How important is exact compliance of tube length?

A forum to ask questions, post setups, and generally discuss anything having to do with photomacrography and photomicroscopy.

Moderators: rjlittlefield, ChrisR, Chris S., Pau

Soki
Posts: 76
Joined: Sun May 03, 2020 5:05 am

How important is exact compliance of tube length?

Post by Soki »

Hey guys,

I shoot via direct projection on the camera sensors. I therefore use finite 160mm Nikon CF/CFN Planapos. It’s pretty hard to get the accurate to 1mm tube length with extension rings. Are Planapos prone to slight deviations of 1mm to 4mm? Especially if you use different cameras with differing sensor planes.
It’s pretty hard for me to see any differences, because I have to refocus the subject after altering.
Is there a computational determination of allowed deviation?

Thanks for your help,
Simon

Pau
Site Admin
Posts: 6052
Joined: Wed Jan 20, 2010 8:57 am
Location: Valencia, Spain

Re: How important is exact compliance of tube length?

Post by Pau »

The sensitivity to changes TL depends of the objective NA. There are two effects of the tube length, both related with NA:

1- Spherical aberration: it increases exponentially with NA, a 0.20 will be very tolerant while a 0.95 wont. Few mm changes are in the eyepiece regulation range, so not likely to be problematic

2- DOF at the image side: it also increases with NA, so high NA objectives are in this aspect more tolerant while low magnification-NA present more parfocality issues. Only important if you change objectives in the same turret nosepiece.

There are papers and graphics on the subject, but, sorry; I can't search them in this moment
Pau

viktor j nilsson
Posts: 423
Joined: Fri Mar 01, 2013 1:43 am
Location: Lund, Sweden

Re: How important is exact compliance of tube length?

Post by viktor j nilsson »

Here's the graph:
Image
(Edit: updated with URL that works)

From:
https://onlinelibrary.wiley.com/doi/abs ... .tb04454.x

(Via: http://janrik.net/RiksLinks.html)
Last edited by viktor j nilsson on Sat Mar 13, 2021 11:33 am, edited 2 times in total.

Soki
Posts: 76
Joined: Sun May 03, 2020 5:05 am

Re: How important is exact compliance of tube length?

Post by Soki »

Hey,

thank you both for the information, that’s what I needed;) . Very interesting graph.

Thanks,
Simon

rjlittlefield
Site Admin
Posts: 23561
Joined: Tue Aug 01, 2006 8:34 am
Location: Richland, Washington State, USA
Contact:

Re: How important is exact compliance of tube length?

Post by rjlittlefield »

viktor j nilsson wrote:
Sat Mar 13, 2021 4:48 am
Here's the graph:
http://www.science-info.net/docs/etc/Tube-Length-na.gif
(Image didn't show when I put it in IMG tags, is the forum not able to show gif files?)
The problem with inline display of this URL is most likely a combination of two things:

1. The browser is automatically promoting the http to https (secure), because the inline display is attempted inside an https connection to photomacrography.net.

2. The browser cannot establish a trusted connection with science-info.net because "This server could not prove that it is www.science-info.net; its security certificate is from *.secure.hostingprod.com." (This message is quoted from my Chrome browser just now.)

It's basically a configuration problem with science-info.net .

--Rik

viktor j nilsson
Posts: 423
Joined: Fri Mar 01, 2013 1:43 am
Location: Lund, Sweden

Re: How important is exact compliance of tube length?

Post by viktor j nilsson »

rjlittlefield wrote:
Sat Mar 13, 2021 11:25 am
viktor j nilsson wrote:
Sat Mar 13, 2021 4:48 am
Here's the graph:
http://www.science-info.net/docs/etc/Tube-Length-na.gif
(Image didn't show when I put it in IMG tags, is the forum not able to show gif files?)
The problem with inline display of this URL is most likely a combination of two things:

1. The browser is automatically promoting the http to https (secure), because the inline display is attempted inside an https connection to photomacrography.net.

2. The browser cannot establish a trusted connection with science-info.net because "This server could not prove that it is www.science-info.net; its security certificate is from *.secure.hostingprod.com." (This message is quoted from my Chrome browser just now.)

It's basically a configuration problem with science-info.net .

