108 replies to this topic

[spokeshave]

In my quest toward fully remote imaging, I needed a flat panel solution that would work with SGP. There is no ASCOM standard for flat panels, so the only flat panel that SGP can control at present is the Alnitak series from Optec. For my Edge 14, I would need the FlatMan XL, which is $1300. I don't mind spending good money on good equipment, but that seems excessive to me, when  DIY approach would seem to be possible at a fraction of the cost. So, I decided to try a more grass-roots approach. I found a light panel that will fully illuminate the 14" scope here:

 

http://www.amazon.co...pf_rd_i=desktop

 

It has very even illumination and adjustable brightness. I mounted it on the wall of the observatory, and set the park position of the scope such that it points directly at the panel:

 

 

 I have tested it by manually adjusting the brightness and aside from being too reflective, it works very well. During testing, I put some t-shirt material over the scope to mitigate the reflections, but that obviously is not a solution for remote flat collection. To permanently eliminate reflections, I frosted the face of the panel with some 220 grit sandpaper and a random orbit sander:

 

 

Of course, I still have the problem with controlling the panel remotely. I found some Arduino code written by one of the SGP developers here:

 

https://github.com/j...ArduinoLightbox

 

It takes the Alnitak flat panel command set and parses them into PWM output on one of the PWM pins on the Arduino. To make this work, I first needed to bypass the control circuitry in the flat panel (no pics of this). I simply cut the + and - wires coming from the 12V input jack leading to the circuit board in the panel, and cut the + and - output wires from the control circuit and then soldered them together. This passes full power to the LEDs, such that their brightness can be controlled by modulating the 12V input with PWM. All that was left was to build the Arduino control box. It couldn't be simpler. It consists of an Arduino Uno ($5) and a simple NPN transistor ($1) in a project box ($3):

 

 

The only thing I had to do the the Arduino code was modify it to tell in which PWM pin I was using on the Arduino. No additional Arduino libraries are needed, and it fired up perfectly the first time. The Arduino connects by USB to the observatory computer, accepts serial commands using the Alnitak command set, and then modulated the 12V input for PWM output to the flat panel. I now have 256 completely repeatable brightness settings that are software controllable. It works brilliantly. Now SGP can perform fully automated flats whenever I want it to by parking the scope, turning on the panel, collecting flats at the precisely determined correct brightness and then turning everything off.

 

Works great and at about 1/10 the price of the FlatMan XL, it is a great bargain.

 

Tim

[pterodyne]

This is awesome! I am needing something similar for ACP automation, tired of doing sky flats and the problems therein.  I was thinking about emulating alnitak's command set, but didn't really have a starting point for that.

 

Have you had chance to get results yet?  I mean real flats for current imaging?

[spokeshave]

I collected some flats yesterday. The panel worked beautifully, and the best part was that I did everything from my easy chair. I haven't applied the flats in calibration yet, but I have used this panel for flats before and they worked very well. I did watch the mean ADU for the flats as they came in for each filter for consistency, and they were all very consistent. I now have all of the correct exposure/panel brightness settings for each filter and binning in my equipment profile ins SGP. So, when I set up a sequence, I can add flats to the end of the sequence, and SGP will point the scope at the panel and then select the correct exposure and panel brightness for the flats. It's going to be a huge time saver.

 

Tim

[pterodyne]

can you give me the model of transistor you are using? I just bought the panel!

 

thanks for posting.  

 

It looks like the arduino code implements flipflat as well.  I noticed that it shows different sizes of the flatman as well.  

 

IN the case of ACP, it uses the following config for a real flatman:

 

LightCtrlProgram      C:\Program Files (x86)\Alnitak Astrosystems\Alnitak Astrosystems Controller\AACmd.exe
LightOnCommand        3 C L B#BRT#       ; Command string to turn light on and set/change brightness
LightOffCommand       3 O D              ; Command string to turn light off
LightOnDelay          60                 ; Time needed (sec) for brightness to stabilize

 

do you use the aacmd.exe as well in SGP pro? Or does it communicate via serial commands only?

[spokeshave]

The transistor is just a TIP3055 NPN that I had in my junk drawer. The base is soldered to pin 3 on the UNO, the ground side of the flat panel ties to the collector, and the emitter is tied to the ground on the UNO. You need a 12V power supply to the UNO, and I soldered a lead from the Vin pin to the 12V side of the panel. No base resistor is needed. Oh - one thing. You will need to specify in the code which PWM pin you're using. I used pin 3. For some odd reason, the code defaults to pin 13, which is not a PWM pin on the UNO. You'll need to change that 13 to a 3 (or whatever pin you use). It is obvious by the comments on the code where you need to do this.

