Fluorescence and filter cubes

Here you can discuss different microscopic techniques and illumination methods, such as Brightfield, Darkfield, Phase Contrast, DIC, Oblique illumination, etc.
Post Reply
Message
Author
Didier_Barbet
Posts: 4
Joined: Sat Sep 14, 2019 8:31 pm

Fluorescence and filter cubes

#1 Post by Didier_Barbet » Sat Sep 14, 2019 9:31 pm

Hello,

I just get an Olympus BX41 equiped with epi fluorescence illumination and with one U-MWB cube (excitation 450-480, barrier filter 515...)
I made several attempts, but without great success, and I have several questions.

1) Which cube is most appropriate to observe vegetables and insects in primary fluorescence (without fluorochrome) ?
Especially lichen and fern spores ?

2) Do I need special objectives for fluorescence ?

3) Are there specific cameras for fluorescence? Is it worth it?

4) Is it possible to check if the barrier filter of the cubes is damaged (that means could be dangerous for the eyes) ?

Thank you for your help
Regards

PeteM
Posts: 2982
Joined: Tue Dec 06, 2016 6:22 am
Location: N. California

Re: Fluorescence and filter cubes

#2 Post by PeteM » Sat Sep 14, 2019 9:56 pm

Didier, others will know much more about fluorescence microscopy, but I can share what little I know.

Regular glass elements will often not pass UV excitation frequencies very well. So, objectives (typically with fluorite glass) are recommended. If your Olympus objectives are marked "FL" (e.g. as in UPlanFL) you're OK. Regular cover slips might also be an issue??

It would help to be sure auto-fluoroscence at the excitation frequency you have at hand is even possible with a specimen while you're learning the ropes. Most research is done with fluorescent tags.

Because the fluorescence response will sometimes be very dim (and researchers are often trying to identify a small bit of protein within a single cell), special cameras are often used. These may even be cooled to reduce "noise" in the image as far as possible. I doubt you'll want to go to that extent?

Messing around with mercury or other powerful UV sources without adequate barriers (watch for scattered reflections from the specimen as well) is an obvious risk. The joy of seeing something auto-fluoresce needs to be weighed against the real possibility of an amateur burning their eyes -- you won't feel it happening -- and losing sight.

I suspect you could use a known good UV capable spectrometer to check transmission of your barrier filter. It sounds like you (like me) are pretty much new to fluorescence, so it might be best to find someone, perhaps at a local university, to show you the ropes?

You also need to be sure your UV source (likely a mercury lamp) is still emitting sufficient excitation. The bulbs are only good for about 100 hours, less if turned on and off frequently as might easily happen lab equipment consigned to surplus. As they age, the bulbs also become somewhat more prone to shattering and the release of mercury. The only indication of age you may have is likely the hours counter on the power supply or to start with a new lamp.

Wearing polycarbonate safety glasses (ones rated for UV blocking) while you're experimenting would be one step I'd take.

Good luck. Take care.

Hobbyst46
Posts: 4277
Joined: Mon Aug 21, 2017 9:02 pm

Re: Fluorescence and filter cubes

#3 Post by Hobbyst46 » Sun Sep 15, 2019 7:09 am

Didier_Barbet wrote:
Sat Sep 14, 2019 9:31 pm
Hello,

I just get an Olympus BX41 equiped with epi fluorescence illumination and with one U-MWB cube (excitation 450-480, barrier filter 515...)
I made several attempts, but without great success, and I have several questions.

1) Which cube is most appropriate to observe vegetables and insects in primary fluorescence (without fluorochrome) ?
Especially lichen and fern spores ?

2) Do I need special objectives for fluorescence ?

3) Are there specific cameras for fluorescence? Is it worth it?

4) Is it possible to check if the barrier filter of the cubes is damaged (that means could be dangerous for the eyes) ?

Thank you for your help
Regards
Hello and welcome. Greetings on BX41 microscope!

If the illumination source is a xenon lamp, the same safety rules as for mercury lamps apply, as mentioned by PeteM above.
Replacement bulbs are generally available from Osram or Sylvania, and are not cheap.
Bulb replacement should be done according to the instructions of the manufacturer of the illuminator. Do not hold the bulb with bare fingers - use cloth gloves or wrapp it with tissue paper while handling. Never look directly at the lamp when it is on. Often alignment of the bulb and reflector-collimator is required, and is done when the lamp house is disconnected from the microscope, and projects a (very very bright!!!) image of the lamp on a nearby wall. And protect your eyes as commented by PeteM.

