# Side bands on diode laser (unexpected wavlength shift.)

Started by November 20, 2012
```So I was putting ~100MHz sidebands on a diode laser (modulating the
current.) to get a frequency 'measure' of the length of a (~20cm)
Fabry-Perot cavity (confocal).  Wavlength ~780nm, FSR ~380 MHz, a
source at 190 MHz would be perfect, but I don't have anything that
high.  So I cranked up the power with 95 MHz and got the second order
sidebands at the frequency I wanted.  (RF voltage of about 1 Vp-p into
50 ohms so modulating the current by maybe +/- 10mA, it's not clear
how much of the RF gets into the diode.)

What was weird (unexpected) was that when I cranked up the RF power
the laser frequency shifted... but I had to *increase* the laser
current to get it back to the wavlength where it started.  The
increase in the DC current was a few mA out of ~50mA.

Anyone know what's going on?  I expected that adding RF would heat up
the diode and thus lead to a lower DC current.

Thanks,
George H.
```
```On Nov 20, 8:39=A0am, George Herold <gher...@teachspin.com> wrote:
> So I was putting ~100MHz sidebands on a diode laser (modulating the
> current.) to get a frequency 'measure' of the length of a (~20cm)
> Fabry-Perot cavity (confocal). =A0Wavlength ~780nm, FSR ~380 MHz, a
> source at 190 MHz would be perfect, but I don't have anything that
> high. =A0So I cranked up the power with 95 MHz and got the second order
> sidebands at the frequency I wanted. =A0(RF voltage of about 1 Vp-p into
> 50 ohms so modulating the current by maybe +/- 10mA, it's not clear
> how much of the RF gets into the diode.)
>
> What was weird (unexpected) was that when I cranked up the RF power
> the laser frequency shifted... but I had to *increase* the laser
> current to get it back to the wavlength where it started. =A0The
> increase in the DC current was a few mA out of ~50mA.
>
> Anyone know what's going on? =A0I expected that adding RF would heat up
> the diode and thus lead to a lower DC current.

Is the diode self-tuned or does it depend on the external F-P
resonator for tuning?

Did the wavelength shift up or down, and did it shift smoothly or
are you looking at mode hopping? If it's internally tuned I'd expect
heating to lengthen the diode's resonator so that you can hop from fa
to fb then back to fa as the resonator length changes by increments of
a half-wavelength.

If it's externally stabilized the front face of the diode's internal
resonator may be forming another resonator with the F-P's input side
which changes length as the diode expands with heating.

Can you monitor the diode's temperature change with wavelength
change to any precision?

(caveat- I'm still on my first cup of coffee)

Mark L. Fergerson
```
```On Nov 20, 3:00=A0pm, "n...@bid.nes" <alien8...@gmail.com> wrote:
> On Nov 20, 8:39=A0am, George Herold <gher...@teachspin.com> wrote:
>
>
>
>
>
> > So I was putting ~100MHz sidebands on a diode laser (modulating the
> > current.) to get a frequency 'measure' of the length of a (~20cm)
> > Fabry-Perot cavity (confocal). =A0Wavlength ~780nm, FSR ~380 MHz, a
> > source at 190 MHz would be perfect, but I don't have anything that
> > high. =A0So I cranked up the power with 95 MHz and got the second order
> > sidebands at the frequency I wanted. =A0(RF voltage of about 1 Vp-p int=
o
> > 50 ohms so modulating the current by maybe +/- 10mA, it's not clear
> > how much of the RF gets into the diode.)
>
> > What was weird (unexpected) was that when I cranked up the RF power
> > the laser frequency shifted... but I had to *increase* the laser
> > current to get it back to the wavlength where it started. =A0The
> > increase in the DC current was a few mA out of ~50mA.
>
> > Anyone know what's going on? =A0I expected that adding RF would heat up
> > the diode and thus lead to a lower DC current.
>
> =A0 Is the diode self-tuned or does it depend on the external F-P
> resonator for tuning?

Well self tuned I guess.  It's got an external grating that does
feedback into the diode
>
> =A0 Did the wavelength shift up or down, and did it shift smoothly or
> are you looking at mode hopping? If it's internally tuned I'd expect
> heating to lengthen the diode's resonator so that you can hop from fa
> to fb then back to fa as the resonator length changes by increments of
> a half-wavelength.