 

You'll need to do a little surgery on the panel to bypass the control circuitry. There is a plastic cover on the back that you'll need to peel back over the corner where the circuit board is. You only need to peel it back over that corner, not the whole panel. There is a red and black wire coming off of the input jack that goes to the control circuit board. You'll need to clip those wires as close to the board as you can. On the other side of the board is the two wires (black and red) that go to the LEDs. You'll need to clip those as close to the circuit board as possible too. There might be some silicone around the wires that you have to carefully cut away to free them. Then you just solder the black to black and red to red. I put shrink tube over the solder joints.

 

SGP simply sends the commands out to the serial port, AACmd.exe is not used. I don't know if AACmd.exe will see my device or not. I may have to try that, though since SGP doesn't use AACmd.exe, it doesn't benefit me if it does.

 

I haven't used ACP, but it seems like it allows you to specify the program that controls the light panel. I suppose it would be a fairly simple matter to write a program that will take the commands sent by ACP, and then interpret the commands and send them out on the serial port. That's all that AACmd.exe does. The serial command set is in the comments in the code.

 

Tim

Edited by spokeshave, 12 May 2016 - 01:23 PM.

[pterodyne]

Got my panel yesterday, and am waiting on an Arduino nano.  I was hoping the nano would be small enough to fit inside the case, but I don't think it will.  I did a little research on the aacmd.exe.  I think all it is looking for is a FTDI USB adapter, possibly not using a virtual com port.  FTDI drivers can be installed as a direct usb device or as a VCP.    I have 3 different arduinos that I put the arduino lightbox firmware on.  None of them detected in the software, but then again none of them have FTDI, they are all cheap arduino clones.  The specific nano I ordered does have the FTDI chip, so we'll see.  If it isn't detected, then Ill still proceed, just have to get ACP to send commands to the serial port directly somehow.  Ive used an old dos app called serialterm in the past, even in scripts.  But maybe cygwin would be a bettter way to write and read the serial port.

[RandallK]

I just ordered my flat panel from Amazon.ca today. I am taking flats in my basement with the windows blocked out when doing flats. I have a Gerd Neumann for my 80mm scope and will be using the larger panel for my 8" SCT. I ordered the smaller version.

If my room is virtually pitch black, and my scope is right up close to the panel, would there be any need to buff it up? Thanks in advance.

[pterodyne]

I don't know the answer to your question, but I can say I buffed mine today.. My first idea was to use a high grit sandpaper so I tried 1500 but just by hand.  It didn't seem to be very even.  Strangely enough I then tried 300 grit on my orbital sander, and it worked really well.  I tried hard to keep the sander perpendicular to the surface, and not to press hard.  Very happy

[sami2012]

In my quest toward fully remote imaging, I needed a flat panel solution that would work with SGP. There is no ASCOM standard for flat panels, so the only flat panel that SGP can control at present is the Alnitak series from Optec. For my Edge 14, I would need the FlatMan XL, which is $1300. I don't mind spending good money on good equipment, but that seems excessive to me, when  DIY approach would seem to be possible at a fraction of the cost. So, I decided to try a more grass-roots approach. I found a light panel that will fully illuminate the 14" scope here:

 

http://www.amazon.co...pf_rd_i=desktop

 

It has very even illumination and adjustable brightness. I mounted it on the wall of the observatory, and set the park position of the scope such that it points directly at the panel:

 

 

 

 I have tested it by manually adjusting the brightness and aside from being too reflective, it works very well. During testing, I put some t-shirt material over the scope to mitigate the reflections, but that obviously is not a solution for remote flat collection. To permanently eliminate reflections, I frosted the face of the panel with some 220 grit sandpaper and a random orbit sander:

 

panel.jpg

 

Of course, I still have the problem with controlling the panel remotely. I found some Arduino code written by one of the SGP developers here:

 

https://github.com/j...ArduinoLightbox

 

It takes the Alnitak flat panel command set and parses them into PWM output on one of the PWM pins on the Arduino. To make this work, I first needed to bypass the control circuitry in the flat panel (no pics of this). I simply cut the + and - wires coming from the 12V input jack leading to the circuit board in the panel, and cut the + and - output wires from the control circuit and then soldered them together. This passes full power to the LEDs, such that their brightness can be controlled by modulating the 12V input with PWM. All that was left was to build the Arduino control box. It couldn't be simpler. It consists of an Arduino Uno ($5) and a simple NPN transistor ($1) in a project box ($3):

 

box inside.jpg

 

The only thing I had to do the the Arduino code was modify it to tell in which PWM pin I was using on the Arduino. No additional Arduino libraries are needed, and it fired up perfectly the first time. The Arduino connects by USB to the observatory computer, accepts serial commands using the Alnitak command set, and then modulated the 12V input for PWM output to the flat panel. I now have 256 completely repeatable brightness settings that are software controllable. It works brilliantly. Now SGP can perform fully automated flats whenever I want it to by parking the scope, turning on the panel, collecting flats at the precisely determined correct brightness and then turning everything off.

