Oscillating signal on spinning disk system

[Niko Ehrenfeuchter]

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

Dear all,

we're recently facing a very strange behavior on one of our spinning
disk systems. One of our users noticed a repeated fluctuation in
recorded timelapse experiments (this was a FRAP experiment, but we don't
think that's related to FRAP).

The timelapse was recorded every 6 seconds, and after background
subtraction a repeating drop of signal is visible in the intensity plot
(please find the corresponding graph under [1], see below). The
fluctuation is also visible by eye when scrolling through the timelapse.

We noticed a frequency of 66 seconds and therefore think we can exclude
any external vibrations as source of this problem (plus the system is
mounted on a Melles Griot optical table). On some experiments the
oscillation almost follows a sine-like function, on others (like the
mentioned one) it looks more rectangular with a very rapid drop and
raise. Also, the frequency does *not* change if we double the
acquisition frequency, it's still on 66 seconds when acquiring every 3
seconds.

One very weird thing is the stability of the background. The plots from
above show the intensity in selected ROI's over cells. The same plot for
an "empty" region in the sample shows a more or less constant signal (at
least we could figure a pattern in it, even after normalization and
using a big scaling factor to visualize the deltas).

Has anyone of you come across a similar behavior?

Any ideas about this would be highly appreciated!

Many Thanks,
Niko

[1]
https://www.biozentrum.unibas.ch/fileadmin/redaktion/Forschung/Research_Groups/IMCF/3i_intensity_fluctuation.png

--
Niko Ehrenfeuchter
Imaging Core Facility
Kragenbau, Room G1055
Biozentrum, University of Basel
Klingelbergstrasse 50/70
CH-4056 Basel (Switzerland)

Office:  [phone]
Email:   [hidden email]

[John Oreopoulos]

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

Hi Niko,

This might seem like an obvious question, but have you checked that your laser power isn't oscillating over the same period you measured? A power meter measurement at the back of the objective should be able to rule this out first. Do other laser wavelengths on your system produce the same effect?

Have you ruled out that this is not the "scan line" artifact sometimes observed with spinning disk systems? See this paper for an explanation of that:

Chong, F.K., et al., Optimization of spinning disk confocal microscopy: Synchronization with the ultra-sensitive emccd, Cogswell, and T. Wilson, Editors. 2004, Spie-Int Soc Optical Engineering: Bellingham. p. 65-76.

It's possible that if you exposure time is not an integer number of disk sector turns, then you might get some kind of intensity beat frequency with you acquisition period of every 6 seconds, but the intensity change you're seeing here seems a bit too big for that. What is your exposure time of the images captured every 6 seconds?

Beyond that, there might be something off with the spinning of the pinhole disk (or pinhole and lenslet disk if this is a CSU) which is a little more difficult to determine. Can you hear the disk(s) spinning? Does it sound like it's turning at a constant speed? Do you know what speed it is spinning at?

Maybe there's a few ideas to troubleshoot there. Let us all know if you solve the problem.


John Oreopoulos
Staff Scientist
Spectral Applied Research
Richmond Hill, Ontario
Canada
www.spectral.ca



On 2013-08-16, at 11:37 AM, Niko Ehrenfeuchter wrote:

