Search the CONFOCAL archive at
http://listserv.acsu.buffalo.edu/cgi-bin/wa?S1=confocal
Hello!
A comment concerning the laser stability issue:
We recently had increasing intensity fluctuations in all 4 lines of the
Ar/ArKr laser on our Leica SL confocal. In the range of 10 to 1 Hz. When it
became so bad that imaging was not possible any more we also heard an
increasing noise from the supply box. It turned out to be a worn out
ventilation fan of the AOTF control box. Replacing the 3-Euro-piece fixed
the problem (the cheapes repair we ever had ;-).
This shows that vibrations in the supply box can have significant influence
on laser intensity. (The fan was running! It was just worn out, causing the
usual vibration everybody knows from old PC CPU-cooling fans.)
In our particular case we had very dramatic fluctuations, intensity going
plus-minus 50% or worse. However, since laser supply boxes of confocals are
usually separated from the microscope stand and are not mounted on optical
tables I am afraid that even weak vibrations from anywhere in the building
can easily have effects on laser intensity fluctuations.
Unfortunately I couldn't make any systematic observations and measurements
on this yet. Maybe someone out there has some more experience with this kind
of "confocal-stability"-issue?
Olaf
Am 31.08.2004 13:53 Uhr schrieb "Arthur Schuessler" unter
<[log in to unmask]>:
> Search the CONFOCAL archive at
> http://listserv.acsu.buffalo.edu/cgi-bin/wa?S1=confocal
>
> Dear Bob,
> as I indicated, we have a kind of an "intrinsic" control, means that we get
> spectra from our samples which we can relatively easily interprete for
> correctness (several peaks, etc.). This is our "standard" - and we didn't
> care about possible 5 nm shifts, as also indicated (since these anyway can
> occur in the samples due to environmental conditions). I also have to say,
> that the measurements were done 1-2 years ago, so the recent discussion was
> not yet "available".
> We also used exclusively PMT1, so no idea about the other ones.
> But this means that we always use the same right or wrong spectra. We can
> say that they have to be "right", since spectra were very (!) similar to
> those from the literature for cyanobacterial pigment fluorescence (PE, PC,
> APC, Chlorophyll), what allowed us to perform quite good spectral unmixing.
> We never had such strange deviations, as you document them.
> About laser-flucuations, I am not sure what causes them (if only
> polarization). But problems were not too big. However, this first was a
> serious concern when measuring spectra in a sequential scan. Therefore I
> think this could cause severe problems if the system is misaligned. This
> could also be a temperature effect. However, our laser fluctuations were
> not too big (488 line).
> Arthur
> PS: I absolutely agree with the need of a standart, and a MIDL lamp
> is for sure a good solution!
>
> At 01:07 31.08.2004, you wrote:
>> Dear Arthur
>> A machine has to be evaluated to have some assurances that your data
>> output is correct. One must run some type of QA standards to access the
>> machine functionality. It is very difficult to interpret an image to
>> access what is wrong with the machine. I use reproducible standards to
>> make a machine assessment. Fro spectral imaging we have found the MIDL
>> lamp to be ideal.
>>
>> To answer your specific questions
>>
>> Laser output power fluctuations appears to be connected to improper
>> fiber polarization. It is important to know how much instability in
>> power exists in your system so it be filtered into your data
>> interpretation.
>>
>> Spectra obtained on a specific PMT may not be accurate if the system is
>> out of alignment or unstable. My three PMTs yield different spectra
>> -which one is correct? You need a standard to insure the spectra you are
>> obtaining on a specific PMT is correct. Abnormal spectra can be obtained
>> by a system that is out of alignment, unstable or not functioning
>> correctly. .
>>
>> We have measured many spectra before and after a technicians visit on an
>> SP1 (many times) and on AOBS (one time) The question is how will you
>> recognize if the system is working correctly. By what parameters are
>> you using to access that your machine is working? We maintain that the
>> MIDL lamp that has defined spectra is an excellent tool to determine
>> proper spectra representation. The MIDL lamp works on all spectroscopic
>> imaging equipment including confocal microscopes from Leica, Zeiss and
>> Olympus. Chroma slides yield broad spectra and are only useful for
>> course spectra. They are not as sensitive as the MIDL lamp for spectra
>> assessment. Putting the slider over a laser line as Leica recommends
>> will determine if the system is reflecting the proper laser light but
>> this is a subjective criteria that Leica has used to address their
>> alignment issues. How much reflection is acceptable and what does that
>> value actually relate to.
