Measuring a Laser for SLM - Suggestions Please

I have this desire to be able to measure the longitudinal modes of lasers. Perhaps this comes from my long term holography hobby :). As desires go, I realise that this isn't going to set the world alight, but i'm interested in learning what different methods are available for carrying out this process.

Ideally, i'd like to point a laser at an instrument and be able to see (perhaps on a screen) whether or not it was operating in single longitudinal mode.

I've heard that a scanning fabry perot interferometer is what is really required, but these don't appear on Ebay too often, even if they did I wouldn't have a clue as to how to spot a good one.. Are there any other methods available that are perhaps more within the hobbyist budget?

Can anyone please explain exactly how the measurement process works so I can better understand?

Does anyone have anything for sale that would fulfil my desires? >:)

Cheers

Jem

For diodes the glass plate trick works well enough, for lasers with longer coherence length you need something with higher resolution, like a scanning interferometer or a long-arm Michelson interferometer.

Tell what kind of laser you have in mind!

Quote:

---

Originally Posted by RedlumX

Tell what kind of laser you have in mind!

---

Well, I guess that's part of the problem as I would like to try many different sorts of lasers. It would also be nice to prove my own lasers were truly SLM.

I have a 200mW Coherent 532 Ring laser, two Coherent C315m's one at 80mW and one at 15mW, two 35mW 632.8nm HeNe's, a 10mW 632.8nm HeNe, a 20mW 488nm Coherent Sapphire and a stupid amount of JDSU 532nm MicroGreen lasers of varying powers. That's just the ones that I believe are SLM and are capable of making holograms. We then start adding the 445nm C*sio diodes the 200mW DPSS 473nm and various red diode lasers, Oh, and the HeCd and the argon and the 2W RGB laser projector... :rolleyes: *retardo

Hehe... I can't believe I just typed that... do I *really* have so many lasers? I must get around to selling some of them, perhaps one day :o

Cheers

Jem

Quote:

---

Originally Posted by Jem

...lots of lasers...

---

Hi,
My name's Jem...
and I'm a Laser-hollic *retardo


I know it's tough Jem; but we can help. The next laser you buy; send over this way and I'll look after it for you. :)

Quote:

---

Originally Posted by Jem

Well, I guess that's part of the problem as I would like to try many different sorts of lasers. It would also be nice to prove my own lasers were truly SLM.

I have a 200mW Coherent 532 Ring laser, two Coherent C315m's one at 80mW and one at 15mW, two 35mW 632.8nm HeNe's, a 10mW 632.8nm HeNe, a 20mW 488nm Coherent Sapphire and a stupid amount of JDSU 532nm MicroGreen lasers of varying powers. That's just the ones that I believe are SLM and are capable of making holograms. We then start adding the 445nm C*sio diodes the 200mW DPSS 473nm and various red diode lasers, Oh, and the HeCd and the argon and the 2W RGB laser projector... :rolleyes: *retardo

Hehe... I can't believe I just typed that... do I *really* have so many lasers? I must get around to selling some of them, perhaps one day :o

Cheers

Jem

---

Well the Compass' don't need testing ;-) Other DPSS lasers won't run SLM unless specifically designed for it, almost not worth to even test. The high power Hene's might have short coherence length, here a scanning FPI would be needed for checking. The argon will work only if equipped with an etalon, and to adjust it you'd need a scanning or Michelson interferometer as well. The HeCd will have very short coherence length, a few mm perhaps, so not worth. The blue Nichia is better in this respect, except for beam quality. But for this and any other diode, you need to very precisely monitor the spectrum and its stability; the glass plate trick should well for these and red diodes since their mode spacing is so large.

Thanks for the reply.

Interestingly the Kimmon HeCd supposedly has a coherence length of around 10cm, as do the HeNe's. When I get around to it I intend setting up a Michelson Interferometer again to test the HeCd. I'd like to test the JDSU's as i'm not certain of their SLM operation, although the one that I have tried does make nice holograms so I guess that's SLM.

@Dan... Yep, it's taken a few years, but at least i can talk about it openly now, I know i'm amongst friends here >:)

Cheers

Jem

Sam's FAQ, Fabry=Perot page:

http://repairfaq.ece.drexel.edu/sam/...a.htm#liasgsf3


one dollar, two dollar, and five dollar versions.