--Rik
Thanks! It indeed seemed to be a problem with the lack of security certificate. Found another version hosted on https://krebsmicro.com and updated my post above.

seta666
Posts: 1071
Joined: Fri Mar 19, 2010 8:50 am
Location: Castellon, Spain

Re: How important is exact compliance of tube length?

Post by seta666 »

This graph helps a lot, thanks!!

I took some measurement (rising the eyepiece to be parfocal with the camera)and my camera is off by 20mm on my BHT/Nikon CFN set up; that is way more than I initially thought, no more tests with high NA lenses until I fix that.

Even removing the phototube from the trinocular head there are just few milimiters to work with, 6mm aprox. And that is with a Sony NEX camera, forget about reflex cameras.

Now the trinocular port is adapted to M58 male with a bunch of step rings I had laying around (25mm) waiting for a 37-M58 female adapter, a thin 58mm polarizer (glass removed, so I can rotate the camera)and a Nex-58 male adapter. That will get me close to those 6mm (around 10-12mm)but I am happy to see that I can still afford to be a few mm off. I could remove the polarizer from the equation but camera rotation is kind of important.

In that case getting the eyepieces parfocal will be very easy, I will just need a couple of spacers (around 5-7mm thick).

The nikon BD plans are other story, I can get the desired tube length within +/- 2mm margin of accuracy

Pau
Site Admin
Posts: 6052
Joined: Wed Jan 20, 2010 8:57 am
Location: Valencia, Spain

Re: How important is exact compliance of tube length?

Post by Pau »

In that case getting the eyepieces parfocal will be very easy, I will just need a couple of spacers (around 5-7mm thick)
This is too much with high NA objectives
Pau

seta666
Posts: 1071
Joined: Fri Mar 19, 2010 8:50 am
Location: Castellon, Spain

Re: How important is exact compliance of tube length?

Post by seta666 »

That is true, but according to the graph up to NA 0.70 Air and NA 1.20 Oil it should be Ok, at least for now; fastest air lens I own now is 0.80 and Oil 1.25
I could get it close to 0 losing camera rotation but I do not really like the idea; if necessary I could just rotate the trinocular head instead of the camera.

Anyway, I am sure with time I will get it close to 0 and keep camera rotation, maybe I will have to order some custom adapters.

I have to say this graph is quite accurate because with the 20mm offset I now have quality with the 40/0.70 looks poor and with the BD 60/0.80 bad; I will have to try that one again with close to 0 offset.

Are there any similar graphs for coverslips? Like the one in nikon CFN brochure but more complete.
Are Oil lenses less prone to quality loss than Air lenses because of coverslip variance?
Image

I bought a cheap biological slide set to have some practice, up to 20x is OK, with the 40X is quite useless; I guess because both TL and coverslip(mounting medium thickness

Pau
Site Admin
Posts: 6052
Joined: Wed Jan 20, 2010 8:57 am
Location: Valencia, Spain

Re: How important is exact compliance of tube length?

Post by Pau »

You have some of them in your email
Pau

rjlittlefield
Site Admin
Posts: 23561
Joined: Tue Aug 01, 2006 8:34 am
Location: Richland, Washington State, USA
Contact:

Re: How important is exact compliance of tube length?

Post by rjlittlefield »

seta666 wrote:
Mon Mar 15, 2021 10:46 am
Are Oil lenses less prone to quality loss than Air lenses because of coverslip variance?
Yes. On pg. 73 of "Optics of the Object Space in Microscopy" (same ref as the tube length graph), it is written that
The tolerance for variation of the cover-slip thickness is about 50 times less sensitive with immersion objectives than with dry objectives for the same apertures. Yet, so powerful is the exponential (or logarithmic) function that at N.A. 1.4, which is the practical N.A. used for most critical work, the tolerance to cover-slip variation by the star test is only about (+-)7μ.
The reason for this difference is that with the oil objective, difference in cover slip variation is compensated by changing the thickness of the oil layer, while with a dry objective, it is compensated by changing the thickness of the air layer. The refractive index of glass is much closer to oil than to air, so trading glass for oil gives a lot less aberration than trading glass for air.

--Rik

seta666
Posts: 1071
Joined: Fri Mar 19, 2010 8:50 am
Location: Castellon, Spain

Re: How important is exact compliance of tube length?