 

Works great and at about 1/10 the price of the FlatMan XL, it is a great bargain.

 

Tim

 

 

Hi Tim

 

If can please share the SCHEMATIC , the Parts List, i would like to do the same DIY

 

best regards

 

sam

[pterodyne]

here is one I did.  Tim correct me if I am wrong..

 

This assumes 12VDC going into the bread board power bus

Attached Thumbnails

• 

Edited by pterodyne, 31 May 2016 - 09:07 AM.

[pterodyne]

I have added the power supply for clarity..

Attached Thumbnails

• 

[spokeshave]

Sorry... I forgot I started this thread and haven't kept up.

 

Randall: I frosted the panel because my panel is a good 4 feet away from the front of the scope. I found that the corrector plate would reflect in the panel causing problems with flats. If you put the panel directly over the front of the scope, you may not have a problem.

 

Sam: I'll try to draw a schematic later but it is dead simple. There are only three parts - an Arduino Uno, a TIP3055 NPN transistor and a 12V LED panel or array. When looking at the transistor:

 

The base is pin number 1. It gets tied to the PWM pin in the Arduino. I used pin #3. Pin number 2 on the transistor (collector) goes to the ground in the Arduino. Any pin on the Arduino marked GND will do. Pin number 3 (emitter) on the transistor goes to the ground on the LED panel. Finally, the positive side of the LED panel gets tied to the Vin pin on the Arduino. Power is supplied by a 12V wall wart plugged into the power port on the Arduino. The only thing left is to change the PWM pin assignment in the Arduino code from 13 to 3.

 

That's it.

 

Tim

[sami2012]

I have added the power supply for clarity..

Thank you

Where is the wire the light panel Connecting in the arduino ????can not figure out the wire input

For arduino code do you any modify in the code

Edited by sami2012, 31 May 2016 - 12:55 PM.

[sami2012]

Sorry... I forgot I started this thread and haven't kept up.

 

Randall: I frosted the panel because my panel is a good 4 feet away from the front of the scope. I found that the corrector plate would reflect in the panel causing problems with flats. If you put the panel directly over the front of the scope, you may not have a problem.

 

Sam: I'll try to draw a schematic later but it is dead simple. There are only three parts - an Arduino Uno, a TIP3055 NPN transistor and a 12V LED panel or array. When looking at the transistor:

 

The base is pin number 1. It gets tied to the PWM pin in the Arduino. I used pin #3. Pin number 2 on the transistor (collector) goes to the ground in the Arduino. Any pin on the Arduino marked GND will do. Pin number 3 (emitter) on the transistor goes to the ground on the LED panel. Finally, the positive side of the LED panel gets tied to the Vin pin on the Arduino. Power is supplied by a 12V wall wart plugged into the power port on the Arduino. The only thing left is to change the PWM pin assignment in the Arduino code from 13 to 3.

 

That's it.

 

Tim

 

Hi Tim

 

for the light panel where is the wire connecting in the arduino??

[spokeshave]

Here is a schematic:

 

 

The single LED bulb represents the LED panel.

 

Tim

[spokeshave]

sami2012, on 31 May 2016 - 1:49 PM, said:
For arduino code do you any modify in the code

In the code, look for this line:

 

volatile int ledPin = 13;      // the pin that the LED is attached to, needs to be a PWM pin.

 

Change the 13 to a 3.

 

That's it.

 

Tim

[pbunn]

Just a question - Why are you using a bipolar transistor rather than a power FET.
The base of the 3055 looks like a short to the Arduino with no current limiting resistor. Does the Arduino have a current limiting resistor on board?

A fet would have a high gate impedance and would not require current limiting and is typically what is used when a computer drives a power device. It would
be likely that the Arduino hasn't got the power capacity to drive the very low impedance base of the transistor and the device may operate in the linear
mode rather than as a switch and might overheat. An MTP 3055V might be a better alternative. It is a FET that has characteristics much like a 2N3055 or TIP 3055
bipolar transistor.

