# Tuning fork oscillator, any ideas?

#### kcin

Joined Oct 19, 2009
2
I'm a visual artist (never studied electronics but can read a circuit diagram and do build stuff) and I work also lot with sound. I'm making a work that uses 2 tuning forks with mirrors on the tines. As the forks vibrate a laser beam reflects off the mirrors and projects lisajous patterns. I have been having problems driving the forks in a reliable way. I used a simple buzzer type circuit before but the contacts burn out over time, then I came across the idea of doing it with 2 coils: a drive and pick up type arrangement using a driver and an amp so the electronics put the forks into self-oscillation. I'm struggling to find an overall solution and a simple circuit that will do this and would welcome any suggestions. The forks are about 20cm tall.

Best Nick

#### MikeML

Joined Oct 2, 2009
5,444
To turn the fork into an oscillator, you need five things: the fork itself, a sensor to convert the instantaneous position of one of the tines into an electrical signal, a power supply, an amplifier, and finally a method of using an electrical signal to impart movement back into the tines...

Seems like you can take these one at a time.

1. Is the fork made of a ferrous material which will interact with a magnetic field?

If yes, then reluctance based sensing would work. So would a solenoid drive system.

If no, then the sensing could be optical (use the laser directly), acoustic (use an electret mic), capacitive (make the tine one plate of a parallel plate capacitor)...
Driving the tine could be acoustic, ...

This might get the discussion started

#### flat5

Joined Nov 13, 2008
403
Perhaps a small motor with a friction wheel could rub the fork. A violin bow technique.

If the fork is steel then I think a coil (or tape head) with an ac signal tuned to approximately the same freq. as the fork placed very close to the fork might excite it.

A fishing line (string) connected to the fork and a transducer (speaker?) might excite it.

Last edited:

#### SgtWookie

Joined Jul 17, 2007
22,201
What is/are the natural frequency/frequencies of the tuning forks?

#### kcin

Joined Oct 19, 2009
2
Thanks for the prompt replies.
The forks are ferrous and 64 hz

#### MikeML

Joined Oct 2, 2009
5,444
Something like this? Pick up coil biased to create a magnetic field. Could be one of those old suction-cup magnetic pickups used to record telephone conversations, or a sensitive relay with the clapper removed. Driver coil= old speaker driver, headphone driver, relay/solenoid coil with clapper/plunger removed. Amp inverting, reverse polarity of either driver coil or pick-up coil to get positive feedback. Gain TBD, depends on air gaps, number of turns in the coils, etc.

#### Attachments

• 43.5 KB Views: 123

#### Wendy

Joined Mar 24, 2008
21,840
Actually, all you need is a stable oscillator tuned to the forks frequency. It can be a square wave, it doesn't matter. I had the same thought as others, but realized the fork does not need to be be the frequency selective component, electronics is much more reliable on that front.

If the fork is ferrous, you can use pulses form a coil (or two, one on each side). If it isn't attach a speaker magnet to the base (basically disassemble a cheap speaker).

#### SgtWookie

Joined Jul 17, 2007
22,201
I like MikeML's idea. I don't think it would be self-starting.

With an independent oscillator, unless you were right on the tuning fork's frequency, you would wind up spending half the time increasing the tuning fork's oscillations, and the other half dampening it.

#### CDRIVE

Joined Jul 1, 2008
2,219
Actually, I don't see the need for the tuning forks at all. Why not just glue the mirrors to speaker cones and feed them 60Hz?

Last edited:

#### MikeML

Joined Oct 2, 2009
5,444
Actually, all you need is a stable oscillator tuned to the forks frequency. It can be a square wave, it doesn't matter. I had the same thought as others, but realized the fork does not need to be be the frequency selective component, electronics is much more reliable on that front.
...
The stability/accuracy of a free-running oscillator would have to be very high. I have a couple of tuning fork resonators (fork, coils sealed inside a can), and they have a Q approaching a thousand. I also have a box of old pager reeds, which work on the same principle. They also have a very high Q.

Starting with an oscillator which free-runs very close to the natural frequency of the fork, where that oscillator is injection-locked to a signal which comes from a sensor near the fork would probably work.