> *****
> To join, leave or search the confocal microscopy listserv, go to:
> http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
> *****
>
> Dear all,
>
> we're recently facing a very strange behavior on one of our spinning disk systems. One of our users noticed a repeated fluctuation in recorded timelapse experiments (this was a FRAP experiment, but we don't think that's related to FRAP).
>
> The timelapse was recorded every 6 seconds, and after background subtraction a repeating drop of signal is visible in the intensity plot (please find the corresponding graph under [1], see below). The fluctuation is also visible by eye when scrolling through the timelapse.
>
> We noticed a frequency of 66 seconds and therefore think we can exclude any external vibrations as source of this problem (plus the system is mounted on a Melles Griot optical table). On some experiments the oscillation almost follows a sine-like function, on others (like the mentioned one) it looks more rectangular with a very rapid drop and raise. Also, the frequency does *not* change if we double the acquisition frequency, it's still on 66 seconds when acquiring every 3 seconds.
>
> One very weird thing is the stability of the background. The plots from above show the intensity in selected ROI's over cells. The same plot for an "empty" region in the sample shows a more or less constant signal (at least we could figure a pattern in it, even after normalization and using a big scaling factor to visualize the deltas).
>
> Has anyone of you come across a similar behavior?
>
> Any ideas about this would be highly appreciated!
>
> Many Thanks,
> Niko
>
> [1] https://www.biozentrum.unibas.ch/fileadmin/redaktion/Forschung/Research_Groups/IMCF/3i_intensity_fluctuation.png
>
> --
> Niko Ehrenfeuchter
> Imaging Core Facility
> Kragenbau, Room G1055
> Biozentrum, University of Basel
> Klingelbergstrasse 50/70
> CH-4056 Basel (Switzerland)
>
> Office:  [phone]
> Email:   [hidden email]

[Nic Cade]

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

Hi Niko,

Which microscope system do you have this problem with? Do you also see a
corresponding change in focus?

We have a very similar problem with our 3i spinning disk system where we see a
periodic oscillation in lateral position, ~250nm every 60s. We think we have
narrowed it down to the heater for the incubation chamber (Okolab) which also
switches on and off with a period of 60s. Another group here with the same
system have found exactly the same thing; turning off the heater stopped the
oscillations.
I'm guessing that in your case the oscillations might be large enough to move
your sample through the focal plane, but which wouldn't affect any featureless
background. You could test this with some fluorescent beads if you haven't
done so already.

Cheers
Nic

On Fri, 16 Aug 2013 17:37:32 +0200, Niko Ehrenfeuchter
<[hidden email]> wrote:

>*****
>To join, leave or search the confocal microscopy listserv, go to:
>http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
>*****
>
>Dear all,
>
>we're recently facing a very strange behavior on one of our spinning
>disk systems. One of our users noticed a repeated fluctuation in
>recorded timelapse experiments (this was a FRAP experiment, but we don't
>think that's related to FRAP).
>
>The timelapse was recorded every 6 seconds, and after background
>subtraction a repeating drop of signal is visible in the intensity plot
>(please find the corresponding graph under [1], see below). The
>fluctuation is also visible by eye when scrolling through the timelapse.
>
>We noticed a frequency of 66 seconds and therefore think we can exclude
>any external vibrations as source of this problem (plus the system is
>mounted on a Melles Griot optical table). On some experiments the
>oscillation almost follows a sine-like function, on others (like the
>mentioned one) it looks more rectangular with a very rapid drop and
>raise. Also, the frequency does *not* change if we double the
>acquisition frequency, it's still on 66 seconds when acquiring every 3
>seconds.
>
>One very weird thing is the stability of the background. The plots from
>above show the intensity in selected ROI's over cells. The same plot for
>an "empty" region in the sample shows a more or less constant signal (at
>least we could figure a pattern in it, even after normalization and
>using a big scaling factor to visualize the deltas).
>
>Has anyone of you come across a similar behavior?
>
>Any ideas about this would be highly appreciated!
>
>Many Thanks,
>Niko
>
>[1]
>https://www.biozentrum.unibas.ch/fileadmin/redaktion/Forschung/Research_Gr

oups/IMCF/3i_intensity_fluctuation.png

>
>--
>Niko Ehrenfeuchter
>Imaging Core Facility
>Kragenbau, Room G1055
>Biozentrum, University of Basel
>Klingelbergstrasse 50/70
>CH-4056 Basel (Switzerland)
>
>Office:  [phone]
>Email:   [hidden email]

[Jorge]

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

Hi Niko
we had a similar problem but additionally observe a pattern of a certain
frequency bands which are more evident when we upgrade to increase the
sampling frequency.
after several trials we identified that the dichroic mirror used to reflect
the laser ablation was defective, it is easy to identify because if you can
see the lines and slightly move the mirror you can see how are you moving
lines.
hope will be of your help
best regards
Jorge


