Re: Preamplifier for fast point-scanning

[Peter Rupprecht]

*****
To join, leave or search the confocal microscopy listserv, go to:
http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
Post images on http://www.imgur.com and include the link in your posting.
*****

Dear List,

this is a follow-up on my earlier post some months ago:
http://confocal-microscopy-list.588098.n2.nabble.com/Preamplifier-for-fast-point-scanning-td7583344.html

Shortly after this post, I tested the ACA-4-40 preamplifier from Becker&Hickl and compared it directly to the Femto DHCPA-100. Amplification and noise seemed to be in the same order of magnitude for both products for my setup. Although performance might be also dependent on bandwidth of PMT and other details that I cannot test properly. I realized that comparing the two products objectively would involve more than just one experimental setup.
The one annoying thing that was also mentioned before by Michael, seems to be the AC-coupling of the preamp. At least this is how I interpret the strange artifacts that I see in the wake of very bright objects. This is very clearly visible when scanning bidirectionally, because then this artifact clearly shows the scanning direction.
Here is an excerpt of an image which shows how this may look like for calcium imaging (this is averaged over a couple of frames); please note the picture titles:
https://www.dropbox.com/sh/bwxxjrw44yxxb0p/AAAM9yWeMMc37NT_VORDEiaBa?dl=0

Based on my limited experience, I would conclude that the B&H ACA is maybe better for photon counting or flourescence lifetime imaging (due to higher bandwidth), but not the appropriate solution for my calcium imaging experiments.

Peter

[Michael Giacomelli]

*****
To join, leave or search the confocal microscopy listserv, go to:
http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
Post images on http://www.imgur.com and include the link in your posting.
*****

Hi Peter,

Those artifacts are indeed caused by the AC coupling.  A bright signal
causes a shift of the zero level that persists for some time
afterwards.  You can however remove them with digital filtering by
treating the data you record as an AC signal modulated at 80 MHz with
the intensity of your image encoded in the envelop.

For what its worth, Thorlabs has begun selling their TIA-60 product
(previously you could get it only with a complete system), which is
specifically designed for fast MPM imaging with the H7422:

http://www.thorlabs.de/newgrouppage9.cfm?objectgroup_id=8862

We have several of these, and while they do roll off a little before
80MHz, their performance is excellent and they are reasonably priced.
They are also DC coupled and, provided you use the included cabling,
are quite stable with an H7422.  I would strongly recommend this part
for an 80 MHz system.  Its not worth paying more unless you have a
very high rep rate system.

Mike

On Wed, May 13, 2015 at 11:29 AM, Peter Rupprecht
<[hidden email]> wrote:

> *****
> To join, leave or search the confocal microscopy listserv, go to:
> http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
> Post images on http://www.imgur.com and include the link in your posting.
> *****
>
> Dear List,
>
> this is a follow-up on my earlier post some months ago:
> http://confocal-microscopy-list.588098.n2.nabble.com/Preamplifier-for-fast-point-scanning-td7583344.html
>
> Shortly after this post, I tested the ACA-4-40 preamplifier from Becker&Hickl and compared it directly to the Femto DHCPA-100. Amplification and noise seemed to be in the same order of magnitude for both products for my setup. Although performance might be also dependent on bandwidth of PMT and other details that I cannot test properly. I realized that comparing the two products objectively would involve more than just one experimental setup.
> The one annoying thing that was also mentioned before by Michael, seems to be the AC-coupling of the preamp. At least this is how I interpret the strange artifacts that I see in the wake of very bright objects. This is very clearly visible when scanning bidirectionally, because then this artifact clearly shows the scanning direction.
> Here is an excerpt of an image which shows how this may look like for calcium imaging (this is averaged over a couple of frames); please note the picture titles:
> https://www.dropbox.com/sh/bwxxjrw44yxxb0p/AAAM9yWeMMc37NT_VORDEiaBa?dl=0
>
> Based on my limited experience, I would conclude that the B&H ACA is maybe better for photon counting or flourescence lifetime imaging (due to higher bandwidth), but not the appropriate solution for my calcium imaging experiments.
>
> Peter

[Craig Brideau]

*****
To join, leave or search the confocal microscopy listserv, go to:
http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
Post images on http://www.imgur.com and include the link in your posting.
*****

According to Thorlabs, cable length is very important with the TIA-60.
(like all HF amps, but I thought it worth mentioning). Keep it short for
less HF roll-off. I have these on a Thorlabs Bergamo and they have the amps
piggy-backed right on top of the PMTs with a very short cable. It seems to
work well as my primary source of noise is LED status lights on equipment
in the room. We spent a lot of time pointing these away from the microscope
and taping over some of them.