Also realize that on the TIP 3055 - the tab is the collector and usually must be insulated from ground.

Edited by pbunn, 14 August 2016 - 01:42 PM.

[pbunn]

A bipolar transistor is a current driven device. To turn it on - you must supply current into the base. The TIP 3055 is a low current gain device and requires quite a lot of current to turn it on
fully. Fully turned on - you will get a voltage drop of about .2 volts from collector to emitter. The amount of current gain is called beta.

The base of the TIP 3055 looks like a diode to ground and has very low resistance. Usually a resistor is placed in series to limit the current that the voltage
output on the computer delivers. Without a resistor - the output of the computer drives into a near short circuit and that will usually blow it out.

A power FET is a voltage controlled device and matches to the computer well. The computer delivers a voltage output (PWM) into a high resistance (the gate) and all works well. Usually a low value series
resistor is placed in the gate lead to suppress high frequency parasitic oscillations. Probably will work OK without one though.

IF you did not isolate the TIP3055 from ground - then likely you shorted the power supply - I don't think your circuit will work with the Arduino driving the base of the TIP 3055.

[spokeshave]

I'm not sure I can be much help - I only know enough about electronics to be dangerous. I do know that the TIP 3055 transistor works just fine for me as drawn above. I have been using the flat panel for a few months now. One difference possibly of note is that my case is plastic and is not grounding the tab.

 

Tim

[brucesdad13]

I hate when I release the magic smoke!    This is a great project and I look forward to making one someday 

[William Mc]

I went ahead and ordered some MTP 3055V Mosfets along with the other stuff.  I assume the wiring will be the same as with the TIP, and it will need to be insulated as well?

[pbunn]

Yes - unless you get the ones with the insulated tab. it will be obvious if they are  - the tap looks like plastic. I have plenty of fets and would have been glad to send you a few.  Also - is all that you are doing is varying the voltage on the display pin for brightness. If so - an Arduino is really, really overkill. A $1 pot would do the trick. I am not familiar with the display so I am not sure. id there is no power involved, then a small signal transistor like 

a 2n3904 with a base resistor would work.

 

AS I said - I probably have several hundred power fets from building VLF transmitters and would be glad to send you a couple.

[spokeshave]

Also - is all that you are doing is varying the voltage on the display pin for brightness. If so - an Arduino is really, really overkill. A $1 pot would do the trick.

The controller used a PWM output pin on the Arduino to vary the pulse width to the LED panel. You can't dim LEDs with a pot.

 

Tim

[pbunn]

Some panels have an analog voltage input to vary the brightness. I don';t think this one does.

 

Unless you are wanting to use a computer to control the display brightness - the Arduino  IS overkill.

 

All that the Arduino is doing is generating a PWM voltage to vary the input voltage to the panel.  This voltage then is filtered to DC. This can be done with any analog variable 1 amp supply or a cheap EBAY regulator.

 

I have used the EBAY switching regulators in many applications after building my own boards. I can't buy the parts for the price of a completed board and the circuit is identical.  It is very efficient because it is a switching supply.

 

http://www.ebay.com/...sd=191831957965

 

 

This analog supply is better suited as it has a digital output supply and a panel mounted type pot.

 

 

 

After a bit more EBAY looking this would be even cheaper ($2) and would work well also: These are also used in DIY dew heaters.

 

http://www.ebay.com/...pYAAOSwubRXKyGn

 

http://www.ebay.com/...tMAAOSwwpdW79GY

 

 

The first one comes from Miami and the second one ships from China. I just bought a panel and the second one. I'll see how it works. I have plenty of analog supplies to test it with also and quite a few of the first type regulator. I use one to regulate my portable LIPO supply to 8.2 volts for my Canon camera and it works very well with no detected noise although I do have a low pass filter between it and the camera for secondary.filtering. No flltering would be required for powering a LED panel

Edited by pbunn, 15 August 2016 - 10:18 AM.

[spokeshave]

Unless you are wanting to use a computer to control the display brightness - the Arduino  IS overkill.

The whole point of this build is to have a controller that simulates an Alnitak panel for which a computer can be used to control brightness and automate the collection of flats. That was fairly clearly stated in the first post.

 

It is a way to use a very large panel, suitable for large scopes, in automation software for a very small fraction of the cost of a comparable Alnitak flat panel.

 

If computer control and automation are not your goal, you're right. An Arduino IS overkill. However, for the intended and very clearly stated purpose, it is the perfect tool for the job.

 

Tim