Fast Stacking and Continuous vs Flash Lighting

Have questions about the equipment used for macro- or micro- photography? Post those questions in this forum.

Moderators: rjlittlefield, ChrisR, Chris S., Pau

newtonsapple
Posts: 11
Joined: Mon Jul 20, 2020 8:39 pm

Fast Stacking and Continuous vs Flash Lighting

Post by newtonsapple »

I am in the process of putting together a stacking setup for .5x-4x work. This is for low volume industrial inspection documentation and I am trying to put together a very efficient workflow. The current setup is a Fuji x-t20 on a Stackshot rail using Laowa 25mm f2.8 2.5x-5x or a Laowa 65mm f2.8 APO. I also have a Nikon Z50 I am trying out as well to see if I can get Helicon Remote to work, but the Beta release seems to still need work.

The goal is to use electronic shutter and run stacks as fast as possible without incurring motion blur. What sort of strategies have been successful for this?

Helicon remote can apparently "burst stack" with the Stackshot, but I haven't been able to try.

I also need to get some decent lighting. Diffusion will likely be done through some sort plastic or paper cone. I think continuous lighting makes sense for use with the electronic shutter, but I am not sure how much light I really need. Would a pair of Godox SL-60W's work? Do I need even more (or less) light? I know this will depend on how close I can get the lights. It seems like they should be as close as possible while evenly covering teh diffuser. I've seen everything here from 3x 150w LED's to stock Ikea LED's. How do I figure out how much light I need in the context of trying to move fast?

chris_ma
Posts: 570
Joined: Fri Mar 22, 2019 2:23 pm
Location: Germany

Re: Fast Stacking and Continuous vs Flash Lighting

Post by chris_ma »

it would mainly depend on your desired speed and thus shutter speed to avoid motion blur, but then also on NA, ISO and distance and diffusion on the lamp.

short story: it would be possible to calculate this, but with all the factors involved you'll likely much better off just to try it.
I'd imagine a couple of godox 60 should work out ok, but probably best to buy it somewhere with sound return option so in case it doesn't you could return it (I'd imagine that if you start to go on higher wattage on close range you'll run into heat problems)
chris

lothman
Posts: 958
Joined: Sat Feb 14, 2009 7:00 am
Location: Stuttgart/Germany

Re: Fast Stacking and Continuous vs Flash Lighting

Post by lothman »

I also do a lot of macro work for technical application but on my job have the luck to use a Keyence digital microscope which is very fast for such application. On the other hand private I do macro stacking using a slider driven by stepper motor. So I know both worlds.

I would say for your scenario (stack-shot +Helicon) that the whole process of positioning the object (cleaning, mounting, controlling light, avoiding or producing reflection....), stacking, then stacking the single frames will not be considerably be shortened by "faster" stacking.

If it takes you 15 Minutes (estimated) to get a ready picture with "standard" stacking, you might get this down to 12 minutes (estimated) when using faster stacking. So not so much gain but therefore dealing with motion blur (vibration blur) and heat problems of high power light sources.

Fast stacking helps if you have moving objects like living insects. Then the chance to catch a stack when they hold still is much higher, but for static technical parts I see no big advantage.

Babylonia
Posts: 92
Joined: Tue Mar 20, 2012 10:56 am

Re: Fast Stacking and Continuous vs Flash Lighting

Post by Babylonia »

As for using flash, you do need quality flash equipment, where each sequence of a flash do have the same flash lighting power and color.
Mostly waiting just a bit longer than minimum time (1.5 - 2x) as for firing next flash is to be recommended.
Greetings from Holland

newtonsapple
Posts: 11
Joined: Mon Jul 20, 2020 8:39 pm

Re: Fast Stacking and Continuous vs Flash Lighting

Post by newtonsapple »

lothman wrote:
Tue Jun 22, 2021 2:22 am
I also do a lot of macro work for technical application but on my job have the luck to use a Keyence digital microscope which is very fast for such application. On the other hand private I do macro stacking using a slider driven by stepper motor. So I know both worlds.

I would say for your scenario (stack-shot +Helicon) that the whole process of positioning the object (cleaning, mounting, controlling light, avoiding or producing reflection....), stacking, then stacking the single frames will not be considerably be shortened by "faster" stacking.

If it takes you 15 Minutes (estimated) to get a ready picture with "standard" stacking, you might get this down to 12 minutes (estimated) when using faster stacking. So not so much gain but therefore dealing with motion blur (vibration blur) and heat problems of high power light sources.

Fast stacking helps if you have moving objects like living insects. Then the chance to catch a stack when they hold still is much higher, but for static technical parts I see no big advantage.
I would like to get a Keyence setup, but they are pretty expensive though, like $60K new, $5K and up for anything less than 10 years old at auction.