Craig

On Wed, May 13, 2015 at 4:44 PM, Michael Giacomelli <[hidden email]> wrote:

> *****
> To join, leave or search the confocal microscopy listserv, go to:
> http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
> Post images on http://www.imgur.com and include the link in your posting.
> *****
>
> Hi Peter,
>
> Those artifacts are indeed caused by the AC coupling.  A bright signal
> causes a shift of the zero level that persists for some time
> afterwards.  You can however remove them with digital filtering by
> treating the data you record as an AC signal modulated at 80 MHz with
> the intensity of your image encoded in the envelop.
>
> For what its worth, Thorlabs has begun selling their TIA-60 product
> (previously you could get it only with a complete system), which is
> specifically designed for fast MPM imaging with the H7422:
>
> http://www.thorlabs.de/newgrouppage9.cfm?objectgroup_id=8862
>
> We have several of these, and while they do roll off a little before
> 80MHz, their performance is excellent and they are reasonably priced.
> They are also DC coupled and, provided you use the included cabling,
> are quite stable with an H7422.  I would strongly recommend this part
> for an 80 MHz system.  Its not worth paying more unless you have a
> very high rep rate system.
>
> Mike
>
> On Wed, May 13, 2015 at 11:29 AM, Peter Rupprecht
> <[hidden email]> wrote:
> > *****
> > To join, leave or search the confocal microscopy listserv, go to:
> > http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
> > Post images on http://www.imgur.com and include the link in your
> posting.
> > *****
> >
> > Dear List,
> >
> > this is a follow-up on my earlier post some months ago:
> >
> http://confocal-microscopy-list.588098.n2.nabble.com/Preamplifier-for-fast-point-scanning-td7583344.html
> >
> > Shortly after this post, I tested the ACA-4-40 preamplifier from
> Becker&Hickl and compared it directly to the Femto DHCPA-100. Amplification
> and noise seemed to be in the same order of magnitude for both products for
> my setup. Although performance might be also dependent on bandwidth of PMT
> and other details that I cannot test properly. I realized that comparing
> the two products objectively would involve more than just one experimental
> setup.
> > The one annoying thing that was also mentioned before by Michael, seems
> to be the AC-coupling of the preamp. At least this is how I interpret the
> strange artifacts that I see in the wake of very bright objects. This is
> very clearly visible when scanning bidirectionally, because then this
> artifact clearly shows the scanning direction.
> > Here is an excerpt of an image which shows how this may look like for
> calcium imaging (this is averaged over a couple of frames); please note the
> picture titles:
> >
> https://www.dropbox.com/sh/bwxxjrw44yxxb0p/AAAM9yWeMMc37NT_VORDEiaBa?dl=0
> >
> > Based on my limited experience, I would conclude that the B&H ACA is
> maybe better for photon counting or flourescence lifetime imaging (due to
> higher bandwidth), but not the appropriate solution for my calcium imaging
> experiments.
> >
> > Peter
>

[Peter Rupprecht]

*****
To join, leave or search the confocal microscopy listserv, go to:
http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
Post images on http://www.imgur.com and include the link in your posting.
*****

Thanks for the replies to Michael, Craig, an off-list - if I'm again in the situation of choosing a preamp, I will certainly have a look again at the Thorlabs model. The input noise level actually seems to be slightly lower than for the DHCPA-100 (used at "high gain", 80 MHz bandwidth). The main advantage of the Femto preamp seems to be that gain and bandwidth can be varied. Also, I sometimes used the possibility to switch to AC-coupling, which is not available for the Thorlabs model.
Peter