So I was looking at the chips in the Radion, AI Sol, and Maxspect Razor units. I figure those chips sets are the best to emulate. Both the Radions and the AI's did NOT use and warm white chips while the Maxspect does. Any reason why Maxspect added warm whites??? It looks like you don't need the warm whites if you have red and green chips in the unit.
LED Spectrum Assistance and Discussion - Moding existing units
- Thread starter kirkwood
- Start date
The razor and mazarra series do not use red or green leds. The warm whites help provide some red and green. I have three mazarra p units and have honestly never missed or wanted green or red leds. I dont like the funky coloring they add and they can easily overpower other colors. Just use sparingly if you do use them.
I have been happy with my current lights. The only change I will probably make is to add another 410/420 to each unit. Easily done as my chips are modular. I think a lot of the issues with coloration/growth people have with leds is the lack of 410/420 and not lack of reds and greens. Plus the "pop" is so much better to me than just the normal florescence that comes from the blues. The pop stays but the actual purple color is easily muted by other colors. Good luck with your builds!
I have been happy with my current lights. The only change I will probably make is to add another 410/420 to each unit. Easily done as my chips are modular. I think a lot of the issues with coloration/growth people have with leds is the lack of 410/420 and not lack of reds and greens. Plus the "pop" is so much better to me than just the normal florescence that comes from the blues. The pop stays but the actual purple color is easily muted by other colors. Good luck with your builds!
So here's an channel issue. We seem to agree the 410-450 need added. We will call them violets. I'm thinking 6 leds in a zig zag every other row down the center. They need to be on the own channel. They run on different specs than 455 blue and white. The only quick easy way is to remove an led and short the pad. Insulate somehow.
Thermo paste a star Base on top. And wire them together in a channel. Most led are close to the same price with the stars. I,don't know what this means for heat management. I have an infrared therm
So many issues to work out.
The 2 Tao white ones I bought have an actual finned heat sink. The MM is a plain sheet of AL
Incidentally 455 blue and cool whites are the exact same LEDs. They spray a coating of phosphorous on the top and the blue light makes the coating glow white. Blues and whites can be run on one channel.
Which means you can exchange all the blues for 400-450 range. Leave the whites and 455 blues on the other channel.
but...
My MM has 2 different drivers. One is 80-100v the other is 55-95v. They look exactly alike and look like the same CBA. So maybe MM are running a higher power on one color over the other. Either one will run the LEDs bright. They are the same base type LEDs. I'm not sure which channel needs to be derated.
Just assume an led takes 3-4 volts to run. 55 LED/2 channels equals approx 27LEDs . 27 * 3 = 81v derated value. 27*4= 98v overdriven value. 27 LEDs *3 w= 71 watts+ rated driver needed.
Maybe the odd extra led made a difference.
All the drivers I seen are internally adjustable with trim pot which means a remote dimming pot can be installed. This makes the driver a voltage limting dimming not current limiting dimming. The effiency of the led will decrease if you don't keep voltage high and limit current instead. Not sure it matters that much.
I still like the DMX led drivers. They come cheap with remotes or ipad software. The only problem is all I have seen Is 36 v . This will only run 8-10 LEDs.
The quick and dirty:
Pull out select blue and white LEDs for good spread, bypass them and paste violet stars on top and a separate dimmable driver. Adjust down the white and blue channel to make up for the missing LEDs. maybe add remote dimming pot to existing drivers.you can opt to install 5 w violets and derate them, and they will last
Disclaimer:
Thermo paste a star Base on top. And wire them together in a channel. Most led are close to the same price with the stars. I,don't know what this means for heat management. I have an infrared therm
So many issues to work out.
The 2 Tao white ones I bought have an actual finned heat sink. The MM is a plain sheet of AL
Incidentally 455 blue and cool whites are the exact same LEDs. They spray a coating of phosphorous on the top and the blue light makes the coating glow white. Blues and whites can be run on one channel.
Which means you can exchange all the blues for 400-450 range. Leave the whites and 455 blues on the other channel.
but...
My MM has 2 different drivers. One is 80-100v the other is 55-95v. They look exactly alike and look like the same CBA. So maybe MM are running a higher power on one color over the other. Either one will run the LEDs bright. They are the same base type LEDs. I'm not sure which channel needs to be derated.
