Flat LED ceiling lighting panels as Flat panels?

[buzz]

Has anyone tried one of these devices as a flat panel?  Some run off 12V and could be dimmed with a simple PWM controller and seem to be bright enough. They are available in very large sizes, from about 500mm square.  They seem to be bright enough but I do not know if the light is uniform.

for example:

http://gb.auroralighting.com/Products/Indoor-Luminaires/Commercial-Lighting/LED-Flat-Panels.aspx

[teleskopjo]

I guess they're actually too bright!

[Physopto]

Has anyone tried one of these devices as a flat panel?  Some run off 12V and could be dimmed with a simple PWM controller and seem to be bright enough. They are available in very large sizes, from about 500mm square.  They seem to be bright enough but I do not know if the light is uniform.

for example:

http://gb.auroralighting.com/Products/Indoor-Luminaires/Commercial-Lighting/LED-Flat-Panels.aspx

Hi Chris,

 I looked into this some time ago. If I remember correctly they are not suitable as flat panels for Flats. LEDs put out the wrong frequencies of light and can be seen as banding or some such. I bought a flat panel from Germany, it is very good but still too bright. So I ended up ordering up a few neutral density filters as well. Not tried it again as yet but should give several different light levels. Then the CCD timing will not be as critical or so short. Some cameras cannot do very short exposures.

Derek

[buzz]

Hi Chris,

 I looked into this some time ago. If I remember correctly they are not suitable as flat panels for Flats. LEDs put out the wrong frequencies of light and can be seen as banding or some such. I bought a flat panel from Germany, it is very good but still too bright. So I ended up ordering up a few neutral density filters as well. Not tried it again as yet but should give several different light levels. Then the CCD timing will not be as critical or so short. Some cameras cannot do very short exposures.

Derek

The frequency is not a function of the LED, but of the power supply.  I was thinking of a PWM controller with a frequency of about 25KHz.  I have some LED's and shall do some experiments and report back.

[Physopto]

The frequency is not a function of the LED, but of the power supply.  I was thinking of a PWM controller with a frequency of about 25KHz.  I have some LED's and shall do some experiments and report back.

We are at cross purposes here.

The frequency I'm talking about is the light wavelength. White light from a white LED comes from Red Green and Blue colours. It is not a white LED as such. A red LED is normally say 652nm so there is really only a specific wavelength or frequency output from that LED. All other wavelengths in the red spectrum are not there. Same sort of thing with Green and Blue. So the light that fools our eyes and brain into thinking it is white is not all there when analyzed. The CCD in the camera is sensitive to all frequencies in the visual spectrum and will therefore show some banding.

Regards,

 Derek

[buzz]

Hi Derek - I did some experiments with daylight balanced LED's.  I was using a QSI 683 with Baader filters:

At a dim light setting:

Filter     Exposure         ADU

L             0.2s               34K

R            0.5s                26K

G            0.5s               37K

B            0.5s                20K

Ha         10s                  30K

OIII         10s                  24K

SII          10s                  23K

SII          0.2s                22K   (light on full power)

With the lamp on full - I had SII, at 0.2s   at 22K  - so, although the spectral output range varies across the wavelengths, the light level adjustment from the panel is such that I could achieve similar exposures through each filter and vary the power (my led array is attach to a dew-heater output )

Might be worth a go.   I could also substitute some of the LEDs for deep red ones too if I chose to.

[Gina]

A white T shirt and daylight fits the bill and is easy and cheap

[Physopto]

Hi Derek - I did some experiments with daylight balanced LED's. I was using a QSI 683 with Baader filters:

At a dim light setting:

Filter Exposure ADU

L 0.2s 34K

R 0.5s 26K

G 0.5s 37K

B 0.5s 20K

Ha 10s 30K

OIII 10s 24K

SII 10s 23K

SII 0.2s 22K (light on full power)

With the lamp on full - I had SII, at 0.2s at 22K - so, although the spectral output range varies across the wavelengths, the light level adjustment from the panel is such that I could achieve similar exposures through each filter and vary the power (my led array is attach to a dew-heater output )

Might be worth a go. I could also substitute some of the LEDs for deep red ones too if I chose to.

