DIY large inexpensive flat field panel

[angryowl]

Having just received my new mount capable of holding my RASA I am now in need of a flat field panel large enough to cover the scope's aperture. Having looked at astro flat field panels I came to the conclusion that paying around £200 for one is a bit steep and I'd rather use that to buy me an LRGB filter set.

So I scoured the net for cheaper alternatives, there's the bare EL panels but for the dimensions I'd need prices are in the range of £60, not exactly cheap for what it is.

Then I found and ordered this http://www.beamled.com/biard-10w-panel-light-300-x-300mm-led.html 

It's a 295mm x 295mm LED ceiling panel which I think has an incorporated diffuser sheet as the light is very uniform. Comes with an LED driver and was only £13. I got the cool white version.

Some pictures of it

The waves seen in the images are a result of the camera's shutter I believe and are not visible in real life

The active area of the panel is only 250mm x 250mm so a bit small for the RASA's 279mm clear aperture so I'll have to figure out a way to cover that entirely. I'm thinking of placing some sort of semi-transparent sheet 50mm in front of the panel and that should in theory expand the beam of light enough so that it would cover the scope.

The panel has a 120 degree beam angle so I think this should work. Also it's VERY bright at 800-1600 lumens so that provides some flexibility in choosing what material goes in front of it.

Has anyone attempted this before? Any ideas for what I could place in front of it? 

 

[The Admiral]

Yes, I constructed a similar flats light but using a smaller circular LED light panel. Internally, they consist of an annular strip of LEDs on a tape (see my posts https://stargazerslounge.com/topic/265275-low-cost-light-source-for-flats/ and https://stargazerslounge.com/topic/274057-flats-light-final-chapter/), and do give what seems to be an even light. And yes, they are very bright, so much so that I had to find a way of reducing the intensity. Commonly used LEDs have a spike in the green part of the spectrum, and so using semi-opaque filters ends up shifting the spectrum towards the green. In the end I did away with the mains driven 'starter' unit and drove the LEDs with a low voltage DC. By adjusting the value of the series resistance it was possible to bring the intensity down to a more sensible level. Good luck with your build.

Ian

[JOC]

Sorry if its a bit Off topic, but I've seen several mentions of these sources of flat field white light on SGL.  I'm assuming that they are a area for the imagers (which I'm not), but I just wondered what (how, even) you use them for?

[angryowl]

35 minutes ago, The Admiral said:

Yes, I constructed a similar flats light but using a smaller circular LED light panel. Internally, they consist of an annular strip of LEDs on a tape (see my posts https://stargazerslounge.com/topic/265275-low-cost-light-source-for-flats/ and https://stargazerslounge.com/topic/274057-flats-light-final-chapter/), and do give what seems to be an even light. And yes, they are very bright, so much so that I had to find a way of reducing the intensity. Commonly used LEDs have a spike in the green part of the spectrum, and so using semi-opaque filters ends up shifting the spectrum towards the green. In the end I did away with the mains driven 'starter' unit and drove the LEDs with a low voltage DC. By adjusting the value of the series resistance it was possible to bring the intensity down to a more sensible level. Good luck with your build.

Ian

Hi Ian

Interesting read your threads. You're right, I think the panel is much too bright and I may just borrow your idea of fitting a current limiting resistor so the flats can have the proper exposure. Or I may go the route of a separate power supply as the panel works on 27 - 42V at 280mA.

On the issue of the emitted light peaking at the green spectrum, I don't think that'll be much of a problem as when taking RGB flats one would probably vary the exposure times for the individual RGB filters anyway. I may be wrong here though....

 

 

 

[alacant]

Hi. At that outlay, certainly worth a try. I had a similar panel but couldn't get the light even. All is not lost however since they're great for giving light over e.g. work benches and they don't get hot. In the end I got something similar to this A3 model which is not too bright and gives even light. The size has the advantage that I think the latter will cover your telescope completely. HTH.

[alacant]

35 minutes ago, JOC said:

I just wondered what (how, even) you use them for?

Hi. They are amongst the most boring photos in the world. It's a snap with the camera pointed at a light source which evenly illuminates the whole of the sensor. They help remove vignetting and dust marks. HTH.

[angryowl]

34 minutes ago, JOC said:

Sorry if its a bit Off topic, but I've seen several mentions of these sources of flat field white light on SGL.  I'm assuming that they are a area for the imagers (which I'm not), but I just wondered what (how, even) you use them for?

