DIY Scienscope fiberoptic light
Moderators: rjlittlefield, ChrisR, Chris S., Pau
- pharamousse
- Posts: 11
- Joined: Thu Sep 21, 2017 7:35 am
- Location: PARIS
- pharamousse
- Posts: 11
- Joined: Thu Sep 21, 2017 7:35 am
- Location: PARIS
I think for the yongnuo twins, I need 6v 3A.
30A ? Maybe i got to make some research.
This video on YouTube can be a beginning for basics in powering speedlights
https://youtu.be/I_5HbVyeQqE
30A ? Maybe i got to make some research.
This video on YouTube can be a beginning for basics in powering speedlights
https://youtu.be/I_5HbVyeQqE
Our heads are round so our thoughts can change direction.
Before had 5v 4a - it was slow for 2 or 3 Nikon SB-27 (depending on the speedlight power settings). Biggest problem is initial current jump. Power supplys on ebay are cheap, it is better to spend few $ more (only) than to be limited during the stacking process - now I can connect all 3 SB-27, ring speedlight at full power, nothing is screaming, heating, overloading, slowing etc. ...pharamousse wrote:I think for the yongnuo twins, I need 6v 3A.
30A ? Maybe i got to make some research.
This video on YouTube can be a beginning for basics in powering speedlights
https://youtu.be/I_5HbVyeQqE
Regarding video - it is little bit too brutal for me
I use dummy batteries, with power supply for studio and batteries outdoors. Before I used sealed acid 6v 7.4A battery (lasted forever), now, when 18650 became cheap - 2x 18650 in series with converter to 6v
Saul
μ-stuff
μ-stuff
As I mentioned before, had an idea to use two separate single FO guides with twee Cree XHP LEDS instead Y type and single LED . Benefit - double light output and possibility to use one or two guides without wasting half of the light output.
Old Y type on the left, new single - at the right
Single - at the left, Y type - at the right
Old Y type on the left, new single - at the right
Single - at the left, Y type - at the right
- iconoclastica
- Posts: 486
- Joined: Sat Jun 25, 2016 12:34 pm
- Location: Wageningen, Gelderland
This discussion rekindled in another thread:
The lamp was placed about 8 cm from the OF entry port and between them I had placed a concave convex lens scavanged from an old 50mm photographic lens.
In this setup, the brightest of the two goosenecks gave me 29500 lux. When I removed the lens, only 10700 lux remained. From the original lamp, a 43mm round piece of glass had remained in the light path. This may or may not be an IR filter, I wouldn't know how to find this out. But the LED lamp is not running hot at all, so I removed it too and was rewarded with an output of now 13200 lux.
Moving the lamp all the way to the FO port showed an output of 1400 to 60100 lux. In this position, the exact placement of the lamp became critical. The best result I got when I placed one of the LEDs exactly against the left half of the FO port.
I would have liked to double the light output thus, but I faild to fix the lamp into a position that yielded a better result than what I started with. So far the nett reult of my tries is negative.
Then I started playing with fresnel lens cards. One against the lens (in a non-optimized position about 10cm back from the FO-port) brought the ouput back in the 20,000s, a second one on front of it popped it up to double that, and a third one in between elevated it to 70,000 lux. Given the raw output of the lamp is a little less than 200,000 lux whiche is divided over two goose necks, this would be an efficiency of about 70% ?!
To do this well, I need to find a way to fix the fresnel cards in an arbitrary position along the z-axis. Don't know yet how to do that.
The focal distance of the cards is roughly 20cm.
I got back to my design that by parallel evolution came to a very similar result as the one discussed here: I removed the original lamp and the power supply, and replaced it by a motor cycle headlight (20W, 4 LEDs behind an acrylic disc with 4 lenses, one in front of each LED) and a surplus laptop power supply. The original potmeter I replaced by a switching frequency PWM LED dimmer.Saul wrote:You have to place the LED close to the FO entrance ( I suspect you used big size LED also - it should be smaller, with smaller angle of the emission).iconoclastica wrote:I have been experimenting with that. In a defunct cold light source with two goosenex, I placed a motor cycle headlight (LED, 20W) and a collector lens. It was not nearly bright enough. Used with an m-plan 60x I had to expose 8 sec and still had a dark picture. Through the eye glasses I could see hardly anything.ChrisR wrote:I query the choice of a gooseneck light source. I appreciate that it's something you have and it works, but designing from scratch, isn't an LED more attractive?
My experience with the Scienscope fiber optic light
No problem at 50x (do not own 60x)
The lamp was placed about 8 cm from the OF entry port and between them I had placed a concave convex lens scavanged from an old 50mm photographic lens.
In this setup, the brightest of the two goosenecks gave me 29500 lux. When I removed the lens, only 10700 lux remained. From the original lamp, a 43mm round piece of glass had remained in the light path. This may or may not be an IR filter, I wouldn't know how to find this out. But the LED lamp is not running hot at all, so I removed it too and was rewarded with an output of now 13200 lux.
Moving the lamp all the way to the FO port showed an output of 1400 to 60100 lux. In this position, the exact placement of the lamp became critical. The best result I got when I placed one of the LEDs exactly against the left half of the FO port.