Just assume an led takes 3-4 volts to run. 55 LED/2 channels equals approx 27LEDs . 27 * 3 = 81v derated value. 27*4= 98v overdriven value. 27 LEDs *3 w= 71 watts+ rated driver needed.
Maybe the odd extra led made a difference.
All the drivers I seen are internally adjustable with trim pot which means a remote dimming pot can be installed. This makes the driver a voltage limting dimming not current limiting dimming. The effiency of the led will decrease if you don't keep voltage high and limit current instead. Not sure it matters that much.
I still like the DMX led drivers. They come cheap with remotes or ipad software. The only problem is all I have seen Is 36 v . This will only run 8-10 LEDs.
The quick and dirty:
Pull out select blue and white LEDs for good spread, bypass them and paste violet stars on top and a separate dimmable driver. Adjust down the white and blue channel to make up for the missing LEDs. maybe add remote dimming pot to existing drivers.you can opt to install 5 w violets and derate them, and they will last
Disclaimer:
DIY optical spectrum analyzer
<div class="gc_ifarem_title">Public Laboratory: Build a $10 USB visible-light spectrometer - YouTube</div>
http://publiclab.org/wiki/spectrometer">http://publiclab.org/wiki/spectrometer</a>
Check out the wiki open source link!,,
I been working on a water proof one
To drop in the tank for testing led units.
The best idea I came up with is a upside down 180 degree Periscope J tube. then the camera doesn't have to be submerged. The opening is at the bottom a J tube facing the leds then two mirrors reflect the light back up and through the slit and up into the camera.
These people at the wiki open source are all about calibrating any version of a reef DIY kit. I sent them some plans and we are working on it so we can have some firm numbers on a kit.
It's a matter of time before led units are self calibrating. I imagine dropping optical sensor pucks in the tank for feed back to the led processor. They are just cheap ars web cams and some software.
<div class="gc_ifarem_title">Public Laboratory: Build a $10 USB visible-light spectrometer - YouTube</div>
http://publiclab.org/wiki/spectrometer">http://publiclab.org/wiki/spectrometer</a>
Check out the wiki open source link!,,
I been working on a water proof one
To drop in the tank for testing led units.
The best idea I came up with is a upside down 180 degree Periscope J tube. then the camera doesn't have to be submerged. The opening is at the bottom a J tube facing the leds then two mirrors reflect the light back up and through the slit and up into the camera.
These people at the wiki open source are all about calibrating any version of a reef DIY kit. I sent them some plans and we are working on it so we can have some firm numbers on a kit.
It's a matter of time before led units are self calibrating. I imagine dropping optical sensor pucks in the tank for feed back to the led processor. They are just cheap ars web cams and some software.
OK, so I don't understand why you think the Violets need a separate driver? I have 3 true UV chips (~405-420nm) in each of my 3 units (2MM 1eshine). All the UVs run on the same channels as my blue 450nm without any issues. I am ordering my chips from Reefbreeders so I will check wiht them if their 420nm violets need a separate channel but I doubt it.
So here is where I'm at - I am currently running 170 3w chips of which 98 are 450nm blue, 63 are 10-12K cool white, and 9 are 405nm UV.
After researching this the past few weeks, my plan is to focus mainly on adding 420nm and 480nm chips. That is really where I feel I'm missing out and everything I've read shows this is the main reef spectrum. So I plan to add 12 420's and 12 480's. The 420's will be added to my blue/actinic channels. The 480's will be added to my white channels because I feel like the 480's will make the tank super blue and diminish the POP of the corals fluorescing.
Then to touch on the green-red spectrum where I also feel I am lacking I will add 7 660nm hyper reds and 7 3500k warm whites. I opted against adding any green chips (a la Maxspect) as the warm whites will cover that range. I don't want to overdo it with these chips and make the color of the tank look funky so I'm just adding what I feel is a minimum amount.
A couple of my final questions. Anyone disagree with what I said about the 480s being better on the white channel? Also, years ago everyone was worried that UV chips could harm coral if there were too many. Now it seems like LED makers see the importance of the 420nm spectrum and hence we see more chips in the newer units. Anyone think I'm adding too much purple??
