XO controlled 480Hz Oscillator

On Feb 2, 9:05 am, Chris <christopher.man...@gmail.com> wrote:



On Feb 2, 8:46 am, Tim Wescott <t...@seemywebsite.com> wrote:

On Tue, 02 Feb 2010 08:29:47 -0800, Chris wrote:
What is the simplest way to get 480Hz from a crystal controlled
oscillator?  Looks like most of the pre-packaged XO's and VCXO, seem to
put out much higher frequencies.  Would a series of dividers be the best
way?

Thanks,
Chris KQ6UP

480Hz what?  Forever constant frequency square wave?  Yes.

No need to vary the frequency.  Yes, square wave 480Hz 12V swing into
a high Z load.

If you want to vary the frequency you may want to use a programmable
divider, and the easiest way to do _that_ may well be to use a small
microprocessor, particularly one with a hardware timer.

For this application the frequency is fixed, but it needs to be
accurate.  I don't think a xtal oven is called for, but I do need the
rock.

If you want a sine wave you could either divide down to a 480Hz square
wave and filter, at the cost of a fairly elaborate analog filter.  Or you
could use that microprocessor again to generate a 480Hz almost-sine wave
to a DAC or PWM, in which case you can use a much less elaborate (and
therefor easier to design and cheaper) analog filter.

Square wave, sorry for not clarifying.

If you can live with a healthy bit of output impedance, you could do this
with PIC, a resistor, a cap, and either a crystal and it's capacitors or
a crystal oscillator that'll be bigger than the PIC.

--www.wescottdesign.com

Thanks,
Chris KQ6UP

I'm old school, so I would build two divider chains out of logic
gates. The first divider gets you to 960Hz, but it probably won't be a
square wave. You use a binary counter and some gates to do the
division. Follow up with a divide by two to get a square wave at
480Hz. You need a level shifter to get the larger 12V signal.

You should investigate if the uP scheme causes jitter in the signal.
In the dark ages when I played with uP, they always had peripheral
timer chips that the uP could control so that the division would be
jitter free. Maybe modern uP have these built in.

On Feb 2, 11:25 am, "m...@sushi.com" <m...@sushi.com> wrote:



On Feb 2, 9:05 am, Chris <christopher.man...@gmail.com> wrote:

On Feb 2, 8:46 am, Tim Wescott <t...@seemywebsite.com> wrote:

On Tue, 02 Feb 2010 08:29:47 -0800, Chris wrote:
What is the simplest way to get 480Hz from a crystal controlled
oscillator?  Looks like most of the pre-packaged XO's and VCXO, seem to
put out much higher frequencies.  Would a series of dividers be the best
way?

Thanks,
Chris KQ6UP

480Hz what?  Forever constant frequency square wave?  Yes.

No need to vary the frequency.  Yes, square wave 480Hz 12V swing into
a high Z load.

If you want to vary the frequency you may want to use a programmable
divider, and the easiest way to do _that_ may well be to use a small
microprocessor, particularly one with a hardware timer.

For this application the frequency is fixed, but it needs to be
accurate.  I don't think a xtal oven is called for, but I do need the
rock.

If you want a sine wave you could either divide down to a 480Hz square
wave and filter, at the cost of a fairly elaborate analog filter.  Or you
could use that microprocessor again to generate a 480Hz almost-sine wave
to a DAC or PWM, in which case you can use a much less elaborate (and
therefor easier to design and cheaper) analog filter.

Square wave, sorry for not clarifying.

If you can live with a healthy bit of output impedance, you could do this
with PIC, a resistor, a cap, and either a crystal and it's capacitors or
a crystal oscillator that'll be bigger than the PIC.

--www.wescottdesign.com

Thanks,
Chris KQ6UP

I'm old school, so I would build two divider chains out of logic
gates. The first divider gets you to 960Hz, but it probably won't be a
square wave. You use a binary counter and some gates to do the
division. Follow up with a divide by two to get a square wave at
480Hz. You need a level shifter to get the larger 12V signal.

You should investigate if the uP scheme causes jitter in the signal.
In the dark ages when I played with uP, they always had peripheral
timer chips that the uP could control so that the division would be
jitter free. Maybe modern uP have these built in.

http://tinyurl.com/yh249wg

This looks like exactly what I need.  Would this have this have a
jitter problem?

