2.94
#...
·
·
1
like
|
|---|
| If you took the image to Aladin you should find out that it is an IR star, possibly a Mira star or a long period variable star (LPV) 😉 |
3.01
#...
·
·
1
like
|
|---|
If you took the image to Aladin you should find out that it is an IR star, possibly a Mira star or a long period variable star (LPV) 😉 Would explain why it was brighter in the 2.5nm OIII compared to the 3nm. Not sure why the 2.5nm Ha and SII showed no change over the 3nm if all of them transmit appreciable NIR at 940nm. |
2.11
#...
·
·
1
like
|
|---|
Jacob Heppell: Wow, I'm sorry! I was hoping it was just a bad product, but apparently, there seems to be a general issue with these filters. Of course, an LED is brighter and there might not be a total cut-off and there might be still residue IR from the LED. But you see a difference between 3nm and 2.5nm, it is certainly about the filter. pete_xl: This was exactly my thought when I first saw this topic. I said to myself, well, that seems like a nice way to find peculiar stars that has a strong emission in NIR. I think this filter 'could' be valuable in that manner. Although, I have to say I don't do observational astronomy. So I uploaded one of your pictures and seems like that brightening star is HD 97671 (or I guess Tycho 2 8959-707-1?). Anyway, guess what I found for HD 97671. Here is a quote from the abstract of the paper: "Four high-luminosity M supergiants, VY CMa, VX Sgr, S Per, and HD 97671, have peculiar energy distributions in the infrared with excess radiation between 1.5 and 9 micron. They show some evidence for excess radiation at the shorter wavelengths as well." Here is the reference: https://adsabs.harvard.edu/full/1974ApJ...188...75H This needs to be double checked as I can't read the annotation on the small image because it overlaps with others and a quick search I couldn't find the spectrum of this star. But at this point, I'd certainly return the 2.5nm's and stick with the 3nm. |
|
3.01
#...
·
·
1
like
|
|---|
Kay Ogetay:Jacob Heppell: Yes it does appear to be a problem with the filters in general. I like your analysis of the infra-red stars! The brightened stars turned out to be quite orange after I acquired some RGB data so I'm not surprised at the high infra-red emissions. |
|
2.11
#...
·
·
1
like
|
|---|
Jacob Heppell: Another piece of evidence is on the pocket. I just noticed that this is not just a brightness increase but they also become egg-shaped. Also in the background, there are faint egg-shaped stars appearing out of nowhere (what you call fuzz patches I guess). This is another explanation of why these are NIR emissions as optics does not focus it properly. I'm still pondering if this could have a scientific use though. |
|
3.01
#...
·
·
1
like
|
|---|
Kay Ogetay:Jacob Heppell: I actually didn't think of those "fuzz patches" as out-of-focus stars. That's a good point actually. The only refractory component of my rig is the coma corrector (I use a Skywatcher 10" Quattro newtonian) so I guess the NIR is too far out of it's designed refractory range. |
|
3.01
#...
·
·
4
likes
|
|---|
|
Next installment! Knowing that my Antlia LRGB dark series and 2.5nm SHO Ultra series transmit in the near infra red (likely between 850-1100nm), as an experiment, I decided to buy a 2" Astronomik L1 luminosity filter, screw it to the back of my coma corrector, then test how much NIR is passed using the remote control (940nm). I've owned this filter before as the clip-in version for my Canon 250D so I'm confident with it. Plus the spec sheet goes out to 1100nm (no NIR transmission seen) and the band pass is just wider than the Antlia Dark Series Lum. The results are screenshoted below  I placed an NIR pass filter (Hoya R72. Passes >720nm) on top of my CC then took a completely blank image and recorded the average ADU (for my camera at gain 56 offset 50, an ADU of ~800-810 is "blank") then I pointed the remote control down the coma corrector and clicked it 10 times then recorded the average ADU. As seen in the table, the ADU increase is very substantial (many thousand!), which proves that the Ultra/Dark series transmit a fair bit of NIR (the Pro version don't have this problem). I then repeated the experiment but with the Astronomik L1 filter screwed to the coma corrector. Suffice to say the Astronomik L1 held up very well as the ADU increased very minimally (13 ADU at most); even better than the Pro series. While not ideal, I reckon I'll just keep the Astronomik L1 screwed to the CC and be done with it! Saves me trying to negotiate returning and/or replacing with the Pro Series. Plus what I like about the RGB dark series is that the bandpass included the whole visible spectrum whereas there's a gap in the orange region with the RGB Pro series. Just a real shame Antlia seem to have overlooked the appreciable NIR passing of their otherwise great Dark/Ultra series. |
|
2.94
#...