#### steveb

Joined Jul 3, 2008
2,431
You could also consider using an ebow, which is used with electric guitars. This is able to drive any of the six strings on the guitar. The usual tuning on guitar is

from low pitch to high pitch. The frequencies are as follows:

82.4 Hz
110 Hz
146.8 Hz
196 Hz
246.9 Hz
329.6 Hz

I see no reason why this will not drive a 64 Hz tuning fork, if it is ferrous.

http://en.wikipedia.org/wiki/EBow

#### MikeML

Joined Oct 2, 2009
5,444
Since the tuning fork is a high-Q resonant system, driving it even +-1Hz from its resonant frequency will not produce hardly any movement of the tines. Tuning the excitation exactly to the fork's resonance will produce large amplitude deflection of the tines. Since the OP attached some mirrors to the tines, likely the resonant frequency will be lower than what is stamped on the fork...

I suggest the following test: take an audio oscillator like a HP 651B (50Ω output) and couple it to a suitable driving coil like I showed in my diagram. Slowly sweep the frequency around 64 Hz. There will be a very narrow range of frequencies that excite the tines. This will give you some idea of how to place the driving coil, and how much signal level it will take to drive the fork.

Just for fun, try exciting the fork at 1/2, 1/3, 1/4 of the resonant frequency, too.

#### SgtWookie

Joined Jul 17, 2007
22,201
Are the tuning forks an essential portion of the design?

CDRIVE's suggestion of just gluing mirrors to speaker cones would certainly simplify the project enormously; then all you would need is a couple of oscillators generating sine waves connected to small audio amplifiers. The frequencies wouldn't be critical at all.

#### THE_RB

Joined Feb 11, 2008
5,430
That's why I suggested a second oscillator to provide a small variation in the freq of the first oscillator. It should sweep the forks and hit the resonance, and the resulting effect of 2 separate tuning forks the OP mentioned and 2 sweeping oscillators should provide some very interesting patterns.

When he said "the forks are about 20cm tall" then it seemed he wants to add something to an existing arts display. So using open-loop oscillators means that they only need to attach to the BOTTOM of the forks.

#### russ_hensel

Joined Jan 11, 2009
825
In the old days some ocscllators used to be controlled by a tuning fork ( unless I made this up ) I think they mounted a coil on the fork ( or 2 coils to make it symetrical ) The added mass will change the frequency. A magnet is inserted into the coils, not touching the fork. The coils can be used as both sensors and drive for the fork, I think it forms a high Q resenator at the tuning fork frequency. And I found this reference http://www.radiomuseum.org/forum/tuning_fork_oscillators.html Google has more.

#### hgmjr

Joined Jan 28, 2005
9,029
This tuning fork technique was the basis for one of the first extremely accurate transistor wrist watches. It was the Bulova Accutron. I still have my Accutron Space-dial model I bought with my summer money ($160) back in 1967. You could see all of the guts. And it made the neatest audible hum. Here is a link to an image of the watch to which I am referring. hgmjr Last edited: #### Bernard Joined Aug 7, 2008 5,228 There were a lot of 400 cps forks used in WWII in avionics. In the 50's we user 50 cps forks to run synchronous motor to add timing lines on ooscillographs. The tynes had small rod magnets about 1/3 way from common juncture. coil driven by 12SL7's & 12SN7's. Checked with WWV once a year-never made a correction. #### THE_RB Joined Feb 11, 2008 5,430 This tuning fork technique was the basis for one of the first extremely accurate transistor wrist watches. It was the Bulova Accutron. I still have my Accutron Space-dial model I bought with my summer money ($160) back in 1967. You could see all of the guts. And it made the neatest audible hum.

Here is a link to an image of the watch to which I am referring.

hgmjr
That's a seriously cool watch!!

#### BillB3857

Joined Feb 28, 2009
2,493
This tuning fork technique was the basis for one of the first extremely accurate transistor wrist watches. It was the Bulova Accutron. I still have my Accutron Space-dial model I bought with my summer money (\$160) back in 1967. You could see all of the guts. And it made the neatest audible hum.

Here is a link to an image of the watch to which I am referring.

hgmjr
The small brass colored "C" clips on the facing pole pieces of the tuning fork were used to fine tune the frequency of the fork. One fork carried a small flex wire with a ruby pawl on it that engaged a gear. A second fixed pawl prevented the gear from rotating in the wrong direction. Jewelers had a small fixture that they laid the watch on and it picked up the fork frequency and compared it to a standard.

#### CDRIVE

Joined Jul 1, 2008
2,219
And it made the neatest audible hum.

Here is a link to an image of the watch to which I am referring.

hgmjr
Seeing the original Accutron again damn near brought tears to my eyes. Back in the 60s my Dad bought one and he would sit in front of his SX42 that he bought in 1946 or 47 and exclaim "Dead balls on with WWV"! I miss him.

Last edited: