# Timing: Beats per hour

Discussion in 'General Clock Discussions' started by gocush, Oct 17, 2016.

1. ### gocush Registered User

Jun 24, 2016
172
3
18
Olympia, Washington
Country Flag:
Region Flag:
#1
I have an app that gives me a current reading of approx. 8325 BPH on a Gilbert clock. How do I find out what the target rate should be?

2. ### Andy Dervan Registered User Gibbs Literary AwardNAWCC Star FellowNAWCC Member

Oct 23, 2002
2,252
4
38
Male
Country Flag:
Region Flag:
#2
It depends upon the pendulum length.

1 meter pendulum is 3600 BPH (typical tall clock)
Timepieces (banjo clocks) are 4680 - 4800 BPH

Andy Dervan

3. ### Willie X Registered User

Feb 9, 2008
11,468
557
113
#3
If its keeping time, you already have your answer. If not, you will need to learn how to calculate the BPH by counting the teeth on the wheels.
Willie X

4. ### Tinker Dwight Registered User

Oct 11, 2010
13,666
63
0
Calif. USA
#4
The exact way is to count the teeth on the pinions and wheels between
the escapement and the minute hand.

Some clock books have a number of common beat rates.

You can carefully time is for some length of time using an accurate
reference clock ( like for 24 hours ).
You then calculate how much fast or slow it is and apply that ratio the
the one you measured with the app.
Tinker Dwight

5. ### shutterbug Moderator NAWCC Member

Oct 19, 2005
39,932
582
113
Male
Self employed interpreter/clock repairer
Iowa
Country Flag:
Region Flag:
#5
If you have or can find an app that simply counts the beats, you can mentally mark a starting place, start the count and let the clock run so the minute hand turns once around the dial. Stop it in the exact place it started. That will be your beats per hour for the clock. It is more accurate if you let it go a few trips around and average the beats by the number of turns.

6. ### Willie X Registered User

Feb 9, 2008
11,468
557
113
#6
Gocu,
A beat counter isn't much use without a book that shows you the beat rates ...
Only thing close in my book is 8277.55 for a Wm. Gilbert Clock Co. Front plate would be stamped 1907 under the handshaft.
Need more info on the clock, like a photo or description.
Willie X

7. ### Tinker Dwight Registered User

Oct 11, 2010
13,666
63
0
Calif. USA
#7
It sounds like the pendulum is missing.
Tinker Dwight

8. ### Willie X Registered User

Feb 9, 2008
11,468
557
113
#8
OP has only spoken once and does not mention a missing pendulum.
Willie X

9. ### harold bain Registered User NAWCC MemberDeceased

Nov 4, 2002
40,850
136
63
Male
deceased
Whitby, Ontario, Canada
Country Flag:
Region Flag:
#9
You must have good ears, I can't hear anything.

10. ### gocush Registered User

Jun 24, 2016
172
3
18
Olympia, Washington
Country Flag:
Region Flag:
#10
No missing pendulum. Here are 2 photos from BEFORE cleaning:

And one after cleaning. The original measurement of 8325 BPH was after cleaning and running well, keeping pretty accurate time. I have tried adjusting the pendulum rod nut down till the measurement reading was near 8277 on my clock tuner app, but it was losing several minutes per day at that rate. So, I am going back to the 8325 rate and calling it good.

BTW several folks have mentioned counting the time train teeth and calculating the target BPH. Can someone point me to the formula? Thanks

11. ### shutterbug Moderator NAWCC Member

Oct 19, 2005
39,932
582
113
Male
Self employed interpreter/clock repairer
Iowa
Country Flag:
Region Flag:
#11
12. ### Tinker Dwight Registered User

Oct 11, 2010
13,666
63
0
Calif. USA
#12
There are special cases with idlers that can confuse and even cases
where it looks like a wheel but for calculations it is a pinion.
I always like keeping track of which way the teeth point.
If the teeth point towards the pallets, you multiply. If the
teeth point towards the minute hand then you divide.
Multiply by 2 when done.
No equations just know that one is direction is multiply and the other
is divide. If you get it wrong, it will be obvious.
Tinker Dwight

13. ### Hudson Registered User NAWCC Member

Jul 19, 2010
168
7
18
Retired
Spicewood
Country Flag:
Region Flag:
#13
Here is a sample of a page from my notes on an Ingraham clock of mine. Hope you can make sense of it.
Calculated BPH on this one is 11,232. It is a tambour style mantle clock with a shorter pendulum than your clock.

Multiply the gears together and also multiply that by 2. Then divide that answer by the result of multiplying the pinions together. Hope this helps. Like others mention it is important not to get confused by gears that don't go into the equation, like the second wheel in the example. That is why I sometimes sketch the time train like in the example.

