Testing Hoya U-340 and U-360 for visible leaks - funding request

[JMC]

All,

 

Hoya U filters are a bit of an unknown quantity when it comes to visible light transmission. As has been mentioned before, the data says 0% transmission for between 407nm and 671nm - https://www.ultravio...dpost__p__38043 - but this seems to be an artificial number rather than an actual measured one.

 

This makes it difficult to compare them with Schott filters.

 

The answer to this is to get some very thin versions of the filters and measure the transmission. The problem is nobody sells them as very thin filters, at least not as stock items.

 

One of my optics suppliers in the UK can do them as a custom job for me (0.5mm thickness and 25mm diameter), however it is not cheap as making such thin filters is very difficult. The price for them would be £125 each plus VAT and delivery. While I am happy to run the spectra for these, I cannot justify buying them just for the test.

 

So, my ask to the forum. Would people be willing to donate £20 each to help fund this experiment and buy the filters? If I can get 10+ people to agree to help fund this, I'm happy to go ahead, absorb the rest of the costs myself and do the measurements.

 

Some of you have already expressed an interest in this and thanks for that, and I appreciate times are especially hard at the moment, but let me know if you can help.

 

I shall let this request run for a couple of weeks, (until August 21st), and if we have enough people committed to it, I'll go ahead and and place to order and let you know my paypal details for the donation.

 

EDIT - just to add, I think it'll be fair to share excel files of the data with those who have paid towards the work. I will of course put graphs up in the forum for everyone to see.

 

If you are happy to help please leave a comment to that effect in this thread.

 

Best wishes, Jonathan

 

 

 

EDIT - People who have expressed an interest (to be updated as needed)

Ulf

Stefano

Colin

Birna

Andrea

Andy P

Anonymous

Dave B (20USD)

SteveE

Steve (Cadmium)

[colinbm]

Jonathan, I will be happy to help out.

One question though, why do the filters need to be sooo thin ?

Col

[JMC]

Jonathan, I will be happy to help out.

One question though, why do the filters need to be sooo thin ?

Col

Thanks Colin. The issue is that the transmission present in the visible is extremely low. Hence going as thin as possible gives the best chance of accurately measuring them.

 

Say for instance that the transmission at 500nm is 0.01% when a filter is 2mm thick. At 1mm this would be 1%, and at 0.5mm it would be 10%, which is a lot easier to measure.

[ulf]

Col,

 

I think Jonathan wrote a more understandable text at the same time as I wrote mine.

 

The thinner the filter is the more light will pass in wavelength regions blocking much, decreasing the difference between the baseline measurement with the light source only and the measurement with the filter included.

Spectrometers have a limited dynamic range, making it difficult or impossible to measure thicker glass with a high OD.

The reference data from Hoya is based on 2.5mm thick glass.

 

 

One important point though when you want to translate the data back to a thicker reference thickness is to be able to measure the the thin filter's thickness with a good precision

[colinbm]

Thanks Johathan, that now makes sense.

I realise there are some people on this forum that are commercial operators, that will benefit greatly from these tests that you will be doing & at a great expense & you appear that you will publish here, & perhaps even your own supplier, have you approached them for major sponsorship ?

Col

[colinbm]

Thanks too Dave.

[JMC]

Thanks Johathan, that now makes sense.

I realise there are some people on this forum that are commercial operators, that will benefit greatly from these tests that you will be doing & at a great expense & you appear that you will publish here, & perhaps even your own supplier, have you approached them for major sponsorship ?

Col

Hi Colin, I think the information will be useful to anyone using these filters, which is why I am comfortable asking the question to everyone on the forum about helping with the costs. I know some of the filter manufacturers on the forum, and if they are able to help out that is great. If not, so be it. As for my supplier in the UK, I have asked, but I can pretty much guarantee that they wont be interested in helping fund it.

[colinbm]

Sure Jonathan, I agree, but just had to say it....hey my money is on the table.

[nfoto]

I'll pitch in. No problem.

[Andrea B.]

Jonathan, please be very transparent about this fund raising on the pages of UVP. Thank you!

We would want to know who are the contributors and who is the maker of the filters.

And of course, some photos of the filters along with a good write-up of your measurement experiment.

 

Also, how do we get the contribution to you? Via Paypal?

 

I will kick in too.

[Andrea B.]

I should add ---- I am curious about how much a small non-violet, non-red visible leak affects a reflected UV photograph? Does it result in any "washout" of UV absorbing areas?

 

For example, we already know that small amounts of visible violet leak really do not affect a UV photo much. There have been several comparison photos on the pages of UVP which support this. And we have seen that even with UV+blue+green filters, the usual UV-signature can be seen although with some alteration of intensity. On the other end, however, an IR leak (or a high red leak) has always resulted in a wash out of UV absorbing areas because -- I think? -- the IR in those areas is almost always reflected.

[JMC]

Thank you Birna, Andrea, I've added you names to the original post.

 

Andrea, I think I've covered your concerns in the 1st post. And yes, the plan is to write up and share the results. As for what the effects would be, that I don't know. My aim here was to try and provide a route to answering the old question about the lack of data in the visible region for the Hoya filters.

 

We're doing well with commitments, but I need at least another 5 really to make this happen.

[Andy Perrin]

You can add me also!

[Stefano]

I should add ---- I am curious about how much a small non-violet, non-red visible leak affects a reflected UV photograph? Does it result in any "washout" of UV absorbing areas?

 

For example, we already know that small amounts of visible violet leak really do not affect a UV photo much. There have been several comparison photos on the pages of UVP which support this. And we have seen that even with UV+blue+green filters, the usual UV-signature can be seen although with some alteration of intensity. On the other end, however, an IR leak (or a high red leak) has always resulted in a wash out of UV absorbing areas because -- I think? -- the IR in those areas is almost always reflected.

