Jump to content
UltravioletPhotography

Mystery IR filter - transmission spectra and images


JMC

Recommended Posts

I heard back from Hoya - they just sent the PDF of the R72 glass transmission spectra which doesn't show the dip. No help what-so-ever.

 

I've bought a new R72 from SRS here in the UK. It arrived today. I've measured it and it behaves like my other one, i.e. like a bandpass, not like a longpass. I'm going to send it to Ulf to measure, so we have data on the same filter.

Link to comment

Ulf can check it, but I am confident enough in both of you, Jonathan and Ulf, that I think he will find the same results. So if you buy a new filter labeled R72, it is (probably) a bandpass.

 

The next question I would like to know is what Cadmium's result will be when he buys "R72" glass directly from Hoya.

Link to comment

The next step is for someone to build or buy an ICP-MS instrument. Then just sand off a gram around the edges of the glass and tell us the true elemental composition. Then we will know the real make up of the real glass formula.

 

There not too hard to build if you have a mass spectrometer, just need a lot of argon.

 

I think yes they are cheating. But does it matter? I would like to see a normal camera image with both real Hoya long pass and Tokina fake bandpass IR filter. Except for Andy's triwave, I bet they will look the same. That is most likely why they are cheating us.

 

Well, if it was that easy. Just think of Brown and Green Glass. It might have the same composition, but the Fe has different oxidation states.So it is a Bit more complicated.

Link to comment

The next step is for someone to build or buy an ICP-MS instrument. Then just sand off a gram around the edges of the glass and tell us the true elemental composition. Then we will know the real make up of the real glass formula.

 

There not too hard to build if you have a mass spectrometer, just need a lot of argon.

 

I think yes they are cheating. But does it matter? I would like to see a normal camera image with both real Hoya long pass and Tokina fake bandpass IR filter. Except for Andy's triwave, I bet they will look the same. That is most likely why they are cheating us.

 

It most defiantly matters.

I know this goes on with eBay sellers, and camera converters, and in my opinion more sellers and converters than you think are using the cheapest glass they can find that works for what they are doing.

And now a well known and accepted standard of IR photography is being made out of some glass that is not what is advertised.

Yes, that matters.

Do you want someone to send you a blue green filter made in China instead of a real S8612 ? Or whatever other filter... you want people to lie to you about what you are really getting?

That matters, and it matters a lot more for some things than it does for most things we do.

Does it matter in IR photos, probably you can not see any difference, but does that mean you want people to lie to you, label something as real Hoya R72, when it isn't even a longpass filter?

It matters.

 

How long has this been going on with Hoya R72 filters? Are all the new R72 filters being made by Kenko, and being sold to and by other photography equiptment sellers, such as B&H, Adorama, and the rest...

are all those new R72 filters really being made out of the non-Hoya-R72 glass?

That doesn't matter to you?

Amazing the things that matter to people, and the things that don't.

Link to comment

Ok I take it back. I do care if I am lied to and ripped off with unknown glass.

I have just been reading so much BS and lie lately in work, that I have maybe become complacent to it.

 

Basically don't trust anything or anyone. Test it yourself and request your money back if its not as desired.

 

Alaun, it will be that "easy" ICP (inductivly coupled plasma) with mass spectrometry determines the exact elemental composition. If its FeO2, FeO3, you will see more oxygen in one than the other.

Link to comment
Ok, that is fine to know, but then some oxides might be relevant at a ppb level, ICP might be able or not to distinguish. I also had in mind, that a heat treatment might change the optical properties. The next step, knowing the composition would be to actually melt this composition. You need to know, how to mix it, what type of crucible works (even using platin is not good enough for some types of UV glass) and so on. It is a lot more, you need to know. But I guess, you know this : )
Link to comment

Ok, that is fine to know, but then some oxides might be relevant at a ppb level, ICP might be able or not to distinguish. I also had in mind, that a heat treatment might change the optical properties. The next step, knowing the composition would be to actually melt this composition. You need to know, how to mix it, what type of crucible works (even using platin is not good enough for some types of UV glass) and so on. It is a lot more, you need to know. But I guess, you know this : )

 

 

Thats why I said "easy" above.

Most likely there is an analytical chem SOP to do this in the actual glass manufacturing location, that may even be more cost effective.

We just taken this out to 5 pages now, so thought we may need a deeper analysis of the glass.

Link to comment
Would this issue with new vs old explain why I can't get fast shutter speeds with the Hoya R72 filter on both my full spectrum converted cameras? I have two Hoya r72 that look more like the new batch (They have the long line in middle of H and it's written 'Made in Japan' instead of the old 'Japan' on the filter. I bought them from the only and reputable distributor in Australia.
Link to comment
Arti currently we don’t know what’s new or old or anything really. We have no way to tell them apart. There is no “new batch” that we know how to delineate by any method except a spectrometer.
Link to comment

I rather doubt the type of glass in the R72 filters has much to do with the exact graphics on the filter ring.

They may change the graphics, and that may not really have anything to do with the filter being 'counterfeit', also the product photos may not be up to date on various seller sites.

I am not saying there are not also counterfeit camera filters, I am only saying that graphics can change.

I doubt there is any way to know what glass is in any R72 camera filter without using a spectrometer.

However, I trust Hoya, I trust their glass and data sheets, and I fully expect a spectrometer scan of a sheet of R72 from Hoya to to be exactly like what they show on their R72 data sheet.

Keep in mind that these R72 camera filters are not being made by Hoya, they are being made by Kenko, or they may even be made for Kenko by a third party,

Hoya has nothing to do with the R72 camera filters being made now, and maybe never did.

If you want real Hoya R72 glass, your going to need to get the glass directly from Hoya, or some Hoya glass distributor, and make the camera filter your self,

or you will have to get a spectrometer and scan some R72 camera filters until you find one that has the correct and true Hoya R72 glass in it.