So after I make the additions I will be running - 74 450nm blue, 49 10-12K cool white, 7 660nm Hyper Red, 7 3500K Warm White, 12 480nm Royal Blue, 12 420nm Purple, and 9 405nm UV.
1)I dont think you are using too much purple. The visibly purple light is easily washed out by the blue and whites. You should end up with an effect similar to the old school MH with actinic supplement look aka great shimmer and "pop". You also seem to be using the reds sparingly, which is good. More than a little red will really affect the visible light color. Great call on no greens imho.
2)on my maxspect mazarra the 410/420nm "uv" are on the same channel as the warm whites. Maxspect claims that you can use any of their replacement or included leds on any of the four channels. Dont know if that applies to or helps you.
3) several units, including mine, seem to use the blue:royal blue in about a 1:1 ratio. Cool white:warm white around 2:1 and total white:total blue about 3:4. That may be a good starting point for you with the purple and reds added in and spaced apart. In those ratios I am very happy with my ability using 4 channels to get great visible light coloration with the ability for very noticeable changes throughout the day. Remember that the whites, especially the cool whites, will easily take over the visible colors at higher settings. I usually have my blues about 10% higher to avoid washout but not give me a "windex" look. Just an observation if you think you need a higher ratio of blue:white. Some units use about 2:1 blue:white.
2)on my maxspect mazarra the 410/420nm "uv" are on the same channel as the warm whites. Maxspect claims that you can use any of their replacement or included leds on any of the four channels. Dont know if that applies to or helps you.
3) several units, including mine, seem to use the blue:royal blue in about a 1:1 ratio. Cool white:warm white around 2:1 and total white:total blue about 3:4. That may be a good starting point for you with the purple and reds added in and spaced apart. In those ratios I am very happy with my ability using 4 channels to get great visible light coloration with the ability for very noticeable changes throughout the day. Remember that the whites, especially the cool whites, will easily take over the visible colors at higher settings. I usually have my blues about 10% higher to avoid washout but not give me a "windex" look. Just an observation if you think you need a higher ratio of blue:white. Some units use about 2:1 blue:white.
I FINALLY put the order in today for the new 3w Bridgelux chips from Reefbreeders. I ordered exactly what I stated in my previous post. Here is a picture of how my lights are arranged in the canopy. Pay no attention to the very thin and long unit in the front - that is my moonlight. I worked out the arrangement today for the placement of the new chips. The main things I wanted was to have only UV 405nm, Purple 420nm, and Royal Blue 450nm on my blue/actinic channels, and the Blue 480nm, cool white 10-12K, red 660nm, and warm white 3500K on the white channels. I also wanted to achieve a 2:1 ratio of blue to white light.
Your feedback is appreciated. Better now than after I start soldering.
<fieldset class="gc-fieldset">
<legend> Attached files </legend>
>
class="gc-images" title="Lights1.jpg[/IMG] style="max-width:300px" /></a> </fieldset>
Your feedback is appreciated. Better now than after I start soldering.
<fieldset class="gc-fieldset">
<legend> Attached files </legend>
LEDs on the way,
I'm still waiting for some. II don't know what the heck my plan is. I just order a bunch and see what's up.
221239054167 eBay number. This is a example of a dimmer that will work. I didn't have time to see if it only uses a pot or will also except PWM input. It seems that since the driver has to run so many series LEDs and has the power supply included the choices are more limited.
I did see some PWM adaptor boards. I didnt know what I was seeing when I came acrossed them. These boards take the place of the pot to provide a PWM input for electronic control. very cool.
I'm curious to see if 450 will completely out shine the lesser violets. The 450 are on the border for matching the violet specs.
Do you have a good volt meter to measure the the channel voltages on the drivers to get a base line before you tear up stuff?
I'm still waiting for some. II don't know what the heck my plan is. I just order a bunch and see what's up.
221239054167 eBay number. This is a example of a dimmer that will work. I didn't have time to see if it only uses a pot or will also except PWM input. It seems that since the driver has to run so many series LEDs and has the power supply included the choices are more limited.
I did see some PWM adaptor boards. I didnt know what I was seeing when I came acrossed them. These boards take the place of the pot to provide a PWM input for electronic control. very cool.