Smooth... (Hmm up or down, I forgot.) So I had to increase the DC
current to get back to the original wavelength.. Increasing the (DC)
current is much the same as increasing the temperature.  Both tend to
make the diode go to shorter wavlengths.
>
> =A0 If it's externally stabilized the front face of the diode's internal
> resonator may be forming another resonator with the F-P's input side
> which changes length as the diode expands with heating.
>
> =A0 Can you monitor the diode's temperature change with wavelength
> change to any precision?

Well not to *any* precision :^)
But I've got 'typical' graphs of how the wavelength changes with
temperature.
>
> (caveat- I'm still on my first cup of coffee)

At 3:00 PM?  An all-nighter?

I've not done a lot of measurements with RF injected currents.  Next
time I look at this I'll see what happens for lower frequencies.  (Say
0.1 to 1MHz.)

George H.
>
> =A0 Mark L. Fergerson- Hide quoted text -
>
> - Show quoted text -

```
```On 11/20/2012 04:20 PM, George Herold wrote:
> On Nov 20, 3:00 pm, "n...@bid.nes"<alien8...@gmail.com>  wrote:
>> On Nov 20, 8:39 am, George Herold<gher...@teachspin.com>  wrote:
>>
>>
>>
>>
>>
>>> So I was putting ~100MHz sidebands on a diode laser (modulating the
>>> current.) to get a frequency 'measure' of the length of a (~20cm)
>>> Fabry-Perot cavity (confocal).  Wavlength ~780nm, FSR ~380 MHz, a
>>> source at 190 MHz would be perfect, but I don't have anything that
>>> high.  So I cranked up the power with 95 MHz and got the second order
>>> sidebands at the frequency I wanted.  (RF voltage of about 1 Vp-p into
>>> 50 ohms so modulating the current by maybe +/- 10mA, it's not clear
>>> how much of the RF gets into the diode.)
>>
>>> What was weird (unexpected) was that when I cranked up the RF power
>>> the laser frequency shifted... but I had to *increase* the laser
>>> current to get it back to the wavlength where it started.  The
>>> increase in the DC current was a few mA out of ~50mA.
>>
>>> Anyone know what's going on?  I expected that adding RF would heat up
>>> the diode and thus lead to a lower DC current.
>>
>>    Is the diode self-tuned or does it depend on the external F-P
>> resonator for tuning?
>
> Well self tuned I guess.  It's got an external grating that does
> feedback into the diode
>>
>>    Did the wavelength shift up or down, and did it shift smoothly or
>> are you looking at mode hopping? If it's internally tuned I'd expect
>> heating to lengthen the diode's resonator so that you can hop from fa
>> to fb then back to fa as the resonator length changes by increments of
>> a half-wavelength.
>
> Smooth... (Hmm up or down, I forgot.) So I had to increase the DC
> current to get back to the original wavelength.. Increasing the (DC)
> current is much the same as increasing the temperature.  Both tend to
> make the diode go to shorter wavlengths.
>>
>>    If it's externally stabilized the front face of the diode's internal
>> resonator may be forming another resonator with the F-P's input side
>> which changes length as the diode expands with heating.
>>
>>    Can you monitor the diode's temperature change with wavelength
>> change to any precision?
>
> Well not to *any* precision :^)
> But I've got 'typical' graphs of how the wavelength changes with
> temperature.
>>
>> (caveat- I'm still on my first cup of coffee)
>
> At 3:00 PM?  An all-nighter?
>
> I've not done a lot of measurements with RF injected currents.  Next
> time I look at this I'll see what happens for lower frequencies.  (Say
> 0.1 to 1MHz.)
>
> George H.
