Anders Wallin writes:

I am measuring the beat-signal between a tunable HeNe laser and a
stabilized one, and seeing something like this:
http://www.youtube.com/watch?v=3arRbVKr9jQ

There is an overall slow drift (probably thermal expansion of the tunable
laser cavity) which I can cope with, but I am not happy with the +/-5 MHz
fast jumps of the beat-note.
I have tried this with two reference lasers, a HP5501A and a MarkTech 7900
and the results are similar - so I believe the tunable laser it at fault,
not the reference laser.

Probably.

The 5501A and 7900 both use an analog feedback loop so they should be quite
clean at least over the short term. However, the 5501A HeNe laser power
supply is generally crap with a lot of ripple. The MarkTech is a common brick.
Neither of these really should result in anything like you're sseing since
the resulting ripple is at 10s of kHz.

All the later HP/Agilent lasers (e.g., 5501B, 5517A/B/C/D etc.) would have this
sort of dance built in as they compare the H and V polarization on a discrete
time scale of a second or so. So, their optical frequency often drifts back
and forth by a small amount. But the 5501A is well behaved.

Any ideas on what could cause this kind of rapid frequency shifting?
I have tried two powersupplies for the tunable laser. With a small brick-
type PSU powered off +12V there is very large intensity modulation at 20-22
kHz. I am now using a better PSU which results in very small intensity
noise.

A few photos of the tunable laser:

https://www.dropbox.com/sh/q4xruaiwlbhru1o/HtTelFz1j_
I believe this is an early version of the 5-color HeNe still on sale, e.g.
this one:

http://www.edmundoptics.com/lasers/g...e-helium-neon-
laser/2550

This may be an early version. It's a similar tube and this may be before
they replaced the nice 4-bar resonator with the cheezy case REO uses now.
That would be quite old, or someone's custom creation.

The Littrow prism has been mounted on a piezo tube which allows electronic
tuning of the laser.

I have tried looking at the beat-note with the piezo amplifier on/off, and
with the piezo grounded - but I still see the same jumpy beat-signal.

You might try looking at the beat between the 5501A and Marktech to further
confirm they are not at fault, or beat one of them with another stabilized
laser like an SP-117/A. You can manual tune the frequency of the 5501A
with the mode balance (bottom) pot on the small PCB on the left side of
the laser. With some luck, you'll be able to match up their optical
frequencies close enough for bandwidth of your photodiode.

With a stable power supply, the tunable should not do anything erratic
like that.

However, don't overlook the mundane - it doesn't take much in the way of
mechanical vibration to disturb the cavity enough to cause a noticeable
frequency shift. A 1 nm change in cavity length is about 1 part in 316
of the FSR of the laser cavity - 300-400 MHz. The fan of the spectrum
analyzer or air currents could even be at fault!

--
sam | Sci.Electronics.Repair FAQ: http://www.repairfaq.org/
Repair | Main Table of Contents: http://www.repairfaq.org/REPAIR/
+Lasers | Sam's Laser FAQ: http://www.repairfaq.org/sam/lasersam.htm
| Mirror Sites: http://www.repairfaq.org/REPAIR/F_mirror.html

Important: Anything sent to the email address in the message header above is
ignored unless my full name AND either lasers or electronics is included in the
subject line. Or, you can contact me via the Feedback Form in the FAQs.