For no logical reason - Rebuilding a horrid old MaxSpect G2 fixture.

[hhaase]

If any of you guys had a MaxSpect G2 when they were being sold, this thread is for you. I figured you guys might get a kick out of a dumb idea of a project that I had.

When I started building my new 150 gallon SCA I still had this pair of old MaxSpect G2-160's sitting in storage from my old 90 gallon. I just didn't have the budget for all the hardware at the time and was looking to save money where possible. There was only 18 months of time on these lights before I tore the old tank down, so they were still comparatively 'new'. But, like pretty much every G2 fixture out there, more than half the blue LED's were dead by then. I had suffered through a bunch of dead drivers too. I knew they could grow some corals though when working, as I had a few decent LPS and clams in the old tank. The original goal was to band-aid my way through about a year while the new tank established, and then get some proper lights when I felt ready for introducing SPS. What can I say. I'm stubborn, occasionally cheap, enjoy DIY, and sometimes just don't know when to let go and move on if I feel I've been wronged.

So these lights are / were my starting point. MaxSpect G2 160's, with the lens kits. 32x 3watt Semi LED's, 14 blue, 14 white, 4 UV/Violet. Plus a pair of 30w cool white's. Three controlled channels (White, Blue, 60W), but no variable power, and each channel only gets a single on and off time per day. Things were pretty basic in 2010. But hey, also has moon-lights, cooling fans, and temp readout for the pendant. That and it came with both ceiling hanging and cross-bar tank mounts. A decent feature set at the time.

At this point, I've done a fair amount of work on these lights and I'm enjoying my trip down the rabbit hole. I had planned on looking for replacements this spring/summer but I think I'm going to see how much further I can push these before I move on.

 

[hhaase]

So, the biggest problem at the start, was the burned out blue LED's. Some people say they were failing because they just had short lifespans, I've also received conflicting info if they were driven at 700mA or 1050mA. Either way, both MaxSpect and Coralview at the time confirmed they were over-driven. I talked with the guys at StevesLED's, and they recommended the Seoul Z5 royal blue's as a good upgrade/replacement, so I bought enough to replace all the blue and UV/Violet.

Out came the old fried LED's, in went the new. Thankfully the hole spacing for the mounting screws was the same, and I opted to wire direct and bypass the original LED power boards in the fixtures. I need to get a newer picture that shows all the wiring soldered in place, and the insulators that keep the mounting screws from shorting out to the heatsink.

I went on to use these as I had previously, including powering up those big 30w units. The conclusion I came to was that the 30w LED's put out too much PAR and fried everything that was under them, but without them it's not enough PAR. Being that I was still getting my head wrapped around things, I didn't acclimate my Monti frag properly and it got absolutely roasted. Bleached out in just a week, and never did recover. Thankfully, after I learned my mistake. I pulled out the lens kit and stopped powering up the 30's.

The few other frags I had at the time were all able to recover and are all surviving, but not thriving. At least they got their coloration back. The Seoul LED's also have a much more blue appearance than the old Semi's. I'm ok with it, kinda like the look, but it's something to be aware of.

This only bought me a few months of lifespan though before the next problems popped up.

mething to be aware of

 

[hhaase]

The other major issue with these lights was the drivers. Inside the control box there were four different types of functional parts. A control board, a power distribution board, a 12v power supply, and a lamp driver for each channel.

The control board is pretty basic but not bad otherwise. Really just a clock and a timer for each channel. Each channel gets one ON and one OFF time per day. It also monitors the temp of the pendant. No color control, no dimming. The power distribution board was also pretty well made. It holds all the drivers and the 12v power supply, plus a few brick relays for turning the drivers on and off. Unfortunately the power board was forced to use these proprietary drivers.

The drivers and 12v power supply were definitely made by a different place though, and they're all just trash. During the initial 1 year warranty I think I had 4 drivers fail and get replaced. God only knows what the current rating on theses drivers was, I've had a few different numbers told to me. Some mix of 700mA and 1050mA is the most likely bet. That would explain why they always burned up the blue LED's.

A few months after I got my new tank running, I finally had a driver fail on me. It started blinking once in a while on startup then eventually just started strobing. Thankfully I had already been working on a plan for this. Like I said, I knew they would fail at some point, so I was already prepping for it to happen. Since I wasn't using the 30w modules any longer I just ran with two drivers instead of three. I swapped them around to keep it all running all while finalizing my long-term fix on this.