2013/8/22 Nic Cade <[hidden email]>

> *****
> To join, leave or search the confocal microscopy listserv, go to:
> http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
> *****
>
> Hi Niko,
>
> Which microscope system do you have this problem with? Do you also see a
> corresponding change in focus?
>
> We have a very similar problem with our 3i spinning disk system where we
> see a
> periodic oscillation in lateral position, ~250nm every 60s. We think we
> have
> narrowed it down to the heater for the incubation chamber (Okolab) which
> also
> switches on and off with a period of 60s. Another group here with the same
> system have found exactly the same thing; turning off the heater stopped
> the
> oscillations.
> I'm guessing that in your case the oscillations might be large enough to
> move
> your sample through the focal plane, but which wouldn't affect any
> featureless
> background. You could test this with some fluorescent beads if you haven't
> done so already.
>
> Cheers
> Nic
>
> On Fri, 16 Aug 2013 17:37:32 +0200, Niko Ehrenfeuchter
> <[hidden email]> wrote:
>
> >*****
> >To join, leave or search the confocal microscopy listserv, go to:
> >http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
> >*****
> >
> >Dear all,
> >
> >we're recently facing a very strange behavior on one of our spinning
> >disk systems. One of our users noticed a repeated fluctuation in
> >recorded timelapse experiments (this was a FRAP experiment, but we don't
> >think that's related to FRAP).
> >
> >The timelapse was recorded every 6 seconds, and after background
> >subtraction a repeating drop of signal is visible in the intensity plot
> >(please find the corresponding graph under [1], see below). The
> >fluctuation is also visible by eye when scrolling through the timelapse.
> >
> >We noticed a frequency of 66 seconds and therefore think we can exclude
> >any external vibrations as source of this problem (plus the system is
> >mounted on a Melles Griot optical table). On some experiments the
> >oscillation almost follows a sine-like function, on others (like the
> >mentioned one) it looks more rectangular with a very rapid drop and
> >raise. Also, the frequency does *not* change if we double the
> >acquisition frequency, it's still on 66 seconds when acquiring every 3
> >seconds.
> >
> >One very weird thing is the stability of the background. The plots from
> >above show the intensity in selected ROI's over cells. The same plot for
> >an "empty" region in the sample shows a more or less constant signal (at
> >least we could figure a pattern in it, even after normalization and
> >using a big scaling factor to visualize the deltas).
> >
> >Has anyone of you come across a similar behavior?
> >
> >Any ideas about this would be highly appreciated!
> >
> >Many Thanks,
> >Niko
> >
> >[1]
> >
> https://www.biozentrum.unibas.ch/fileadmin/redaktion/Forschung/Research_Gr
> oups/IMCF/3i_intensity_fluctuation.png
> >
> >--
> >Niko Ehrenfeuchter
> >Imaging Core Facility
> >Kragenbau, Room G1055
> >Biozentrum, University of Basel
> >Klingelbergstrasse 50/70
> >CH-4056 Basel (Switzerland)
> >
> >Office:  [phone]
> >Email:   [hidden email]
>

--
Jorge Toledo H.
Ph.D. student
Biomedical Neuroscience Institute (BNI)
Laboratory of Scientific Image Analysis (SCIAN-Lab)
Faculty of Medicine
University of Chile
www.scian.cl | www.couvelab.org

[Oliver Biehlmaier-2]

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

Dear Nic,

Our system indeed is a 3i-spinning disk as well and we actually narrowed it down to the same origin as you did: the heater of the incubation chamber.

We contacted Okolab yesterday and they are very helpful in helping us to find a solution for our problem.
We will try to change some things in the setup and restore the default parameters of the thermal controller. We hope this will help.

