Apollo Reef LED's... anyone have them?

[TanksNStuff]

I'll be making a trip to Lowes myself. I need to pick up a multimeter to check if I'm getting the proper signals from the RA controller. I also seem to have misplaced my wire strippers and had to carefully use a pair of wire cutters to do my stripping. Two things good to have around the house anyway.

I spent a few hours last night trying to get the lights to come one with no luck. I emailed Joe from Apollo and he told me my wiring should be correct and suggested I check if I'm getting the proper signal from the controller with a multimeter. He also suggested to: "make a simple tester with AA batteries or a 300Ma Enercell from Radio Shack ( about 19.00 ) , use them to supply the DC signal to the light to ensure your wiring is correct. Test it as well with the multimeter to ensure you are getting a DC signal between 1 and 10V". I think I'm going to try Mike's suggestion and just hook the ethernet wires to a 9v battery, just to see if the lights power up. Joe replied to my last email just now and said that should work, at least in theory... so I'll save the $19 and just use a 9v battery.

Here's a copy/paste from my post on the RA forums to show what I did. I'm hoping someone over there can help me figure out my next step before I get home tonight.

*NOTE* I didn't use any RA expansion modules at all for this test. I didn't install any of the jumpers on the mainboard either. I connected the single relay box to the head unit via the VGA cable. I attached the Serial TTL cable to the head unit and the USB end to the relay box port. I then plugged the relay box power cable into a live wall outlet. The head unit booted up and was in the Demo mode. I didn't hook it up to the PC at all yet and all I did was changed the date/time setting via the menu and joystick.

1. I plugged both white and blue power cords into the fixture, then to a live wall outlet. Fans went on when either (or both) the white or blue rocker switches were in the "on" position (they lit up green). This meant that the fixture definitely had power to it.

2. I plugged a CAT5 UTP patch cable into the fixture's ethernet port and cut/stripped the wires on the other end.

3. I connected the PWM-to-Analog lead wires into my controller relay box and proceeded to connect the ethernet wires to the leads.

4. I used the following website as a reference to figure out which pins were which colored wires... using the second chart ("Straight-thru Ethernet Cable Pin Out for T568B") as this matched the color pattern on my patch cable plug. http://www.ertyu.org/steven_nikkel/ethernetcables.html

5. So, for the "whites" I used pins 1 & 2 [(white/orange) for the + and (orange) for the - contacts]. For "blues" I used pins 5 & 6 [(white/blue) for + and (green) for - contacts.] After not seeing any lights, I reversed the + / - for both and still didn't see the lights go on.

6. I checked the controllers demo program for the LED's and the signals were set to 48% Daylight, 48% Actinic. That was eliminated as an issue.

7. Since no lights came on after several attempts (even tried both fixtures with all of the above) I'm guessing that it's either an issue with my wiring or the controller is on a time schedule and since it was after 10 PM just didn't send the signals at all. So, I changed the controllers time/date setting to make it 12:30 in the afternoon. Still, no lights came on.

It was very late at that point so I just packed it all up and went to bed. I'd like to know if I wired the ethernet cable correctly to the lead wires, and also if there's anything else I did wrong that would have prevented the lights from coming on?

While I'm at Lowes, I'm going to look around for alternate options for the mounting pegs too. If I find anything that will work well, I'll let you know Mike. Cool that yours are on the way already too!

 

[Mike S]

I'm sure it is something simple like a little tweak when loading the internal memory code. Someone will probably get back to you quickly. Ill be sure to follow along. I should have my lights today and the pwm to analog cables a little later this week. By then hopefully you have all the kinks worked out.

 

[TanksNStuff]

I was up til 2am trying last night. So I'll get it working or it will kill me... one or the other, lol.

I just started browsing the manuals and tutorials and I think it may be something with the Dusk/Dawn settings on the RA.

At least at some point tonight I should know at least whether my wiring was correct and if so, someone on the forums should be able to help me figure out the RA side.

 

[Mike S]

I feel your pain. I had a few long nights when i got it trying to figure out how to get the basic functions going. Its a very flexible controller but just takes extra effort to get comfortable with it. Your doing the right thing by taking one step at a time. I spent a lot of time spinning my wheels because I was trying to absorb everything at once.

