Abstract

We demonstrate the removal of the dither modulation from an iodine-stabilized He–Ne laser by using a frequency-modulated acousto-optic modulator and feed-forward techniques. This procedure reduces the linewidth of the beat between this laser and a flywheel He–Ne laser from 6 MHz to 8 kHz, the undithered beat linewidth being 7 kHz. Dither suppression greatly reduces counter errors during beat measurements from stroboscopic effects between the counter’s gate and the frequency of the dither modulation and increases the utility of the already formidable array of dithered laser frequency standards by making locking to them an easier task.

© 2000 Optical Society of America

Full Article  |  PDF Article

References

  • View by:
  • |
  • |
  • |

  1. T. J. Quinn, Metrologia 30, 524 (1994).
    [CrossRef]
  2. A. J. Wallard, J. Phys. E 5, 926 (1972); J. L. Hall, G. Kramer, and R. L. Barger, in Conference on Precision Electro Magnetic Measurement Digest 1972 (Institute of Electrical and Electronics Engineers, New York, 1972), p. 75.
    [CrossRef]
  3. Early digital frequency counters typically accumulate the number of positive zero crossings in a set time, giving correct counts but limited precision. A newer design (HP53132A) has vastly higher resolution by using a random sampling. At carrier frequencies above 50 MHz this strategy works even with the 6-MHz peak-to-peak FM discussed in this Letter. But, at much below this carrier frequency, such deep FM causes erratic operation, giving errors of many megahertz.
  4. M. L. Eickhoff and J. L. Hall, IEEE Trans. Instrum. Meas. 44, 155 (1995).
    [CrossRef]
  5. See Ref. 1. For example, the Comité Consultatif du Longeur (CCL, formerly CCDM) has recommended values for five He–Ne laser lines and one argon-ion laser line locked to several lines in I2 by dither-locking techniques.
  6. The unit used was Model 925A from Berkeley Nucleonics. This information is provided for technical communication purposes only and does not constitute an endorsement by the National Institute of Standards and Technology or the University of Colorado.
  7. Feedback alone could in theory cancel such FM and was in fact tried with the goal of inherently correcting for the nonlinearity of the AOM driver. However, the noise performance of feedback was found to be far inferior to feed forward, because to have a useful gain of G at modulation frequency f a required feedback servo bandwidth of greater than Gf brings added noise across this entire frequency region. Feed forward can provide exact cancellation of a known perturbation with essentially no bandwidth and hence practically zero added noise.
  8. M. S. Taubman and J. L. Hall, “Precise removal from a laser beam of large frequency modulations generated either internally or externally to the laser,” , May19, 1994.

1995

M. L. Eickhoff and J. L. Hall, IEEE Trans. Instrum. Meas. 44, 155 (1995).
[CrossRef]

1994

T. J. Quinn, Metrologia 30, 524 (1994).
[CrossRef]

1972

A. J. Wallard, J. Phys. E 5, 926 (1972); J. L. Hall, G. Kramer, and R. L. Barger, in Conference on Precision Electro Magnetic Measurement Digest 1972 (Institute of Electrical and Electronics Engineers, New York, 1972), p. 75.
[CrossRef]

Eickhoff, M. L.

M. L. Eickhoff and J. L. Hall, IEEE Trans. Instrum. Meas. 44, 155 (1995).
[CrossRef]

Hall, J. L.

M. L. Eickhoff and J. L. Hall, IEEE Trans. Instrum. Meas. 44, 155 (1995).
[CrossRef]

M. S. Taubman and J. L. Hall, “Precise removal from a laser beam of large frequency modulations generated either internally or externally to the laser,” , May19, 1994.

Quinn, T. J.

T. J. Quinn, Metrologia 30, 524 (1994).
[CrossRef]

Taubman, M. S.

M. S. Taubman and J. L. Hall, “Precise removal from a laser beam of large frequency modulations generated either internally or externally to the laser,” , May19, 1994.

Wallard, A. J.

A. J. Wallard, J. Phys. E 5, 926 (1972); J. L. Hall, G. Kramer, and R. L. Barger, in Conference on Precision Electro Magnetic Measurement Digest 1972 (Institute of Electrical and Electronics Engineers, New York, 1972), p. 75.
[CrossRef]

IEEE Trans. Instrum. Meas.