The cube you have transmits excitation light at 450-480, and fluorescence light at >515nm. The excitation is somewhat less than the optimum for fluorescein (490-495nm), but is probably adequate, so the cube is useful for fluorescein. This would be reasonable, since fluorescein is a very ubiquitous stain. The problem is, that it does not reside in organisms, unless artificially introduced. It is not "auto-fluorescence", in which you are interested.

So, will you see auto-fluorescence with this very cube ? possibly, but it will be quite weak. The strongest fluorescence from green plants originates from chlorophyll and its relatives. These fluoresce at 600-660nm (red). Optimum excitement is at 400nm (violet) asa well as 425nm (violet-blue). UV is not required. There are other natural occuring chemicals, that fluoresce at 500-600nm, when excited at 365nm, namely UV. It might be worthwhile to Google for "insect fluorescence". I know nothing about it; fireflies AFAIK do not fluoresce, they emit light due to a different mechanism. You might also want to search Goggle for fluorescent lichen and spores etc.

If the fluorochrome of interest to you requires UV, then UV-specific objectives from Olympus are advantageous, but I would try and start without them. Brand "UIS" objectives are great (and UIS-2 even greater) so might suffice. When no fluorescence at all is visible, I would not blame the objectives. If the excitation will be visible light only, UIS is sufficient in my opinion.

Fluoresecence, as you are probably aware, is very weak in general. So, a light sensitive camera is a must, but possibly a modern DSLR or similar camera can do the job, at least for a start. Dedicated, cooled cameras from Olympus exist, such as models DP-70...DP-74, but are very costly (thousands of $$). If you buy such, consider to add a ~0.5X reducing lens, to increase the FOV seen by the camera.

You can inspect the cube you have with visible light. Blue, green, red laser pointers. Remove the illuminator from the microscope. Wear eye protection against laser light. Shine light into the excitation port. Place a piece of white paper on the stage, and a piece on the output port (where the camera will be placed). A blue beam will yield a blue spot on the stage. Then place a flat mirror on the stage. No blue spot should be visible anywhere. Replace again the mirror by a white paper. Shine a red beam - no spot should be visible. With a green beam, a mixture of down and up spots might be visible when a mirror is on the stage.

Didier_Barbet
Posts: 4
Joined: Sat Sep 14, 2019 8:31 pm

Re: Fluorescence and filter cubes

#4 Post by Didier_Barbet » Sun Sep 15, 2019 11:35 am

Thank you for your answers. It will help.
Regards
Didier

MicroBob
Posts: 3154
Joined: Sun Dec 25, 2016 9:11 am
Location: Northern Germany

Re: Fluorescence and filter cubes

#5 Post by MicroBob » Sun Sep 15, 2019 11:54 am

Hi Didier,
here you can see an application of fluorescence suitable for amateur use: https://www.mikroskopie-forum.de/index. ... ic=32609.0
I myself would be hesitant to use UV light sources. In professional use the microscope will be used only for this single purpose and professionally seviced. As an amateur one uses a wide variety of ever changeing setups and it is easy to forget one important filter in the setup process. When I see used fluorescence filters on ebay they are often in bad condition so this is a point to look at closely.
With the use of LEDs some safe fluorescence methods have become possible. There are fluorescence objectives with especially high apertures for optimized light gathering - probably not necessary for amateur use with todays digital cameras.

Bob

Hobbyst46
Posts: 4277
Joined: Mon Aug 21, 2017 9:02 pm

Re: Fluorescence and filter cubes

#6 Post by Hobbyst46 » Sun Sep 15, 2019 12:33 pm

MicroBob wrote:
Sun Sep 15, 2019 11:54 am
Hi Didier,
here you can see an application of fluorescence suitable for amateur use: https://www.mikroskopie-forum.de/index. ... ic=32609.0
I myself would be hesitant to use UV light sources. In professional use the microscope will be used only for this single purpose and professionally seviced. As an amateur one uses a wide variety of ever changeing setups and it is easy to forget one important filter in the setup process. When I see used fluorescence filters on ebay they are often in bad condition so this is a point to look at closely.
With the use of LEDs some safe fluorescence methods have become possible. There are fluorescence objectives with especially high apertures for optimized light gathering - probably not necessary for amateur use with todays digital cameras.