·
·
1
like
|
|---|
|
An impressive result, Jacob! Btw I just saw my example images are gone. do you have an idea why? Pete |
|
3.01
#...
·
|
|---|
An impressive result, Jacob! I wonder if I should put this up on Cloudy Nights as well. I'm a long time stalker of that site but rarely post! I can still see your pic from my end so not sure what's up. |
|
2.11
#...
·
·
1
like
|
|---|
Jacob Heppell:An impressive result, Jacob! Very good testing, thanks for sharing! Please post it to Cloudy Nights as well. Your findings are quite important. They shouldn't be selling these filters at all. |
|
3.01
#...
·
·
1
like
|
|---|
Kay Ogetay:Jacob Heppell:An impressive result, Jacob! I just put added a post to CN! |
1.51
#...
·
·
1
like
|
|---|
|
Hi Jacob, very interesting. Would you mind sharing the Cloudy Nights link with your post? Thanks, Rob |
|
3.01
#...
·
|
|---|
Rob Kiefer: See here. https://www.cloudynights.com/topic/925125-antlia-darkultra-series-pass-near-infrared/ |
|
2.11
#...
·
·
1
like
|
|---|
Jacob Heppell:Rob Kiefer: Replied there as well. Hopefully, this will take more attention, thanks again for your efforts! |
1.20
#...
·
|
|---|
| Hello and thanks for sharing I think it has helped many to choose their next purchase. I would also be interested in these filters, having to switch from my Antlia 3nm 1.25" to the 2" ones. Leaving aside the problem with the OIII filter, if you have time you could share your impressions and tests on the HA and SII filters and also OII, to understand if you are experiencing any other problems and if and how much they are better than 3nm filters. Thank you. |
0.00
#...
·
·
1
like
|
|---|
| I have seen this what looks like a haloing/diffraction effect on my OIII 2.5nm as well! Not a function of star brightness but on specific orange-y stars... I've been picking my brains on what could be causing this for a while now but this investigation comes closest to explaining it... Thank you for looking into this! |
|
3.01
#...