The new plan was that I made a new power board that uses easily avaialable meanwell drivers!

Top half here is the original max-spect power board and drivers..... you can see I've got one removed. The bottom half of the photo is my new power board with the much larger meanwell drivers. Currently I'm only using two channels due to size constraints, but the board does support that third driver. I just need to get creative in mounting it. The 12v RECOM power supply is probably serious overkill, but I want reliable, so it meets my needs. Everything just barely fits in the housing. But it does fit without obstructing airflow, and the original control board runs it all flawlessly.

The lighting LOOKS nice in the tank, I'm happy with the color to my eyes. But I haven't got a clue how much light I'm putting out. I'm using 500mA drivers, which are less than the originals, but the LED's are far more efficient. If I change a few blue lights into the white circuit I can run a 700mA drivers for both, but at this point I'm just going with guesswork as to what's the best option. This is what the status is for the lights right now on the tank.

They're reliable, which is a first for a G2 MaxSpect, but PAR is a mystery. I'm leaving some capabilities un-used with that empty channel. So the next step is going to be PAR testing and there is a meter enroute already. Once I get some numbers I can start planning a path forward.

If PAR is stupid low I may decide to just upgrade to new lights. If it's only KINDA low, I'll start testing the effects of the lens kit and maybe look at options for boosting PAR either through bigger drivers or maybe adding back in that third channel. I've got a fun idea about that which fits into the original design concept of the lights too, and building up a variation of the original PAR cannons.

 

[hhaase]

Good News! PAR meter is arriving today for me. It'll be some time before I can do any comprehensive testing, but I do want to get at least a couple quick numbers tested today if I can.

So, anybody want to venture any guesses on PAR numbers for the following locations?

1: Favia @18" depth
2: Kenya Tree @ 12" depth
3: Damicornis @ 16" depth
4: Porites @ 14" depth
5: Montipora. @ 20" Depth

Just as a quick recap.... each fixture is 18x Seoul Z5 royal blue's and 14x Semi warm-whites. 500mA drivers. Tank is 24" high.

 

[hhaase]

Numbers were pretty surprising, I was expecting some garbage low PAR across the entire tank. I wasn't expecting numbers that are almost workable. Still low but nowhere as long as expected. Should be a viable situation already for softies and LPS, plus some low-light SPS in a couple spots. Man, when those big PAR cannons were running and had the lenses on, it must have just been abusive for that poor Monti which was up at location #2.

Time for some planning!

#1 - 110
#2 - 240
#3 - 120
#4 - 200
#5 - 110

 

[hhaase]

One big thing I wish I knew was the specs on the original Semi 3w white LED’s in these fixtures. Just a big mystery to me what they really are.

I know I have about 15v available in the blue circuit, which means I could maybe shoehorn in four more LED’s. Realistically because of how the reflector is built, I’d have to do multi-LED stars to do that. Would be pretty easy actually. I could use the opportunity to add in some other wavelengths too, maybe a pair of green and pair of red in each pendant?

The white circuit has 4 less LED’s mounted. I could potentially get 8 more in there. Or I can replace all the whites and have known numbers to plan accordingly (Nah, too spendy). Wouldn't be adding more white in though, I'd be doing probably a mix of royal blues and cool blues.

This doesn’t even include the empty channel. I’m considering doing a half dozen hyper violets between those two locations. Combine it all together and I can really throw some PAR.

One big thing to factor into this though is budget. Adding in the extra LED's into the white circuit will be just shy of $100. Adding into the blue circuit another $60. Plus shipping. Is it worth sinking another $175 into these fixtures to get maybe another 30% out of them? If I do the hyper-violet stars that'll be another $100, so we're at almost another $300 into a pair of dead-end fixtures. To maybe boost par by 60%. Doesn't have to be all at once though.

But that's still less than one Radion XR15, and I'd eventually probably want three or four of those on this tank.

Hmmmmmmmmmmmm It's a lot to think about.

I think I'll get those detailed PAR readings first.

 

[hhaase]

Really none of the PAR meters come with anything to hold the sensor. When you're good with a 3D printer, you don't need to spend $60 for a handle.

This mount attaches to 3/4" Schedule-40 PVC and routes the USB cable up through the pipe. Nice and clean, and holds the sensor very solid while protecting the cable. Easy to install and remove too.

Now I can start measuring for real.