I would have liked to double the light output thus, but I faild to fix the lamp into a position that yielded a better result than what I started with. So far the nett reult of my tries is negative.
Then I started playing with fresnel lens cards. One against the lens (in a non-optimized position about 10cm back from the FO-port) brought the ouput back in the 20,000s, a second one on front of it popped it up to double that, and a third one in between elevated it to 70,000 lux. Given the raw output of the lamp is a little less than 200,000 lux whiche is divided over two goose necks, this would be an efficiency of about 70% ?!
To do this well, I need to find a way to fix the fresnel cards in an arbitrary position along the z-axis. Don't know yet how to do that.
The focal distance of the cards is roughly 20cm.
--- felix filicis ---
Very useful/interesting test !
https://www.ebay.com/itm/2PCS-PMMA-Plas ... DGDUxDgC-g
If I understood correctly, it is useful for the uniform illumination:
https://www.nikon.com/products/microsco ... lv100npol/
https://www.nikoninstruments.com/Produc ... clipse-Ti2
Because I wanted to keep my design very compact, I contacted manufacturer, asking them do they have smaller diameter lenses . They replied, that they can cut from existing ones for the same price (it means 4pcs from the 70mm diameter).
Other hand, 2x flye eye fresnel lens setup for the uniform light is still on my list ( for my Optiphot modular light source ), still waiting for the parts/materials to arrive ...
Question for Nikon Eclipse owners - any idea what is the size of the single lens (in the Nikon fly eye fresnel lens) ?
BTW, how many W your LEDs are ? Make sure that your wires are thick enough . And your power supply can supply enough current for your setup.
How you are measuring it ?In this setup, the brightest of the two goosenecks gave me 29500 lux.
In some post I mentioned about fly eye type fresnel lens:Then I started playing with fresnel lens cards.
https://www.ebay.com/itm/2PCS-PMMA-Plas ... DGDUxDgC-g
If I understood correctly, it is useful for the uniform illumination:
https://www.nikon.com/products/microsco ... lv100npol/
https://www.nikoninstruments.com/Produc ... clipse-Ti2
Because I wanted to keep my design very compact, I contacted manufacturer, asking them do they have smaller diameter lenses . They replied, that they can cut from existing ones for the same price (it means 4pcs from the 70mm diameter).
Are these fresnel lenses credit card size ? I have couple of them, but, after playing with condenser & different lenses (and after reading Pau's and other member's tests) decided do not go this route. Now I'm pretty satisfied with my "quadruple" setup, very rarely I have to use 100% output....The focal distance of the cards is roughly 20cm..
Other hand, 2x flye eye fresnel lens setup for the uniform light is still on my list ( for my Optiphot modular light source ), still waiting for the parts/materials to arrive ...
U-shaped slide-in frame with set screws, sliding in the small 8020 aluminum type extrusion along Z-axis ?To do this well, I need to find a way to fix the fresnel cards in an arbitrary position along the z-axis. Don't know yet how to do that.
Question for Nikon Eclipse owners - any idea what is the size of the single lens (in the Nikon fly eye fresnel lens) ?
BTW, how many W your LEDs are ? Make sure that your wires are thick enough . And your power supply can supply enough current for your setup.
Last edited by Saul on Tue Nov 27, 2018 8:38 pm, edited 1 time in total.
Any idea what is the size of the single cell lens ("ommatidia") ?houstontx wrote:If I remember correctly on a nikon 90i the "fly eye" lens is about 1.5 to 2 inches in diameter placed in the light path inside the unit, directly after the lamp house mounting. I sold the 90i unit to aidanmoore a member on the forum. Maybe he could help.
- iconoclastica
- Posts: 486
- Joined: Sat Jun 25, 2016 12:34 pm
- Location: Wageningen, Gelderland
Several years ago I bought a couple of credit card size fresnel lenses from a cheap source at ebay, after I read that someone had used them successfully to flash monkeys high up in the tree. I have never used them for that purpose, nor pbotagraphed any monkey in a tree. They just sat at the bottom of my drawer until yesterday.
The light meter I used is a lux-meter, with the sensor hidden behind a white done, like a photographic incident light meter. Not the best thing for tiny light sources, but with the end cap of the goose neck pressed against the dome it turned out pretty consistent.
The lamp is 20W, the power supply 30W, the cable is the one that originally connected the Ps to its laptop. Also the timer is well overrated. (the first thing I learned is that you can't simply throttle this with a potmeter )
The light meter I used is a lux-meter, with the sensor hidden behind a white done, like a photographic incident light meter. Not the best thing for tiny light sources, but with the end cap of the goose neck pressed against the dome it turned out pretty consistent.
The lamp is 20W, the power supply 30W, the cable is the one that originally connected the Ps to its laptop. Also the timer is well overrated. (the first thing I learned is that you can't simply throttle this with a potmeter )
--- felix filicis ---
Haha, already done thathoustontx wrote:The lens looked just like this. It even had the solid edges like this one does. Maybe you can make a guess based off this image assuming 1.5 or 2 inches from point to point or side to side? I'll let you count the points
If you remember - can it be approximately 2mm ? More ? Less ? I'm not looking for the exact number ....
-
- Posts: 104
- Joined: Thu Sep 24, 2015 8:09 am
- Location: Toronto