>>
>> How are you going to check your system to determine that the wavelength
>> registration is correct and that your system is not picking up stray
>> light. We use a MIDL lamp to access spectra and determine if there is
>> proper sensitivity and accuracy in our system.
>> Other suggestions to check spectral accuracy are welcome.
>>
>> Best wishes
>> Bob
>>
>> Robert M. Zucker, PhD
>> U.S. Environmental Protection Agency
>> Office of Research and Development
>> National Health and Environmental Effects Research Laboratory
>> Reproductive Toxicology Division, MD 72
>> Research Triangle Park, North Carolina, 27711
>> Tel: 919-541-1585; fax 919-541-4017
>> e-mail: [log in to unmask]
>>
>>
>>
>> |---------+-------------------------------->
>> | | Arthur Schuessler |
>> | | <[log in to unmask]
>> | | RMSTADT.DE> |
>> | | Sent by: Confocal |
>> | | Microscopy List |
>> | | <[log in to unmask]
>> | | FFALO.EDU> |
>> | | |
>> | | |
>> | | 08/17/2004 07:11 PM |
>> | | Please respond to |
>> | | Confocal Microscopy |
>> | | List |
>> | | |
>> |---------+-------------------------------->
>>
>>> ----------------------------------------------------------------------------
>>> ----------------------------------|
>> |
>> |
>> | To: [log in to unmask]
>> |
>> | cc:
>> |
>> | Subject: Re: Confocal Instability-shocking
>> observation |
>>
>>> ----------------------------------------------------------------------------
>>> ----------------------------------|
>>
>>
>>
>>
>> Search the CONFOCAL archive at
>> http://listserv.acsu.buffalo.edu/cgi-bin/wa?S1=confocal
>>
>> Hi Bob,
>> I am just leaving for hollidays ... so will not get any answers.
>> I can only say that we had no obvious problems, but as I wrote, we
>> always
>> used the same PMT. Slight shifts (5 nm range) we would not have
>> detected,
>> with our biological samples. We often used settings suitable for in vivo
>>
>> measurements over longer times, so we had to adjust to low laser power,
>> and
>> often used 10 nm width, with 5 nm steps, as I wrote. We measured the
>> power
>> at object level with a photodiode, but I don't have the value handy,
>> now.
>> We also see laser power variation, just by recording changing
>> fluorescence
>> intensity of a sample over time. In so far, I agree that this should be
>> checked, but since we anyway cannot make it more stable ...
>> I wonder why no real approach is used by Leica (and others?) to regulate
>>
>> and stabilize the laser power. Should be easy by using the AOTFs? At
>> least
>> the changes we saw were not too fast not to be corrected, I think.
>> But as I said - our SP1 has no service contract and works quite well !!
>> And: we didn't make pretty pictures in the noted project, we measured
>> many,
>> many spectra, mainly in vivo ...
>> I also know the Olympus/Zeiss/Leica comparison (I mean that one with the
>>
>> Leica before and after visit of technician, I have a pdf-file). I really
>>
>> wondered (wasn't it a SP2 ? don't remember) how the shown extremely (I
>> would say) bad spectra (before re-adjustment by Leica stuff) were
>> recorded
>> - ours look like those after the tech-visit since long. I think there
>> was
>> something VERY seriosly wrong with your machine? Were the spectra
>> "before
>> re-adjustment" just caused by a completely missaligned system?
>> We unfortunately do not have the money to buy a Lightform, although this
>>
>> for sure makes sense. But I also never felt a realy need for it, yet.
>> Arthur
>>
>> At 22:34 17.08.2004, you wrote:
>>> HI Arthur
>>> You must have a fantastic machine for it to work 5 years without
>>> problems. What is your secret??
>>>
>>> How are you measuring stability of your machine? Is it subjective or
>>> have you applied a specific test procedure to insure that it is stable.
>>> Most biological samples have too broad of a spectra to be accurate. We
>>> have had laser power drifts in excess of 25% and spectral shifts of in
>>> the range of 5-20nm. This can only be detected by measuring the system
>>> with a specific test procedure. If you do not QA the machine then your
>>> endpoint of a pretty picture becomes the gold standard. When excessive
>>> noise comes into the system, it can be observed as bad images. Most
>>> problems can be masked with extra averaging to remove the noise. If
>> you
>>> set large detection regions I do not think you will observe spectral
>>> instability properly.