I worked through the rough calcs and it seems like you can get the shutter times into reasonably fast at moderate magnifications and a little iso bump.
Capture.JPG
I'll just have to play with it ultimately. I think some COBs assembled onto Loc Line might be the way to go ultimately as they can be positioned very close. The goal is to be able to run the rail continuously while doing a burst. This should save 2+ minutes for a 60 frame stack.

Scarodactyl
Posts: 1617
Joined: Sat Apr 14, 2018 10:26 am

Re: Fast Stacking and Continuous vs Flash Lighting

Post by Scarodactyl »

Keyence setups are cool, but I don't think they produce very high resolution images. Maybe that's more reflective of older models.

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

Re: Fast Stacking and Continuous vs Flash Lighting

Post by rjlittlefield »

Scarodactyl wrote:
Tue Jun 22, 2021 7:49 pm
Keyence setups are cool, but I don't think they produce very high resolution images. Maybe that's more reflective of older models.
There was a recent article in National Geographic, "Bugs In The Trees", April 2021, HERE (subscription only), first two pages currently shown at https://www.hcgilliland.com/features-1 . I checked with the author who did the photographs; he said they were from a Keyence VHX-5000. The product brochure for VHX-5000 says that it stitches up to 20,000 by 20,000 pixels, but suggests that the native resolution is 1600x1200 and "high resolution" is 4800x3600. I do not know what was used for the National Geographic article.

--Rik

Scarodactyl
Posts: 1617
Joined: Sat Apr 14, 2018 10:26 am

Re: Fast Stacking and Continuous vs Flash Lighting

Post by Scarodactyl »

Crunching some numbers, they quote a max NA of 0.82 for their fluorite objective on the vhx-5000, which at max mag has a diagonal FoV of 0.076mm. The camera has a 1/1.8" sensor with a ~8.9mm diagonal, so that's 117x optically speaking, but that understates things a bit given the small field coverage. That would be like a 290x objective with a strict 22mm fn, more like 350x with a wider 26.5mm fn to cover uw eyepieces or aps-c.
They have some spectacular features and I'm not surprised they're so popular, but I don't think they'd be a very efficient way to take high resolution pictures.

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

Re: Fast Stacking and Continuous vs Flash Lighting

Post by rjlittlefield »

Scarodactyl wrote:
Tue Jun 22, 2021 9:34 pm
high resolution pictures
That opens another question that I don't think has been asked.

newtonsapple wrote
This is for low volume industrial inspection documentation and I am trying to put together a very efficient workflow.
I wonder how much resolution is needed for this purpose. That matters a lot, because it affects the required NA, which affects the step size as 1/NA^2, using the quarter-lambda rule. So, if you keep everything optimized, and reduce the required resolution by 2X on axis, then you cut the image size by 2X on axis (4X by pixel count), and the number of frames by 4X. The acquisition time drops by 4X, the computational cost by 16X -- it's a big deal.

newtonsapple, I know that it's a diversion from your title, but have you worked through the resolution requirements yet?

--Rik

lothman
Posts: 958
Joined: Sat Feb 14, 2009 7:00 am
Location: Stuttgart/Germany

Re: Fast Stacking and Continuous vs Flash Lighting

Post by lothman »

Scarodactyl wrote:
Tue Jun 22, 2021 9:34 pm
They have some spectacular features and I'm not surprised they're so popular, but I don't think they'd be a very efficient way to take high resolution pictures.
this is a newer Youtube clip of two guys looking at watches using a Keyence microscope.

https://www.youtube.com/watch?v=hKrJOMa ... sWatchGang

Since you can do automated stitching (even with HDR) you crank up the magnification, stich and have your high-Resolution image - ate least for technical reports. And this goes within some seconds up to a minute.

Of course with a high end optic on my Sony A7riv I can squezze out more, but it takes me 50x of the time. And the Keyence solution can be teached to a noob within 15 minutes.

newtonsapple
Posts: 11
Joined: Mon Jul 20, 2020 8:39 pm

Re: Fast Stacking and Continuous vs Flash Lighting

Post by newtonsapple »

rjlittlefield wrote:
Tue Jun 22, 2021 9:49 pm
Scarodactyl wrote:
Tue Jun 22, 2021 9:34 pm
high resolution pictures
That opens another question that I don't think has been asked.

newtonsapple wrote
This is for low volume industrial inspection documentation and I am trying to put together a very efficient workflow.
I wonder how much resolution is needed for this purpose. That matters a lot, because it affects the required NA, which affects the step size as 1/NA^2, using the quarter-lambda rule. So, if you keep everything optimized, and reduce the required resolution by 2X on axis, then you cut the image size by 2X on axis (4X by pixel count), and the number of frames by 4X. The acquisition time drops by 4X, the computational cost by 16X -- it's a big deal.

newtonsapple, I know that it's a diversion from your title, but have you worked through the resolution requirements yet?