 

[Mike S]

Just a thought. When you set it so that the time was 1230 pm did you give it a few minutes to see if the lights came on? Im not sure if the demo code uses the slope function but if it does it slowly brings the lights up over a period of time. I've read were there have been instances where the lights don't come on until it ramps up to 15 or 20%.

There is also a delay function if the relay is set up as metal halides vs stardard lights. This tripped me up a few times.

 

[TanksNStuff]

Yea, that's why I just stuck with the demo until I get the lights to work. Once I get that hashed out, I'm going to start hooking up the wifi and all the other modules and then work on making my own pde for everything I have hooked up. I should probably do those one at a time also... just to make sure any errors are related to one piece only.

 

[TanksNStuff]

Just a thought. When you set it so that the time was 1230 pm did you give it a few minutes to see if the lights came on? Im not sure if the demo code uses the slope function but if it does it slowly brings the lights up over a period of time. I've read were there have been instances where the lights don't come on until it ramps up to 15 or 20%.

There is also a delay function if the relay is set up as metal halides vs stardard lights. This tripped me up a few times.

Hmm, I hadn't thought of that. That's probably what the issue was. When I didn't see lights right away, I started switching wires around. I do recall Roberto saying the drivers don't work until around 15%... so, if the PWM was set at 48%, it would have likely taken at least 10 minutes or so until it ramped up from 0% to 15% (where the lights would register).

Thanks for the tip Mike. I'm going to add that to my post on the RA site as a possible issue.

 

[Mike S]

It seems to be more of an issue when the drivers aren't used to capacity which given the number of leds and drivers doesn't seem to be the case with the Apollos. Its worth a shot though.

 

[TanksNStuff]

I figured it out! Somehow I must have changed the preloaded settings in the Dawn and Dusk menu when I was browsing all the menus at first boot.

I checked this menu again and it showed:

Dawn: 04:30 PM
Dusk: 11:00 PM

As soon as I changed the Dawn to 04:30 AM and hit OK.... the lights popped on!

Now I'm going to start setting up the other features of the controller one at a time. I think in the next couple days I might setup a temporary hanger to get my LED's going. Might just wait until the weekend when I have a good time slot to get that done.

Anyway, will update as I go, then add some pics of everything in next couple days. I'm too focused on getting things working to post pics right now.

 

[Tazmaniancowboy]

Cool, glad to hear it! Can't wait to see both toys in action.

Sent from my PG86100 using Tapatalk

 

[redfishbluefish]

Good stuff George. Can’t wait! You’ll be very happy….I’m sure.

 

[Mike S]

Awesome, how do they look? My got delivered today now I just need to patiently wait for the cables. In the meantime I'm trying to get my code squared away and trying to figure out the slope function.

I looked at those studs and I don't think they are going to be easy to replace. They are bolted to the chassis with a much smaller bolt and I'd rather not enlarge the hole. I'm hoping lowes has metric threaded rod couplers.

 

[TanksNStuff]

By the time I got them going, it was already pretty dark so I didn't want to put them over my tank... but you know I did for a brief moment. I only had one hooked up at a time since I only had one spare ethernet cable to splice and use for the lights. I still had my T-5 fixture on my tank, so I couldn't hold it directly over center, but I sort of held it at an angle from about 1' from the surface and it had great penetration and looked about equal to my T-5s while only being set to 50% intensity (white and blue).

I think by the end of the weekend I'll be able to make a temporary support to hang them from, remove the T-5 fixture, and then take some pics.

Glad you're working on the coding for the PWM slope function. I'm going to need to steal that from you since I'm still kind of clueless with the coding part so far. I had some issues with installing all the software programs they have to edit the code and check the status, etc... and I learned today that the reason it wasn't working was due to a pathing error to the file locations. I have instructions on how to manually fix that, so I'll get to play some more tonight with getting all the modules working.

I was able to get the 2 temp probes and the pH probe registering no problem. But I needed to get that other stuff going to setup WiFi and the other modules.

