Building Bernard Tekippe’s Precision Regulator

[Chris Raynerd 2Participant @chrisraynerd2]

I don’t know if any of you are familiar with the late Bernard Tekippe, but in the 1990s he designed a lovely high precision regulator with a fairly simple design! He dispensed of the need of the pallet arbor and crutch by mounting the pallets directly on the pendulum. He made a number of these clocks which have shown to keep higher accuracy than other regulators, keeping accuracy of 0.008 sec/day – comparable to other regulators running in a vacuum chamber.

Despite this amazing accuracy and in my opinion, lovely looking design, there is very little info online!

I’m no expert, I just like attempting to build clocks! There is a great thread over on the NAWCC and I’ve been in contact with Steve on there who is building this clock and who has been super helpful and provided me with lots of drawings and information from his work so far. He has given me permission to use some of his CAD models in my videos – once again, I only do them for fun because I enjoy making them! I’m no expert in any of this.

This is definitely a tribute / hommage to Bernard’s fantastic design – if I can get a working clock keeping good time, I’ll be delighted. If I can get it keeping accurate time then that’s a bonus and something to aspire to.

I hope there is some interest in this clock as it seems a great design with such little available information and discussion!

Here is my first video in my efforts in building this clock:

Cheers, Chris

[Michael GilliganParticipant @michaelgilligan61133]

A great tribute to a very interesting design, Chris

… Keep it coming !

MichaelG.

[Andrew CrowParticipant @andrewcrow91475]

I’ll second that, always good to see an interesting clock design being made.

Andy.

[Chris Raynerd 2Participant @chrisraynerd2]

Thanks guys, I’m looking forward to the rest of the build. Onto the pendulum support this afternoon.

[Chris Raynerd 2Participant @chrisraynerd2]

I’ve just continued making the pendulum support bracket in the video below.

I’m on to the pendulum next. Does anyone know a UK supplier of Invar that isn’t going to totally destroy my wallet?

Also, just curious if anyone has any experience with bearings and possibly ceramic bearings in clocks? Quite a few people seem to be using them now and I’m considering that as an option, certainly on the great wheel arbor and possibly the escape wheel arbor.

[John HaineParticipant @johnhaine32865]

Have you considered using carbon fibre pultruded tube? Far cheaper than invar and may have lower tempco. There have been some articles about using small ball bearings in clocks in HJ.

[bernard towersParticipant @bernardtowers37738]

Ihave a friend who is a clockmaker and and he uses bearing exclusively. and he makes a run of ten normally!!!

[bernard towers]

On bernard towers Said:

Ihave a friend who is a clockmaker and and he uses bearing exclusively. and he makes a run of ten normally!!!

Hi, if you could by any luck find out what make and type of bearings, I`m be very greatful. I can see from the research people have done that some are by far better than others.

<em>QUOTE: John Haine: Have you considered using carbon fibre pultruded tube? Far cheaper than invar and may have lower tempco. There have been some articles about using small ball bearings in clocks in HJ.</em>

 

Someone else mentioned this to me but the pallets are mounted to a milled out section of the invar rods. Therefore the design wouldn`t work in tube due to the interesting arrangement of the pallets!

 

Thanks

Chris

[bernard towers]

On bernard towers Said:

Ihave a friend who is a clockmaker and and he uses bearing exclusively. and he makes a run of ten normally!!!

Could do better Bernard!  Please go the extra mile by asking why your friend uses bearings, where he applies them, and what type(s) he uses?

Dave

 

[John HaineParticipant @johnhaine32865]

https://www.bocabearings.com/general/ball-bearings-in-clocks?srsltid=AfmBOooMcgAa4DDcBr4ly5joRJidB7JNnEIUOT1vJdJmxmr7KqWZIKpT

This republishes articles by Rex Swenson with useful information on using ball bearings in clocks.

[bernard towersParticipant @bernardtowers37738]

As you wish SOD, he uses bearings as he say they they offer a train with less friction and the bearings are according to him top quality stainless 2mm id x 5mm od. They are used all through the train with larger bearings on the winding shafts.

[bernard towers]

On bernard towers Said:

As you wish SOD, he uses bearings as he say they they offer a train with less friction and the bearings are according to him top quality stainless 2mm id x 5mm od. They are used all through the train with larger bearings on the winding shafts.

I have done the same with my regulator clock.  I did some comparative tests of friction on the bearings of as supplied versus all lubricant washed out and found that the bearings had significantly less friction when run dry.  My regulator has been going continuously for ten years with no sign of problems with the bearings.

Russell

[John Haine]

On John Haine Said:

Have you considered using carbon fibre pultruded tube? Far cheaper than invar and may have lower tempco. There have been some articles about using small ball bearings in clocks in HJ.