I'm curious to see if 450 will completely out shine the lesser violets. The 450 are on the border for matching the violet specs.
Do you have a good volt meter to measure the the channel voltages on the drivers to get a base line before you tear up stuff?
I expect the 450 to completely outshine the violets. However where the violet will show is in coral fluorescence in some different visible spectrums. There is a very interesting graph (Figure 8) in this article that shows the different colors that fluoresce in the various spectrums of LED light. http://www.advancedaquarist.com/2012/10/aafeature">http://www.advancedaquarist.com/2012/10/aafeature</a>
I've got no volt meter. I spoke with another member that did this mod a year ago with no ill effects. The basics being that all the LEDs are run at a maximum of 2w so the violets will run fine on the same channel as the royal blues.
darrrenjmartin;912935 said:221239054167 eBay number. This is a example of a dimmer that will work. I didn't have time to see if it only uses a pot or will also except PWM input. It seems that since the driver has to run so many series LEDs and has the power supply included the choices are more limited.
I did see some PWM adaptor boards. I didnt know what I was seeing when I came acrossed them. These boards take the place of the pot to provide a PWM input for electronic control. very cool.
QUOTE]
Marty, take a look at this. I don't think the manufacturers of the China LEDs invested extra $ in more drivers than absolutely necessary which is why we see most units have 2 channels and only 2 drivers. I think they just wired the LED chips in parallel, which it sounds to me doesn't over run the driver. Sounds like the Meanwell ELN 60-48P would work? They sell them at RapidLED.
My chips arrived today from Reefbreeders. I picked up the soldering iron and some solder at Lowes. Now I'm waiting for my arctic white thermal paste and thermal surface cleaner to arrive from amazon. Ready to get this project started.
I also made some final modifications to the spectral layout. I am happy to hear your feedback.
<fieldset class="gc-fieldset">
<legend> Attached files </legend>
>
class="gc-images" title="led.jpg[/IMG] style="max-width:300px" /></a> </fieldset>
I also made some final modifications to the spectral layout. I am happy to hear your feedback.
<fieldset class="gc-fieldset">
<legend> Attached files </legend>
This project went WAAAAY quicker than I thought it would and my only regret is that I didn't do it sooner. Here are the major visual differences as it will be about 6 months before I can report on growth and increased coral coloration.
1) It seems like I took a lot of the light out of the water column, now the coral and rockwork look a lot brighter with better contrast while the water column looks crystal clear
2) the light "color" now looks more realistic
3) I forgot how gorgeous my fish are - the colors of the fish are fluorescing like never before.
4) I definitely didn't add too many reds or warm whites as the light temp looks great with no hotspotting.
5) There is a BIG difference between the UV chips that came with the units and the 420nm I got from Reefbreeders. The 420's have much more visible light. Def a good add.
The first MM box I modded took me awhile because I was learning my technique. The eShine unit I flew through in 2 hours from start to finish and that included replacing about 25 chips. I did the final MM box last night. I now have a wealth of extra 450nm Royal Blue chips. I could use more 480's and 420's.
First pic is the MM box modified. Second pic is the blue channel of the eshine - you can see all the purple and uvs I added. Third pic is the white channeld on the eshine - I added 3x660nm red, 3x3.5K warm white, and 4x480nm Blue.
1) It seems like I took a lot of the light out of the water column, now the coral and rockwork look a lot brighter with better contrast while the water column looks crystal clear
2) the light "color" now looks more realistic
3) I forgot how gorgeous my fish are - the colors of the fish are fluorescing like never before.
4) I definitely didn't add too many reds or warm whites as the light temp looks great with no hotspotting.
5) There is a BIG difference between the UV chips that came with the units and the 420nm I got from Reefbreeders. The 420's have much more visible light. Def a good add.
The first MM box I modded took me awhile because I was learning my technique. The eShine unit I flew through in 2 hours from start to finish and that included replacing about 25 chips. I did the final MM box last night. I now have a wealth of extra 450nm Royal Blue chips. I could use more 480's and 420's.
First pic is the MM box modified. Second pic is the blue channel of the eshine - you can see all the purple and uvs I added. Third pic is the white channeld on the eshine - I added 3x660nm red, 3x3.5K warm white, and 4x480nm Blue.