 

[hhaase]

I ordered what I feel is the most “bang for the buck” for updating these lights further. This is to re-implement that third channel. I ordered a a pair of triple LED stars for each fixture, which will give them a half dozen hyper-violets each. These will be driven by 700mA drivers instead of the 500’s in the first
two channels. Should be a significant PAR boost all around. I’m hoping for about a 30-50% increase. It also keeps with the theme of the PAR cannons in the original design. I’ll have to be gentle in adding these into the light schedule though. Plus it’s a wavelength that’s currently lacking in the spectrum.

Adding another couple more royal blues to channel one really wouldn’t be a huge boost. I just don’t want to spend more on this channel until I know the fixtures are capable of getting into my desired range. And I won’t know that without the big increase in channel 3. Just didn’t seem like smart money right now.

Rebuilding channel two is even more confusing at the moment. Yes there’s more room to grow than with channel one, but it’s going to get me even less par per dollar since I should probably be replacing all the existing LED’s in order get known quantities. This is where my next phase will likely be if I keep going on modifying these fixtures.

 

[hhaase]

I'm not going to go crazy with a dense PAR map on this project, a quick and dirty one is already telling me all I need to know. These are the baseline numbers which I'll be using to judge future updates, I'll share the spreadsheets later on when I have more results to compare.

For instance, at 6" under the water (+ 2.5" of air), I get single digit PAR around the edge of the tank. This is completely expected as I only have a pair of fixtures that are only 400mm wide. I did take some measurements, but all it does is reinforce that I do eventually need more fixtures for more area coverage. BUT, directly under the lights at 6" is a good measuring point for sheer power output. Both fixtures give me 100 PAR from the white LED's, 300 from the blue, and 400 total. Too bad none of my rock work goes that shallow. The key point is the white SEMI LED's aren't the major PAR contributor, and the UV purchase is justified.

For the sand bed I did a 3x5 grid, leaving out the two center spots that are covered with rock. 2.5" of air, then 20" of water depth.

The white LED's give me anywhere from 10-25 par, depending on the location. 10 around the edges and 25 in the high points. Closer to the center lights of course is higher, and a surprising bump in the dead-center of the tank. So I guess there is a bit of overlap between the fixtures. Some expected shadowing as well.

The Blue LED's had a bit more swing. Left edge had around 10 PAR, right edge around 20 PAR, peaking at 45-55, patterning essentially the same except for that higher number on the right edge which is curious.

With all the currently installed LED's running I'm looking at 15-30 on the left edge, 20-45 on the right edge, peak readings were about 70 PAR. So no clams on the sand bed for me, I'd even struggle with Zoa's down deep right now. But I maybe I'll shift my Favia and Porites around a bit to get them into the 200's range until the next batch of LED's go in.

 

[hhaase]

I decided my final plan going forward on these lights. I'm not going to try and eek out every last possible PAR, that's just not worth the effort. After I add in the violets, I'm going to rebuild the white LED circuit to try and kinda match the color patterning in XR15's.

So the 14 cool white SEMI LED's in each fixture will be replaced by luxeons all around. 6 cool blue, 4 warm white, 2 cyan, 1 deep red, and 1 lime. Overall this should give me an approximation of the AB+ lighting pattern, but without the same level of control.

That's about all I'm willing to put into these lights and the end of the upgrade path on these. Any further light changes will be by additional or replacement fixtures.

 

[hhaase]

Ok, good day on the lighting situation! Which is great, because the past few days have had some other ideas not pan out as well as I had hoped. I'm going to stick to the good and fun today, because this exceeded expectations.

The new LED's for the white channel arrived the other day. All Luxeon 3w stars, which are super easy to retrofit into this fixture. For all that I bash it, this fixture does have its good points, there are very few on the market that I could have done this kind of retrofit/rebuild on. And honestly I think the mounting options on this set were awesome, and only needed a touch of refinement. I wish there were ANY lights currently on the market that had this style crossbar, works great for eurobraced tanks.

Anyway, here's those new LED's, though I didn't have the triple-stars in this picture. This is all the singles for the white channel.

My plan with these last two channels was to do a pseudo-copy of the AB+ spectrum. I of course don't have any brightness control, but I can at least choose the colors and locations of the LED's. So here's the full color map of each fixture. Red letters are channel 1, black letters are channel 2, green is channel 3. I was also able to lay these out and solder all the main wiring together outside the fixture, which saved a heap of time.

RB = Royal Blue
B = Cool Blue
W = 5k full-spectrum white
C = cyan
L = Lime
R = deep red
HV = hyper violet.

