brighten 12v light (LED?) over 2 minutes; low voltage trigge

This is the place for any magazine-related discussions that don't fit in any of the column discussion boards below.
remotecamera
Posts: 5
Joined: Sat Jul 05, 2008 6:08 am
Contact:

brighten 12v light (LED?) over 2 minutes; low voltage trigge

Post by remotecamera » Sat Jul 05, 2008 7:34 am

I wanted to ask you for some direction regarding a project I need help with.

I am using a motion-activated video camera system to record wildlife in the wilderness. The animals are fearful of sudden bright light. One solution I heard about is for the light to gradually brighten over a 1-2 minute time period. I am using a PIC chip-based circuit I purchased to control my camera in response to a heat/motion sensor. The board controls 12vDC light up to 3 amps. I want to use an LED light which should be below the 3amp limit of the motion circuit.

I am not experienced with building electronic circuits, although I can do some basic wiring. Can someone suggest what I would need or can someone offer to help me build a light controller that I can use to control my 12volt light (or LED if I switch to that)? I understand a phase-wave modulator would help with the dimming, but I don't know where to begin to start. I saw a summary of two Nuts & bolts articles on this kind of thing, but I have not purchased these articles yet. I am willing to pay expenses and time for help with this project.


Thanks

Bryan

User avatar
Viking
Posts: 87
Joined: Mon Apr 12, 2004 1:01 am
Location: Aylesbury, England
Contact:

Post by Viking » Mon Jul 07, 2008 10:20 am

Hello Bryan,
I should be able to design/build this for you, FOC.
As I understand it, your 'PIC' board supplies a switched 12VDC supply, when the sensor is activated? Is this correct.
Then you want a circuit operating on this switched supply to gradually bring on a 12V light. Please give specification of this light.
Also you mention a 12V LED light. Is this something that you are intending to build or do you have one already? If you have one already, give specification.
You say that the current limit of the controller is 3A, which I take to be the rating of the relay or switching device within the controller. What is the power source for the equipment, in other words will the power source deliver 3A?
Just one thought...have you considered infra-red illumination? Would obviously be invisible to the wildlife so no problem with sudden switch-on, but would produce essentially mono-chrome pictures, which might be the reason for you having discounted this approach.
Regards
Rob

Engineer1138
Posts: 458
Joined: Thu Feb 05, 2004 1:01 am
Location: Minneapolis, MN
Contact:

Post by Engineer1138 » Mon Jul 07, 2008 1:32 pm

Hello Bryan:
Have you selected a light yet? I have a general purpose control circuit board that I can reprogram for you if the power requirement falls within certain parameters. It would be a pretty simple software-only mod, so as long as you don't mind waiting "until I have time" there would be no charge except for the hardware.

I think my profile here has the wrong email address (and I can't get in to change it!) so if you don't get a response to a Private Message, you can leave me a message at http://softwarefromthefarm.blogspot.com

Every time I leave my email in a public forum I get spammed, sorry.

--lyndon

remotecamera
Posts: 5
Joined: Sat Jul 05, 2008 6:08 am
Contact:

Post by remotecamera » Mon Jul 07, 2008 5:52 pm

Yes, the PIC board supplies a switched 12VDC supply when the sensor is activated.

I want to use LED lights instead of the current automobile driving lights. I don’t have the LED lights yet. I was looking at several options of 1-3 watt LEDs in water-tight housings or I could put them into my water-tight case I already have built for the relay I am using now. The new circuit could be mounted into this housing along with some LEDs. I want to try and get about as much light as 110 watts of regular incandescent lights, so I was thinking of 6 or so of the extra-bright LEDs. I would be open to your suggestions if you have any ideas for what would be best to use.

I said the controller is 3A, but it is actually 4.2A, under 60 volts. I am not sure what the switch is. It is something inside the card that holds the PIC chip. The only power source I have is a 12V, 22amp-hour battery. As long as I stay under the 4.2A limit, I can use the switch in the PIC chip circuit board. The specs on the card I am using are here:

http://www.pixcontroller.com/PixRF/PixRF_Specs.htm

I used Infrared with my older analog camcorders by modifying the camcorders to support daytime monochrome recording and night-time IR. But the new HD camcorder is not so easy to modify and would somewhat defeat my purpose of going with HD to get the best quality color photos I can get.

And now that I think about the time to go from darkness to full brightness, I will probably be best off if we can make it adjustable. This is because I usually set the system to record for 2.5 minutes, and I don’t want to waste ½ of my time in darkness or low light to wait for the light to get bright enough for the camera to see it. It may be 30 seconds is long enough. I may not know until I actually have some critters visit my site before I will be able to set the fastest time possible while not scaring the critters away.