 

I don't think you will be able to see any difference in photos between a fake Hoya R72 with bandpass type glass and a real R72 filter with genuine Hoya R72 longpass glass in them.

So far we have two "Hoya R72" camera filters that have been tested by Jonathan, and as far as I know he has no Hoya R72 camera filter that has tested to be real longpass Hoya R72 filter glass,

so no one has one of each, verified real and fake, to compare photographically.

Until we have a fake and a real version to compare photographically we can not say what possible difference there may be in actual photos.

Link to comment

My real problem is this. I bought a canon 6d that a guy removed the hot filter himself and I know that the camera behaves like a full spectrum camera. When I put the Hoya r72 I can't get fast exposures using even 35mm f1.4 fast glass in the middle of the day. Many people claim to get 1/200 f8 and more at iso 100 with the same filter. The difference betwen the full spectrum and normal camera during the day is about 5 stops maximum faster. Would love if anyone give me an idea why is this possible?

Here is an example of my problem. Same camera canon 6d ,(full spectrum above normal spectrum below) custom white balance in the grass, same lens same filter.

https://www.dropbox.com/s/slqs7yu6wm3arzf/70711653_784122262006590_3169542451195019264_n.jpg?dl=0

Link to comment

That is hard for me to say. Probably the easiest way to test that is to get some other filter and compare any difference in exposure time.

35mm f/1.4?

Anyway, try some other longpass IR filter, in the same general range and see if there is much difference.

Link to comment
It might be a bit hard to duplicate, given I have a different camera, lens, etc... but if I can test one of my R72 filters in some way, settings and similar scene, then let me know.
Link to comment

Arti,

Move over to the "UV/IR gear talk" forum and post a new topic with exposure settings and some photos.

Best to in clude same image with UV/IR block filter, no filter, then IR filters.

What may have been used to correct the infinity focus may be the problem. Cheap UV blocking glass on sensor could be an issue.

Link to comment

Update and new information (also updated the first post).

 

I've been working to get to the bottom of this with Ulf, and I am ready to share what I think is happening (Ulf will also add his perspective on this too).

 

I bought a new R72 filter, measured it and sent it to Ulf - and we got different results. My scans showed the bandpass behaviour and Ulf's didn't - they showed the expected longpass profile.

 

So I tested a few different filters on my system - B+W091, B+W092, Heliopan 715, Heliopan 780 and Heliopan 830. In theory all these should be longpass. This is what I got;

post-148-0-04938400-1569686569.jpg

 

Very strange. The B+W091 behaved as expected, but the 092 started to drop above 830nm. Very odd. The Heliopan 715 dropped more and sooner at the long wavelength end, and the effect was even more severe for the Heliopan 780 and 830. Although not shown here, tests with things like a UV lens protecting filter, and a Tiffen Yellow, showed no drop at all at the long wavelength end.

 

Even though the graphs for these IR filters look smooth all the way up to 880nm, it isn't real. The filters aren't bandpass, they are longpass, and this is artifact of how they have been measured. Ulf has some ideas about what specifically is happening and I'll leave the detail to him.

 

It just goes to show that getting data is one thing, getting reliable data is another thing entirely, and knowing the limits of your equipment is the most important thing.

 

Needless to say, I now no-longer believe any of my own data above 800nm, and my graphs on here must be viewed with that in context. There is no evidence from my measurements that the Hoya R72 or my mystery IR filter (whatever it may be) are bandpass filters.

Link to comment
Wow, so it was all a mirage! Well that makes me feel better about my R72. But now I would like to know why your equipment is malfunctioning sometimes but not others.
Link to comment

Cadmium,

I wouldn't edit anything. Our opinions are thoughts based on what we know. Not facts. Facts are real elements that seem to be disappearing over time. Becoming over run by opinion and common sense.

 

Jonathan, interesting that your system seems to selectively break down after 800nm. Well 200nm to 800nm is still a decent range. It is funny how 800nm to 1100nm does seem to be important to us UV photographers. But maybe we focus too much on the leak. Eka has shown that the UV+ images are quite nice.

To be selective I wound if its a counting problem. Not absorption limiting the signal, but the counting of the light is off time somehow. Does the apparent absorption go away if you conduct more scans? More counts at higher wavelengths?

 

Link to comment

I won't change anything. Jonathon warned us... my fault. :-)

Jonathan, what is different between your spectrometer and Ulf's? I thought you both had the same brand and model, no?

I still don't see why some of the longpass filters show as longpass and others don't. Makes no sense.

Link to comment

Actually I also saw a similar, but not identical phenomenon at longer wavelengths with some initial configurations of my spectrometer setup.

I'll return later in a post showing my experiments and describing my conclusions from them.

 

The invaluable help from Jonathan running a differently configured spectrometer with a different measurement software on several filter types made it possible to understand this problem better.

 

When measuring with our kind of spectrometers we are doing experiments with optical bench setups with unknown limitations.

The measurements results we have presented before are disturbed unexpectedly much by internal crosstalk in the spectrometers in the NIR end of their range.

At first both Jonathan and I was fooled by the polished output that Jonathan gets from his Ocean Optics measuring software that guid the user through the measurement step by step and finally delivers a nice graph.

 

I'm now sure that we understand the limitations causing this type of measurement error and also how to bypass it.

We have learned that we cannot trust wide-band transmission measurements above some wavelength.

In Jonathans case around 800nm and in my setup 900-925nm.

With reconfigurations of the measurement setup I'm convinced that both Jonathan and I can reach higher.

Link to comment

Please sign in to comment

You will be able to leave a comment after signing in



Sign In Now
×
×
  • Create New...