Fast two-photon imaging: Constant fraction discriminator and variable delay for Laser pulse-synch

[Tobias Rose]

*****
To join, leave or search the confocal microscopy listserv, go to:
http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
*****

Dear all,

I am trying to sync the sample-clock of our 125Ms FPGA Daq-board to the pulse rate of our 80Mhz Ti:Sa MaiTai for fast (resonant scanning) 2p imaging. To be independent of the small, jittery, and wavelength-dependent sync output of the MaitTai I thought to use a constant fraction discriminator (CFD) and combine the latter with a variable delay. The idea would be to maximize fluoerescence readout during the peak emission time of our fluorophores by tuning the delay to a detection sweet spot during acquisition.

Does this even make sense for analog integration (not photon counting) - or am I trying to solve a problem that I don't even have?

What equipment would you suggest? For the CFD this looks promising on paper:
http://www.phillipsscientific.com/preview/6915pre.htm

I could not find any of-the-shelf variable delay that would span ~20ns... I know that the STED people usually build their own but this is beyond my expertise. Can anyone suggest a vendor and device?

Thanks!

Best,T

[Mark Cannell]

*****
To join, leave or search the confocal microscopy listserv, go to:
http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
*****

Sounds like you want to implement lock-in detection. Is background a problem -or why else are you trying to do this? Nanosecond delay devices exist, google nanosecond digital delay.

Mark


On 17/11/2011, at 11:04 AM, Tobias Rose wrote:

> *****
> To join, leave or search the confocal microscopy listserv, go to:
> http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
> *****
>
> Dear all,
>
> I am trying to sync the sample-clock of our 125Ms FPGA Daq-board to the pulse rate of our 80Mhz Ti:Sa MaiTai for fast (resonant scanning) 2p imaging. To be independent of the small, jittery, and wavelength-dependent sync output of the MaitTai I thought to use a constant fraction discriminator (CFD) and combine the latter with a variable delay. The idea would be to maximize fluoerescence readout during the peak emission time of our fluorophores by tuning the delay to a detection sweet spot during acquisition.
>
> Does this even make sense for analog integration (not photon counting) - or am I trying to solve a problem that I don't even have?
>
> What equipment would you suggest? For the CFD this looks promising on paper:
> http://www.phillipsscientific.com/preview/6915pre.htm
>
> I could not find any of-the-shelf variable delay that would span ~20ns... I know that the STED people usually build their own but this is beyond my expertise. Can anyone suggest a vendor and device?
>
> Thanks!
>
> Best,T

[Tobias Rose]

*****
To join, leave or search the confocal microscopy listserv, go to:
http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
*****

Yes - that would be lock-in detection. And also yes: background is a big problem for us since we image during visual stimulation of animals. But shouldn't a good Ti:Sa pulse lock-in of our DAQ board also improve signal itself? If we sample at the beginning of the lifetime decay of our fluorophores this should be better than random sampling at 125MHz, or not? I have no experience in doing this so I might be on the wrong track here... especially since I don't know if our slow amplified PMT signal would smear out these timing issues anyways...

Tobias

________________________________________
Von: Confocal Microscopy List [[hidden email]]" im Auftrag von "Mark Cannell [[hidden email]]
Gesendet: Donnerstag, 17. November 2011 12:42
Bis: [hidden email]
Betreff: Re: Fast two-photon imaging: Constant fraction discriminator and variable delay for Laser pulse-synch

*****
To join, leave or search the confocal microscopy listserv, go to:
http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
*****

Sounds like you want to implement lock-in detection. Is background a problem -or why else are you trying to do this? Nanosecond delay devices exist, google nanosecond digital delay.

Mark


On 17/11/2011, at 11:04 AM, Tobias Rose wrote:

> *****
> To join, leave or search the confocal microscopy listserv, go to:
> http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
> *****
>
> Dear all,
>
> I am trying to sync the sample-clock of our 125Ms FPGA Daq-board to the pulse rate of our 80Mhz Ti:Sa MaiTai for fast (resonant scanning) 2p imaging. To be independent of the small, jittery, and wavelength-dependent sync output of the MaitTai I thought to use a constant fraction discriminator (CFD) and combine the latter with a variable delay. The idea would be to maximize fluoerescence readout during the peak emission time of our fluorophores by tuning the delay to a detection sweet spot during acquisition.
>
> Does this even make sense for analog integration (not photon counting) - or am I trying to solve a problem that I don't even have?
>
> What equipment would you suggest? For the CFD this looks promising on paper:
> http://www.phillipsscientific.com/preview/6915pre.htm
>
> I could not find any of-the-shelf variable delay that would span ~20ns... I know that the STED people usually build their own but this is beyond my expertise. Can anyone suggest a vendor and device?
>
> Thanks!
>
> Best,T

[Johannes Helm]

*****
To join, leave or search the confocal microscopy listserv, go to:
http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
*****

Thank you, Mark, for your good answer.