One of these days, I might have to do the "sound card interface" Fabry Perot.

Steve

Thanks Steve. Can't say I understand it all, but i'll try and digest it a bit more later.

So, I guess my dream of just buying a secondhand piece of equipment that will do these measurements is a no-goer, what a shame...

Let me know when you get the "sound card interface" Fabry Perot off the ground ;)

IF you want to try Sam's 3$ red FP interferometer, I have the same mirrors from Ebay, about 20 of them, and would be glad to pop a pair into the mail for the cost of postage and shipping materials. &^%$ shipping materials are not cheap!

I'd need some one to cut the code for the sound card.

Steve

Thanks for the offer Steve, it's much appreciated. However, I think this project is beyond my skill level, besides it looks like it's only any use for HeNe's - correct?

I'm sure something will eventually become available 'off the shelf' :)

Quote:

---

Originally Posted by Jem

Thanks for the offer Steve, it's much appreciated. However, I think this project is beyond my skill level, besides it looks like it's only any use for HeNe's - correct?

I'm sure something will eventually become available 'off the shelf' :)

---

The problem is you need to change mirror sets for each wavelength, unless you want to buy a super expensive mirror set. I have 550$ in the required set in year 2000 dollars.

If you could find two Omnichrome 632 mixed gas high reflectors at 99.9% or better, you'd be all set for 430 to say 670 nm.

The reflectance of the mirror determines the FP finesse which is the resolution. Anything less then 98% reflective is not worth messing with. So your restricted to dielectric mirrors. The curvature is important too, You can do it with flats, but the required alignment is a pain in the A$$, so you can do Fabry-Perot and have cheap and great resolution, or do Michelson and have inexpensive, but pain in the butt to build over long distances.

Most people replace the mirrors on with a "V" folded mirror pair or a retroreflector for long distances to reduce alignment problems on the Michelsons.

The old saying cheap, accurate, or easy to build, pick any one or two, applies.

Steve

Hi Steve

Read it and (pretty much) understood it all :)

Idea is now firmly placed on the back shelf... For now ;)

Thanks for your explanations, as I said before, much appreciated.

Jem

Easy test, shoot gratings. The better the diffraction, the better the coherence, and you only need 2 mirrors or one of these things:

http://www.delphiglass.com/annual-be...vel-box-of-100


The 4 bevels form a nice cross grating if you flood it with a diverging lens. Easy way to measure the coherence distance.

Steve

I forgot to mention, I have listed some infos here:

http://redlum.xohp.pagesperso-orange...asers.html#SLM

Thanks, interesting stuff :)

RedlumX,

Quote:

---

Originally Posted by RedlumX

I forgot to mention, I have listed some infos here:

http://redlum.xohp.pagesperso-orange...asers.html#SLM

---

Nice summary.
Regarding the Fizeau interferometer, you state this:

Quote:

---

This method does not work for lasers with longer resonators (longer than the thickness of the plate, roughly)

---

Will this interferometer still work for ECDL setups with cavities of a few centimeters?

Quote:

---

Originally Posted by Roy G. Biv

Will this interferometer still work for ECDL setups with cavities of a few centimeters?

---

Hi Roy,

in fact, as a matter of observation, it does.

Recall that an ECDL is a complicated system from the resonator point of view. There are two relevant scales, the size of the diode and the size of the extended cavity. In a sense these form two coupled resonators and the combined effect is some complicated superposition of the modes; this is a bit similar to a conventional resonator with an extra etalon. However an ECDL also has an grating which even more complicates matters. So it's hard to tell beforehand what the resulting laser spectrum would look like.

But I noticed that mode jumps in an ECDL are quite large, and their spacing seems to be determined by the size of the chip and not of the external cavity; so the jumps can be resolved by a thin (ca 1mm) glass plate interferometer. However I don't know whether this is generally true or not, so far all lasers I investigated behaved in this way.

The movie which is displayed on this page is for an ECDL with a HL6385 diode, and you see very clearly that the jumps displayed on the spectum analyzer are visible in the interferometer as well.