File size:
230 KB
Views:
280
14. ### gocush Registered User

Jun 24, 2016
172
3
18
Olympia, Washington
Country Flag:
Region Flag:
#14
LaBounty says:
The formula is:

Ratio = Mw/Cp x Hw/Mp

where:

Ratio = # of revolutions of the minute hand for one revolution of the hour hand.
Mw = # of teeth on the Minute Wheel
Cp = # of leaves on the Cannon Pinion
Hw = # of teeth on the Hour Wheel
Mp = # of leaves on the Minute Pinion

This is explained very well in an article by Archie Perkins, CMW, FBHI, FNAWCC, FAWI, in the September 1988 "Horlolgical Times", pg. 29 - 31.
==================

For clarification, is the "leaf" of a pinion the open slot in between the bars?

Hudson, I like your drawing since it illustrated the need to bypass the "transfer" wheels.

Tinker, can you show us with a photo or drawing how it works for you? I'm a graphic learner and like to see things. Thanks.

Paul

15. ### Tinker Dwight Registered User

Oct 11, 2010
13,666
63
0
Calif. USA
#15
Last edited: Oct 19, 2016
First, this particular quote from LaBounty just shows the motion works ratio
that is normally 12:1 but is some cases is 24:1.
We can use Hudson's drawing. The second wheel acts like an idler in the
path from the escapement to the minute hand and could be left out of the equation.
Still, for demonstration of my method, I'll leave it in so that the path is complete.

The teeth on the escapement point to the pallets. These we multiply.

The teeth on the escapements pinion point to the minute hand ( following the train ).
These we divide.

The teeth on the 4th wheel point to the pallets so multiply.

The teeth on the 4th wheel's pinon point to the hour hand so divide.

The teeth on the 3th wheel point to the pallets so multiply.

The teeth on the 3th wheel's pinon point to the hour hand so divide.

Now, that second wheel.
It has teeth pointing to the pallets and teeth pointing to the minute
hand so for completeness we will both multiply and divide ( a math student
would note that these cancel ).

The teeth on the center wheel point to the pallets so we multiply.

To put that all together:

(((( 48 / 7 ) * 42 / 7) * 42 / 8 ) * 60 / 60 ) * 26 = 5616
or a ratio of 5616:1
Now, each tooth of the escapement it used twice so
2 * 5615 = 11232 ticks per hour.
So for each rotation of the minute hand of 1 hour there are 11232 beats or BPH.

It is easier to do without remembering the formula if you just keep two multiplied list
of numbers of teeth pointing each way in the train. It does help to make a drawing.

Towards the pallets: 48 * 42 * 42 * 60 * 26 = 132088320
Towards the minute hand: 7 * 7 * 8 * 60 = 23520

if we now do the calculation wrong, by dividing wrong ( 23520 / 132088320 ),
we get 0.000178062678..... An obviously wrong number.
If we divide the other way, ( 132088320 / 23520 ) we get 5616. A much better looking number.
Just don't forget to multiply by two because each tooth of the escapement
is used twice, once for the tic and once for the toc.

As for leafs, it is just another word for teeth of the pinion. It doesn't make
any difference if you count teeth or spaces between teeth as the number will
be the same.
Tinker Dwight

16. ### gocush Registered User

Jun 24, 2016
172
3
18
Olympia, Washington
Country Flag:
Region Flag:
#16
So, I decided to make a spreadsheet to illustrate this and create a template which can easily be applied to each clock as I work on it.
Here is the example that Hudson showed in his drawing:

I haven't figured out how to apply LaBounty's formula which seems to come from the opposite direction. Any help in understanding his method would be appreciated. Thanks. And if anyone wants this spreadsheet, send me a PM.

Paul

File size:
80.7 KB
Views:
143
17. ### Tinker Dwight Registered User

Oct 11, 2010
13,666
63
0
Calif. USA
#17
Last edited: Oct 19, 2016
The part you show in your post is not useful for determining the beat rate.
Look at the second half of that post by LaBounty. It basically describes
what I've posted but slightly differently.
Read from where he says:
Now, on to your second question of how to determine beat rate...on
down and ignore the part above, he was talking about a different problem
there.
The reason I describe it as I do is because there are often different routing
from the center shaft to the escapement.
In some clocks, power from the spring ( or whatever ) goes through the
center wheel.
In other cases, there are more or less wheels in the path.
Following the path and noting the direction the teeth point along the path
always works and is the least confusing. As I've shown, you don't even need
to remember some formulas.
LaBounty starts at the center wheel and goes to the escapement. He doesn't
state how to deal with idler wheels. He only talk about wheels and pinions
while other interesting teeth count sizes exist. If a tooth count is larger
than the wheel it is meshed with, is it a pinion or a wheel?
Looking at the direction the teeth points along the path always works,
with no confusion.
The path from the escapement to the center wheel is almost easier to follow.
Tinker Dwight

18. ### gocush Registered User

Jun 24, 2016
172
3
18
Olympia, Washington
Country Flag:
Region Flag:
#18
Tinker, we're saying the same thing, only I'm doing it in graphic and table format and you are describing in text form.
SumProduct of the Blue Wheels on my graphic (EW to Min Wheel): 2x48x42x42x26 divided by the
SumProduct of the Black Pinions on my graphic: 7x7x8

Likewise, LaBounty is the same except he starts at the "Center Wheel" and moves to the EW. I just didn't see that his Center Wheel was what I called the Min wheel.