>>
>>    Mark L. Fergerson- Hide quoted text -
>>
>> - Show quoted text -
>

At a guess, the RF is confusing the DC bias supply.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs
Principal Consultant
ElectroOptical Innovations LLC
Optics, Electro-optics, Photonics, Analog Electronics

Briarcliff Manor NY 10510

hobbs at electrooptical dot net
http://electrooptical.net
```
```On Nov 23, 12:30=A0pm, Phil Hobbs
<pcdhSpamMeSensel...@electrooptical.net> wrote:
> On 11/20/2012 04:20 PM, George Herold wrote:
>
>
>
>
>
> > On Nov 20, 3:00 pm, "n...@bid.nes"<alien8...@gmail.com> =A0wrote:
> >> On Nov 20, 8:39 am, George Herold<gher...@teachspin.com> =A0wrote:
>
> >>> So I was putting ~100MHz sidebands on a diode laser (modulating the
> >>> current.) to get a frequency 'measure' of the length of a (~20cm)
> >>> Fabry-Perot cavity (confocal). =A0Wavlength ~780nm, FSR ~380 MHz, a
> >>> source at 190 MHz would be perfect, but I don't have anything that
> >>> high. =A0So I cranked up the power with 95 MHz and got the second ord=
er
> >>> sidebands at the frequency I wanted. =A0(RF voltage of about 1 Vp-p i=
nto
> >>> 50 ohms so modulating the current by maybe +/- 10mA, it's not clear
> >>> how much of the RF gets into the diode.)
>
> >>> What was weird (unexpected) was that when I cranked up the RF power
> >>> the laser frequency shifted... but I had to *increase* the laser
> >>> current to get it back to the wavlength where it started. =A0The
> >>> increase in the DC current was a few mA out of ~50mA.
>
> >>> Anyone know what's going on? =A0I expected that adding RF would heat =
up
> >>> the diode and thus lead to a lower DC current.
>
> >> =A0 =A0Is the diode self-tuned or does it depend on the external F-P
> >> resonator for tuning?
>
> > Well self tuned I guess. =A0It's got an external grating that does
> > feedback into the diode
>
> >> =A0 =A0Did the wavelength shift up or down, and did it shift smoothly =
or
> >> are you looking at mode hopping? If it's internally tuned I'd expect
> >> heating to lengthen the diode's resonator so that you can hop from fa
> >> to fb then back to fa as the resonator length changes by increments of
> >> a half-wavelength.
>
> > Smooth... (Hmm up or down, I forgot.) So I had to increase the DC
> > current to get back to the original wavelength.. Increasing the (DC)
> > current is much the same as increasing the temperature. =A0Both tend to
> > make the diode go to shorter wavlengths.
>
> >> =A0 =A0If it's externally stabilized the front face of the diode's int=
ernal
> >> resonator may be forming another resonator with the F-P's input side
> >> which changes length as the diode expands with heating.
>
> >> =A0 =A0Can you monitor the diode's temperature change with wavelength
> >> change to any precision?
>
> > Well not to *any* precision :^)
> > But I've got 'typical' graphs of how the wavelength changes with
> > temperature.
>
> >> (caveat- I'm still on my first cup of coffee)
>
> > At 3:00 PM? =A0An all-nighter?
>
> > I've not done a lot of measurements with RF injected currents. =A0Next
> > time I look at this I'll see what happens for lower frequencies. =A0(Sa=
y
> > 0.1 to 1MHz.)
>
> > George H.
>
> >> =A0 =A0Mark L. Fergerson- Hide quoted text -
>
> >> - Show quoted text -
>
> At a guess, the RF is confusing the DC bias supply.
>
> Cheers
>
> Phil Hobbs
>
> --
> Dr Philip C D Hobbs
> Principal Consultant
> ElectroOptical Innovations LLC
> Optics, Electro-optics, Photonics, Analog Electronics
>
> 160 North State Road #203
> Briarcliff Manor NY 10510
>
> hobbs at electrooptical dot nethttp://electrooptical.net- Hide quoted tex=
t -
>
> - Show quoted text -

Hmm, That's an interesting idea.  So the RF goes in via an sma right
on the laser head, and the 'DC' current drive is in a separate box,
connected by a 6 foot computer cable with DB9 connectors on the ends.
The current drive is mostly copied from Jan Hall, RSI '93.  There's a
big inductor before the real DC part.  But there's this opamp ~1MHz
current modulation circuit that's on the laser side of the inductor.
I've no idea how that behaves with 100 MHz leaking in from the back
end.

Oh! I've also got these ESD protection diodes across the laser diode
in the laser head.  One Schottky in reverse, and three pn's in series
in the forward direction.... (?)

Thanks,  I've got a few things to look at.

George H.

George H.
```