RedlumX,

In your movie the fringes are only crisp when the OSA shows the laser to be in SLM. Are the fringes always blurred with the laser in MLM? Why I ask is because I think that with a Michelson interferometer you can get fringes even though the laser is in MLM, but when you move one of the arms you see that the coherence length is very short, like a few millimeters. But, I guess this is only the case when the spectrum is composed of a few well separated spikes. Your movie clearly shows that when the spectrum is broad with overlapping spikes, the fringes are blurred. Or is the spectrum "hill" on the OSA just a lack of resolution?

So, will the Fizeau interferometer show MLM behavior or only mode hops? I mean, if you switch on the ECDL and see crisp fringes, can you tell whether the output is SLM or MLM without changing the driver current or diode temp?

Here is a nice MIT movie of a HeNe with single or multi line spectrum in a Michelson interferometer and an SFPI (27MB).

Here are lots of interesting video demonstrations on lasers and optics from MIT.

Quote:

---

Originally Posted by Roy G. Biv

In your movie the fringes are only crisp when the OSA shows the laser to be in SLM. Are the fringes always blurred with the laser in MLM?

---

Essentially yes, but things are more complicated. For example, if the laser has two modes, one at 99% strength, say, the other at 1%, then there will be only a slight blurring. In a hologram you may get a slight banding then, or if there are many weak side modes, a slight loss of contrast; perhaps noticeable, perhaps not. So this is a gradual issue. In fact, many diodes will run weak side modes, and to what extend they are harmful, well, I don't know a good rule of thumb. I would say if the interferometric fringes blur noticeably, you'd see this in terms of a diminished contrast of a hologram as well.

Moreover, even sharp fringes do not exclude multimode behavior. See below (*).

Quote:

---

Originally Posted by Roy G. Biv

Why I ask is because I think that with a Michelson interferometer you can get fringes even though the laser is in MLM, but when you move one of the arms you see that the coherence length is very short, like a few millimeters.

---

Yes, if the two path lenghts of the interfering beams are equal, there will always be fringes. Therefore they should not be made equal. For the glass plate interferometer, where the two beams come from reflections of the front and back side of the plate, the path difference is 1mm or so. So it can detect only coherence lengths of less than one mm; which is relevant for diodes. It is useless for other lasers where the "worst case" coherence length is more than a 1mm.

Let me remind you that if there are two isolated modes of equal strength, then there is a periodic structure. That is, if you change one of the paths, then the fringes will blur and after a while become sharp again; and this repeats indefinitely (similar for several modes). So when you see a sharp fringe pattern, you actually cannot exclude that there are several modes, if the path lenghts of the interferometer are by accident such that that the fringes happen "to be sharp again". This is what I meant above (*). This would be a rare accident and I never noticed this to happen. The strict way to exclude this would be to check the fringes for different path lengths, but I guess the effort wouldn't be worthwhile.

These matters had been discussed already a few times in the holography forum (where this thread would actually belong). See also here, esp look carefully at the figure.
http://www.holowiki.com/index.php?ti...herence_Length

Quote:

---

Originally Posted by Roy G. Biv

Or is the spectrum "hill" on the OSA just a lack of resolution?

---

Yes, it was set low.

Quote:

---

Originally Posted by Roy G. Biv

So, will the Fizeau interferometer show MLM behavior or only mode hops? I mean, if you switch on the ECDL and see crisp fringes, can you tell whether the output is SLM or MLM without changing the driver current or diode temp?

---

If the fringes blur much, you can tell immediately that there is strong MLM behavior. If the fringes are just very slightly blurred, it may be hard to tell by naked eye. Then the hologram my suffer from slightly reduced contrast.

The only completely safe way to tell is to use a high resolution scanning interferometer, which can detect line widths of a few Mhs, or coherence lengths of tens of meters.

RedlumX,

Thanks for the explanation. So, if I understand you correctly, when the diode (even as an ECDL) is in MLM the coherence length is always shorter than the plate thickness (due to the short diode cavity), and therefore there will be a blurring of the fringes. The blurriness strength is dependent on the strength of the side modes.

Quote:

---

Originally Posted by Roy G. Biv

RedlumX,

Thanks for the explanation. So, if I understand you correctly, when the diode (even as an ECDL) is in MLM the coherence length is always shorter than the plate thickness (due to the short diode cavity), and therefore there will be a blurring of the fringes. The blurriness strength is dependent on the strength of the side modes.

---

Right! Though I am not sure whether this applies to ECDLs under all circumstances; I simply noticed this as a general rule, without a good explanation.