Post by seta666 »

Pau wrote:
Mon Mar 15, 2021 11:15 am
You have some of them in your email
rjlittlefield wrote:
Mon Mar 15, 2021 12:49 pm

Yes. On pg. 73 of "Optics of the Object Space in Microscopy" (same ref as the tube length graph), it is written that...
Thanks to both of you..

Today I went for another water sample; every drop is full of diatoms and other interesting creatures..This is fun..really fun!!
I will try to post some tomorrow

For now with the 40/0.70 I close the condenser Iris more than I would like to, but my system will stay too far off until I receive the right adapters

seta666
Posts: 1071
Joined: Fri Mar 19, 2010 8:50 am
Location: Castellon, Spain

Re: How important is exact compliance of tube length?

Post by seta666 »

A few more questions...

- Does this graph also apply to infinite systems? EG, mitutoyo 20/0.42 would have little tolerance but lower NA lenses would allow quite a lot changes in extension to make framing tighter or give it some air.

- Adding extension to the TL could also work to use no coverslip lenses with slides with a coverslip; just the opposite one could use lenses made for cover slip for subjects without one, right? Maybe a focusing helicoid could work as objective correcting collar sort of..

- What is the best way to check TL is spot on? Using a high NA lens and make camera parfocal with viewing the eyepieces diopter adjustment set to 0?

viktor j nilsson
Posts: 423
Joined: Fri Mar 01, 2013 1:43 am
Location: Lund, Sweden

Re: How important is exact compliance of tube length?

Post by viktor j nilsson »

seta666 wrote:
Thu Mar 18, 2021 2:18 am
A few more questions...

- Does this graph also apply to infinite systems? EG, mitutoyo 20/0.42 would have little tolerance but lower NA lenses would allow quite a lot changes in extension to make framing tighter or give it some air.

- Adding extension to the TL could also work to use no coverslip lenses with slides with a coverslip; just the opposite one could use lenses made for cover slip for subjects without one, right? Maybe a focusing helicoid could work as objective correcting collar sort of..

- What is the best way to check TL is spot on? Using a high NA lens and make camera parfocal with viewing the eyepieces diopter adjustment set to 0?
I understand less about infinity systems so I won't comment on that.

In fact, it is better to check tube length compliance using a low-power objective. Low power objectives have less depth of focus at the sensor plane than high power objectives, so using low magnificiation objectives gives you much higher precision.

Depending on your eyes, it is not certain that the 0 adjustment will be ideal. You should begin by making sure that your objectives are as close to parfocal to each others as possible. Set the eyepiece diopters to zero. Begin at a high magnification, set focus at specific feature. Switch back to a lower magnification, set focus again at the exact same feature using the diopter settings - not the focus knob. You may need to go back and forth between high and low mag a few times before you get it perfect.

After that, you can make your camera parfocal with your viewing eyepieces.
Last edited by viktor j nilsson on Thu Mar 18, 2021 3:21 am, edited 1 time in total.

Pau
Site Admin
Posts: 6052
Joined: Wed Jan 20, 2010 8:57 am
Location: Valencia, Spain

Re: How important is exact compliance of tube length?

Post by Pau »

1- I think so. If you use the nominal tube lens focal length focused to infinite, you can think in the inf. objective+tube lens being equivalent to a finite corrected objective

2- Yes, as you well know there is a recent thread on the subject: https://www.photomacrography.net/forum/ ... 95#p259395, although likely more experiments and clarifications would be needed

3- Nope, your method only would be good if you put exactly the right working distance or parfocal distance but both are more difficult to measure with enough precision.
IMO the best method for right parfocality is:
. First put the eyepieces at O or at your diopter eyesight (if your microscope head doesn't change focus when changing the distance between eyes)
. Be sure that the objectives are parfocal between them* (just a bit of refocusing with the fine focus knob) If not, correct it with the diopter rings
. Then with a low power** objective like a 10X set the camera parfocal with the eyepieces and redo the second point with camera

* use objectives of the same brand and similar series and cover correction
** low power/NA objectives have smaller DOF at the camera/eyepiece side while high NA ones have smaller DOF at the subject side but bigger DOF at the camera/eyepiece side.
Pau

Post Reply Previous topicNext topic