Hi Joc

Yes, you are correct, flat field generators/panels are exclusively used with imaging setups and their purpose is to provide flat frames used to calibrate images taken with the scope. They correct for vignetting caused by the scope/camera combination and any accessories one may have in the light path and also for any dust bunnies there may be on any of the surfaces in front of the camera.

[angryowl]

2 minutes ago, alacant said:

Hi. They are amongst the most boring photos in the world. It's a snap with the camera pointed at a light source which evenly illuminates the whole of the sensor. They help remove vignetting and dust marks. HTH.

Oops beat me to it 

[angryowl]

10 minutes ago, alacant said:

Hi. At that outlay, certainly worth a try. I had a similar panel but couldn't get the light even. All is not lost however since they're great for giving light over e.g. work benches and they don't get hot. In the end I got something similar to this A3 model which is not too bright and gives even light. The size has the advantage that I think the latter will cover your telescope completely. HTH.

Thanks for the link, that's a nice panel. I'm assuming it's just an EL panel and appears to be driven through USB which is rather nice.

[alacant]

7 minutes ago, angryowl said:

driven through USB

Yes. It comes with a cable. It really is a simple affair which just works.

**for your RASA, you will almost certainly need to interpose (interspace?) layer(s) of paper; with my 150/f8 I get about 1s, but on my 200/f3.9 it needs dimming, otherwise dark bands appear as the shutter is out of sync with the refresh of the panel. Or something like that...

[JOC]

I am now wiser - thanks folks.  I always reckon its easier to ask than to sit and wonder.

[The Admiral]

1 hour ago, angryowl said:

Hi Ian

Interesting read your threads. You're right, I think the panel is much too bright and I may just borrow your idea of fitting a current limiting resistor so the flats can have the proper exposure. Or I may go the route of a separate power supply as the panel works on 27 - 42V at 280mA.

On the issue of the emitted light peaking at the green spectrum, I don't think that'll be much of a problem as when taking RGB flats one would probably vary the exposure times for the individual RGB filters anyway. I may be wrong here though....

I'm not sure how effective it would be to use a series resistance when using the driver, as the drivers are constant current devices I believe. I eliminated the driver and used a low voltage dc source. By all means try adding absorptive material, with RGB it could well work, but I use a OSC camera.

1 hour ago, alacant said:

... it needs dimming, otherwise dark bands appear as the shutter is out of sync with the refresh of the panel. Or something like that...

With my set up it is a dc supply so the light doesn't (shouldn't) cycle. A bit puzzled by the linked one giving banding though, as it's driven by the USB. On the other hand, if it's an electro-luminescent panel, these operate at a significantly higher voltage I believe, so perhaps it contains an hf transformer to step up the voltage.

Ian

[alacant]

41 minutes ago, The Admiral said:

these operate at a significantly higher voltage

Hi. You plug it into your phone charger. HTH.

[Horwig]

2 hours ago, alacant said:

**for your RASA, you will almost certainly need to interpose (interspace?) layer(s) of paper; with my 150/f8 I get about 1s, but on my 200/f3.9 it needs dimming, otherwise dark bands appear as the shutter is out of sync with the refresh of the panel. Or something like that...

I use a diy panel made from an LED ropelight and two diffuser sheets. I used a combination of sheets of white paper and a PWM dimmer to give me exposures in the seconds range on an f3.5 scope

 

 

Huw

[martin_h]

I used a ND filter to dim my el panel..... https://www.stagesuperstore.co.uk/lee-frosts--diffusionsspun-243-c.asp

[angryowl]

3 hours ago, Horwig said:

I use a diy panel made from an LED ropelight and two diffuser sheets. I used a combination of sheets of white paper and a PWM dimmer to give me exposures in the seconds range on an f3.5 scope

 

 

Huw

Huw, looking at your designs, have you used opal acrylic sheets for the diffusers? This looks promising...

[angryowl]

Lots of great input so far, thanks to all that replied...

Have a few designs in my head atm but nothing definitive

1. Scrap the LED driver that came with the panel and use one that would reduce the light output considerably and then maybe just use a diffuser sheet in front

2. Keep the supplied driver and use some sort of semi-opaque sheeting to reduce intensity of light 

3. Using an ND sheet to dim the panel also seems promising

4. Cheating a bit and purchasing http://www.beamled.com/biard-40w-panel-light-600-600mm-led.html the 600mm x 600mm version which would cover the scope and then some 

I know the panel is too bright on its own but anyway I'm going to try and do a few tests with the scope to see roughly how much it needs to be dimmed by using sheets of paper so we'll see how that goes...

