Jump to content
  • Sign Up to reply and join the friendliest Watch Forum on the web. Stick around, get to 50 posts and gain access to your full profile and additional features such as a personal messaging system, chat room and the sales forum PLUS the chance to enter our regular giveaways.

Recommended Posts

I've designed and built a clock with wooden gears but without applying excessive torque on the chain wheel the gear train will not run.

A little bit of background.

I wanted the clock to have a pendulum, have a second finger and require winding every eight days. The escapement wheel has 30 teeth and is on the shaft for the second hand. An arrangement of gears with a ratio of 1:60 drives the minute hand and a pair of gears with a ratio of 1:12 drives th hour hand. In order to achieve the 8 days between winds a further set of gears with a ratio of about 1:24 drives the hour hand. This makes the gear ratio between the chain wheel and escapement wheel 1:1440; quite a ratio to drive!

With all the gears in place the chain wheel torque to drive the mechanism is too high and would require too much weight on the chain.

However, if I lift the final pinion on the escapemant shaft so it is disengaged from the gear train the gears run smoothly with virtually no torque on the chain wheel. With the exception of the connection of the gear train to the escapement wheel shaft all the other gears are in place and run as required.

I just don't understand why engaging this one gear almost stops the mechanism so dramatically. There are a total of 18 gear wheels and pinions and adding one more element of friction shouldn't have the effect it does. I realise the last straw brakes the camel's back 

I'm using PTFE bearings on silver steel shafts. (I originally used oilite bushes and changed to PTFE to reduce friction but there has been little benefit.)  I have polished the shafts with 1000 grit wet and dry and brasso and all individual gears on their shafts spin extremely freely. I have used a lubricating wax on the gear teeth and polished them. All the involute gears and the support frames are CNC machined and are therfore pretty accurate.

My next step is to strip the wax from the gears and replace it with a dry PTFE lubricant.

Interestingly, I made a proof of concept mechanism using a 3D printer. I wasn't too bothered about accuracy so the bearing fits are loose and the PLA runs on wooden shafts but it runs quite freely. I sprayed the teeth with furniture polish but PLA is quite slippy anyway. So, I conclude that if i can reduce bearing or gear friction all should be well but any advice would be welcome. I've spent a few hyndred manhours getting to this stage and wouldn't like to see them wasted.

A few images would have helped me describe the problem better but I can't find a way of uploading them.

Link to post
Share on other sites

Hi, and welcome to the forum, Bob. Sounds an interesting project, but sadly, I can't help you with your problem, however uploading pics is something I can tell you about. You will need to upload your pics to a third party hosting site such as 'Imgur', 'Flickr', 'Post Image' etc. The forum requires a URL in order to find and display your pics, and your device (phone/tablet/PC) won't generate that by itself. Alternatively, you can join the 'Gallery' here which will do the same. :thumbsup:

Link to post
Share on other sites

Hi Bob,

Yes, some pictures will help to understand where is the problem. But there is one thing that is bothering me: I red that there are 18 gears in the train??? Is this a typing mistake? See, 8 gears is too much, 18 is absolutelly out of range...

Also, not only the material of pivots and bushes is important. Their size is werry important. And, in watchmaking, no polishing with sandpaper is used on pivots - thus the surface will charge with abrasive particles. You need to burnish pivots instead...

Link to post
Share on other sites

There are 18 gear wheels; being 7 pinion/wheel combinations and four separate pinions. The shafts rotate in plain bearings with the steel shafts running in PTFE plain bearings. After using 1000 grit paper on the shafts i further polished them with brasso, a mild abbrasive used for polishing metal.

Could you explain what you mean by, "Their size is very important"?

I'll upload some photos when i have mastered the technology.


Link to post
Share on other sites

OK, this is better. You have 7 wheels, I was thinking they are 18... But still, only 5 wheels are needed for building of 8-day clock. More wheels means more transformations of energy, and more losses…

My English is not so good, but as I know, what You mean by 'shaft', in watchmaking is called 'pivot'.

The wheel on which the chain seats, rotates very  slowly, but the torque and force in bearings are great. Tre escape wheel rotates fast, the torque and force in bearings are small. In order to reduce friction losses, they do pivots and bearings on different wheels with different diameters. Usually diam. of esc. wheel pivots is about 0.6mm, and diam. of chain wheel pivots – about 6 mm. The friction in 0,6mm bearing is much less than the friction in 6mm bearing. And 6mm bearing will bear much more force than 0.6mm… The bushes of the bearings in watchmaking are made usually of brass, sometimes of phosphorous bronze. This materials have lo friction with steel, but need oiling.  Plastic bushes are not traditional, their use is possible, but have in mind that most of lubricants make plastics to shrink in long periods of time. In watchmaking, lubrication is applied ONLY to the pivots, and never to teeth of wheels and pinions.

No polishing by hard abrasive particles it used for pivots, as particles may charge in surface of bearings. Burnishing is the process they use on pivots. Burnishing is done by burnisher – something like a very hard steel file, but with no teeth. Burnisher surface is prepared by grinding, it must have traces of grinding, parallel to the direction of it’s movement. In the process of burnishing, almost no material is taken from the pivot, but the surface of pivot is rather ‘smashed’ and plastic-deformed, and thus, surface hardness is significantly increased. Only oil is applied on burnisher surface.  

To  show pictures here, You need to upload them in some picture site, like postimage or imgur or so. Then copy the link to the image and paste it in the message…

  • Like 1
Link to post
Share on other sites

Thanks for your reply;much appreciated. As you have probably realised i know very little about clocks but suddenly decided to make one.

Some images of the clock here: https://www.flickr.com/photos/188874329@N06/?