Bob
May I add some thoughts. I assume (am I correct ?) that the microsope in question is configured for fluorescence as well as brightfield illumination ? Here I assume that this is indeed the case.

Filters for fluorescence are contained within two regions on the BX41:
a) the filter cube itself, which is a closed box and is either in or out of the light path above the objective nosepiece, by means of a rotating wheel (a cube turret). So switching from fluorescence to brightfield and vice versa is straightforward.
b) between the lamp house and the nosepiece, in filter slots. These (typically neutral density filters) can be left in place, even when doing brightfield since they are only effective for the fluorescence light source (epi-illuminator).
Filters for brightfield are (as far as I remember) inserted in slots in the base, in the path of the trans-illuminator, so they also can be left in place, no matter which illumination is used.

I do not know how other filters (polarization, DIC etc) are mounted on such a setup, but the Olympus brochure no doubt shows how.

To avoid the safety issues and other issues involved in mercury/xenon arc lamps, LEDs can be used, but they need to be high power (say, 15W at the minimum). And, every fluorophore requires its specific excitation, namely a separate LED lamp (in addition to the specific cube). Although, if using LEDs, the excitation filter might sometimes be omitted. If a UV LED is to be used, the same safety concern about UV radiation is valid as for mercury/xenon arc lamps. I would definitely suggest to get gain experience with visible light LEDs first - for example a 425nm lamp to see green plants fluoresce red.

Please note, that whatever the light source, Kohler illumination should be established for uniform brightness of the FOV.

MicroBob
Posts: 3154
Joined: Sun Dec 25, 2016 9:11 am
Location: Northern Germany

Re: Fluorescence and filter cubes

#7 Post by MicroBob » Sun Sep 15, 2019 2:01 pm

For a long time only incident fluorescence is used because of the better image quality.

microb
Posts: 729
Joined: Wed Mar 27, 2019 6:39 am

Re: Fluorescence and filter cubes

#8 Post by microb » Sun Sep 15, 2019 4:55 pm

I'll add this to my to-read list:

https://depts.washington.edu/keck/handbook3.pdf

PeteM
Posts: 2982
Joined: Tue Dec 06, 2016 6:22 am
Location: N. California

Re: Fluorescence and filter cubes

#9 Post by PeteM » Sun Sep 15, 2019 4:58 pm

I'd just like to reiterate that UV from a LED (or any other) source is not necessarily safer than a mercury or xenon lamp. It's the intensity that counts and while some LEDs are weak emitters, others have more than sufficient power to damage the eye.

microb
Posts: 729
Joined: Wed Mar 27, 2019 6:39 am

Re: Fluorescence and filter cubes

#10 Post by microb » Sun Sep 15, 2019 6:42 pm

Eye-pieces on a research microscope in a lab environment should serve only as a prop for photo-ops where the scientist takes one of two poses:

1) Proudly looking into the camera while next to a famous symbol of science -- the microscope. The eye-pieces really help establish so much, especially if you see the boring appearances of today's digital microscopes, like a Keyence or Olympus DSX1000. Let's face it. They look like KitchenAid blenders. But slap some eye-pieces on there, and you have yourself the international symbol for science. Snap that picture now.

2) The second pose, used almost exclusively for raising second round VC for your microfluidics lab-on-a-chip entrepreneurial venture, where you pretend to meaningfully stare into the eye-pieces as if science is actually done that way today. It's a classic. If only you could point at something while eager colleagues look on in envy. But alas, the eye-piece world is a lonely one. So purse the lips a little bit like you're thinking -- thinking really really hard -- very important when impressing VC. Plus you can use the picture to make one of those large lobby prints to occupy the lost glazed over eyes of people in the waiting area. Oh yeah, maybe hold the fine focus knob, because you're clearly the one about to find the cure for cancer. Now click the picture.

3) The third usage - oh wait. There isn't one - not for a lab or any serious research. So remove that triocular head and throw it into a trash can -- or sell it on e-bay for people who don't know the dangers.

Basically if someone is in a lab, especially with grad students or worse yet, other professors, and they look into some eye-pieces; they're not thinking things through. You might want to question the quality of their work and their thinking process. Because, they're an idiot.

