Kundo kundo clock upgraded. for better swing.

[Raymond101]

Hi I a got a early kundo electronic clock as everyone knows someone had attempted a repair. Got it cheap in really bad condition.
So I decided it needs a TLC and would be a fun project. I have watched many videos of these clocks and they all seem to have what I call a fish tail swing. After entirely stripping it down & cleaning. The suspension spring was totally destroyed. I made a new 1 from a pattern of the original. With the T shape top bar and double leafs 0.03 .
This I found oscillatored from the coil pulses . So I remade it just a straight flex . BTW I buy 1mt roll of stainless springsteel foil 0.03 good for a thousands of suspension springs .
This would keep good time but also had fish tail effect.
So I decided to make a damping copper ring in the coil .left side . This is the single wound coil type 2000mh , 3100ohm . & single magnet. And it had been opened before. So after fitting the copper damping ring. The pendulum rode now swing perfectly symmetrical to center.
The original electronic board just need cleaning as all the components were still within 1% tolerance, amazingly.
Put an oscilloscope on the coil pulse and took a second reading from end of swing with IR sensor. To see if there was any losses in the suspension spring vs power. Very happy the drift different was less than u /sec.
So thought this might help others with this type of clock .
1 beat pulse = 1.5s .. (1.5hz)
90 beats Per minute
5400 BPH .

 

[Raymond101]

Just a data correction the suspension spring plate size is 0.05mm not as 0.03 above typo . The fixing holes should be 10mm center to center apart .

Raymond
The suspension plate in the picture was a first attempt. And was scraped
And the new one built to the drawing.
Which is now perfect
90 peats per minute
Check with frequency counter
Has a time error of 0.0009Us .
I guess it's not quite to NASA STANDARDS. but I think I can get it a little less .

 

[praezis]

5400 BPH

Rather 5400 full swings /min what equals 10800 BPH. In this special case however only one "beat" / swing.

Frank

 

[Raymond101]

Rather 5400 full swings /min what equals 10800 BPH. In this special case however only one "beat" / swing.

Frank

Thanks Frank I would expect that on a mechanical clock.
It's an electronic impulse so the swing to the left I classified as the forward pulse the return is the back impulse trigger to reset the circuit. Timing.
The count was done with an oscilloscope with a reference frequency of 1.5hz the actual time pulse minus spring loss. 1.499989hz .
These clocks do have a problem if the over swing is moving the magnet too far out of the coil the circuit becomes unstable. They are a fun clock to play with .
Sorry I'm a precision nerd if it's not to atleast 10-6 or better it's not good

 

[praezis]

Sorry I'm a precision nerd if it's not to atleast 10-6 or better it's not good

Me too
But 10 -6 is hardly possible with such Kundo.
That‘s why I use to synchronize my electric pendulum clocks with DCF77.

Why my BPH comment? Because it is usually related to a certain pendulum length.

Frank

 

[Raymond101]

Me too
But 10 -6 is hardly possible with such Kundo.
That‘s why I use to synchronize my electric pendulum clocks with DCF77.

Why my BPH comment? Because it is usually related to a certain pendulum length.

Frank

Thanks nice to know I'm not the only time nut on the planet..
But sorry iv now got it to 10-7 .
I make my own suspension spring from high grade foil that I buy in meter roll .
I ground the inside side edges of the suspension spring so to get the precision swing . The foil trends to twist as the pendulum swings though the coil. The damping collar reduces the eddy current making the pulse drop more evenly. Getting the exact swing has to be done with an oscilloscope to get the precision I required .
More of over kill . And I was into nothing is impossible. What surprised me the most I check all the original components and they were spot on . The resistors were 0.05% I used the HP lab meter to double check. & the 2 ge transistor were perfect.
So I decided to make run to its peak max possibly. .
The only week point is, are Duracell up to their name? . If not I have some 10 year battery ok they are even more Over kill as they are $30 a peice . .
Or photo cell charging a supercap.
As the power requirements are low .
Now I'm look for a custom case to build
Totally hermetically sealed glass tube.

Raymond.

 

[praezis]

But sorry iv now got it to 10-7 .

0.009 s/day? For how long do you expect it? Let's talk about 10-7 in 1 month

The only week point is, are Duracell up to their name? . If not I have some 10 year battery ok they are even more Over kill as they are $30 a peice . .

A C-cell will last much longer than 10 years. At least with an Ato, that is not different from a Kundo.

Frank

 

[Raymond101]

How long will it last at that accuracy. Not long I guess as there are the mechanics of this clock .dirt dust etc magnetic interference from lighting etc .
If I get to stay for a month I guess I have achieved a goal .
It's more of a fun thing. With this clock.
My old brocot made in 1850 keeps within 1 minutes a week . But these clock were made to last .
Kundo mechanics are good but cheap design. Made as a more novelty clock which I know I am just see how far I can push it . So far it's spot . But if I turn on the ac unit I see it spike..
As for the battery I'm using a C Duracell which is the fitting it had .
The Duracell say they have a shelf life of 10 years. I had a few of them in storage and when I opened the sealed packing they were leaking. About 5yrs
Yes in theory it should last 10yrs in practice the casing break down..
I have a commercial batteries that are over 30 yrs in my Hp scope . So far still showing good on last cal.
Best regards
Raymond

 

[Raymond101]

Well I finely finish it refurbish the case .
Replaced 3 broken glass sides the back Replace with bronzed mirror though it make it bit different.
Front Glass is the original.
The hands I chromed as the black was hard to see .
Even though I look for an all glass case made way to expensive. So I settle on the next best.