User avatar
Viking
Posts: 87
Joined: Mon Apr 12, 2004 1:01 am
Location: Aylesbury, England
Contact:

Post by Viking » Tue Jul 08, 2008 9:08 am

Hello Bryan,

Had a look at the control board specs, and although it doesn't explicitly say so, it implies that there is a 'volts free' switch connected to J14. The annoying thing is it doesn't specifically say so, and I cannot see anything resembling a relay on that board. I thought the white rectangle to the left of the connect might be it, but on further inspection this looks like a silkscreen trace where you are supposed to write the revision number of the board.
So the controlling element may be a surface mount MOSFET, which means that the switch is probably not ‘volts free’. Not a problem if it is, but you need to connect the power supply for the external light in the way shown in the article.

Making the brighten-up period adjustable will not be a problem.

Regarding LED replacement of incandescent bulbs, I would have to do some research in order to establish just how many and what type would be needed to replace the bulb with the equivalent light output.
One thing occurs to me, you would probably have to adjust the color balance of the pictures produced by the LED light because it is spectrally different than say incandescent light. This is because there is no such thing as a ‘white LED’. Led’s produce a single wavelength and the white light is produced, I think by using a IR LED and a white phosphor screen, in the same way as fluorescent strip lights.

Regards
Rob

Engineer1138
Posts: 458
Joined: Thu Feb 05, 2004 1:01 am
Location: Minneapolis, MN
Contact:

Progress

Post by Engineer1138 » Wed Jul 09, 2008 8:05 am

After talking to Bryan I did some tests last night.

My initial idea was to use a constant-current charged capacitor to get a linear slope and use this voltage as the setpoint of a current controller to brighten the LED. It worked like a charm first time out the gate (yay, me!!), but the problem was that I forgot vision is logarithmic, not linear. Duh! So while the current increased smoothly, to the eye it didn't look quite "right."

Next try was replacing the linear voltage with an exponential rise. Didn't look better; it really wants to be logarithmic. I considered adding a log amplifer stage since I am using a quad op-amp, but that would be too much work.

So it looks like either the linear rise, or (most likely) I'll implement a log fader with a microcontroller. Sometimes it's nice to have the hardware already designed and built and waiting for application :-) Although this project seemed like it was crying out for an analog solution.

Rob:
I would be interested in seeing how you approached this problem.

lyndon

User avatar
Viking
Posts: 87
Joined: Mon Apr 12, 2004 1:01 am
Location: Aylesbury, England
Contact:

Post by Viking » Wed Jul 09, 2008 9:42 am

Hello Lyndon/Bryan,

Yes, the linear/analog approach is the simplest, as you correctly point out, vision (at least human) is not linear. I suspect that you would need to have small steps at the beginning, gradually increasing towards the end, for a human to perceive this as a linear increase in brightness. Is this so, did you establish this during your experiments?

But our target audience is not human, so who knows how say a beaver would perceive a lamp with a linear current change. Also the object of the exercise is to bring on the bulb without alarming the wildlife, not to win marks for the most pleasing effect. If needed, I think I could modify the constant current approach to have several changes in the constant current source value, so the current would initially increase at a rate A to a certain output duty cycle, then the current would switch to rate A*0.5.
But I suspect that the simplest constant current approach would be sufficient for this application.

Some thing that occurs to me. If we are driving a incandescent lamp using the constantly varying duty cycle drive, the thermal mass of the filament will smooth out any possibility of the light flickering. But if we are driving a LED lamp, then it will respond to the PWM signal and appear to flicker. Us humans do not perceive this if the rate is above about 25pps, but again we are not the target audience. If the pigeon looks a motion picture at 25 frames/second, it sees the individual frames separately and so as far as it is concerned, it is looking at a slide show. You would need to increase the frame rate to around 250 frames per second before the pigeon thought it was looking at constant motion.
There could be other critters out there who perceive even higher frame rates as individual frames, maybe birds of prey?

I did a quick look at LED’s v incandescent lights and came up with the following.
A 100W incandescent lamp has an output of about 1600 lumens.
A white, 3000mcd 5mm LED has an output of about 1 lumen and a 1W Luxeon assembly has an output of about 25 lumens and a 5W Luxeon Star about 120 lumens.
So to replace the 100W filament bulb you would need 64 Luxeon assemblies, 13 Luxeon Stars or 1600 5mm white LED’s. I could have gone horribly wrong here, maybe a lighting expert can comment here and point out where I’ve gone wrong.

