Homemade microscope illuminator

dennisua · Post by **dennisua** » Fri Sep 21, 2012 4:54 am

Hi everyone!
I own Chinese (or probably Indian) microscope sold by AmScope. It's main illuminator contains 6W 20Wt halogen lamp and simple two lens condenser with ground glass in-between. The ground glass eats away lots of light needed for high power work or things like dark-field or phase contrast. But its removal greatly degrades image quality. There also no field diaphragm to cut stray light. So I was thinking about rebuilding illumination system from scratch. Is it possible to use high power LED (something like XM-LT 60 or SSC P7) as a light source and camera lens (say 50mm F2.8) as a condenser. Putting LED into focal point of the lens will produce parallel beam from the lens entrance. And most lenses are achromatic, aplanatic and anastigmatic so the beam should be aberration free. Will it allow me to have Koehler type illumination? What do you think?
Thanks!

g4lab · Post by **g4lab** » Fri Sep 21, 2012 5:28 am

The only slight problem I see with your idea , which certainly would work OK for starters (Aside from the fact that I much prefer quartz halogen light to LEDs since LEDs are not really white) is that microscope condensors are very fast short focal length lenses. Using a nice beautiful free 50mm F2 will probably end up being the limiting aperture in the system which you might like to be larger.

This is a question for Rik to answer I believe. But I think I am close.

Pau · Post by **Pau** » Fri Sep 21, 2012 7:40 am

May be this one can fit your scope allowing Kohler:
http://www.ebay.com/itm/Kohler-Illumina ... 19d5610cd6

dennisua · Post by **dennisua** » Sat Sep 22, 2012 4:16 am

Thanks Pau, It does look like drop-in replacement but I don't really want to spend $150 without trying other (cheaper) options first.

dennisua · Post by **dennisua** » Sat Sep 22, 2012 4:34 am

g4lab wrote:The only slight problem I see with your idea , which certainly would work OK for starters (Aside from the fact that I much prefer quartz halogen light to LEDs since LEDs are not really white) is that microscope condensors are very fast short focal length lenses. Using a nice beautiful free 50mm F2 will probably end up being the limiting aperture in the system which you might like to be larger.

Where this limitation comes from? What if LED image will fully illuminate the substage condenser diaphragm? I will probably need to move the lens a bit further (something like 50cm from the substabe condenser). Bulky but doable.

rjlittlefield · Post by **rjlittlefield** » Sat Sep 22, 2012 10:29 am

dennisua wrote:Putting LED into focal point of the lens will produce parallel beam from the lens entrance.

dennisua wrote:
g4lab wrote:The only slight problem I see with your idea , which certainly would work OK for starters (Aside from the fact that I much prefer quartz halogen light to LEDs since LEDs are not really white) is that microscope condensors are very fast short focal length lenses. Using a nice beautiful free 50mm F2 will probably end up being the limiting aperture in the system which you might like to be larger.
Where this limitation comes from?

Perhaps it will help to review what a condenser does.

The main purpose of a condenser is to allow every part of the objective to contribute equally to image formation. This requires that every part of the objective must receive light from each point on the subject. Even if you look at a subject that is just a single pinhole, say 10-100 microns in diameter, still the whole aperture of the objective should be uniformly illuminated.

A parallel beam of illumination absolutely will not accomplish this. Instead the pinhole will simply create a thin beam of light that will strike only a small area of the objective. Instead of your 10X objective having say NA 0.25, it will act as if it had NA 0.01 and the resolution will be correspondingly bad.

To work properly, the condenser must produce a broad cone of light, with an angle at least as wide as the entrance cone of the objective. To maintain full resolution with a 20X NA 0.40 objective, the condenser must be at least NA 0.40 also. But an f/2 lens is only NA 0.25. Even f/1.4 = NA 0.36 is not quite enough.

In general, to fully utilize the objective you would need a condenser output lens with f-number = 1/(2*NA), for example f/1.25 at NA 0.40, f/0.83 at NA 0.60, and so on. Anything smaller than this will effectively limit the aperture of the objective.

If you want to use a highly corrected lens as a condenser, then it would be better to use a microscope objective instead.

The ideal situation is to have matched objectives for condenser and imaging, for example 20X NA 0.40 as a condenser to go with 20X NA 0.40 for imaging. This can give you uniform illumination across the field, while also giving you enough NA to fully utilize the imaging objective.

--Rik

Pau · Post by **Pau** » Sat Sep 22, 2012 11:04 am

dennisua wrote: It's main illuminator contains 6W 20Wt halogen lamp and simple two lens condenser with ground glass in-between. The ground glass eats away lots of light needed for high power work or things like dark-field or phase contrast. But its removal greatly degrades image quality. There also no field diaphragm to cut stray light...

dennisua,
I think there may be some confusion between the condenser and the illuminator collimating optics (Rik answer is about the former one and mine is about the last one)

A picture of your microscope light source and condenser with some indications about the parts you are thinking to change would be adequate to clarify the topic.

rjlittlefield · Post by **rjlittlefield** » Sat Sep 22, 2012 11:44 am

Pau wrote:I think there may be some confusion between the condenser and the illuminator collimating optics (Rik answer is about the former one and mine is about the last one)

Good point, thanks for catching that.