·
·
1
like
|
|---|
|
So I got a decent response from Antlia after I told them about my most recent tests (plus they saw my post). Was much more comprehensive than the very short responses I received up until now and they included the filter transmission up to 1100nm. See below. LRGB Dark Series  OIII  Ha I think. Wasn't given SII.  There is certainly excellent blocking with a slight uptick close to 1100nm. Curiously, the OIII shows a gentle slop from ~1030 onwards Antlia's main criticism of my test is that I was using a point light source (the remote control), which is not representative of the incident light from a telescope that is coming to focus. This is a very fair point as the NB filters are designed for near parallel light. With that in mind, I decided to do one last test. I swapped the Antlia Lum Dark series in my filter wheel for the Astronomik L1, then tested the NIR transmission with and without the Antlia Dark series Lum screwed to the CC (also with the Hoya R72 pass filter). As seen in the table below, the Astronomik L1 transmitted appreciable NIR that was blocked when I put the Antlia Lum Dark series on the coma corrector. My guess is that, the remote control light shining directly down (low angle) onto the Lum filter screwed to the CC is transmitted with no NIR (so reached the FW filter with no NIR) but light that hits the said filter as a high angle contains a lot of NIR but is baffled by the CC  This certainly clarifies A LOT of confusion I've had regarding my results so far such as why all the LRGBSHO Dark/Ultra filters appeared to transmit appreciable NIR but it was only 2.5nm OIII that had an obvious problem. Unfortunately, the weather hasn't permitted me to test the replacement 2.5nm OIII filter I got from Antlia. If those fuzz patches disappear then I can put the matter to rest! I'll post the result here when I get the chance to do the test. So to sum up: This all started because of the very-real fuzz patches in my images taken with the 2.5nm OIII filter that disappeared when I put a lum filter on the CC, and didn't exist in the 3nm OIII Pro version. This rightfully led me to conclude that there must be out-of-band transmission. Those fuzz patches were associated with orange/red stars likely rich in NIR. The eventual conclusion here is that the fuzz patches were out-of-focus NIR from those stars. The fact that Antlia's spec sheets stopped around 850nm led me to speculate that there is NIR transmission in the 850-1100nm region. The fact that another astrophotographer had a similar experience only further advanced this hypothesis. My tests with the remote control seemed to further validate the NIR transmission but I agree they are invalid now as I was using a point source of light. However, there is still no doubt in my mind that my 2.5nm OIII filter was transmitting NIR but I don't know if it was simply a faulty filter or something else. I'm also pondering if the speed of my Newtonian (F4.6 with the CC) is resulting in more NIR transmission compared to the more parallel light of slower scopes (presumably Antlia's transmission test was done with parallel light). Unfortunately, I've got no way of testing this. Under the stars, I've only observed a problem with 2.5nm OIII. I haven't noticed any problems with the LRGBSH Dark/Ultra filters either on their own or compared with the Pro series. My main criticism for Antlia is that they should start issuing spec sheets out to 1100nm (not stop at 850nm), which will avoid all speculation about NIR transmission and conclusively show that they effectively block to 1100nm. |
|
2.94
#...
·
·
2
likes
|
|---|
|
Hi Jacob, this is interesting news from Antlia. With all (self-)criticism regarding the remote control as an IR light source - we should not reverse the causalities. It was the anomalies in the astro images that led you to do the test with the remote control, not the other way around. As described in my post above, I can fully confirm your results on the basis of the OIII 2.8nm ultra filter that I bought at the time. We also got unspecific answers from Antlia when my dealer asked them for me. As for the LRGB dark filters, the Astronomik Deep Sky series had no problem with a point light source, as you can see in my test image above. For me, at any rate, extending the spec sheets is not enough as long as the filters produce ugly artifacts on infrared stars under real skies, which have to be laboriously searched for and manually removed. CS Pete |
|
2.11
#...
·
·
1
like
|
|---|
Hi Jacob, I agree with Pete here. Their criticism about point source of light is valid, however that doesn't change the facts. What they say does not explain why you had those fuzz patches, why others had similar experiences, and why they are associated with specific IR emission stars. Also, if you take the remote quite far, it is practically a point source of light. But I'm more interested in your current filter's astro result. If that doesn't show the issue you had, we know the issue. |
|
3.01
#...