The minute hand is on gear4 and there are two other gears between this and pinion 1, the second hand. Unfortunately the second had rotates anticlockwise so there is another gear to reverse the rotation. Between gear4 and the winding drum there are a further two gears which drive wheel 7 which is locked to the winding gear. There are then another set of gears to drive the hour hand. I could send the spreadsheet with the details but i don't think it will add any value.

The gears range from about 25mm to 140mm diameter with modules of between 2 and 4. All the shafts are 6mm silver steel finished to a mirror finish. I used PTFE as it has the fhird lowest coefficient of friction of all solid materials. PTFE if one of the few plastics that remains stable and does not shrink with time or moisture. Under no load the gears spin freely on the shafts and i'm bacoming convinced that the problem lies in the friction between the gear teeth.

I first assembled the gears with untreated gears but it was even more difficult to spin the gears. The lubricating wax is by Liberon and is used for reducing friction on cutting tools used on woodworking machines.

I have tried the the dry PTFE lubricant on a piece of wood and it doesn't seem to reduce the friction.

I think the gears just reach a critical point were the force applied to the winding gear is such that the friction on the bearings and gear faces increases non-linearly to the point were the gears lock, or nearly.

I could reduce the gearing between the winding gear and the minute hand by four which would make it a 2 day clock but i'd rather find a solution to the arrangement i have.

Any ideas on reducing gear friction would be welcome.

Link to post
Share on other sites

Hi Bob,

OK, there is something wrong in the design of this thing...

You have built it in the way they build some reducers, but this type of reducers usually work exactly in opposite direction – the fast rotation of weak and small electric motor is reduced to slow rotation with big torque.

As I see, You use shafts that stay still, and the bearings are in the gears. Thus the size of the bearings is the same for all the gears, and gears can press each other…  OK, this is not so bad, but price is too much loses. You gears are with involute teeth profile, which ads some more loses.

But the main reason for You problem is that the shaft on which seats ‘scape wheel is connected hard to it and must rotate with it. Further more, on this shaft there is another wheel, that rotates with different speed. This wheel is pressed by the next wheel and creates great friction forces in shaft’s bearings in the frame and in it's own bearing… Believe me, this will never work!

The most simple thing You can do is to make this shaft still as the others, and put bearing in the ‘scape wheel like in the other wheels. You must put a spring washer on the shafts to prevent the wheels from pressing each other, especially  the ‘scape wheel. This means no seconds hand… But You can paint a seconds hand directly on the table of ‘scape wheel. It will rotate anticlockwise… In normal clocks, there is only one wheel between center wheel and ‘scape wheel, this way rotation direction is normal.


Link to post
Share on other sites

I think that i understand your reasoning.

Gear 3 currently runs on the shaft to which pinion 1 is fixed and whilst they both run in the same direction they run at different speeds. 

I can fix the shaft on which they currently run and only gear 3 will run on this shaft. I will then have to put pinion 1, fixed to the escape wheel,running on a new seperate shaft.

Is this correct?

Driving the gear box low to high was a concern early in the design but i could see no way around this; hence making the plastic assembly. I used involute gears because i thought they would be the best but from what you say obviously not. What gear form would you recommend?

Thanks once more for your help.


I've taken pinion one off the shaft on which it is currently fixed and had it driven by wheel 2 whilst on a seperate shaft and it seems to work well. I just held the shaft so the gear engaged but will have to make a better trial. I'll keep you posted. Loss of the second hand is no problem.

Link to post
Share on other sites

Hi Bob,

If the loss of the second hand is no problem, then just make the pinion of the escape wheel to rotate free on the shaft and fix the shaft to the frame. No need to put another shaft on different place. But You must put on the shaft some kind of separator to keep the two wheels little away from each other. This separator must be fixed to the shaft.

Involute teeth have many advantages. There is no friction in teeth, they rather roll on each other. The torque is transmitted absolutely smoothly. But there is a disadvantage – they create force that wants to increase distance between gears. This force acts to the bearings and thus leads to increased loses due friction in bearings.

In watches and clocks, they use cycloid teeth.  They create this kind of force too, but much, much less then in involute teeth. Also, historically, cycloid gears appeared much earlier and earliest watchmakers didn’t know involute gears. And, cycloid gears are easier to make (without CNC machines)



In clocks and watches, a Wheel always consist of Table and Pinion. Table is the big gear, usually made of brass. Pinion is the smaller gear, phrased of steel. The pinion has two smaller diameter cylindrical pivots on both ends.  The pivots, together with the holes in the plates, form the bearings. This is the traditional way of clock building, and it has many advantages. At least, loses because of friction in bearings this way are minimized. But, building of clocks with plywood gears is not traditional at all, so no rules here…

Link to post
Share on other sites
  • 2 months later...


I've now made the following changes and things have significantly improved; more later.

  1. I remade all the gears using cycloidal tooth profiles. There is much more freedom in them and hence any meshing friction is minimised.
  2. I fixed all the gears on their own pinions and placed their bearings in the plates. I'm still using PTFE bearings but as one of them is threaded I can adjust the end float.
  3. I have not polished or lubricated the teeth
  4. I had the pinion for the second hand far too long and it projected from the front plate by about 100mm. As it was made from steel it applied differential loads on the bearings in the two plates. I shortened it and made it from wood.

With the initial design it required 3.5kg on the chain to turd the gears but with the above modifications it requires just over .5kg. I believe the main problem was 4. above. 1 unit of frictional torque on the second shaft requires 1500 units of torque on the chain wheel to overcome it.

nevenbekriev; without you i would still have been swimming around in a sea of uncertainty. A big thanks.

I'm now setting up the esacape mechanism and whilst i'm nearly there i'm having some problems which I'll post sepearately under "Dead beat escapement problem".

If you could help on that i'd be grateful.

Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

  • Recently Browsing   0 members

    No registered users viewing this page.

  • Create New...