[redtail]

I built one of these and it works very well and also works in sgpro.https://www.cloudynights.com/topic/536533-diy-alnitak-flat-panel/

[billyharris72]

I used one of these for my base LED as it's dimmer than the lighting panel and also has brightness adjustment.

https://www.amazon.co.uk/gp/aw/d/B06W5N7J6P/ref=mp_s_a_1_1?ie=UTF8&qid=1504764198&sr=8-1&pi=AC_SX236_SY340_QL65&keywords=tracing+panel+led+a3&dpPl=1&dpID=51D9AaTF%2BGL&ref=plSrch

I used that with a few sheets of A3 paper on top and then a translucent perspex sheet (about £5 from Amazon). Seems to work quite well so far.

[angryowl]

So last night tried the bare panel in front of the dew dew shield and got the following results with my narrowband filters and Atik 414EX:

OIII 0.001s exposures 44000 ADU 
SII 0.01s exposures 41000 ADU
Ha 0.006s exposures 43000 ADU

I could get both SII and Ha down to about 10-15k ADU by reducing the exposures. That's what I was using for my flats with my previous setup with the Atik and they seemed to work great.

Now there's a few opinions on the duration of flats as far as I'm aware, some say between 10-15k ADU, others swear by 25-30k and some even prefer 40-45k. I know the duration of flats mainly depends on the camera's full well, gain factor and ADC levels, but I remember a while back when I started imaging I calculated all this somehow and the sweet spot for my Atik 414EX was between 10 and 15k ADU.

So just wondering if I could actually get away with this as I if I needed lower ADU flats I'd need to find a way to reduce the light output of the panel as 0.001s is the camera's minimum exposure time.

Also on this note I know the ideal flat duration is one where you expose for long enough so that the sensor read noise is overpowered but not so long that you start inducing unnecessary dark noise in the image. So my question is, if hears some use exposures in the range of seconds for their flats, would that not introduce enough dark noise into the flats that would require more processing later on to remove. Of course in a proper workflow one would calibrate and optimize the individual flats by subtracting a scaled down version of the master dark to match the dark current in the flat frames.

Here's some individual flat frames just stretched in PI

OIII

SII

Ha

And here's the unmodified FITS for OIII

OIII_1.fit

[Horwig]

On 06/09/2017 at 19:51, angryowl said:

Huw, looking at your designs, have you used opal acrylic sheets for the diffusers? This looks promising...

Yes indeed, opal plastic, two layers of it, just to be sure:

http://www.ebay.co.uk/itm/White-Acrylic-Perspex-Sheet-Colour-Cut-to-Size-Panel-Plastic-Matt-Satin-Gloss-/172107516062?var=&hash=item281268489e:m:m7HwpRF6NxxTSFYywY_IoBA

 

Huw

[angryowl]

Thanks Huw

I realise there are many cheaper, less time consuming and more effective DIY flat field panels ideas but seeing I've bought this I might as well give it a go and log my progress here.

I have some acrylic mirror sheets and I plan on cutting four strips and place them at a 45 degree angle on the face of the panel just where the active area ends. These would work by reflecting light coming off the side of the panel (120 degree beam angle) upwards. Some sort of acrylic diffusing sheeting would then sit on top of the strips of acrylic mirror and hopefully light reflecting off the mirrors and that from the active area should be close enough in intensity so that the diffuser can then blend the two together.

This would provide the coverage I need whilst at the same time reducing intensity by means of the diffuser sheet. If further reduction in intensity is needed an additional sheet can be added or some other semi-opaque sheeting before the diffuser. An advantage to this design is the diffuser can then be drilled to fit some mechanism or legs to hold the finished panel on the scope.

[coatesg]

I'd just looked at these as well so interesting to hear some first hand experience. I'm slightly surprised by the extra short length of exposure needed here, but wouldn't it be fairly simple to add an extra diffuser or two in front of the source to further even up and attenuate the light output?

 

Cheers

[angryowl]

Just to wrap this up I will no longer be building a flat field generator with this LED panel as it'll end up being much to heavy to hang on the end of my flexible dew shield. Instead I'll probably be purchasing a large EL panel as it's much lighter and dimmer requiring no modifications whatsoever. 

[Dr_Ju_ju]

That's what I went with in the end, bought one of these https://elwirecraft.co.uk/product/el-panel-25-x-25cm/, yes they are more expensive, but they are extreemly light and even illumination etc.