However, as Lock In Detection also might be a solution, I would like to
mention that there has been done a lot of work in this field at the Royal
Institute of Technology years ago with the IMS technique, of which there
exist two variations. The later version aims at using the lifetimes of
fluorophores together with a phase tuned Lock-In-Amplifier for detection.
Relevant web pages on this issue:

http://www.biox.kth.se/staff/faces/anders.liljeborg/quadra.htm

And some publications:

K. Carlsson, A. Liljeborg, "Confocal fluorescence microscopy using
spectral and lifetime information to simultaneuosly record four
fluorophores with high channel separation", J. Microsc. 185, 37-46 (1997)

K. Carlsson & A. Liljeborg, "Simultaneous confocal lifetime imaging of
multiple fluorophores using the Intensity-modulated Multiple-wavelength
Scanning (IMS) technique," J. Microsc. 191, 119-127 (1998)

R. M. Andersson, K. Carlsson, A. Liljeborg and H. Brismar,
"Characterization of Probe Binding and Comparison of Its Influence on
Fluorescence Lifetime of Two pH-Sensitive Benzo[c]xanthene Dyes Using
Intensity-Modulated Multiple-Wavelength Scanning Technique," Analytical
Biochemistry 283, 104-110 (2000)

K. Carlsson, A. Liljeborg, R. M. Andersson, and H. Brismar, "Confocal pH
imaging using fluorophore lifetimes and the intensity-modulated
multiple-wavelength scanning (IMS) technique," J. of Microscopy 199, Pt 2,
106-114 (2000).


The first version of the method had been tested for measuring
[Ca2+](x,y,z,t) with Fura-2:

P.J. Helm, O. Franksson, and K. Carlsson (1995),
A confocal scanning laser microscope for quantitative ratiometric 3D
measurements of [Ca2+] and Ca2+ diffusions in living cells stained
with Fura-2,
Pflügers Archiv - European Journal of Physiology, Vol. 429, pp. 672-681

and

PJ Helm, A Patwardhan, and EMM Manders (1997),
A study of the precision of confocal, ratiometric, Fura-2 based measurements,
Cell Calcium 22(4):287-298



Best wishes,
Johannes

> *****
> To join, leave or search the confocal microscopy listserv, go to:
> http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
> *****
>
> Sounds like you want to implement lock-in detection. Is background a
> problem -or why else are you trying to do this? Nanosecond delay devices
> exist, google nanosecond digital delay.
>
> Mark
>
>
> On 17/11/2011, at 11:04 AM, Tobias Rose wrote:
>
>> *****
>> To join, leave or search the confocal microscopy listserv, go to:
>> http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
>> *****
>>
>> Dear all,
>>
>> I am trying to sync the sample-clock of our 125Ms FPGA Daq-board to the
>> pulse rate of our 80Mhz Ti:Sa MaiTai for fast (resonant scanning) 2p
>> imaging. To be independent of the small, jittery, and
>> wavelength-dependent sync output of the MaitTai I thought to use a
>> constant fraction discriminator (CFD) and combine the latter with a
>> variable delay. The idea would be to maximize fluoerescence readout
>> during the peak emission time of our fluorophores by tuning the delay to
>> a detection sweet spot during acquisition.
>>
>> Does this even make sense for analog integration (not photon counting) -
>> or am I trying to solve a problem that I don't even have?
>>
>> What equipment would you suggest? For the CFD this looks promising on
>> paper:
>> http://www.phillipsscientific.com/preview/6915pre.htm
>>
>> I could not find any of-the-shelf variable delay that would span
>> ~20ns... I know that the STED people usually build their own but this is
>> beyond my expertise. Can anyone suggest a vendor and device?
>>
>> Thanks!
>>
>> Best,T
>

--
P. Johannes Helm

Voice: ([phone] (office)
Fax: [phone] (office)

[ChrisWilms]

*****
To join, leave or search the confocal microscopy listserv, go to:
http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
*****

Regarding the delay: if you do not need it to flexible in timing,  
simply getting a coaxial cable of the correct length (20 cm  
corresponds roughly to 1 ns) is one option. A bit more flexible and  
still affordable are variable delay lines such as the Ortec 425 <http://www.ortec-online.com/download/425A.pdf

Best, Christian

[Tobias Rose]

*****
To join, leave or search the confocal microscopy listserv, go to:
http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
*****

The Ortec 425 is an excellent tip, Christian! That's what I need - they even have a nice CFD. I was googling around for digital nanosecond delays but the price for the ones with precise enough timing was a little steep...
A simple BNC surely would do the trick but I'd like to find the best delay empirically (I still don't know if better pulse lock-in of the sampling will make a difference at all). Given the many meters of cabling, convoluted optical path, and different fluorophores I'm using at my setup I wouldn't even know were to start calculating the correct delay and cable length.  The Ortec spans the right range of delays... so let's hope it's cheap enough.
Thanks everybody for the replys!

Best, Tobias
________________________________________
Von: Confocal Microscopy List [[hidden email]]&quot; im Auftrag von &quot;Christian Wilms [[hidden email]]
Gesendet: Samstag, 19. November 2011 01:50
Bis: [hidden email]
Betreff: Re: Fast two-photon imaging: Constant fraction discriminator and variable delay for Laser pulse-synch

*****
To join, leave or search the confocal microscopy listserv, go to:
http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
*****

Regarding the delay: if you do not need it to flexible in timing,
simply getting a coaxial cable of the correct length (20 cm
corresponds roughly to 1 ns) is one option. A bit more flexible and
still affordable are variable delay lines such as the Ortec 425 <http://www.ortec-online.com/download/425A.pdf

Best, Christian