With today's flat panel displays and cameras, there is no reason to stare into some of the brightest culminated light sources in human history only topped by staring at the sun -- and it's even dumber to do so while hoping that everyone respected the sticker on the microscope that says: all filter cube positions must be occupied. Because, let's face it, a grad student might have just wanted to try one little quick thing -- a simple test check, a tiny harmless change removing a pesky filter blocking the data. Nothing big. Oh wait! Genealogy has pizza downstairs and it's late Friday afternoon! All worries can be forgotten about what just got altered on this turret cartridge of six filter cubes spun like a revolver loaded for Russian roulette till next Wednesday. Yes! It's pizza! OMG! They even have Corona!

It's been centuries, and eye-pieces still suck. Blinking, eyelashes, crinking the neck and head just right, plus the extra money spent on nonsense ergonomic tilting garbage. Why do it?

Put it on a screen and share with people. You can even point as eager colleagues look on in envy. It's not a lonely world anymore!

See the image clearly, and maybe add some image processing to remove distortions which seem to be an idea with so much potential. But even today the camera software still doesn't do it. What's a lut? Barrel distortion? Calibration wha?

If it's serious work, you're capturing the video or images anyways. If you're using a micromanipulator setup, they put a flat panel in front of your face nowadays -- just above some silly eye-pieces because that's how Leica sells inverted microscopes. You can't make them stop. Other than small hobby microscopes, the cost in putting in the extra prism-sliders and lenses is done because manufacturers feel this pressure to appease a market taught with an apprenticeship style that holds onto dogmatic principles far too long.

No research microscope should sell today with eye-pieces. If you spent the money on a used Olympus BX##, put in a single-port tube lens (U-TLU) onto that epi. Edmunds and Thorlabs even sell Olympus matching TLU's too if you desire brand new. Or get one of those harder to find dual port setups so some confocal attachment can get in on the action. But don't spend thousands new or $700-$1200 e-bay used on a BX triocular that's going to burn your eyes out. Burn out a camera instead. A month later when the CCD seems to need some massive color correction -- then you'll know that ripping the eye-pieces out of all the lab setups was a good idea. Why risk it?
Last edited by microb on Tue Sep 17, 2019 6:53 am, edited 2 times in total.

MicroBob
Posts: 3154
Joined: Sun Dec 25, 2016 9:11 am
Location: Northern Germany

Re: Fluorescence and filter cubes

#11 Post by MicroBob » Sun Sep 15, 2019 6:53 pm

PeteM wrote:
Sun Sep 15, 2019 4:58 pm
I'd just like to reiterate that UV from a LED (or any other) source is not necessarily safer than a mercury or xenon lamp.
Hi Pete,
this is true of cause. The source of the light doesn't matter, it is wave lenght and intensity that makes the damage. I plan to stay away from UV LEDs too.

In the post in the german forum I linked Hans-Jürgen used a 455nm LED and got a nice fluorescence effect. This is what I would look into: Incident light fluorescence with LED excitation in the visible range. This will excluse many objects and methods but it is a simple way to stay on the safe side.

Bob

Hobbyst46
Posts: 4277
Joined: Mon Aug 21, 2017 9:02 pm

Re: Fluorescence and filter cubes

#12 Post by Hobbyst46 » Sun Sep 15, 2019 9:21 pm

microb wrote:...
May I ask, which, in your experience, is the affordable DSLR/Mirrorless camera that is as fluorescence-sensitive as the Olympus DP-73 (say) ?

microb
Posts: 729
Joined: Wed Mar 27, 2019 6:39 am

Re: Fluorescence and filter cubes

#13 Post by microb » Sun Sep 15, 2019 9:52 pm

Hobbyst46 wrote:
Sun Sep 15, 2019 9:21 pm
May I ask, which, in your experience, is the affordable DSLR/Mirrorless camera that is as fluorescence-sensitive as the Olympus DP-73 (say) ?
Someone else here has to be way more into cameras than me.

I've never tried it, but this guy set up a Peltier cooled Canon body in a very simple manner: http://dslrmodifications.com/rebelmod450d16c.html

wstenberg
Posts: 291
Joined: Sun Jul 23, 2017 6:11 pm
Location: Dallas TX
Contact:

Re: Fluorescence and filter cubes

#14 Post by wstenberg » Tue Sep 17, 2019 3:18 am

I agree with microBob.
We have eyepieces on several of the microscopes in my lab at work, but we don't use them. We always work through the monitor. Especially with fluorescence.
We do still use the eyepieces on the stereomicroscope for procedures on mice. The assistant watches through the monitor though.
At home, I mostly use the monitor also.

We use monochromatic cameras for epifluorescence. You will get much better resolution than a color camera.

