>
> Could you provide more information on your problem:
> Is the non-uniformity you speak of, a bright signal in the centre of the
> field of view, and darker at the edges of the screen? If not could you
> describe the pattern of non-uniformity.
>
On PMT1 and PMT2 the pattern is sloped from left to right. On PMT3 it
is noticably peaked in the center and falls off to the sides.
I will make an image available on my web site begining next week.

> What lens are you using? Is it a Plan or Plan Apo? Dry, oil, water
> immersion?
>
Sorry about that. It was a Plan Apo 64x/1.4 oil objective.

> Are you "looking at the same wavelengths in each of the PMTs? This would be
> quite difficult given the fixed dichroic for PMT3.
>
That did not come out right. I meant that the excitation is constant, and
I am looking at the same target. Wavelength ranges are approx. 505--555
for PMT2, 555--605 for PMT1, and 605nm and lower for PMT3. I dont think this
is explained only by chromatic aberation.

> I think what you may be seeing is chromatic aberration. If you use a Plan
> Apo lens the non-uniformity across the screen should be reduced. A zoom of
> 1.5 or 2 should ensure that you use the flattest part of your objective and
> produce flatter optical sections.
>
Zoom of two clearly helps, but I still think it is out of spec, and the
large difference on PMT3 bothers me.

> The effect of chromatic aberration by it's very nature must be different if
> you are looking at different wavelengths. Presumably you are splitting the
> fluorescent signal using the standard A2 block (565LP) so green fluorescence
> will be reflected to PMT-1 , orange-red to PMT-2 the MRC-1024 has a fixed
> dichroic (640 short pass) so that red fluorescence is reflected to PMT-3.
>
I altered my dichroics so that the ranges are as mstated above.

> These different wavelengths must be effected by chromatic aberration by
> differing degrees. Unfortunately, where chromatic aberration is a problem
> the signals in the three detectors should not match in Z. This is "normal".
> You need to choose a lens system that minimises this aberration, ie. Plan
> Apo and minimise Refractive Index (RI) mismatch. This may be difficult if
> the RI of your plastic is unknown, although imaging very close to the
> surface of your block should help.
>
I dont image right by the surface because in that case small variations in
parallelism of the stage will cause large differences. 25 micron is
Biorad's procedure, but I find similar problems 5 and 10 micron
into the block.

> If you need to repeat the test I suggest that you would need to place beam
> splitter block in and compare PMT-1 to PMT-2 with the open position chosen
> for the band pass filter. A suitable band pass filter would then have to be
> screwed into the rear of you first filter cube. This way PMT-1 and PMT-2
> will "see" the same colour, except that you still have the 640SP infront of
> PMT-1. To circumvent this problem the band pass filter you placed in the
> first filter cube could exclude red wavelengths. Comparison of PMT-3 under
> such controlled conditions is more difficult as the dichroic is fixed, and
> so is probably not worth trying.
>
I have used the 515 Low pass filter in PMT1, so that I get at least some
of the same signal as PMT2 (the dichroic isn't the sharpest filter.
Your suggestion is good, but again, since PMT1 and PMT2 are pretty well
correlated, I may not gain much from this test.

thanks for the detailed reply.

--aryeh
Aryeh Weiss                          | email: [log in to unmask]
Department of Electronics            | URL:   http://optics.jct.ac.il/~aryeh
Jerusalem College of Technology      | phone: 972-2-6751146
POB 16031                            | FAX:   972-2-6751275
Jerusalem, Israel                    | ham radio: 4X1PB/KA1PB