Chris

On Feb 2, 11:38 am, Chris <christopher.man...@gmail.com> wrote:



On Feb 2, 11:25 am, "m...@sushi.com" <m...@sushi.com> wrote:

On Feb 2, 9:05 am, Chris <christopher.man...@gmail.com> wrote:

On Feb 2, 8:46 am, Tim Wescott <t...@seemywebsite.com> wrote:

On Tue, 02 Feb 2010 08:29:47 -0800, Chris wrote:
What is the simplest way to get 480Hz from a crystal controlled
oscillator?  Looks like most of the pre-packaged XO's and VCXO, seem to
put out much higher frequencies.  Would a series of dividers be the best
way?

Thanks,
Chris KQ6UP

480Hz what?  Forever constant frequency square wave?  Yes.

No need to vary the frequency.  Yes, square wave 480Hz 12V swing into
a high Z load.

If you want to vary the frequency you may want to use a programmable
divider, and the easiest way to do _that_ may well be to use a small
microprocessor, particularly one with a hardware timer.

For this application the frequency is fixed, but it needs to be
accurate.  I don't think a xtal oven is called for, but I do need the
rock.

If you want a sine wave you could either divide down to a 480Hz square
wave and filter, at the cost of a fairly elaborate analog filter.  Or you
could use that microprocessor again to generate a 480Hz almost-sine wave
to a DAC or PWM, in which case you can use a much less elaborate (and
therefor easier to design and cheaper) analog filter.

Square wave, sorry for not clarifying.

If you can live with a healthy bit of output impedance, you could do this
with PIC, a resistor, a cap, and either a crystal and it's capacitors or
a crystal oscillator that'll be bigger than the PIC.

--www.wescottdesign.com

Thanks,
Chris KQ6UP

I'm old school, so I would build two divider chains out of logic
gates. The first divider gets you to 960Hz, but it probably won't be a
square wave. You use a binary counter and some gates to do the
division. Follow up with a divide by two to get a square wave at
480Hz. You need a level shifter to get the larger 12V signal.

You should investigate if the uP scheme causes jitter in the signal.
In the dark ages when I played with uP, they always had peripheral
timer chips that the uP could control so that the division would be
jitter free. Maybe modern uP have these built in.

http://tinyurl.com/yh249wg

This looks like exactly what I need.  Would this have this have a
jitter problem?

Chris

All electronics has jitter. A logic gate is an analog circuit after
all. But a uP could have issues if not programmed correctly,
especially if interrupts are used. [Interrupts have to be serviced in
a timely manner.] There is some skill in doing real time programming,
though the divider task doesn't sound complicated.

Assuming you could find that part, it would be part of the first half
of the circuit. A ripple counter needs to be clocked on one cycle,
then read on the next. [I'm thinking of ripple counters I've done in
silicon, not off the shelf, so you need to read the datasheet.] You
clock it, let the dividers ripple down the chain, then "sample" the
result on the other half of the cycle. That is, you do the comparison
with logic gates after the flip flops have settled. This will yield a
short pulse when the countdown condition has been met. You then follow
up with a divide by two to square it up.

I think in terms of gates, but I rarely build PCB products. You may
want to investigate the uP, or even some programabe logic chip. What
you do on a PCB is different than custom IC design.

On Feb 2, 9:58 am, Silvar Beitel <silverbee...@net1plus.com> wrote:



On Feb 2, 11:29 am, Chris <christopher.man...@gmail.com> wrote:

What is the simplest way to get 480Hz from a crystal controlled
oscillator?  Looks like most of the pre-packaged XO's and VCXO, seem
to put out much higher frequencies.  Would a series of dividers be the
best way?

Thanks,
Chris KQ6UP

Would a CD4060B + 7.86432 MHz crystal + 1 resistor (maybe 2) + 2 caps
running at 12 volts work for you?  Output (rail-to-rail square wave)
from Q14.

--
Silvar Beitel

That sounds pretty easy divide by 2^14.  Would that be a custom
crystal, or a standard freq?