I bought a Glo-Fish from the pet store the other day. They are Danio rerio zebrafish that have been genetically modified. Cheaply available all over the US. I looked under the microscope with the GFP filter cube at work. Amazing fluorescence with GFP. Nothing with our Texas Red filter cube or the DAPI cube (blue). I am still preparing some photos; I'll post them someday.
William
Astoria, Oregon

Zeiss Axiomat
Zeiss Stereomikroskop
Zeiss Tessovar

microb
Posts: 729
Joined: Wed Mar 27, 2019 6:39 am

Re: Fluorescence and filter cubes

#15 Post by microb » Tue Sep 17, 2019 5:13 am

wstenberg wrote:
Tue Sep 17, 2019 3:18 am
I agree with microBob.
Oh. That's why everyone recently has been calling me Bob. Maybe the admin can change my ID. It's too close to Micro Bob's.

Hobbyst46
Posts: 4277
Joined: Mon Aug 21, 2017 9:02 pm

Re: Fluorescence and filter cubes

#16 Post by Hobbyst46 » Tue Sep 17, 2019 3:08 pm

wstenberg wrote:
Tue Sep 17, 2019 3:18 am
...I looked under the microscope with the GFP filter cube at work. Amazing fluorescence with GFP. Nothing with our Texas Red filter cube or the DAPI cube (blue)....
This is expected, given the relatively short excitation wavelength for DAPI (~360nm) and the relatively long excitation wavelength for Texas Red (~590nm), far away from the GFP (~470nm). A bandpass filter in the cube will block wavelengths shifted by 100nm.
Side note: the USB camera software Toupview, among its other advanced and scientifically oriented options, includes a list of fluorescent dyes and their spectral data. Cool !

MicroBob
Posts: 3154
Joined: Sun Dec 25, 2016 9:11 am
Location: Northern Germany

Re: Fluorescence and filter cubes

#17 Post by MicroBob » Tue Sep 17, 2019 5:28 pm

Here: https://www.photomacrography.net/forum/ ... microscope
Pau shows a nice modern fluorescence setup.

And here: https://www.chroma.com/knowledge-resour ... microscope
Is a link to a good description of fluorescence use.
I think Leice has a table somewhere that shows dies and suitable filter sets.
Some fluoescence dies are extremely expensive.

Bob

Leitzcycler
Posts: 254
Joined: Mon Sep 09, 2019 11:55 am

Re: Fluorescence and filter cubes

#18 Post by Leitzcycler » Tue Sep 17, 2019 6:28 pm

[/quote]
To avoid the safety issues and other issues involved in mercury/xenon arc lamps, LEDs can be used, but they need to be high power (say, 15W at the minimum).
[/quote]

Do you know where can I buy 15W leds? In Ebay there are only 3W leds with proper wavelengths for fluorescence microscopy.

Hobbyst46
Posts: 4277
Joined: Mon Aug 21, 2017 9:02 pm

Re: Fluorescence and filter cubes

#19 Post by Hobbyst46 » Tue Sep 17, 2019 6:35 pm

There is a brand of LED illuminators for microscopy, called PE-100 or PE-300, they are single and multi-LED lamps respectively. They provide adequate intensity ( I guess, about the same order of magnitude of Hg/Xe lamps but I never tried to verify). They include dimmers. But, they are directed at the research market, and priced in the hundreds of $$ or more.
Also, I am GUESSING that their power is ~15W.

Perhaps, take a look at the DIY LED illuminator for fluorescence built and posted by Pau on the photomacgoraphy.net forum. I think he was happy with the performance on his microscope.
I made several attempts, but without great success, and I have several questions.
Perhaps you want to tell us in more details about the attempts and their results.

Leitzcycler
Posts: 254
Joined: Mon Sep 09, 2019 11:55 am

Re: Fluorescence and filter cubes

#20 Post by Leitzcycler » Tue Sep 17, 2019 7:41 pm

Thanks! I looked photomacrography.net as you suggested and found leds being 3-5W. I have also built an fluorescence illuminator myself using Ebay 3W leds. The system works somehow, however would need more light. With this experience I would say 15W is a good quess for commercial light sources (which are far too expensive).