Oh, I stopped at Lowes and tried to find a solution for replacing the mounting pegs. I ended up grabbing a 6-32 all thread (about 3' long) that I think will fit in the hole that's there. I picked up a bag of same sized nuts (also had short bolts that I won't need) to fasten the rod inside the case. I figured I could just cut the rod to the length I need, drill a hole in the 2x4 (or whatever crossbeam I use... might be aluminum angle) and then slide the rod through it and use a washer/nut on top to hold it up. How's that sound? {Pans the room for Paul}

 

[Mike S]

I've got the code written up and ready to go. Ill let you know how the slope function works and send you the code once I know its working OK. It seems pretty simple but ill see what happens when i load it. Might give it a shot tonight if i can finish hanging it.

 

[Xavier613]

George, you can use sammies to secure the 6-32 to a 2x4 if you wanted something easier than figuring out how to attach it to metal. Just screw it in and thread into the end of the Sammie.
Link if you haven't used em. I might have a few in the garage if you'd want to try it
http://www.itwbuildex.com/sas_wood.shtml

 

[TanksNStuff]

Mike can you pm me the code? I'd like to try it tonight.

Xavier, thanks for the idea but not sure those will work. First off I didn't see any in that link that fit 6-32 rod. Secondly, I wouldn't be able to screw the rod into those after I fasten them inside the fixtures case. Well maybe if I left enough play in there I could. It's worth trying if I can find them locally to see.

 

[Mike S]

No problem. I'LL send it to you in a little while. I've asked Roberto to confirm one or two things for me before I try it. I posted a question over on the RA forum about it.

I did get the lights hung last night but I need to run out today and get some extension cords because the power cords won't reach to my relay box. After that I just need to hook of the pwm cables to the Ethernet cords, load the code and keep my fingers crossed that everything works.

I did try the moonlight's over my tank and they are bright! Might have to get my fish some sunglasses so they can get to sleepnat night.

 

[TanksNStuff]

Cool, thanks Mike. It might be just as easy for me to check the RA forums and find your post... and see what Roberto says about it.

FYI, the light power cords don't need to plug into the relay box if you don't want to. They can be plugged into any wall outlet or power strip, etc. But then you won't have the controller handling when each socket gets powered up... meaning you wouldn't have control of the fans so they'll be on 24/7 (unless you have timers setup for them.)

Did you ever get the PWM-to-Analog Lead wires? It might still work using the straight PWM signal, but it's advised to use the leads to convert the signal to analog.

Yes, moonlights are bright. And that's them running at 1W each. The rest of the blues will be just under 2w each at 100% intensity. This is another reason why I was considering replacing the ML drivers with dimmable drivers.

Anyway, here's another big update on my overall progress. Before doing any of the following, I connected all the RA hardware that I have except the WiFi module and the PWM expansion module. It was basically the head unit, connected to the main relay box (first 8 outlets), connected to the expansion hub (basically a 10-port USB hub that lets you connect more than one extra module), then connected my first relay expansion box to that (this is my second 8 outlets... or outlets 9-16). I didn't hook up the PWM module yet because it requires a 12V power supply which they included... but had a round jack on the end. I didn't feel like cutting the end off and figuring out the +/- and all that. Also, my 2 temp probes and my pH probe were connected to the head unit to provide data for those parameters.

1. I didn't get started on this stuff till late so I didn't get anywhere near what I wanted done. I didn't even hook my lights up again because I figured my top priority was reinstalling all the software so that my controller was able to use something other than the Demo code.

2. First thing I did was release all the "in use" COM ports via Device Manager. The only one shown in the manager was COM1, so I went into the settings for it and attempted to change it to the other "in use" ports. After changing from one to the next, the "in use" designation was removed on them all. I then went back to COM1 for this port.

3. Next, I reinstalled the RA Arduino program, and the controller was now designated as COM3 (which was what I was looking for).

4. From here, I installed the RA Gen program. For those who don't know, this is a GUI applet that helps you create a custom code by selecting features you want to have via check boxes. Ok, so now this installed correctly, but I had to manually change the paths of some of the file locations in the settings. That was apparently the issue that caused my problems the first 7 or 8 tries.