I will be trying to repair a 480Hz standard tuning fork.  It is
actually a 960Hz fork that had it's divide by 2 flip flop die.  It is
from a 70's vintage piece of equipment.  The fork is hermetically
sealed with solder in a can.  I try to get in there tonight, and if
the dead part is some obsolete IC or something, I will probably
replace with a crystal and divider to get my 480Hz.  A crystal would
be more quiet and stable I believe.

On Feb 2, 7:34 pm, Chris <christopher.man...@gmail.com> wrote:



On Feb 2, 9:58 am, Silvar Beitel <silverbee...@net1plus.com> wrote:

On Feb 2, 11:29 am, Chris <christopher.man...@gmail.com> wrote:

What is the simplest way to get 480Hz from a crystal controlled
oscillator?  Looks like most of the pre-packaged XO's and VCXO, seem
to put out much higher frequencies.  Would a series of dividers be the
best way?

Thanks,
Chris KQ6UP

Would a CD4060B + 7.86432 MHz crystal + 1 resistor (maybe 2) + 2 caps
running at 12 volts work for you?  Output (rail-to-rail square wave)
from Q14.

--
Silvar Beitel

That sounds pretty easy divide by 2^14.  Would that be a custom
crystal, or a standard freq?

Farnell doesn't stock it, so it probably isn't a standard frequency,
but getting a custom frequency in that range is no big deal.

I will be trying to repair a 480Hz standard tuning fork.  It is
actually a 960Hz fork that had it's divide by 2 flip flop die.  It is
from a 70's vintage piece of equipment.  The fork is hermetically
sealed with solder in a can.  I try to get in there tonight, and if
the dead part is some obsolete IC or something, I will probably
replace with a crystal and divider to get my 480Hz.  A crystal would
be more quiet and stable I believe.

Probably. Look at the 74HC401013 from NCXP or TI if you want to use a
standard crystal - it gives you a programmable 8-bit divider in a
single package, and you can cascade them if you want more bits

http://www.nxp.com/documents/data_sheet/74HC40103.pdf

You will have to buffer the output to get a 12V swing, or use the
slower HEF401013BEY from ST Microelectronics.

--
Bill Sloman, Nijmegen

On Feb 2, 12:43 pm, Bill Sloman <bill.slo...@ieee.org> wrote:







On Feb 2, 7:34 pm, Chris <christopher.man...@gmail.com> wrote:

On Feb 2, 9:58 am, Silvar Beitel <silverbee...@net1plus.com> wrote:

On Feb 2, 11:29 am, Chris <christopher.man...@gmail.com> wrote:

What is the simplest way to get 480Hz from a crystal controlled
oscillator?  Looks like most of the pre-packaged XO's and VCXO, seem
to put out much higher frequencies.  Would a series of dividers be the
best way?

Thanks,
Chris KQ6UP

Would a CD4060B + 7.86432 MHz crystal + 1 resistor (maybe 2) + 2 caps
running at 12 volts work for you?  Output (rail-to-rail square wave)
from Q14.

--
Silvar Beitel

That sounds pretty easy divide by 2^14.  Would that be a custom
crystal, or a standard freq?

Farnell doesn't stock it, so it probably isn't a standard frequency,
but getting a custom frequency in that range is no big deal.

I will be trying to repair a 480Hz standard tuning fork.  It is
actually a 960Hz fork that had it's divide by 2 flip flop die.  It is
from a 70's vintage piece of equipment.  The fork is hermetically
sealed with solder in a can.  I try to get in there tonight, and if
the dead part is some obsolete IC or something, I will probably
replace with a crystal and divider to get my 480Hz.  A crystal would
be more quiet and stable I believe.

Probably. Look at the 74HC401013 from NCXP or TI if you want to use a
standard crystal - it gives you a programmable 8-bit divider in a
single package, and you can cascade them if you want more bits

http://www.nxp.com/documents/data_sheet/74HC40103.pdf

You will have to buffer the output to get a 12V swing, or use the
slower HEF401013BEY from ST Microelectronics.

--
Bill Sloman, Nijmegen

I found a source of dirt cheap CD4060B's.  There are no example
circuits in the data sheet.  Would just use a very generic xtal
oscillator?