I made the pendulum rod for my regulator from pultruded carbon fibre  tube, mainly because I couldn’t find a source of invar rod in France.  It is claimed to have a slight negative temperature coefficient  which can be compensated for by brass fittings at either end and the suspension spring.  However the tolerance on the expansion coefficient is rather loose and it seems to creep a bit under the weight of the pendulum.

Russell

[BazyleParticipant @bazyle]

Slowly reading the NAWCC link you gave – a page a day! so only on page 3.
Interesting that he CNC cut the 240 tooth great wheel which inevitably meant the profile was made of facets and the surface of each facet would have micro steps from the stepper motor movement, yet nobody suggested either polishing the teeth in the usual way nor ‘running in’ the train initially under electric power for say 1 year equivalent rotation.
I would also suggest making a note of the orientation (ie mark on wheel) wrt X & Y if cnc machined as such a high precision clock might show up the effect of even minute out of round errors.

[Chris Raynerd 2Participant @chrisraynerd2]

Interesting you say about the polishing the wheels – that comes later in the thread 😂

[John HaineParticipant @johnhaine32865]

Chris, a very interesting clock build and it seems to give excellent performance.  After a bit of digging I posted something on the HSN forum here:

https://groups.io/g/Horological-Science-Newsletter/topic/clock_b_vs_tekippe_s_clock/112278009

I’d be interested in seeing some long-term performance data for the clock.

[Chris Raynerd 2Participant @chrisraynerd2]

Hi John

Thanks for the post and you made a good point in your link. It does seem a very simple design to get such great results.

I’ll be honest in that the first stage is to get the clock up and running. It If I can do that, then I’ll worry about performance and improvements. I’ve built a few clocks now but I’ve never built a regulator so looking at compensation for different variables is all quite daunting. I think I need to do a lot more reading. However, that said, I’d like to try and get it as accurate as I can and I’ll certainly post microset or similar data once I am up to that point.

Chris

[Chris Raynerd 2Participant @chrisraynerd2]

The pendulum is now on and fitted!

[bernard towersParticipant @bernardtowers37738]

Chris, you will find that if you turn the end of the slots into holes of at least double the slot width you will find it easier to tighten the suspension screws. IMHO

[bernard towers]

On bernard towers Said:

As you wish SOD, he uses bearings as he say they they offer a train with less friction and the bearings are according to him top quality stainless 2mm id x 5mm od. They are used all through the train with larger bearings on the winding shafts.

I was hoping for more detail and a rationale!   Perhaps the friend chooses not to explain because it’s a trade-secret!

I guess the value of fitting bearings depends on the size of the clock?  Due to their mass ball bearings have considerably more inertia than pivots running in a bush, and pivots are also low-friction, especially when run in a jewel.  However, as bearings can take more load than pivots, I feel they make sense in large clocks that turning big hands and are powered by  hefty weights.  Bearings being easier to fit than pivots may be another benefit.

But I note watches and most domestic clocks have pivots rather than bearings.  The reasons are not clear cut.

I’ve no doubt bearings will work in clocks, but it may be unwise to assume they are more effective than pivots, especially if the bearings are chosen carelessly.  When comparing bearings, good to know what the moving part weighs and the coefficient of friction. Bearings are another aspect of clocks that gets more complicated the closer one looks.  There must be a point at which the advantage switches from pivots to bearings.  I’d like to know roughly where it is and shall have to do more digging.

Ta,

Dave

PS.   My attempt at a precision clock eliminates stiction, friction, inertia, and wear by replacing the mechanical gear train with an electronic counter.  Still difficult!  Now I have new worries: sunlight on the sensor, electronics, computer issues, spikes on the mains, power cuts etc etc…

 

 

 

[John HaineParticipant @johnhaine32865]

Dave, look at my link above to Rex Swenson’s articles on using ball races in clocks on the Boca Bearing site.

[Nigel BennettParticipant @nigelbennett69913]

You have a PM, Chris!

Nigel

[Chris Raynerd 2Participant @chrisraynerd2]

I do – sorry I’ve replied 🙂

[SillyOldDuffer]

On SillyOldDuffer Said:
[…] There must be a point at which the advantage switches from pivots to bearings.  I’d like to know roughly where it is and shall have to do more digging.[…]

This might help you define one boundary, Dave

https://waitingtrain.blogspot.com/2008/08/

Have a look at the text on the knurled disc on the pendulum support 🙂

It’s there for the avoidance of Brinelling and such.

MichaelG.

[Chris Raynerd 2Participant @chrisraynerd2]

A little bit more progress… cut one of the big wheels and it went well so a little more relaxed now 🙂

Clock Wheel Cutting – Tekippe Regulator Build Part 4 [Horology / Clockmaking]

[Author]

Posts