5. Once the file locations were fixed, I then used the Library updater utility (separate program) to update all the development libraries to the most current.

6. OK, now I went back to RA Gen and created my internal memory file. This is something that has to be done on initial startup to provide certain info to the firmware on the controller. The GUI lets you change light cycle times and some other things I don't remember at the moment. I don't know much more about this other than it needs to be done and it apparently has something to do with memory.

7. After that, I used RA Gen to create my first custom code. It was very basic and minimalistic (I selected the Simple Menu option) so I could basically start with only the coding I needed for the hardware I had. I did select the options for the WiFi and PWM modules even though they were not physically connected yet. My screen just showed the graph (it's a tracker for your temps, pH, etc in a low res format), then it showed the digital readout for the main parameters (Temp1, Temp2, Temp3, pH, Actinic %, Daylight %), then at the bottom was the main relay (numbered 1 - 8) which showed them as a red background for "off" and a green background for "on", at the very bottom was the Date and Time.

8. Since this screen was not even close to what I was looking for, I began to steal bits and pieces from others posted on the forums as a trial to learn what I can do with it. Here's a brief rundown on what I changed:
- First I removed the graph entirely. It takes up a big chunk of the small screen and I really don't think I wanted it due to the low-res look anyway.
- Then I added the second relay outlet box to the bottom. I had to manually adjust the position (x-y coordinates for where top-left of the data being displayed) goes so that nothing overlapped.
- Then I wanted to make use of the room at the top after removing the graph, so I added in a banner heading to the top (It said "George's 75G" in a teal color... or it could have been aquamarine or something like that.)
- When I borrowed the code for the banner, I found examples that made the temp1 display in bigger numbers and other parameters to the right of it... all in different colors to look fancy and stuff. So I added that change in. This required me to comment out the original code for displaying this data, otherwise it would have written over top of each other. This also required me to edit the ReefAngel_Features.h file to include some #define FONT_xxxxxx lines to be able to use the bigger font sizes.
- Was about to get into adding the PWM signal data but couldn't decide which one I like best. I also figured I'd wait to see if Mike included that in his coding. Besides that, the PWM expansion module wasn't connected anyway.

Through all of that, I uploaded at each change to make sure the code was good and to see what it looked like on the screen. I had to redo some parts several times to move the positions and make things look good. It's a work in progress but it's a start!

Oh, I did at some point try to get my WiFi module running. I was able to connect it to my local home network and it has an IP address... but I was too tired to go any further than that. I know I need to add in some code to my .pde to get that to communicate and send data from controller to the web app, etc.

Another day or two of working out the setup kinks and I'll start taking some pics of things in working order, etc.

 

[Mike S]

Looks like you've been busy. I havent bothered to tweak the display on the controller menu but might look at that down the road. Most of the time I just check my parameters on my tablet or phone. Was it difficult to do?

I did the the analog to pwm connectors and am going to get that all rigged up tonight. I'm going to run the power cables to the relay even though it takes up have ofbthe outlets. I may just ended up getting another relay box down the roadbif I find that I'm missing the extra outlets.

Good idea about replacing the moonlight drivers with a dimmible version. I'll open up the case tonight and get the model numbers off it. With only only 3 led per fixture we might need to use one driver to control the moonlights on both.

 

[TanksNStuff]

Yes, I've been busy and time flies by when messing around with this stuff. It isn't really hard, it's just time consuming and a bit of trial and error to get the positioning. Most of it is just copy/paste from others work and then some minor mods to the positioning data.

Yea, I've already considered putting both fixtures moonlights on one driver but for now I'm not going to bother with it. Too much work and the possibility of voiding my warranty are ruling that out for now.

I do have some possible bad news though. I've been in contact with Joe from Apollo who recently received a RA controller setup to test his lights with. Going to just copy/paste it here so I don't have to type it all... but he doesn't think the controllers dimmer output voltage is high enough to control these lights properly. Here's the entire convo...

having issues. We are only getting 3.12 volts out of the Reef Angel at 100%. The Drivers need 10V for full power.

it takes us 23% output to get one light to turn on. 59% to get 3 lights to turn on. The output is way off.