Hi Buzz,

That looks interesting. There is quite a difference in the ADU output for each colour but that is just a case of timing I would guess. The high numbers at low exposures is the very reason I also bought the neutral density filters for my light pannel. They just allow more control over the exposure times. What is the full range of ADU values. I know from what I've read they recommend about half full values if I remember correctly. It is now about a year since I was able to spend some time taking mine. Work has gotten in the way! I have also changed cameras since then. I'll have a look back and see what values I had.

Regards,

Derek

[Tinker1947]

A white T shirt and daylight fits the bill and is easy and cheap

I brought a pair of pillar cases from Asda, sliced one down the side and end it would cover the Hubble Telescope all for 2 quid......

[buzz]

A white T shirt and daylight fits the bill and is easy and cheap

I'm not so sure.... Something in the back of my mind says sticking a diffuser directly in front of the lens, floods the entire optical system and you get a lot of stray light. In terms of imaging, the light coming into the scope is almost collimated. In my mind, I would want the quality of light to be the same.

Anyway - enough of the theory -  I'm doing some tests to compare a diffuser directly over the lens hood or taking an image of a matt white wall from several meters away. I shall do calibrated flats using both systems and then do a PixelMath comparison   0.5+ (Experiment1 - Med(Experiment1))-(Experiment2-Med(Experiment2))  - that will give me a difference image to evaluate.

[Tinker1947]

It has been since the beginning of time been the T-Shirt over the end of the scope, technology will have a hard job changing this.....Good Luck

[Gina]

Works for me

[buzz]

A full set of flats takes 3 hours for me, 50x exposures for 8 filters and 2 binning levels. I cannot rely on consistent sky conditions over that time and for that reason, I do them indoors. Currently I'm using a bathroom light, (50W Tungsten Halogen Bulb with opal glass), illuminating a white emulsioned wall. Plenty of Ha and SII.

Anyway it is Easter, it is cloudy and I'm bored and it is surprising what happens when you challenge the norm! In the back of my mind, a friend and I both have some flat calibration issues on our images - leaving a very slight halo affect - as if the vignetting on the image has a very slight different radius to the calibrated flats. He's using Maxim, I'm using SGP/PixInsight but the faint effect is very similar. It would be good to nail it down.

[buzz]

I completed the comparison between a direct diffuser over the dew shield and imaging a white wall at 3 m. The difference is very subtle. I ran both integrated flat frames with the PixelMath equation 0.5+(flat1/med(flat1))-(flat2/med(flat2).

This normalizes the flat frames to the median value and subtracts them, adding 0.5 to make it mid grey.  There was a faint narrow ring where the main vignetting edge was, which could account for the pale arc I saw on my backgrounds.

This only affected the vignetting outline- the one dust spot on the filter disappeared when the two flat frames were subtracted in this way, proving this is not a registration or scaling issue.

It is easy to see how most will be fine with a simple diffuser over the dew shield.  

[Joseki]

I built myself a flat field panel using an EL panel:

http://www.ebay.co.uk/itm/EL-Panel-White-Neon-Back-Light-Board-12V-DC-Inverter-for-Advertising-Board-IDE-/111349271792?pt=LH_DefaultDomain_3&hash=item19ecef0cf0

Stuck it to some aluminium sheeting for rigitity and a sheet of opal acrylic light diffuser in front (total cost about £10-£15), it looks a lot like the Gerd Neuman flat fields http://www.gerdneumann.net/english/astrofotografie-parts-astrophotography/aurora-flatfield-panels/uebersicht-aurora-flatfield-panels-overview.html   (other than I made sure the wires wouldn't fall apart) and seems to work okay.