Then it's simply a matter of removing the old MaxSpect / SEMI-LED chips, and installing the new. All the singles got thermal grease underneath and are screw mounted, with some fiber washers to prevent shorts to ground. The hyper-violets aren't powered yet but are wired and mounted using thermal adhesive, since the screw patterns in my desired location don't match this size/style of star mount.

The original red circuit boards in here are doing almost nothing now. Initially the original LED chips plugged into it to make them easy to swap, but I bypassed that completely. Channel 3 never used those boards even from the factory. So now they're just there for the moon lights, and as a connection point for the fan and thermistor. Theoretically I could engineer them out somehow. But I don't see a need to modify the fixtures any further other than maybe some upgraded fans.

Button it all up and power it up, things seem to be doing well! I've only done the right side fixture so far, I want to give it a day to burn-in before doing the left side fixture. It's appreciably more blue, but not unpleasantly-so. Give an impression of a deeper water look. It also matches my expectations for how it would look and I'm quite happy with it.

The only real surprise is that initial PAR testing is showing an appreciable bump all around. I really didn't expect much of an increase. I got maybe 25% more. 6" under the water surface went from 400 to 500. That's quite respectable to be honest. I'm anxiously looking forward to getting both fixtures finished and doing a spot-for-spot PAR map to compare to the pre-upgrade numbers. It's also given a fair amount more visual POP to my chalice, which is the only stony coral under that side. I think this is going to work nicely.

 

[hhaase]

Some spot numbers on the PAR to show the changes. Interestingly, the right side fixture seems to have increased more than the left. Interesting because it's the same models/power in the drivers, and the same LEDS pattern. The very left wall of the tank shows no increase at all, but there is some shadowing at the extreme left and right due to the shape of the mounts, so I'm not concerned with that at all.

All lights - 550 PAR at 6" depth under the right fixture, up from 400 (+37%). Left fixture went to 450 from 400 (+12%)

Locations 1-5
1: From 110 to 145 (+31%)
2: From 240 to 305 (+27%)
3: from 120 to 145 (+21%)
4: From 200 to 250 (+25%)
5: From 110 to 155 (+41%)

Peak on the bed, in the front-right, went from 60 to 125 (+108%), dead-center in the tank went from 70 to 85. (+21%)

Last stage to come is going to be turning on channel #3 at some point, but for now I'm going to let everything acclimate to the new lighting for a bit. My Favia only saw a slight increase, which is fine since it's already in a happy range. Chalice saw a fair bump, small increase for the Damicornis. Once I get some nutrient things sorted out though I'd be more than willing to do some acropora on the top of the left side rockwork. Particularly knowing the PAR is going to be increasing once all the hyper-violets are introduced.

But man, all of my important locations increase from 20%-40% on the rockwork. Had a doubling on the substrate on one location. This is without any increase in the wattage of the drivers, the only difference was 1:1 swapping of the LED's on the less populated channel to a more appropriate spectrum.

I think this stage was well worth the effort.

 

[hhaase]

Hmmmm, I'm working on the violet channel, and I may have an issue with it. I don't think the PAR meter is going to really show what's going on there. The spec sheet shows that it under-reports by 50%-60% in the 415-425nm range these LED's use.

But, the third channel is installed. I've only got it programmed to run for 5 minutes a day and I'll slowly expand on that in the future to acclimate things. I had to get a bit creative to fit the driver, because there wasn't room in the case, and the mounting tabs don't make for a clean install onto just one half of the exterior either. It's a rare cosmetic compromise I had to make. I'll try to fix it at a future time with a bracket or something, to straighten it all out.

Because that light frequency is so drastically under-reported, I can't just shoot the PAR numbers and call it good. So I'm ONLY measuring the PAR for these LED's, doubling it, and then adding that to the previous numbers. But only for the 5 coral locations. Here's the progression from the initial base numbers, then the channel 2 rebuild, then the calculated final numbers.

#1 110 to 145 to 195 = +77%
#2 240 to 305 to 405 = + 68%
#3 120 to 145 to 195 = + 62%
#4 200 to 250 to 320 = + 60%
#5 110 to 155 to 205 = + 86%

As you can see, I exceeded my +50% goal by some substantial amounts. Depending on location, the rockwork is getting from 195 to 405 PAR. I'm confident that, with proper placement, I can sustain any coral I want with these lights. I'm getting an estimated 700 PAR at 6" depth. Not bad for some fixtures that last year were almost complete junk ready to be trashed.