Probably nectar guides are still visible in deep violet, as they usually appear orangish on yellow flowers (for example). Infrared just removes everything, as you said. Everything appears white (or light-colored). The same for red, most flowers (red, yellow, white, maybe purple if they are red+blue) reflect all red light. I wonder if the green leak of thin U-340 can be visible in images, and how it contaminates them. I am interested in Jonathan's measurements, especially about that green leak (but also about the transmissions in general). Maybe I will finally get an answer to this: https://www.ultravioletphotography.com/content/index.php/topic/3849-how-much-green-does-hoya-u-340-leak

[Cadmium]

Your not going to get any visible range leaks with U-360 at any practical thickness for use.

U-360 2mm is totally solid, no visual transmission.

U-360 1mm will add a small amount 400nm+ visible violet, but the rest of the visual range is suppressed.

U-340 however has a little bit of upper visual range leak starting in the mid 500nm range, but that is only a factor when using thin versions like 1mm, even at 1.5mm it will be suppressed, and at 2mm it is totally solid.

 

If you are going to do this kind of test, then you would be best to compare directly with Schott UG1 and UG11 using the same thickness.

That is the best way to compare any differences. Otherwise you are comparing your Hoya scans to Schott data. Would be best to compare scans of both with each other,

just as an added check and balance.

[JMC]

Steve, I'd love to test Schott UG filter as well, but the cost for this is going to escalate rapidly if I do that. This at least fills a gaps in our knowledge, providing some data where there is currently none. If in the future there is more a push for more testing, then the UG filters are an option.

 

Thanks Andy, name added.

[Andy Perrin]

I don’t see why Schott would be necessary. We have good data for Schott and we can rescale it to any thickness for comparison purposes. (Unless you think Schott’s data is wrong for some reason?)

[JCDowdy]

I can measure thicker filters if you want to compare.

[Cadmium]

Schott UG11 data is 1E-05 and has 1mm reverence thickness.

 

Here is a graph comparing:

1) Schott UG11 0.5mm thick plot (1mm reference thickness).

2) manually entered Hoya U-340 0.5mm thick plot (2.5mm data sheet reference thickness),

3) and pasted from Hoya program output U340 0.5mm thick plot (also 2.5mm reference thickness). Pardon the 0.00 drop outs using the Hoya program output,

just an example of how the Hoya program doesn't output as good of data as one might expect, direct manual entry from data sheet is better for U-340.

 

[dabateman]

Jonathan,

Don't you have 1mm UG11 at least to compare with? Not ideal, as not the same thickness but might help.

What will happen to the filters when done with the measurements?

How would final ownership look like?

 

I wonder if John Dowdy could be convinced to run them as well. He has a very sensitive, optimized spectrometer for this type of thing.

 

I may be willing to help out as long as it doesn't get messy. I might be able to just through in $20 US, not expecting anything but the spreadsheet.

[JMC]

David,

 

I hadn't gotten as far as determining final ownership. I thought it would be prudent to send them to Ulf to measure after I had finished with them (given his setup), but what happens next, that I hadn't thought about.

 

The whole idea behind this was to try and answer a question which seems to crop up on the forum - the issue of actual data for the Hoya UV filters in the visible part of the spectrum. My filter supplier can make them, but the cost is high, too high for me to try an absorb myself, hence asking for donations to help answer that question.

 

As it stands at the moment I have received 'pledges' of support mounting to 200GBP, which when I factor in VAT and delivery for them, leaves me to cover about 120GBP. To answer the question about the filters, I can justify that, but not over 300GBP. However what then - I cannot really cut the filters up and send everyone a piece? The filters themselves wont be much use for UV imaging given they are so thin.

 

As I have said above, if people are able to help that's great, if not then its not a problem. Happy to put you on the list if you're ok with that (for 20USD) it all helps :)

[colinbm]

Another problem with filters, that I think I have read here, is that they are far from an exact science.

The differences from batch to batch can be tolerated by the manufactures if the transmittance curves only meet a few critical points.

[Andy Perrin]

colinbm - I think that's more of a problem with the lower cost filters? That is, it's probably true that they only check a few points for Schott/Hoya also, but their control is much superior so there seems to be less variation. You have only to look at the results on our board — by and large, when two of us take the same photo with the same nominal filter glass and thickness, we see the same things, even though it's highly doubtful that all our filters came from the same batch. The curves that we work with from the companies also have the same issue (it is noted on the data sheets) but we use them successfully.

[dabateman]

Andy,

Remember this post:

https://www.ultravioletphotography.com/content/index.php/topic/3784-some-unusual-ug-filters-uk-supplier-h-v-skan/page__view__findpost__p__34575

 

I wonder if Jonathan might have a 0.5mm ug11 or ug5. I am atill unsure.

[Cadmium]

I think I have some 1mm of all 4 types, but nothing thinner ever.

Setting the UG1 and U-360 aside for the moment, the fact is that at 1mm is where the visual range suppression with both U-340 and UG11 start to become a little weak.

You don't see the visual ~500~ range leak with UG11 like you can with U-340, and even with U-340 it is a little hard to detect, but UG11 is not far behind, and it can be detected with UG11 if you really try hard.

Either one will work, but I would not use U-340 at anything less than 1.5mm to play it safe, and 2mm works great.

I have not noticed any visual range leaks from UG1 or U-360 even at 1mm thick.

Those can be used as 1mm for slightly extended exposure + slightly added 400nm+ violet, and also for dualband IR instead of the usual 2mm thick versions.

Also, UG1 has a much better tolerance to surface oxidation, and I have not seen such problems with U-360 either.

I prefer U-360 to all other U glass for UV pass stacks, personally, although if you want to cut off lower than 400nm, then UG11 or U-340 is the option.