M. L. Eickhoff and J. L. Hall, IEEE Trans. Instrum. Meas. 44, 155 (1995).
[CrossRef]

J. Phys. E

A. J. Wallard, J. Phys. E 5, 926 (1972); J. L. Hall, G. Kramer, and R. L. Barger, in Conference on Precision Electro Magnetic Measurement Digest 1972 (Institute of Electrical and Electronics Engineers, New York, 1972), p. 75.
[CrossRef]

Metrologia

T. J. Quinn, Metrologia 30, 524 (1994).
[CrossRef]

Other

Early digital frequency counters typically accumulate the number of positive zero crossings in a set time, giving correct counts but limited precision. A newer design (HP53132A) has vastly higher resolution by using a random sampling. At carrier frequencies above 50 MHz this strategy works even with the 6-MHz peak-to-peak FM discussed in this Letter. But, at much below this carrier frequency, such deep FM causes erratic operation, giving errors of many megahertz.

See Ref. 1. For example, the Comité Consultatif du Longeur (CCL, formerly CCDM) has recommended values for five He–Ne laser lines and one argon-ion laser line locked to several lines in I2 by dither-locking techniques.

The unit used was Model 925A from Berkeley Nucleonics. This information is provided for technical communication purposes only and does not constitute an endorsement by the National Institute of Standards and Technology or the University of Colorado.

Feedback alone could in theory cancel such FM and was in fact tried with the goal of inherently correcting for the nonlinearity of the AOM driver. However, the noise performance of feedback was found to be far inferior to feed forward, because to have a useful gain of G at modulation frequency f a required feedback servo bandwidth of greater than Gf brings added noise across this entire frequency region. Feed forward can provide exact cancellation of a known perturbation with essentially no bandwidth and hence practically zero added noise.

M. S. Taubman and J. L. Hall, “Precise removal from a laser beam of large frequency modulations generated either internally or externally to the laser,” , May19, 1994.

Cited By

OSA participates in CrossRef's Cited-By Linking service. Citing articles from OSA journals and other participating publishers are listed here.

Alert me when this article is cited.


Figures (4)

Fig. 1
Fig. 1

Dither-cancellation experiment. The Dithered I2:HeNe reference laser produces a nominal 70 µW of power at 633 nm. Its internal servo unit applies a dither signal at 8.333 kHz to a piezo element on one of the laser cavity mirrors, producing a dither modulation depth of 6 MHz peak to peak. The Fly Wheel He–Ne laser, producing 1 mW of power, is heterodyned on a fast photodiode with the reference via an isolating AOM and then offset locked to the reference laser by 193.6 MHz by use of a loop that does not transfer the laser dither (loop not shown). The dither signal passing through a variable attenuator and phase shifter is then fed forward to the AOM either through the FM input of the AOM driver in the analog case or via a DDS, a rf frequency converter, a filter, and an amplifier in the digital case. A detection rf frequency converter is also added to facilitate the use of a FFT analyzer. Each frequency converter consists of a double-balanced mixer (DBM) and a local oscillator (LO). The feedback loop shown is used for the slow correction of the amplitude and phase settings of the feed-forward pathway. OSC’s, oscillators; other abbreviations defined in text.

Fig. 2
Fig. 2

Results from the analog experiment, showing the beat note between the two lasers for cases of Dithered Linewidth (dither on but cancellation off); Dither Off, showing the narrow linewidth of the pure beat note; and finally Dither Canceled, showing the performance of the cancellation mechanism. Detection resolution beat width, 30 kHz.

Fig. 3
Fig. 3

Results from the digital version: Dithered Linewidth and Dither Canceled spectra. The narrow peaks in the latter show much better dither cancellation but evidence an undesirable multiplicity because of the insufficient sampling rate of 200 kHz of the DDS external FM input. Detection resolution beat width, 10 kHz.

Fig. 4
Fig. 4

Three central peaks in Fig. 3. A linear scale Gaussian fit reveals an optical power linewidth of 8 kHz, close to the minimum possible with the lasers used.

Metrics