--Rik
I have not worked through the resolution requirements yet. I can very likely drop resolution to speed things up. This will be as big of deal on the processing side as on the shooting side.

I think this thread is an example of what I am going for: viewtopic.php?f=27&t=43579&p=274128&hilit=burst#p274128

They were shooting 10fps, ISO 500, 1/200 at 7.5x and 10x. So this does seem possible. Edit - I just realized that this was Lothman who posted the other thread. Is there any motion blur at 1/200th in single frames? Are you shooting with a mechanical or electronic shutter?

newtonsapple
Posts: 11
Joined: Mon Jul 20, 2020 8:39 pm

Re: Fast Stacking and Continuous vs Flash Lighting

Post by newtonsapple »

lothman wrote:
Wed Jun 23, 2021 2:41 am
Scarodactyl wrote:
Tue Jun 22, 2021 9:34 pm
They have some spectacular features and I'm not surprised they're so popular, but I don't think they'd be a very efficient way to take high resolution pictures.
this is a newer Youtube clip of two guys looking at watches using a Keyence microscope.

https://www.youtube.com/watch?v=hKrJOMa ... sWatchGang

Since you can do automated stitching (even with HDR) you crank up the magnification, stich and have your high-Resolution image - ate least for technical reports. And this goes within some seconds up to a minute.

Of course with a high end optic on my Sony A7riv I can squezze out more, but it takes me 50x of the time. And the Keyence solution can be teached to a noob within 15 minutes.
Yeah, no question I would just go with a Keyence setup if I could justify the price. I can put together a nice stacking setup for less than $2k and do much better than shooting with a dinocam or camera on my stereo scope. That's sort of the point I am at. I am just looking to speed up the process by addressing both the shooting and processing aspects.

I am a little timid about a used Keyence setup too because they have rolled through so many variants over the years that I assume support isn't great for something 10 years old.

Lou Jost
Posts: 5944
Joined: Fri Sep 04, 2015 7:03 am
Location: Ecuador
Contact:

Re: Fast Stacking and Continuous vs Flash Lighting

Post by Lou Jost »

How deep are your stacks? You are in the magnification range where you do not need a rail; you can stack far faster without one, using a camera with a large buffer and automatic focus bracketing. A stack can take less than two minutes. I do this with MFT.

For m<=1, there is no limit to the depth of the subject; this method always works. For m >>1 the depth of the stack will be limited by the available focus throw of your tube lens. By far the most convenient tube lens for MFT is the Panasonic 45-175mm zoom because of its long focus throw, low price, and internal zoom and internal focus.

lothman
Posts: 958
Joined: Sat Feb 14, 2009 7:00 am
Location: Stuttgart/Germany

Re: Fast Stacking and Continuous vs Flash Lighting

Post by lothman »

newtonsapple wrote:
Wed Jun 23, 2021 8:52 am
They were shooting 10fps, ISO 500, 1/200 at 7.5x and 10x. So this does seem possible. Edit - I just realized that this was Lothman who posted the other thread. Is there any motion blur at 1/200th in single frames? Are you shooting with a mechanical or electronic shutter?
electronic shutter, I would say no blurr

newtonsapple
Posts: 11
Joined: Mon Jul 20, 2020 8:39 pm

Re: Fast Stacking and Continuous vs Flash Lighting

Post by newtonsapple »

Lou Jost wrote:
Wed Jun 23, 2021 1:35 pm
How deep are your stacks? You are in the magnification range where you do not need a rail; you can stack far faster without one, using a camera with a large buffer and automatic focus bracketing. A stack can take less than two minutes. I do this with MFT.

For m<=1, there is no limit to the depth of the subject; this method always works. For m >>1 the depth of the stack will be limited by the available focus throw of your tube lens. By far the most convenient tube lens for MFT is the Panasonic 45-175mm zoom because of its long focus throw, low price, and internal zoom and internal focus.
So at the extreme end of things, I think something like 5mm FOV with 3mm depth. One of the step calculators shows this at .021mm steps or 143 images @f2.8. Now playing around a bit, I can likely drop some resolution in a lot of instances and back off to a 10mm FOV and crop down which requires .050mm steps at 60 images. I need to digest the resolution tradeoffs between cropping (or shooting MFT or higher pixel density) and increasing the aperture value with regard to final resolution.

Lower mag (1-2x) will be limited to a depth of 10mm.

Post Reply Previous topicNext topic