On Tue, 02 Feb 2010 11:54:21 -0800, m...@sushi.com wrote:
On Feb 2, 11:38 am, Chris <christopher.man...@gmail.com> wrote:
On Feb 2, 11:25 am, "m...@sushi.com" <m...@sushi.com> wrote:

On Feb 2, 9:05 am, Chris <christopher.man...@gmail.com> wrote:

On Feb 2, 8:46 am, Tim Wescott <t...@seemywebsite.com> wrote:

On Tue, 02 Feb 2010 08:29:47 -0800, Chris wrote:
What is the simplest way to get 480Hz from a crystal controlled
oscillator?  Looks like most of the pre-packaged XO's and VCXO,
seem to put out much higher frequencies.  Would a series of
dividers be the best way?

Thanks,
Chris KQ6UP

480Hz what?  Forever constant frequency square wave?  Yes.

No need to vary the frequency.  Yes, square wave 480Hz 12V swing
into a high Z load.

If you want to vary the frequency you may want to use a
programmable divider, and the easiest way to do _that_ may well
be to use a small microprocessor, particularly one with a
hardware timer.

For this application the frequency is fixed, but it needs to be
accurate.  I don't think a xtal oven is called for, but I do need
the rock.

If you want a sine wave you could either divide down to a 480Hz
square wave and filter, at the cost of a fairly elaborate analog
filter.  Or you could use that microprocessor again to generate a
480Hz almost-sine wave to a DAC or PWM, in which case you can use
a much less elaborate (and therefor easier to design and cheaper)
analog filter.

Square wave, sorry for not clarifying.

If you can live with a healthy bit of output impedance, you could
do this with PIC, a resistor, a cap, and either a crystal and
it's capacitors or a crystal oscillator that'll be bigger than
the PIC.

--www.wescottdesign.com

Thanks,
Chris KQ6UP

I'm old school, so I would build two divider chains out of logic
gates. The first divider gets you to 960Hz, but it probably won't be
a square wave. You use a binary counter and some gates to do the
division. Follow up with a divide by two to get a square wave at
480Hz. You need a level shifter to get the larger 12V signal.

You should investigate if the uP scheme causes jitter in the signal.
In the dark ages when I played with uP, they always had peripheral
timer chips that the uP could control so that the division would be
jitter free. Maybe modern uP have these built in.

http://tinyurl.com/yh249wg

This looks like exactly what I need.  Would this have this have a
jitter problem?

Chris

All electronics has jitter. A logic gate is an analog circuit after all..
But a uP could have issues if not programmed correctly, especially if
interrupts are used. [Interrupts have to be serviced in a timely
manner.] There is some skill in doing real time programming, though the
divider task doesn't sound complicated.

Assuming you could find that part, it would be part of the first half of
the circuit. A ripple counter needs to be clocked on one cycle, then
read on the next. [I'm thinking of ripple counters I've done in silicon,
not off the shelf, so you need to read the datasheet.] You clock it, let
the dividers ripple down the chain, then "sample" the result on the
other half of the cycle. That is, you do the comparison with logic gates
after the flip flops have settled. This will yield a short pulse when
the countdown condition has been met. You then follow up with a divide
by two to square it up.

I think in terms of gates, but I rarely build PCB products. You may want
to investigate the uP, or even some programabe logic chip. What you do
on a PCB is different than custom IC design.

The 4060 has a nice square output.

--www.wescottdesign.com

On Feb 2, 11:54 am, "m...@sushi.com" <m...@sushi.com> wrote:

On Feb 2, 11:38 am, Chris <christopher.man...@gmail.com> wrote:




On Feb 2, 11:25 am, "m...@sushi.com" <m...@sushi.com> wrote:

On Feb 2, 9:05 am, Chris <christopher.man...@gmail.com> wrote:

On Feb 2, 8:46 am, Tim Wescott <t...@seemywebsite.com> wrote:

On Tue, 02 Feb 2010 08:29:47 -0800, Chris wrote:

What is the simplest way to get 480Hz from a crystal controlled
oscillator? Looks like most of the pre-packaged XO's and VCXO, seem to
put out much higher frequencies. Would a series of dividers be the best
way?

Thanks,
Chris KQ6UP

480Hz what? Forever constant frequency square wave? Yes.

No need to vary the frequency. Yes, square wave 480Hz 12V swing into
a high Z load.

