I converted my NEX-5, how to test it?

[bostwickenator]

Hi everyone,

 

I recently decided I would convert a NEX 5 to full spectrum. I pulled the camera apart and replaced the sensor glass with a slice of a fused quartz UV microscope slide. I'm now coming across a problem I didn't expect. How can I verify I'm actually getting UV sensitivity? I maybe foolishly expected this to be completely obvious.

 

I'm clearly getting plenty of IR sensitivity as you would expect

 

But for UV this is a pinhole test with a UV bandpass filter from iosoptics.com. I painted two stripes across my right arm with sunscreen as a potential target and you can see the faintest hint of them. But this is not the level of contrast I was expecting, it appears the majority of the sensitivity is IR as the shorts I'm wearing in the photo are dark green to the human eye.

 

This is a shot using the same filter on a Sony SEL55Z18, same sunscreen smeared on my hand. This is a pretty complex modern lens with Zeiss T* coating so I wouldn't expect any UV pass.

 

This is manufacturer specified bandpass for the filter. So the manufacturer is claiming a 160x delta between the transmission of UV in the 300-400nm band and the VIS and IR spectrum. Does that sound at all reasonable? This doesn't seem to be the case or my quartz is cutting most of my UV.

 

I'd appreciate any ideas on how to address debugging my setup, I have no idea what I'm seeing, sorry there are so many unknowns.

[Andy Perrin]

This looks exactly like what I would expect from the given graph (which is on a linear scale). The problem is that you are forgetting the rapid fall-off in sensitivity to UV of the sensor, and the relatively large leaks of that filter. 160x may sound like a lot, but it actually isn't -- typically we look for attenuation by a factor of OD4 or so, which is 10,000!

 

When you buy a filter, you want to find a logarithmic graph of the absorption, or else one of Schott's "diabatic" transmission graphs which are kind of a mix of logarithmic and linear scales. That is the only way to estimate the true size of a leak.

 

Stacking that with some S8612 2mm will fix the IR leak. I am less certain about the visible leak near 550nm.

[dabateman]

Great job on the conversion. With tight fitting electrical components that can be nail bitting to avoid ripping a tiny ribbon cable.

 

Yep filters and lenses will be your next rabbit hole to fall into. Cameras are really IR sensitive, just as Andy said. Since you have one filter, the 2mm S8612 will most likely be the best purchase. Then you will probably get a U360 or U340.

Or just jump to a Baader venus U filter, but really expensive.

 

Since you have a NEX, see if you can get a Sigma 30mm f2.8 art lens with smooth focus ring for under $100, they work well.

 

Since you saved about $300 on the camera conversion, you can spend that on a lens and some filters right?

[colinbm]

Since you have a NEX, see if you can get a Sigma 30mm f2.8 art lens with smooth focus ring for under $100, they work well.

These smooth focus ring Sigma 30mm lenses, some are black & some are silver ?

Are the optics the same for the Sony mount & the 4/3rds mounts ?

Col

[dabateman]

 

These smooth focus ring Sigma 30mm lenses, some are black & some are silver ?

Are the optics the same for the Sony mount & the 4/3rds mounts ?

Col

 

As far as I know. They should be the same. Only the mount is different. Sigma originally designed it for the Nex and then released them for m43rds. That why the trio has odd m43rds focal lengths, 19mm, 30mm and 60mm. The 19mm and 60mm where also reported here to be good for UV, but up to 1 stop slower in UV.

The Sigma 30mm is actually quite a nice sharp UV lens.

[colinbm]

Thanks Dave

"The Sigma 30mm is actually quite a nice sharp UV lens." Good

I have just got a silver one m4/3, it has a loose rattle.....I am told this is normal...?

[dabateman]

Thanks Dave

"The Sigma 30mm is actually quite a nice sharp UV lens." Good

I have just got a silver one m4/3, it has a loose rattle.....I am told this is normal...?

 

Yes I think that is normal. But I rarely shake my lenses.

 

Do you have a m43rds camera now? I have seen used Olympus Em1mk1 for $230 and was almost tempted to buy a second. The stock camera without conversion is great for run around UV photography. Just need to buy the Hld7 grip as the dials can fail. Also being the last with tilt inaxis touch screen, might be it for me. I have come to hate the flippy screen these days. I have that on my original E3 and Em5mk2.

Only sacrifice I see on the Em1mk1 is that 1/8 electronic shutter limit. Its got the same EVF and back screen pannel as all following Olympus cameras, which is surprising. I haven't used high resolution shot mode, enough to place that as a need. But I have photoacute which will do that with Em1 files, so no loss. Just have to shot at least 5 images.

[colinbm]

Thanks Dave

You don't need to shake the lens, just turning it front to back, you can feel the internal weight dropping forwards & back.

I have had a Panasonic Lumix DMC G3 for a number of years, that I converted & replaced the hot mirror with fused silica.