MicroBob
Posts: 3154
Joined: Sun Dec 25, 2016 9:11 am
Location: Northern Germany

Re: Fluorescence and filter cubes

#21 Post by MicroBob » Tue Sep 17, 2019 8:25 pm

Hi together,
when comparing an LED with a mercury bulb one would have to take the efficiency and the spread of the output over the spectrum into account. I think a 3W LED of just the right wavelength can be quite strong compared to a mercury lamp where most of the output is held back by the exciting filter. Peter Höbel once showed a nice arranged diatom slide into which be burned a discoloration with an UV LED. 8-)

Bob

Hobbyst46
Posts: 4277
Joined: Mon Aug 21, 2017 9:02 pm

Re: Fluorescence and filter cubes

#22 Post by Hobbyst46 » Tue Sep 17, 2019 9:06 pm

MicroBob wrote:
Tue Sep 17, 2019 8:25 pm
Hi together,
when comparing an LED with a mercury bulb one would have to take the efficiency and the spread of the output over the spectrum into account. I think a 3W LED of just the right wavelength can be quite strong compared to a mercury lamp where most of the output is held back by the exciting filter. Peter Höbel once showed a nice arranged diatom slide into which be burned a discoloration with an UV LED. 8-)

Bob
Although I agree in principle, my experience is that stronger LEDs are much better, at least with regards to visible light. Yet, the issue is the lack of a common standard for fluorescence, that shows the absolute minimum illumination intensity. One reason is that the units of fluorescence are arbitrary.
A well known strongly fluorescent material for microscopy is fluorescent polymer (PSL for example) beads. They are sold by several suppliers as a suspension in water, and cover a diameter range of 0.2-10um at least. They are labeled with several different dyes, so emit light at various wavelengths of excitation, under several excitation wavelengths.

A quite strongly auto-fluorescent liquid, cheap and easy to use, is virgin olive oil or virgin hemp oil. Use an air-sprayer to spray tiny drops of oil (say, 10-50um diameter) on a slide. Try to focus on the drop. Shine light at 400-430nm. If you can easily see the red fluorescence of the drops, chances are that the setup is OK, at least for a start.

Leitzcycler
Posts: 254
Joined: Mon Sep 09, 2019 11:55 am

Re: Fluorescence and filter cubes

#23 Post by Leitzcycler » Wed Sep 18, 2019 8:49 am

One thing I don't understand, could someone explain: why filter cubes might not be ok and safe if there are no visible delamination, scratches or missing parts. What can go wrong? Some ageing process destroying filtering capacity? Sounds really strange...

Hobbyst46
Posts: 4277
Joined: Mon Aug 21, 2017 9:02 pm

Re: Fluorescence and filter cubes

#24 Post by Hobbyst46 » Wed Sep 18, 2019 11:59 am

Leitzcycler wrote:
Wed Sep 18, 2019 8:49 am
One thing I don't understand, could someone explain: why filter cubes might not be ok and safe if there are no visible delamination, scratches or missing parts. What can go wrong? Some ageing process destroying filtering capacity? Sounds really strange...
The safety concerns as expressed by some fellows above are not so much about cubes per se. The cube protects the eyes (and camera and other sensors) against strong stray excitation light. Three interference filters are used in the cube. Admittedly, these deteriorate with time, even when safely kept within the cube, and are also affected by intense light, so should be routinely inspected (I would say on a yearly or half-yearly basis). I have seen many aged filters (over years, though, not short periods of time!!!) where the coating apears to be damaged. Especially problematic, from the protection aspect, are the dichroic mirror and emission filters. If the dichroic is defective, and reflects only part of the short wavelength light, then that portion of hazardous excitation light that is reflected back from the specimen passes directly to the emission filter. And if the emission filter is defective as well, the protection level drops.

However, the main concern is the fluorescent lamp, and its interface with the microscope. Potential issues are (in my experience):
1) UV radiation spill from openings in the lamp house (say, if somewhat broken or damaged or misses a part;
2) UV radiation spill from openings in the connection of the lamphouse (or collimator) to the microscope frame (rare);
3) UV radiation spill from an open port in the filter cube carrier; the carrier is typically either a linear slab (in older scopes) or turret (in newer scopes, like the Olympus BX models) multiport unit; if not all ports are properly populated with cubes, including dichroic beamsplitters, that might create danger;
4) Reflected UV radiation from misplaced shiney metal surfaces (possible though rare);
5) Strong UV radiation upon handling a new lamp during bulb replacement and/or alignment, when the lamp house is disconnected from the microscope;
6) Explosion of aged bulbs when touched while still very hot;

All these are easily avoided if the microscopist is careful and responsible at his home lab, not one of the students that microb described above...

Post Reply