By the way, do you have any use for a few million gallons of water? There is plenty spare available here, it’s been raining solidly all day!

Regards
Rob

Engineer1138
Posts: 458
Joined: Thu Feb 05, 2004 1:01 am
Location: Minneapolis, MN
Contact:

Post by Engineer1138 » Wed Jul 09, 2008 12:37 pm

Yeah, we could use the rain: it's been unusually dry for the last month.

There's nothing wrong with your analysis, but except for flood illumination, light is not normally needed in all directions so LEDs are more "efficient" than they seem. MrAl can probably expand on this. IIRC, a 1W white LED is roughly equivalent to a 5W incandescent when measured appropriately. And that efficiency is increasing almost daily as new LEDs come out.

I have one more idea I want to try with the linear circuit before I give up on it; I'll see if I have time tonight.

ashleycameron
Posts: 1
Joined: Wed Jul 09, 2008 1:04 pm
Contact:

Post by ashleycameron » Wed Jul 09, 2008 1:09 pm

Viking wrote:Hello Lyndon/Bryan,

Yes, the linear/analog approach is the simplest, as you correctly point out, vision (at least human) is not linear. I suspect that you would need to have small steps at the beginning, gradually increasing towards the end, for a human to perceive this as a linear increase in brightness. Is this so, did you establish this during your experiments?

But our target audience is not human, so who knows how say a beaver would perceive a lamp with a linear current change. Also the object of the exercise is to bring on the bulb without alarming the wildlife, not to win marks for the most pleasing effect. If needed, I think I could modify the constant current approach to have several changes in the constant current source value, so the current would initially increase at a rate A to a certain output duty cycle, then the current would switch to rate A*0.5.
But I suspect that the simplest constant current approach would be sufficient for this application.

Some thing that occurs to me. If we are driving a incandescent lamp using the constantly varying duty cycle drive, the thermal mass of the filament will smooth out any possibility of the light flickering. But if we are driving a LED lamp, then it will respond to the PWM signal and appear to flicker. Us humans do not perceive this if the rate is above about 25pps, but again we are not the target audience. If the pigeon looks a motion picture at 25 frames/second, it sees the individual frames separately and so as far as it is concerned, it is looking at a slide show. You would need to increase the frame rate to around 250 frames per second before the pigeon thought it was looking at constant motion.
There could be other critters out there who perceive even higher frame rates as individual frames, maybe birds of prey?

I did a quick look at LED’s v incandescent lights and came up with the following.
A 100W incandescent lamp has an output of about 1600 lumens.
A white, 3000mcd 5mm LED has an output of about 1 lumen and a 1W Luxeon assembly has an output of about 25 lumens and a 5W Luxeon Star about 120 lumens.
So to replace the 100W filament bulb you would need 64 Luxeon assemblies, 13 Luxeon Stars or 1600 5mm white LED’s. I could have gone horribly wrong here, maybe a lighting expert can comment here and point out where I’ve gone wrong.

By the way, do you have any use for a few million gallons of water? There is plenty spare available here, it’s been raining solidly all day!

Regards
Rob

Thanks for your information mate..
by laptop battery dell

remotecamera
Posts: 5
Joined: Sat Jul 05, 2008 6:08 am
Contact:

Post by remotecamera » Wed Jul 09, 2008 5:55 pm

This reminds me of another idea. Perhaps you can rig something where the lights come on one after another. So if we have say, 6 LEDs (or 25 or whatever), then the first one might ramp up quickly, but if its only one LED, then it may appear to be pretty dim to the animal. Then at some point (at 50% brightness perhaps), you start to ramp up the 2nd light using the same fast attack. If you did that in a series, applying the rapid attack curve to each light independently so at the end of 1 minute, all the lights are at 100% brightness, it might be just fine. Would this make it any easier? Maybe you could also just turn them on to 100% one at a time. Its hard to say at what point it would be noticed.

I had not thought about the perceived frame rate an animal sees things. And I wonder what the effect of flicker would be. I think we want to simulate the sun rising (sort of). The animal is bound to notice the light at some point, but if it is "natural" to them, then maybe they wont be surprised. If nothing else, it will be an interesting experiment to be able to put this into the woods and capture on video what the reaction of the light effect is. Perhaps we can try a few different timings/curves/frame-rates and see how the animals response changes.