Even better is holy smokes does that hyper-violet channel brighten up the colors. The phosphorescence is pretty dramatic with the Chalice, Favia, and Capnella. I'm super happy with how things are looking right now. As long as no functional issues pop up, I think I'm just about to close this project and call it a complete success.

About all that's left to do is some cosmetics and maybe a final PAR mapping of the sand bed. That and I'll have to start learning how to really get some good pictures of coral, because suddenly I've got color pop that wasn't there earlier.

 

[hhaase]

Ok, there's ONE more thing I'd like to do on these lights, which is find a replacement fan. The originals are getting a bit long in the tooth and are a bit 'clicky' on startup.

But I can't seem to find frameless fans of this size. I measure about 120-125mm in size. Biggest I can find for sale is around 90mm. I could always put a 140mm standard fan on the outside of the housing, but I'd prefer to keep it internal if I could

I find it hard to believe that MaxSpect would have designed their own fan for these things. There's got to be a source for them.

 

[oreo54]

Ventilation Fan - Dongguan Xing Dong Electronics Co., Ltd. - page 1.

ID-Cooling ND-8015M12B2 Server - Frameless / GPU Fan 12V 0.25A, W120x 2x2, D83xC42, White

12V 0.25A, 83 mm Dia. 42mm c.t.c ND-8015M12B2 ID-Cooling ND-8015M12B2 Server - Frameless / GPU Fan

www.pchub.com

 

[hhaase]

Two months after the last update I am happy to report that everything is running stable and I've had zero issues with the rebuilt lights. Corals are happy, tank looks great, and I hit the PAR numbers I expected, so I think I can call the rebuild process a success.

The only actual "Problem" is that I wish I had a third unit for more coverage, as two of these just aren't broad enough for a 60" tank and I've got some shadowing. If I could get another unit for absurdly cheap, I'd rebuild it for install onto the tank, but I'm not going to go crazy trying to find one either. Eventually I do plan on modernizing the lights, it's just something quite far down my priority list right now.

Still never got around to swapping the fans, a good cleaning seems to have helped keep the originals working quietly and temps are fine. I'll likely end up with externally mounted 140mm fans in order to get more flow and not have to worry about non-standard fan sizes. Eventually.

 

[hhaase]

I'm resurrecting this thread for a short time, as the old MaxSpect franken-lights are about to be replaced finally. Some new 57" AI Blades are enroute and I'll finally have some sleek modern lighting. But I'm not completely done on this rebuild, as I want to clean up the last missing parts of the retrofit as a matter of pride.

Here's a recent full-tank shot showing the current mounting. I did end up adding a repaired Kessil to cover the mid-tank gap, but that was a different discussion. What you can see here is that I did finally swap on a pair of 140mm external fans as the originals had finally spun their last spin. Harder to see is that I also replaced the original mounting cross-bars with some maker-rail extrusions and 3D printed bracketry. I'll give a closeup on that in a future post.

For a report on how well these lights can grow LPS and SPS..... the two large acros and the grapefruit sized damicornis on the left side rockpile all grew from inch long frags under these lights. My duncans went from 2 heads to 14, my chalice went from a postage stamp to about the size of my hand. Not bad for something I pieced together from scraps. But two of these units just doesn't have the coverage for a tank of this size and I am dealing with increasing shadow issues as the corals grow out. These lights would be a good match for something in the range of a 55 or 75 gallon. Not so much on a 60" wide 150.

Anyway, a couple more posts coming up to close the story on the last few improvements. Then, after some cleanup, these lights will go up for sale if some enterprising tinkerer wants to take over the story.

 

[hhaase]

Man, it irks me how much easier it is to get a good clear picture with my freaking phone than when I try to use my older DSLR.

Anyway, here is a shot that shows the various 3D printed bracketry I'm using. I can't wait to get the Kessil removed, as I never was happy with the sag in those red brackets.... but it got the light mounted. However I am very happy with the way the mounts on the G2's came out. Far more sturdy than the originals and let me slide them front to back quite easily. Very smooth. Held them in place more solid than the originals did too.

These lights were originally black anodized when new. All that time with the reflected light out of the tank really bleached them out considerably. That'll be getting painted up. I want these lights to look sharp when they are finally retired.

 

[ROB2005]

How many of the AI blades are you getting?

 

[hhaase]

I picked up two, and to be honest I don't see a need for a third. PAR output on just the two blades is right where I wanted.