--Rik

rjlittlefield · Post by **rjlittlefield** » Sat Sep 22, 2012 12:38 pm

After several re-reads, it does seem clear that dennisua is talking about just the parts before the main condenser system. I think that removes the issue of NA but raises other questions about exactly what the condenser expects as input. With the condenser in front of me right now, from an aus Jena Laboval 2 marked only "1,2", feeding in a perfectly collimated beam would not work well. I'm worried by the observation that removing the ground glass in dennisua's current system greatly degrades image quality. But there's probably no way to tell except by trying it.

--Rik

ChrisR · Post by **ChrisR** » Sat Sep 22, 2012 2:01 pm

I suspect denisusa's

(say 50mm F2.8) as a condenser.

has been corrupted by the cool smilie.

In this post I have disabled smilies.

g4lab · Post by **g4lab** » Sat Sep 22, 2012 2:28 pm

The purpose of the ground glass in critical illumination iirc, is to increase the na of the illumination beam. By scattering.

I think that in Kohler illumination the na of the system can be restricted by whatever the limiting aperture is. I think all the lenses need to be as fast or faster than the condensor lens.

I am sorry I have not thought about any of this for several decades. So I might be confused or in error.

But I believe, that is why , for example, when you are collecting light from the proverbial HBO103 for fluorescence, or XBO 75 for incident light metallurgy, that the first glass encountered by the light is one or more very thick, fat and very fast collecting lenses.

Your overall idea probably is not bad. But you probably need to go to Edmund Scientific and buy an f/0.5 aspheric.

You can also find pictures of where different images of the light source are to fall in the optical chain. These vary depending on whether you are doing critical or Koehler illumination. You should read the requisite pages at microscopy U where I am sure there are nice diagrams.

dennisua · Post by **dennisua** » Sat Sep 22, 2012 3:34 pm

Sorry for the confusion, I wish I could explain things better.

You are right, I'm talking about condenser that goes right after the lamp but before the substage condenser. I attached the images of the condenser.

dennisua · Post by **dennisua** » Sat Sep 22, 2012 4:29 pm

ChrisR wrote:I suspect denisusa's
(say 50mm F2. as a condenser.
has been corrupted by the cool smilie.

In this post I have disabled smilies.

Exactly, corrected my initial post. I checked the lens it's actually 1.8, but i think it's too low anyway.
I'll probably have better luck If I just buy something like OI-19 (http://www.microscopy-uk.org.uk/mag/art ... html#OI-19). It has 2 lens 0.65NA condenser.

BTW. Someone has already did what I wanted to do.
http://www.microscopy-uk.org.uk/mag/ind ... arger.html

rjlittlefield · Post by **rjlittlefield** » Sat Sep 22, 2012 4:31 pm

dennisua wrote:Sorry for the confusion, I wish I could explain things better.

I apologize for the misinterpretation! I think everybody else understood perfectly and I just wandered off into La La Land.

g4lab wrote:I think that in Kohler illumination the na of the system can be restricted by whatever the limiting aperture is. I think all the lenses need to be as fast or faster than the condensor lens.

This doesn't sound quite right.

With Köhler we can think of the whole illumination chain as being a single very thick lens that projects an image of the light source onto the aperture of the objective. The exit NA of the chain has to match the objective, but the chain's exit NA and entrance NA will be related by the magnification of the chain. So it doesn't seem like there's any problem getting large exit NA even with a small entrance NA -- just adjust magnification as required.

However, there's an interaction with the size of the light source. If you have to crank down the magnification of the chain in order to get large exit NA, then you also require that the light source be sufficiently larger to make up the difference. So there's an advantage to larger entrance NA in that you can get by with physically smaller light sources. Larger entrance NA also means higher efficiency, another advantage.

I suspect the diffuser comes into play to make a virtual light source that is sufficiently large even when the real one isn't. It has the effect of increasing the available NA, but the logic is a little different from a simple limit on the lenses.

I hope I've got this right. It always makes me nervous to reason about optics without some sort of check, and I don't have tools or time to do that right now.

--Rik

Pau · Post by **Pau** » Sun Sep 23, 2012 12:37 am

dennisua,
- The part you marked in the picture isn't the condenser (or substage condenser) but the illuminator optics (also called collecting lens o collimating lens).

- Your scope looks very similar to a chinese model I have at the high school where I work that in its turn is a copy of the Olympus CH. That one has an M42x1 thread to mount the illuminator optics over the halogen lamp and I'm almost sure that the part I linked in my first reply will match.
Most school microscpes in this range do not allow Kohler but some sort of critical illumination, and I think this is not a bad thing, it works well for low and moderate magnification and is much easier to set up. The frosted glass is an important part of this system, not surprising that it doesn't work well without it.
Even many better microscopes like my Zeiss Standards have a frosted glass, sometimes removable, in the light path to homogenize the illumination (I call this "semi Kohler")

- Be aware of the lengh of the light path in the Ted Clarke's Lomo adaptation, not doable in a compact frame without an external optical bench as he did.

Homemade microscope illuminator

Homemade microscope illuminator

Re: Homemade microscope illuminator

Re: Homemade microscope illuminator