Powerful Aquarium Lighting
#81
Posted 07 June 2012 - 05:31 AM
#82
Posted 09 June 2012 - 02:53 PM
Thanks for keeping me up to date with the NCL driver and its progress, you've saved me a lot of time.
I feel your initial driver solution sounds the most reliable, so I think I am going to follow suite and have a go at making that one.
I will get a shopping cart together and post it for you to give it a once over if you don't mind.
Also just to refresh your drivers provide 3 channels per attiny, and each of these can be independantly controlled? Is this done by having 3 PWM signals going in to the attiny?
Finally we discussed inductors in length and I have some coil craft samples but am a little unsure of which are which, the attiny require 1 inductor @ 100uh right? Is that likely to be the smaller ones on the right in this picture with 101 written on them? These samples were the 100uh for the attinys and the 33uh for the NLCs (the larger ones on the left have 333 written on them.
Thanks in advance,
Andy
Andre21,
Welcome to the forum!
This project is for an aquarium lighting enclosure, Magpie who is a key contributor is already sucessfully using one of the drivers discussed in his own projects for low power domestic lighting, so this information should be useful for most LED projects.
I would like to say this project is much further on, but in reality I've have spent a fair bit of time investigating the LED drivers and the dimming control, also with the release of the netduino GO I have decided to completely rethink my project in a more modular format, so it is still in the testing phase at the moment.
If you are interested in aquarium lighting/ control I would recommend having a go a building Magpies first driver from this forum using the AtTiny chips, you can look at dimming with the chips in various ways and this really comes down to the complexity you want in your aquarium. For example a simple aquarium lighting controller could have 30 3W Leds (go for decent cree LEDs for efficientcy and lumenums) you would then have a mix of royal blues and bright whites say 15 white 15 blue. Technically you could split these in strings of 5 accross 2 drivers and you would get a really efficient aquarium lighting enclosure with 6 channels of control. However I never seem to make life easy (or cheap) for myself and I am going for a less efficient more expensive approach of 1 LED, 10 drivers, 30 channels. This will allow me to do some effects such as cloud cover and lightening.
If you want to start a project like this, I would suggest getting some LEDs and build a driver and some way of controlling the brightness. Once you are happy with the setup start looking at driving all the LEDs and how you are going to power them. In terms of power the main thing to consider is voltage drop and protecting your sensitive LEDs from over current. One way to do this was suggested by Magpie, by building an additional circuit. If you have string of 5 LEDs and each LED had a voltage drop of 3v you would look for something like 18v -24v power supply.
hth
andy
#83
Posted 11 June 2012 - 11:14 PM
#84
Posted 16 June 2012 - 12:30 PM
The details are on my domestic lights project . It is based on the Attiny but with some improvements.
I hope this helps.<br>
Edited by Magpie, 17 June 2012 - 06:46 AM.
#85
Posted 16 June 2012 - 09:30 PM
Thanks again,
Andy
EDIT: Found the link http://www.eevblog.c...ighting-project
#86
Posted 16 June 2012 - 10:03 PM
#87
Posted 17 June 2012 - 01:44 AM
#88
Posted 27 June 2012 - 11:48 PM
#89
Posted 28 June 2012 - 12:22 AM
Yes the 5v/12v sounds fine, it gives me a little option if and when (its inedvitable really) that I get another larger tank!
I have an avr ISP mk2 will that do the trick?
I noticed on your schematic you have the 861As, is your program too big for the 461As?
Diode and fuse sounds better and cheaper than a FET.
I was contemplating the same dip or soic and I thought until I have something up an running theres no point complicating things with surface mount.
I look forward to your parts list.
In terms of connectors that is a good question :
smaller than jst
jst
polarised
screw
my favourite
#90
Posted 28 June 2012 - 01:12 AM
But the Attiny will be Dip, just so I am sure that I can program it, I have never done an ISP header before and a cock up here would be catastrophic, the big inductors will be dip, and the connectors and fuse holder.
I think the program will be very small, probably the tiny261 will be fine.
I assume so but I really don't know.I have an avr ISP mk2 will that do the trick?
Those connectors look good (your favourite).
I think I want the screw in onthe flying lead and the L shaped black header pins, probably I will go for 5mm.
I will draw the pcb at 5mm.
#91
Posted 04 July 2012 - 11:09 PM
#92
Posted 04 July 2012 - 11:11 PM
Attached Files
#93
Posted 05 July 2012 - 03:48 PM
#94
Posted 05 July 2012 - 06:20 PM
#95
Posted 05 July 2012 - 11:00 PM
I know I am jumping in a little late here, but I just looked at the picture with the inductors. Aren't the ones on the left actually 33mH instead of 33uH?
I looked at this link coilcraft series
I must admit I don't know.
that's right.The bit in the bottom right corner is the black regulator right?
ps. I changed the logo, it will come out on the silkscreen.
I almost mounted the torroid footprints upside down, that would've been a disaster.
I just sent the gerbers off to seeed studio last night.
If you want I can send you some, I will test them first though.
#96
Posted 06 July 2012 - 10:30 AM
Attached Files
Edited by Magpie, 07 July 2012 - 01:57 PM.
- mcinnes01 likes this
#97
Posted 07 July 2012 - 07:58 AM
Attached Files
#98
Posted 07 July 2012 - 01:54 PM
Thanks for pointing out the missing caps from my BOM, it was also missing R1.
All the caps are just decoupling or filtering caps of 100n 0805 and the Resistor was just a 10k 1206.
I cant believe I made this mistake, It just goes to show how easy it is. I am definitely not a computer.
I haven't gone exhaustively through your list as it seems your not quite decided the best way to go anyway.