Cheers,
Oliver


----------------------------------------------------------------

Oliver Biehlmaier, PhD

Head of Imaging Core Facility

Biozentrum (Kragenbau, Room G1054)
University of Basel
Klingelbergstrasse 50/70
4056 Basel
Switzerland

Office: [phone]
Lab:        [phone]
Email:     [hidden email]
http://www.biozentrum.unibas.ch/imcf
----------------------------------------------------------------




Am 22.08.2013 um 19:01 schrieb Nic Cade:

> *****
> To join, leave or search the confocal microscopy listserv, go to:
> http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
> *****
>
> Hi Niko,
>
> Which microscope system do you have this problem with? Do you also see a
> corresponding change in focus?
>
> We have a very similar problem with our 3i spinning disk system where we see a
> periodic oscillation in lateral position, ~250nm every 60s. We think we have
> narrowed it down to the heater for the incubation chamber (Okolab) which also
> switches on and off with a period of 60s. Another group here with the same
> system have found exactly the same thing; turning off the heater stopped the
> oscillations.
> I'm guessing that in your case the oscillations might be large enough to move
> your sample through the focal plane, but which wouldn't affect any featureless
> background. You could test this with some fluorescent beads if you haven't
> done so already.
>
> Cheers
> Nic
>
> On Fri, 16 Aug 2013 17:37:32 +0200, Niko Ehrenfeuchter
> <[hidden email]> wrote:
>
>> *****
>> To join, leave or search the confocal microscopy listserv, go to:
>> http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
>> *****
>>
>> Dear all,
>>
>> we're recently facing a very strange behavior on one of our spinning
>> disk systems. One of our users noticed a repeated fluctuation in
>> recorded timelapse experiments (this was a FRAP experiment, but we don't
>> think that's related to FRAP).
>>
>> The timelapse was recorded every 6 seconds, and after background
>> subtraction a repeating drop of signal is visible in the intensity plot
>> (please find the corresponding graph under [1], see below). The
>> fluctuation is also visible by eye when scrolling through the timelapse.
>>
>> We noticed a frequency of 66 seconds and therefore think we can exclude
>> any external vibrations as source of this problem (plus the system is
>> mounted on a Melles Griot optical table). On some experiments the
>> oscillation almost follows a sine-like function, on others (like the
>> mentioned one) it looks more rectangular with a very rapid drop and
>> raise. Also, the frequency does *not* change if we double the
>> acquisition frequency, it's still on 66 seconds when acquiring every 3
>> seconds.
>>
>> One very weird thing is the stability of the background. The plots from
>> above show the intensity in selected ROI's over cells. The same plot for
>> an "empty" region in the sample shows a more or less constant signal (at
>> least we could figure a pattern in it, even after normalization and
>> using a big scaling factor to visualize the deltas).
>>
>> Has anyone of you come across a similar behavior?
>>
>> Any ideas about this would be highly appreciated!
>>
>> Many Thanks,
>> Niko
>>
>> [1]
>> https://www.biozentrum.unibas.ch/fileadmin/redaktion/Forschung/Research_Gr
> oups/IMCF/3i_intensity_fluctuation.png
>>
>> --
>> Niko Ehrenfeuchter
>> Imaging Core Facility
>> Kragenbau, Room G1055
>> Biozentrum, University of Basel
>> Klingelbergstrasse 50/70
>> CH-4056 Basel (Switzerland)
>>
>> Office:  [phone]
>> Email:   [hidden email]

[alexia ferrand]

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

Dear all,
We now figured out that the problem is coming from the temperature
controller (Oko-Lab). This periodicity of 66 sec seems to be the exact time
that the temperature/CO2 controller box takes to have the fan ON, then OFF,
then ON again. So every minute, the temperature is checked and if a
variation is detected, the chamber receives warm air (at about 48°C). And
somehow, this seems to influence the intensity/laser during the imaging. If
the controller box is turned off, the oscillations immediately stop. We
figured out that the user actually forgot a few times to fill in the water
dish containing the probe, and then the temperature controller was trying to
compensate. Once the probe was reimmerged and the system fully
re-equilibrated, I could not observe the oscillations anymore.
I contacted Oko-Lab while we were trying to figure out the problem, and they
have been very responsive. They pointed out a couple more things but the
probe not being in water still seems to be the main culprit.
Best Regards,
Alexia