>>>
>>> We have developed a procedure using a Lightform lamp in which fixed
>>> wavelengths of light are measured across the spectra. These peaks
>> should
>>> come into the same area on ALL spectral imagine machines including
>>> confocal microscopes of Leica, Zeiss and Olympus. If you do not test
>>> for spectral accuracy there is no assurance that the machine is
>>> measuring spectra correctly. Do you have a test ? It also appears
>> that
>>> other machine problems can be detected when we observe bad spectra
>> from
>>> the lamp that may indicate problems with the machine alignment . all
>>> scientists using spectral imaging systems should consider this
>> Lightform
>>> test system.
>>> Best wishes
>>> Bob
>>>
>>>
>>> Robert M. Zucker, PhD
>>> U.S. Environmental Protection Agency
>>> Office of Research and Development
>>> National Health and Environmental Effects Research Laboratory
>>> Reproductive Toxicology Division, MD 72
>>> Research Triangle Park, North Carolina, 27711
>>> Tel: 919-541-1585; fax 919-541-4017
>>> e-mail: [log in to unmask]
>>>
>>>
>>>
>>> |---------+-------------------------------->
>>> | | Arthur Schuessler |
>>> | | <[log in to unmask]
>>> | | RMSTADT.DE> |
>>> | | Sent by: Confocal |
>>> | | Microscopy List |
>>> | | <[log in to unmask]
>>> | | FFALO.EDU> |
>>> | | |
>>> | | |
>>> | | 08/16/2004 04:27 PM |
>>> | | Please respond to |
>>> | | Confocal Microscopy |
>>> | | List |
>>> | | |
>>> |---------+-------------------------------->
>>>
>>>
>>> -------------------------------------------------------------------------
>> -------------------------------------|
>>
>>> |
>>> |
>>> | To: [log in to unmask]
>>> |
>>> | cc:
>>> |
>>> | Subject: Re: Confocal Instability-shocking
>>> observation |
>>>
>>>
>>> -------------------------------------------------------------------------
>> -------------------------------------|
>>
>>>
>>>
>>>
>>>
>>> Search the CONFOCAL archive at
>>> http://listserv.acsu.buffalo.edu/cgi-bin/wa?S1=confocal
>>>
>>> Our confocal is very stable.
>>> We have a SP1 (one of the first ones), used by all the biology
>>> department,
>>> without any "real" control (means no technician or scientist really
>>> responsible for the instrument, used by many different groups).
>>> Our group regularly did spectra with 488nm exication and PMT1 for
>>> detection
>>> 530-730nm, and also with 2-photon excication (detection 400-700nm). We
>>> often "oversample" spectra, to have better signal and a good spectral
>>> resolution (e.g. 10 nm width with 5 or 7nm steps). But we also use 5 or
>>> 7
>>> nm width with same step sizing. Depends on the samples.
>>> So, I cannot answer about the different PMTs, but the many spectra we
>>> recorded over 2 years were all OK. (Although we probably would NOT have
>>> detected a 5nm shift, since we assume this to be in the range of error
>>> (temperature changes, physiological changes - we did many in vivo
>>> measurements of cyanobacterial pigments, pH changes in slightly
>> buffered
>>>
>>> material, etc.).)
>>> So, we e.g. have the pigment-fluorescence spectra as references, no
>>> calibration lamp or so. But these spectra are originating from
>>> fluorescence
>>> of 2-4 different pigments, and show 2-3 peaks (dependent on the sample)
>>> between 450-680 nm, so well suited to compare. Since the spectra are
>>> mixed
>>> ones (and they change in different developmental stages) we also did
>>> spectral unmixing by iterative approaches, using the individual (known)
>>> spectra of the different pigments. The fits worked very well, so the
>>> spectra are OK.
>>> Our SP1 was running perfectly for 5 years "as delivered", without
>>> realignment. Now it only was readjusted after changing the lasers, and
>>> again everything is OK.
>>> We of course work at 23°C all the time. We also only used water
>>> immersion
>>> lenses (20x and 63x).
>>> Arthur
>>>
>>>
>>> At 19:39 16.08.2004, you wrote:
>>>
>>>> Confocal Users
>>>> How stable is the alignment in your confocal microscopy system? Does
>>> the
>>>> beam wander or are the mechanical components affected by heat?