If you want to vary the frequency you may want to use a programmable
divider, and the easiest way to do _that_ may well be to use a small
microprocessor, particularly one with a hardware timer.

For this application the frequency is fixed, but it needs to be
accurate. I don't think a xtal oven is called for, but I do need the
rock.

If you want a sine wave you could either divide down to a 480Hz square
wave and filter, at the cost of a fairly elaborate analog filter. Or you
could use that microprocessor again to generate a 480Hz almost-sine wave
to a DAC or PWM, in which case you can use a much less elaborate (and
therefor easier to design and cheaper) analog filter.

Square wave, sorry for not clarifying.

If you can live with a healthy bit of output impedance, you could do this
with PIC, a resistor, a cap, and either a crystal and it's capacitors or
a crystal oscillator that'll be bigger than the PIC.

--www.wescottdesign.com

Thanks,
Chris KQ6UP

I'm old school, so I would build two divider chains out of logic
gates. The first divider gets you to 960Hz, but it probably won't be a
square wave. You use a binary counter and some gates to do the
division. Follow up with a divide by two to get a square wave at
480Hz. You need a level shifter to get the larger 12V signal.

You should investigate if the uP scheme causes jitter in the signal.
In the dark ages when I played with uP, they always had peripheral
timer chips that the uP could control so that the division would be
jitter free. Maybe modern uP have these built in.

http://tinyurl.com/yh249wg

This looks like exactly what I need. Would this have this have a
jitter problem?

Chris

All electronics has jitter. A logic gate is an analog circuit after
all. But a uP could have issues if not programmed correctly,
especially if interrupts are used. [Interrupts have to be serviced in
a timely manner.] There is some skill in doing real time programming,
though the divider task doesn't sound complicated.

Assuming you could find that part, it would be part of the first half
of the circuit. A ripple counter needs to be clocked on one cycle,
then read on the next. [I'm thinking of ripple counters I've done in
silicon, not off the shelf, so you need to read the datasheet.] You
clock it, let the dividers ripple down the chain, then "sample" the
result on the other half of the cycle. That is, you do the comparison
with logic gates after the flip flops have settled. This will yield a
short pulse when the countdown condition has been met. You then follow
up with a divide by two to square it up.

I think in terms of gates, but I rarely build PCB products. You may
want to investigate the uP, or even some programabe logic chip. What
you do on a PCB is different than custom IC design.


The chip that I posted the link to does not look like a programmable
chip. It just looks like 14 cascaded binary dividers.

Regards,
Chris

Dunno if one would go down to 480Hz, but if it did it'd be the way to go.

The simplest way is MCU with external crystal. For 480 Hz output, the
crystal is likely to be a multiple of 3 MHz. The 12.0 MHz would be the
most common.

What is the simplest way to get 480Hz from a crystal controlled
oscillator? Looks like most of the pre-packaged XO's and VCXO, seem
to put out much higher frequencies. Would a series of dividers be the
best way?

Thanks,
Chris KQ6UP

Vladimir Vassilevsky schrieb:

The simplest way is MCU with external crystal. For 480 Hz output, the
crystal is likely to be a multiple of 3 MHz. The 12.0 MHz would be the
most common.


Hello,

but for 480 Hz out of 12 MHz, the divider is 25. If a square wave is
necessary, better use a 24 MHz crystal and divide two times by five and
then by two.

On Feb 2, 11:29 am, Chris <christopher.man...@gmail.com> wrote:

What is the simplest way to get 480Hz from a crystal controlled
oscillator? Looks like most of the pre-packaged XO's and VCXO, seem
to put out much higher frequencies. Would a series of dividers be the
best way?

OK this is probably a stupid idea. But could you phase lock to the
8th harmonic of the 60Hz line frequency?

????

On Feb 2, 11:29 am, Chris <christopher.man...@gmail.com> wrote:
What is the simplest way to get 480Hz from a crystal controlled
oscillator?  Looks like most of the pre-packaged XO's and VCXO, seem
to put out much higher frequencies.  Would a series of dividers be the
best way?

Thanks,
Chris KQ6UP

OK this is probably a stupid idea. But could you phase lock to the
8th harmonic of the 60Hz line frequency?

George H.