I certainly do appreciate the information and the help to get this together. I will be checking my system again this weekend after a week of being in the field. I just put out the new RF sensors so I have 3 RF sensors watching the target area which gives me much more coverage than the original wired-sensor I was using. The RF sensors are like those X10 home motion sensors that send the detection signal to an RF receiver (the board I provide a link to above). I am still using the 100watt light, but so far, almost nothing has come in at night but one porcupine and a snow shoe hare. And with the hare, when the light came on, it sat there for about 10 seconds, and then bolted. It was really surprised by the sudden light. I think the coyote may be in the area, but they are not even coming close so far. I am planning to cover my gear with a hide-scented material so it will be less obvious and may smell less too.

remotecamera
Posts: 5
Joined: Sat Jul 05, 2008 6:08 am
Contact:

Post by remotecamera » Wed Jul 09, 2008 6:27 pm

Here is a video clip of that hare that ran fast when the light came on.

With this, you can get an idea of how much light the camera sees with 110 watts of incandescent light. I can brighten this up a lot in my editing program, but this is the raw footage of my site to give you an idea of how it looks recorded on HD. I slowed this down by about 25% so you have time to actually see the hare run away. It went from total darkness to what you see here. The camera came on as soon as the light came on.

http://remotecameras.blip.tv/posts/?top ... =fast-hare

User avatar
Viking
Posts: 87
Joined: Mon Apr 12, 2004 1:01 am
Location: Aylesbury, England
Contact:

Post by Viking » Thu Jul 10, 2008 3:58 am

Hello Bryan,
It would not be difficult to daisy-chain a load of linear drivers together, with the first one as the master. Actually this would make quite a good lighting effect, which could be usable in other applications.

I suspect that the light needs to come on very slowly , and then speed up towards the end. After all, the biggest change as far as the animal is concerned is the change between there being no light and there being some light. Once the light is on, it could brighten up relatively quickly.

I think a field test is called for. Rather than speculating about this, just build and test a linear setup for the lamp you are using at the moment.

Actually talking of the lamp you are using, if it is a 110W, 12V auto lamp, then it will be drawing about 9A. So how are you controlling it if the control board will only allow a maximum of 4.2A?

Rob

Engineer1138
Posts: 458
Joined: Thu Feb 05, 2004 1:01 am
Location: Minneapolis, MN
Contact:

Post by Engineer1138 » Thu Jul 10, 2008 5:11 am

A thought:
Have you considered using infrared LEDs so the animals won't notice at all, or do you need white light because you want the recording in color?

User avatar
Viking
Posts: 87
Joined: Mon Apr 12, 2004 1:01 am
Location: Aylesbury, England
Contact:

Post by Viking » Thu Jul 10, 2008 8:36 am

Hello Lyndon
We have covered this one in post 2 and 3:-
"Just one thought...have you considered infra-red illumination? Would obviously be invisible to the wildlife so no problem with sudden switch-on, but would produce essentially mono-chrome pictures, which might be the reason for you having discounted this approach."
Regards
Rob

remotecamera
Posts: 5
Joined: Sat Jul 05, 2008 6:08 am
Contact:

Post by remotecamera » Thu Jul 10, 2008 4:24 pm

On the power limits issue; I am using a relay that actually powers the light. The relay can handle 30amps or some such amount. It works well. I have the lights positioned so one is aimed onto the foreground and the other is aimed directly out to the target area. It might be fine with just one on the target area, so perhaps 50watts would be all I need. I felt with the 22ah battery, I could afford to go the extra wattage.

I agree; the most important part of the light is at the beginning. WE want to make it so the animal has no awareness of the light coming on. Once its on and they are busy sniffing the lure or eating the bait, then they hopefully wont notice it.

Lyndon has offered to put something together for me to on the circuit stuff. I am good with software, but not very good with hardware other than basic soldering and connecting things together. Lyndon thinks he can put something together that will give the effect I need.

My motivation for the LEDs is I want to leave the unit unattended for a month or so at a time, because in the winter, I would have to travel several miles on snow shoes with a sled and a laptop to service the camera where I have it located. In the summer, its easy access so I can trouble shoot it now and get ready for longer term deployment this winter.

AS it is now, with the 110 watts and relay, its only good for a 7-10 days. Once the voltage gets below a certain level, it seems like the relay no longer triggers, so I don't get the dimmer lights like I have without the relay. I could also drop one of my lights and go with 55 watts as an alternative.

Here is a bear I had visit my camera two weeks ago:

http://remotecameras.blip.tv/file/1052926/

I slowed it down at the end because the camera shut off too early. I hope the new sensors I am using will fix that problem. I often wonder if the next time I am there, when the bear will decide to visit the site. So far, I have not been face-to-face with a bear yet! They are usually afraid of people so they stay clear.

Post Reply

Who is online

Users browsing this forum: No registered users and 14 guests