Thanks,
Paul

Oct 11, 2010
13,666
63
0
Calif. USA
#19
20. ### Hudson Registered User NAWCC Member

Jul 19, 2010
168
7
18
Retired
Spicewood
Country Flag:
Region Flag:
#20
Last edited: Oct 20, 2016
OK, here is another page from my notebook showing the Time Train on a grandfather clock of mine. In my notes the "minute" gear is the gear that turns at the same rate as the minute hand (once per hour). Perhaps my nomenclature is incorrect, but my math is correct! Anyway, it helps me to sketch it out the path from the escape wheel to the minute hand.

File size:
194.3 KB
Views:
105
21. ### Tinker Dwight Registered User

Oct 11, 2010
13,666
63
0
Calif. USA
#21
The math is correct. It clearly shows the reason to draw the
path out.
Notice that the size at the center wheel, one is 40 and the other is 30.
The 40 is what one would normally call the pinion while the 30 is what
would be called the wheel.
Tinker Dwight

22. ### KurtinSA Registered User NAWCC Member

Nov 24, 2014
3,014
72
48
Aerospace Engineer (Ret.)
San Antonio, TX
Country Flag:
Region Flag:
#22
The equation at the bottom of that figure doesn't compute out to 3600 BPH. The figure shows 70 for the wheel but 71 is used in the equation. Just sayin'!

Kurt

23. ### Tinker Dwight Registered User

Oct 11, 2010
13,666
63
0
Calif. USA
#23
You can see in the drawing where he'd erase the first number and put 70
there.
Tinker Dwight

24. ### Hudson Registered User NAWCC Member

Jul 19, 2010
168
7
18
Retired
Spicewood
Country Flag:
Region Flag:
#24
Last edited: Oct 20, 2016
Good catch guys. I caught it myself when I did the math, but didn't correct my note. The clock does run at 3600 bph. Corrected sketch:

File size:
195.3 KB
Views:
53
25. ### shutterbug Moderator NAWCC Member

Oct 19, 2005
39,932
582
113
Male
Self employed interpreter/clock repairer
Iowa
Country Flag:
Region Flag:
#25
It's still WAY easier with a beat counter

26. ### Willie X Registered User

Feb 9, 2008
11,468
557
113
#26
If you want an exact number, that will have to come from a calculation. Electronics can only give you an approximate number, which is usually good enough.
The only thing you can take-to-th-bank is what the minute hand says after a 7 day run. â˜º
Willie X

27. ### harold bain Registered User NAWCC MemberDeceased

Nov 4, 2002
40,850
136
63
Male
deceased
Whitby, Ontario, Canada
Country Flag:
Region Flag:
#27
If it isn't slipping

28. ### shutterbug Moderator NAWCC Member

Oct 19, 2005
39,932
582
113
Male
Self employed interpreter/clock repairer
Iowa
Country Flag:
Region Flag:
#28
Not so, Willie. The modern beat counters are pretty accurate, and counting the beats required for the minute hand to travel once around the dial is also very accurate. Once you determine the BPH, you STILL rely on the electronics to set it, right? No difference. Otherwise, counting the teeth is futile as well.

29. ### KurtinSA Registered User NAWCC Member

Nov 24, 2014
3,014
72
48
Aerospace Engineer (Ret.)
San Antonio, TX
Country Flag:
Region Flag:
#29
Uh, there's still another "71" on the page...nearer to the top!! Not that anybody's really making a big deal of it or anything!!

Kurt

30. ### Tinker Dwight Registered User

Oct 11, 2010
13,666
63
0
Calif. USA
#30
Beats can be funny numbers. Unless you know what to expect, you might never
guess from the method you propose.
As an example, I have a clock in front of me with a beat of 11567.02041 and another
with 10462.5.
I doubt any thing your beat counter would give would not make any sense of these.
Tinker Dwight

31. ### Willie X Registered User

Feb 9, 2008
11,468
557
113
#31
Bugs,
How would you account for the extraneous clicks and pops a movement would make in an hour? Also, your stop/start point will always be guesswork.
I will stand by my last post. â˜º
Willie X

32. ### shutterbug Moderator NAWCC Member

Oct 19, 2005
39,932
582
113
Male
Self employed interpreter/clock repairer
Iowa
Country Flag:
Region Flag:
#32
The start/stop point is compensated for if you allow several trips around the dial and average the beats. The extraneous clicks and pops are eliminated by turning the sensitivity down so that only the "real" beat is detected. It really is quite accurate.
Tinker - your fraction measurements are just as hard for you to set at the end as it would be for a beat counter to detect. I don't see a big issue.
The bottom line is getting the clock close enough to be acceptable. It takes a LOT of fiddling to get a clock to keep perfect time ..... if possible at all.