As far as boards go I would go the modular approach, so if a board fails you can have spares, x channels per board.
Maybe 3, maybe 8, maybe 16. Design decision.
Also that way you could have some boards at 12 volts some at 5.
Actually on the subject of design decisions, the way just about every project I have worked on has been is that you never design it right first time anyway.
So there is no point trying to perfect the initial design. The moment the first working version comes out, somebody sees things that they hadn't noticed before.
So I think,
- Get something out there,
- Get something out there that isn't impossible to change or fix.
- Keep changing it till you think it's right.
I am writing as I think so the conclusion of what I think you should do is 3 channel boards (wait till I confirm that the design works) start with 5 or 12 volts, (your decision) and see how things go.
As far as 5 or 12 volts for 1 led.
5 volts can be slightly more efficient but, will it always be one led, because you can't easily change it.
If you go 5 volts you may have a noisy supply to the micro, this may or may not be a problem for the Current sensing on the Attiny.
If you go 5 volts you don't need the step up bc847 and it's two resistors. But you do need a different, logic level FET, just like AHellene's circuit.
5v you can use a smaller inductor.
I would use 12 volts but there could be a case for 5v and event a case for 7 volts.
If you want to use my circuit with 12v
just use a different TVS (12v operating voltage).
and remove everything including C3 to L2 reading left to right on the schematic.
Jumper the 24 volt rail to the 12volt rail.
If you want to use my circuit with 5 volts.
just use a different TVS (5v operating voltage).
and remove everything including C3 to U1 reading left to right on the schematic.
Jumper the 24 volt rail to the 12volt and the 5 volt rail.
you don't need the 3 x step up bc847 and short r13 to r15 and open circuit r16 to r18
It gets a bit harder as you may need a clean supply for the micro. I don't know.
Personally I wouldn't bother. It probably would be ok, but I think 12v will give you what you need.
No the netduino can only source less than 1 amp.Do you think this side of the circuit can be powered practically off the netduino
I would power the netduino from my 12v supply. Maybe via a resistor to just take away a bit of the heat.
That is a mistake, it was left there from some other current sense resistor that I had considered, please disregard.Also one of the resistors is marked as a bit wide, is this the foot print or the value and does it matter?
ps. You probably don't want to populate r28,r29, r30, I just put the footprints in for emergencies. I think 100mOhms if the lowest we could go anyway.
Also you don't need R12, this is also a contingency footprint.
You will loose very little running 1 led at 12v vs 1 led at 5. Only really in the higher series resistance of the inductor because you need a bigger inductor.Perhaps even provide a jumper option, where by you can select 5v or 12v and swap out the power supply accordingly?
Just run at 12v.
Steff shield is currently running 3 channels of 7 leds at 24 volts and at about 500mA. (XTE warm white) No heating problems on the shield and very little heat sink on the leds.How many ma are your running your Crees at in this circuit?
I will do 7 leds as well with my 24 volt prototype.
I want to move to 24 - 48. but will probably have to settle on a 24-36v version and 36-48v version, the black regulator doesn't have a wide input range. I just need to change a resistor value to switch between the two variations.
One thing that will work for you is the input circuitry on the Dimmer channels, you should be able to put your square wave straight in.
The input filters use 10k and 0.1uF which gives 166 Hz. If you can give it a pwm above 15KHz then your filtering is done.
BOM is corrected, thanks for finding the errors.
Attached Files
#99
Posted 14 July 2012 - 11:43 AM
Attached Files
#100
Posted 14 July 2012 - 01:03 PM
I made some changes, added to new column that flags the changes.Please can you confirm that I haven't made any mistakes?
I left the inductors at 100u you can probably go smaller as well, probably down to around 10u. So many variables to choose these. 100 will be fine if you have them but say 33u might be better, but only marginally.For the 5v version, you mentioned I need a smaller inductor, can you recommend any that would be good for the job? Or will either of the ones I all ready have do the trick which seem to be 100uh and 330uh?
Look for operating voltage of 5v, but since you are providing a regulated low noise supply off board, the only thing this will do is protect against reverse polarity.You also said I need a different TVS diode, what would you recommend for this? PS I am probably going to order from Mouser in the UK.
Also bear in mind that you need to provide the regulated low noise supply off board because this is what the 12->5v part of the supply was doing.
No the sense resistor is about right, it will be ridiculously efficient but may be prone to noise.I guess the current sense will change in value right? Do you have the equation for 700ma from a 5v input?
You may want to try 200mOhms. I tried to layout my board inputs as "kelvin" inputs. Unless you do this well your current sense inputs will be prone to noise.
Any prototypes you may want to use up to 500mOhms.
Calcs: 100mOhms at around 700mA the current sense should read .7 x .1 = 0.07 volts at full load.
We are not getting all that 700ma through the sense resistor so say we get 500mA. ( Some goes through the Shottky). = 0.05v
then we multiply by 32 in the internal gain stage = 1.6 volts. our full scale may be 3.3 volts so we get a reading of 500/1024 at full current.
Your pwm outputs should feed straight in to these as long as they are 5v full scale. Dont connect the 5v but connect the rest.Finally the parts for the dimmer channels, are these already on the BOM and do the values still stand for 5v?
Apparently they are in transit.Have you recieved your boards yet and got one soldered up?
I have the parts from Digikey. The murata inductors look a mess, they must've been made by the apprentice. The last ones looked great.
I have also got some really fine prismatic diffusing glass on order, it breaks up the leds nicely. The plastic ones are good too but they look shit when they are turned off.
Good luck
Attached Files
0 user(s) are reading this topic
0 members, 0 guests, 0 anonymous users