much appreciated.
Best regards, BerryG