It ha worked OK & looks promising with the Sigma 30mm.

I just never in the past had any good UV capable lenses.

[Andrea B.]

Bostwickenator,

 

(1) The ideal and easily found test subject for UV is a flower such as Sunflower (Helianthus species), Black-eyed Susan (Rudbeckia species) or Dandelion (Taraxacum) all of which have a UV-dark central bullseye and UV-bright rays. If your gear is working well for reflected UV photography, you will easily see those dark bullseyes in the photo.

 

(2) To best bring out the UV-dark and UV-light contrasts, you might want to white balance your photo by a white-click on the magenta area or by setting an in-camera white balance against an asphalt road. (Later you can learn more about white balancing in UV work.)

 

(3) Your chosen filter definitely has IR leakage and might have a bit of visible green leakage. But not to worry, just stack an S8612 filter (as suggested above) over it to block the IR.

 

(4) Your filter peaks around 360 nm. Your lens probably doesn't pass UV much beyond 380 nm. You might want to look for one of the classic old 35/3.5 lenses we talk and write about so much here. See the Lens Sticky in the reference section.

 

Welcome to UVP !!! Hope you enjoy your UV adventures.

Andrea B.

[enricosavazzi]

Thanks Dave

You don't need to shake the lens, just turning it front to back, you can feel the internal weight dropping forwards & back.

I have had a Panasonic Lumix DMC G3 for a number of years, that I converted & replaced the hot mirror with fused silica.

It ha worked OK & looks promising with the Sigma 30mm.

I just never in the past had any good UV capable lenses.

The rattle is normal with the Sigma 30 mm f/2.8 (both the old type with sculptured focus ring and the new one with smooth focus ring). The Sigma 19 mm f/2.8 (both versions) also rattles. Both the old and new versions of these lenses are roughly equally good in UV, but the 30 mm is a little better than the 19 mm.

 

The good thing with these lenses is that they are cheap and they are among the few UV lenses capable of autofocus, but they don't go very deep into the UV. If you prefer (or can live with) manual focus, there are better UV lenses available, like the several 35 mm f/3.5 manual focus lenses.

[Andrea B.]

Added to my preceding post.

I white-balanced your hand foto in Photo Ninja. The result shows the IR and green contamination but also shows that some UV is getting through. Remember that there is lots more IR and Visible light in sunlight, so even tiny leaks can contaminate a UV photo.

(The white-balance ideally is done only in the raw file for best results.)

 

[colinbm]

Thanks Enrico for the confirmation of the Sigma m4/3 lens rattle.

I have some legacy 35 mm f/3.5 manual focus lenses to try out.

[bostwickenator]

Thanks for the valuable input everyone!

 

This looks exactly like what I would expect from the given graph (which is on a linear scale)

I kind of can't believe that I forgot we are talking about logarithmic responses here. As soon as I read that first sentence I realized my error. I was optimistic I could go a cheaper route as I hadn't seen this manufacturer talked about here before. I really should have spent a little more time studying the filter stack lists. Time to get back on eBay

 

You might want to look for one of the classic old 35/3.5 lenses we talk and write about so much here. See the Lens Sticky in the reference section.

I've already got a Kaligar 35/3.5 in the mail based on one of your reviews. Thanks for taking the time to do those write-ups

[Andrea B.]

I'm very happy to hear it is helpful.

[bostwickenator]

Good news. Double stacking that same filter glass generates great results, luckily I had two of them due to an ordering error. I 3d printed a little filter holder stuck them on the front of the Kaligar 35 and got this. Total cost of the setup was: $80 for the camera, $10 for the fused silica coverglass, $35 for the lens, and $50 for the two filters.

 

[Andrea B.]

That is a very nice looking UV gear set you have put together for such a great low price! I think you'll have some fun with it for sure. BTW, I think it is *so cool* that you and other members are 3D printing stuff for your lenses and cameras.

 

Try the sunscreen streak on the hand again to determine whether it now looks UV-black. Double stacking an IR-leaky filter can help, but some IR will/might still get through. So what we are hoping here is that it is only a little bit which won't lighten UV-dark areas too much. You can always order an Uviroptics S8612 on Ebay to stack on the UV-pass. It won't be too much more $$.

[Stefano]

Double stacking an IR-leaky filter can help, but some IR will/might still get through.

Yes, but this happens if your single filters have at least a 1% leak (assuming they are equal). 1% is OD 2, and double stacking will bring it down to OD 4, which may still show up (I personally think that "safety" is above OD 4.5, even if I actually have 0 experience in various OD filters).

 

In his case, the green leak (the worst one) seems to be at ~0.2-0.3% or even less, so he should be above OD 5. Still, a sunscreen or flower test would make sure his filter stack is leak-proof.

[bostwickenator]

Sunscreen test on a fingertip. Apologies for the blur it's pretty hard to hand hold this camera with 4 second exposure times