I put out an email to Roberto to call me. Maybe we are missing something here in the settings but I dont see it.

Joe

Joe,
If what you're saying is true, then I'm a bit worried. From what Roberto told me, the controllers PWM signals can support upto 4 drivers each. Hopefully this was an isolated incident or just some kind of setting error or something. I'll be real upset if there isn't a fix and don't get the full voltage required out of this controller! But, on the other hand, my plan is to have only 1 driver per PWM signal channel so I might be ok when setting it up this way. How many drivers did you have on one channel when noticing the insufficient voltages?

In reading your previous email, I did think of one possible issue. How long did you leave the lights on before testing the voltage? From what I understand, the controllers are set to ramp up the intensity from 0% to whatever you set the "max" voltage, over a specified period of time. Not sure what the Demo code is set to for the ramp-up period, but I think it might be 1 hour?

If that's what the controller is doing, then that could be why there's lower voltage initially.


I'll definitely do some more testing tonight and check the signal voltage, as well as take those pics I promised you.
George

George,
We tested first with two lights, then when we didnt see the right responses we went to one single light, so one driver per channel. On the manual setting it didnt light until the RA said 23%.
The Meanwell drivers are designed for 10% , using 1V to output 10% of the capacity.

It is very odd, when we test the voltage coming out of the RA with nothing connected it is 9.5V, close enough to 10. But when we connect a single light which works fine on Apex and Profilux, , one driver per channel, the voltage drops off to only 3V or so. Very confusing. Roberto asked me this am if I had a light I could send him but I am running very low on stock at the moment so cannot spare one, it will be another 2-3 weeks before we have more inventory so I dont know where this is going to land at the moment.

Stay in touch

Joe

Hi Joe,
Hmm, that is odd that it drops from 9.5V to 3V by just connecting to a single light. I wonder if it has anything to do with the PWM to analog converter? I think there's a resistor or something tied into the middle of that wiring. I wonder how it would work with a straight PWM signal (ie. wiring the ethernet wires right to the PWM pins? Would that hurt the meanwell drivers do you think?

Yea, I'll check back with you often and anytime I make any progress I'll keep you in the loop. Thanks for doing the same.
George

No you cannot put PWM to analog drivers, it is completely different.

I notice on RA website the mention of meanwell booster for the PWM meanwells, I wonder if he is going to need something like that here. As I said, the lights I tested on worked fine with Apex and Profilux, in fact the Profilux I had 3 lights piggy backed to one dimming port, driving 3 drivers each with one channel each for a total of 6 drivers and it worked like a champ.

Apex I am able to use splitters to share he jacks and expand the VDM module. Ive tested up to two lights on one jack with no issues, dont know the total capacity how many I could get away with on Apex but again here, with one fixture it is falling on its face. Ill be curious to see what results you get when you get back to testing, like I said, it took 23% before the light turned on at all which is not right. Theoretically that would be 2.3V from the RA. See if that is where your light turns on as well.

I too wonder about that inline converter, but lets see what happens when you get back to testing, if you get the same results we know there is an issue with the RA supporting more than one driver which will never work. This is the same type of issue we are aware on with Reef Keepers which is why we do not list those as an approved device.

The part in red is the part that gives me some hope. It seems like the voltage being put out is correct... but when the LED wiring gets connected it drops. If you review my previous post on how I wired the ethernet patch cable to the converters... you can understand how this could be happening. My gut feeling is that the PWM to Analog converter is causing the issue of drop in voltage because there are some kind of resistors inline that convert the signal types. I'm thinking that maybe the wrong size were used? I'm going to test my output with a multimeter tonight to see if I get the same results. I'm also going to test at what % the lights come on.

This has jumped to top of my priority list as this could be a major issue. If mine does the same thing (maybe he got defective converters?) and Roberto can't come up with a fix... I'm going to ask Joe if it would be possible to replace my drivers with the PWM version to eliminate the converter.

I'll keep you posted.