·
·
3
likes
|
|---|
this is interesting news from Antlia. With all (self-)criticism regarding the remote control as an IR light source - we should not reverse the causalities. It was the anomalies in the astro images that led you to do the test with the remote control, not the other way around. As described in my post above, I can fully confirm your results on the basis of the OIII 2.8nm ultra filter that I bought at the time. We also got unspecific answers from Antlia when my dealer asked them for me. Yes I've got no doubt you and I had issues with the Antlia 2.5/2.8 OIII transmitting off-band. My assessment based on the star tests still stands, but I probably took it all too far by asserting that all the filters had a problem based off the remote control tests. Antlia weren't too happy about the Cloudy Nights post; particularly as I thought is suspicious that the spec sheets stops at 850nm, which they took as an accusation of data falsification. No not exactly! I've been a research scientist long enough to know that what is not included or shown can speak volumes, but could just as easily mean nothing. My logic here is that if Antlia knew their filters have great blocking out to 1100nm (which they do based on the updated spec sheets they sent me) then why not include that in the supplied spec sheet instead of stopping short at 850-900nm. Certainly could have helped me if I had that data. Last thing I want to do is bash down Antlia. They bought 3nm filters to market that were nigh on par with brands like Chroma but at a fraction of the price. A lot of my fellow astro mates use Antlia filters and love them. I dare say they've taken a large chunk of market share in recent years. Kay Ogetay: Yes I agree. I'll let you know when the sky is clear enough. Probably in a few days. Been wet and overcast here for a few weeks now. |
0.00
#...
·
·
2
likes
|
|---|
|
we are very confident that our filters cut off at 1100nm and above. Attached please find our data. Additionally, professional photometers are essential to provide the most accurate measurements data for the transmission/cutoff of optical coating filters. If you still have doubts, please scan the filters in your local third-party such as a optical lab, university or an optical company who are using a photometer with a resolution as accurate as 0.1nm, we believe that they will give you convincing scaning data. Antlia has always been serious about product manufacturing and maintains integrity to the publics.     |
12.28
#...
·
·
2
likes
|
|---|
Jacob Heppell: In the end, it seems the filter has to work in the application. If the filters are creating artifacts in images and that is a real and reproducible observation - that is good for the general buying public to know. If a graph of transmission versus wavelength does not show the issue, it would be incumbent, I think, upon the manufacturer of the filters to devise a test that does and either fix the problem or advise the buying public that the filters will behave this way. In the end, we buy filters to put in front of our cameras and take images, not generate graphs of %T vs wavelength. If a filter shows a problem in the intended application, what good does it do to show a graph taken in the lab under different conditions that does not? |
|
3.01
#...
·
|
|---|
Arun H: Yes I would agree with that statement. Lab vs real world can produce different results for sure. |
|
3.01
#...
·
·
3
likes
|
|---|
|
Well I had a beautiful clear night for the first time in weeks so finally had the opportunity to test the replacement 2.5nm OIII filter that Antlia sent me. Much to my gratification, the replacement filter showed no fuzz patches (out of focus NIR) when I imaged NGC3572 (original target where I noticed the problem). For good measure, I took another image but with a lum filter on the coma corrector. There was no difference with or without the lum filter!! Replacement 2.5nm OIII Filter only  Replacement 2.5nm OIII Filter + Lum filter.  Couldn't see a difference when blinking the two. So I can only conclude that the original 2.5nm OIII filter was indeed defective. The spec sheet I received for the new non-defective filter (250-850nm) was identical to the original defective filter. So presumably the spec sheet it not for individual filters but is generic (unless the filters are incredibly consistent). I certainly have mixed feeling about that. A spec sheet for individual filters would be ideal but not sure if it's practical for the factory to do that. If the 2.5nm OIII filters showed excellent reproducibility then a generic sheet would be acceptable but clearly there is a least some variability as shown by my defective filter that wasn't picked up in QC. Perhaps testing a statistically appreciable sample to monitor for variability and pick up problems early would be the way to go (Antlia may already do this). What proportion of the 2.5nm/2.8nm OIII filters show this is hard to say. From this thread, we now know of two. Are we a 1/100 or 1/1000 case or is it more like 1/10 but others just haven't noticed a problem yet? My advice is by all means buy the 2.5/2.8nm OIII, but watch out for this issue of potential NIR transmission and seek a replacement if you notice the issue. |
|
12.28
#...
·
·
1
like
|
|---|
| It would be interesting to redo the remote test with the new filters. That might indicate if there is a problem with the old batch of filters sent to you AND if your test is indicative. Also - I remember Chroma sending me a transmission graph specific to my filters. I will have to check how far out in wavelength it went. |