>>>>
>>>> We Tested our Leica SP1 spectral system with a LightForm lamp as
>>>> described our MM 2004 abstract and in a tutorial review on
>>> spectroscopic
>>>> imaging to be published in Cytometry (in press) â Calibration and
>>>>> V>Validation of Confocal Imaging System.â The ttest meaasures
>> spectral
>>>> registration of defined peaks between 400-650 using the LightForm
>>>> Spectral Lamp.
>>>>
>>>> Morning: The system showed that PMT 1and PMT 2 were excellent, with
>>>> sharp narrow wavelength reference spectra peaks. PMT 3 had less
>>>> resolution than the other two for an unknown reason.
>>>> Late Afternoon: Strange image data recorded in PMT 2 below the
>>>> 647-excitation line while measuring TOPRO-3 (PMT3) and Alexa 568 (PMT
>>> 2)
>>>> -Possible reflections were occurring in the detecting region assigned
>>> to
>>>> PMT 2.
>>>>
>>>> We next tested the three PMTs in system with LightForm lamp. It was
>>>> observed that PMT 1 and PMT 3 were identical to the morning values.
>> PMT
>>>> 2 shifted 5nm to lower wavelength values and the FWHM of the peaks
>> were
>>>> now very broad. This suggests that resolution was lost using this PMT
>> 2
>>>> (our best PMT in the morning). We put sliders over the laser line and
>>>> found the unusable range below the laser lines to be 8nm (488)
>>> 16nm(568)
>>>> and 32nm (647). They supposedly should be 7nm or less.
>>>>
>>>> How is this occurring? Why is the system going out of calibration?
>>> Must
>>>> we test at multiple times in the day to insure a calibrated system?
>> How
>>>> would you test for laser or machine drift on your confocal machine?
>> It
>>>> is a shocking observation that other confocal users need to be aware
>>> of,
>>>> as the ramifications are very serious for good reproducible data.
>>> Letâs
>>>> di>discuss this confonfocal instability on this user group.
>>>>
>>>> PSSame effecct observeed about 5 different timees on our machine over
>> a
>>>> 2-year period and at least twice on another machine located in a
>>>> different geographical area. We do not check for this problem all the
>>>> time. Do You?
>>>>
>>>> Best wishes
>>>> Bob
>>>>
>>>> Robert M. Zucker, PhD
>>>> U.S. Environmental Protection Agency
>>>> Office of Research and Development
>>>> National Health and Environmental Effects Research Laboratory
>>>> Reproductive Toxicology Division, MD 72
>>>> Research Triangle Park, North Carolina, 27711
>>>> Tel: 919-541-1585; fax 919-541-4017
>>>> e-mail: [log in to unmask]
>>>
>>>
>>> Dr. Arthur Schuessler
>>> Institute of Botany, FB10, TU Darmstadt
>>> Schnittspahnstrasse 10
>>> D-64287 Darmstadt
>>> GERMANY
>>>
>>> Fax: ++49 1212 66151 164568
>>> e-mail: [log in to unmask]
>>> http://www.geosiphon.de/
>>> http://amf-phylogeny.com/
>>
>>
>> Dr. Arthur Schuessler
>> Institute of Botany, FB10, TU Darmstadt
>> Schnittspahnstrasse 10
>> D-64287 Darmstadt
>> GERMANY
>>
>> Fax: ++49 1212 66151 164568
>> e-mail: [log in to unmask]
>> http://www.geosiphon.de/
>> http://amf-phylogeny.com/
>
>
> Dr. Arthur Schuessler
> Institute of Botany, FB10, TU Darmstadt
> Schnittspahnstrasse 10
> D-64287 Darmstadt
> GERMANY
>
> Fax: ++49 1212 66151 164568
> e-mail: [log in to unmask]
> http://www.geosiphon.de/
> http://amf-phylogeny.com/
--
---------------------------------------------------------------------
| Dr. Olaf Selchow |
| Central Imaging Facility - Microscopy & Image Analysis |
| Institute of Cell Biology and Immunology - SFB 495 |
| University of Stuttgart, Allmandring 31, 70569 Stuttgart, Germany |
| T: +49 711 685 8209, F: +49 711 685 7484 |
| email: [log in to unmask] |
---------------------------------------------------------------------
|
