Light box too bright for ST8300

[Ivor]

I've discover my light box is too bright for the SBIG ST8300M, to avoid the shutter impacting the exposure the recommended duration is 4 seconds however the image even at 0.1 seconds is over-exposured.

The light source is a 69cm strip of these LEDs

http://www.ebay.co.uk/itm/12v-LED-STRIP-LIGHT-3528-SMD-300-LEDS-5M-UK-/320681128958?pt=UK_Home_Garden_Night_Lights_Fairy_Lights&var=&hash=item4aaa15b3fe

Inside one of these http://www.tesco.com/groceries/Product/Details/?id=277902897 (chocolates removed)

I tried covered intermittent LEDs to reduce the brightness however I only start seeing an impact on the image when there are too few LEDs leading to an uneven illumination on the diffuser.

I'm now going to try a different approach and add a potentiometer however I think my maths might be wrong in calculating the size of resistor.

The LED strip power is consuming 4.8W per metre so my strip should be using 3.3W

It has a 12V power supply so Power=IV so the current should be 3.3/12 = 0.276 A

Therefore based on V=IR the current resistance of the circuit is 12/0.276 = 42.5 Ohms.

As I'm trying to reduce the Power based on P=I²R, I need to reduce the resistance to reduce the power so I need a potentiometer 0-50 Ohms, is this right?

[Mac]

Yup should work if you wire them up in series, so that the potentiometer takes half the voltage at full resistance, you also need to
consider that when the pot is at 43 ohms it will be disapating .8 watt, so you might need to get at least a 1W pot.
The problem you will run in to is that as you increase the pot so the voltage to the leds will dim, to a point where they will
probably not turn on at all, this might not allow them to be as dim as needed,

The other way i would do it is to build a variable mark space ratio timer using a 555 chip,
design it so its running at 1000hz,

very simple, and then use this to power the strip, that way you have an adjustable dimmer for it from 0 to 100%,

and as all you are doing is turning the leds on and off fast, its fully controlable,

The leds will always light as they are getting the full supple voltage.

Mac

[Ivor]

Thanks Mac. The smallest pot Maplins have is 1K but it is only .4W, so I'm not that will work then based on your calc

when the pot is at 43 ohms it will be disapating .8 watt, so you might need to get at least a 1W pot.

BTW how did you calculate this?

Not sure where else to buy this locally.

[Mike]

The other way i would do it is to build a variable mark space ratio timer using a 555 chip,
design it so its running at 1000hz,

Otherwise known as Pulse Width Modulation (PWM). Like Mac said, easily done with a timer chip very cheaply too. Read up on 555 and PWM.

Ivor do you own a tablet computer? I have found a completely white screen and minimum brightness on my iPad makes a perfect flat panel.

[Mac]

Calculations.

You are correct in that your strip is 43.6 ohms.
If you get a 50 ohm pot and run in in series, then when the pot is at 43.6 ohms the voltage drop is exactly half the supply
the other half being over the 43.6 ohm strip.

You can the calculate the power as W = V²/R
which is 6*6 / 43.6 = 0.82 W
As the pot  resistance is reduced so the voltage across it is also reduced.
So when its 10 ohms, the voltage will be 2.2V
The power will be 2.2*2.2 /10 0.48W

Mac.

[Rocket Pooch]

Or you can buy some ND sheet from fleabay.

http://www.ebay.co.uk/itm/Neutral-Density-Filter-Sheet-299-1-2ND-/110601846109?pt=UK_Sound_Vision_Other&hash=item19c0623d5d

I have some of this for my flat panel which was to bright.

[Ivor]

Last night I placed some white card over the diffuser which made a different but the image duration was still only 0.75s which is likely to lead to the shutter impacting the shot. In addition this approach only gave a time range of 0.75 - 0.78s to play with so I need to look at either adding the pot or the PWM to give sufficient control over the brightness.

Trying to find a pot which is 100ohms and 1W rating is quite hard, Farnell have some but it will be £10 including delivery.

Can anyone think of a local store that might sell components other than Maplin?

My wife has an iPad and it is just big enough to fit over the FLT110, however I found that the touch screen appeared to respond to the telescope touching it, so I need to investigate some more. I don't really do Apple products so I found it quite strange being in an alien world of IT when I normally familiar with most things.

The only consolation is the poor flat I did last night highlighted that the problem I'm seeing in the light subs is caused by something else. Two problems great!!

[Mike]

Ivor, put a sheet of paper in front of the iPad if that is happening.

[Mac]

You can definitely build a pwm switch for less then £10.

Here's a quick circuit straight off of google.



You would need to probably have the PWM controll a darlington pair for the control  of the LED strip.


Instead of feeding an output, you would just have your strip instead of the resistor.

Mac.

[MarkS]

I've used these PWM controllers from Maplin for controlling dew heater bands and also for controlling Peltiers.  Easily mounted in a plastic "project box".
http://www.maplin.co.uk/panel-mounted-speed-regulator-module-30310
Switches up to 3 amps continuous

[Rocket Pooch]

Hi,

You do know you can cut some of the LED off the chain if needed? 

Also I'm not sure all LED like to be dimmed either, but that will be easy to test, as for the 555 idea, you could end up with inconsistent flats if the flicker is not consistent between flats.

I'm having a negative day...

[Ivor]

The LEDs I'm using can't be dimmed so does this mean the pot and PWM solutions are ruled out?

I know I could cut some of the LEDs out, however as the current strip has the LEDs in parallel and the strip goes around the inner circumference. to get around this I covered 1 LED every 3 to ensure even illumination.

This solution produced a lovely flat for the canon http://forum.orpington-astronomy.org.uk/index.php?topic=8941.0 it's just the ST8300 is too sensitive.

I will play with the ipad solution some more tomorrow and I will see if a 100ohm resistor added to the circuit adjusts the brightness before investigating the pot or PWM.

I think I has most of the bits for the PWM in the garage from my failed attempt to built a dew heater last year so it would be pretty cheap if it was viable.

[Mike]

Why can't they be dimmed?

[Mac]

if you use the pwm you are not dimming them, you are just turning them on and off fast using a variable mark space ratio.

the wave form output looks something like this (per cycle).

on for 10%
__                                     
   |                                         
   |                                         
   |____________                                         
                                           
on for 50%

_______                                     
          |                                         
          |                                         
          |________

on for  90%
____________                                     
                  |                                         
                  |                                         
                  |__


so at 10% they are only illuminated at the full voltage for 10% of the time, therefore their brightnes will only be 10%
at 50% they are only illuminated at the full voltage for 50% of the time, therefore their brightnes will only be 50%

you get the picture.

Mac.

[Ivor]

Why can't they be dimmed?

It an assumption I'd admit, I'm basing this on a LED needing a minimum voltage to fire I'm not sure what the voltage range is for this type of LED, maybe I'm being too pessimistic.