Nick,
I guess I could be considered your local expert on this subject. I met you a couple months back when I stopped in killing time while Sabic was cutting some plex for me, we sat in the back room while you were doing PWCs
Here is the RC thread I help run on scrubbers (it's long)
http://www.reefcentral.com/forums/s...?t=1977420 and a few reference posts
Basics Write Up - Post #1
Updated Summary - Post #2001-2010 (this is also reposted on here, but it takes forever to load. It is also reposted on several other sites, but RC is the most active)
Quick Troubleshooting - Post #1902
Alternate/Updated Sizing - Post #2723
Here's the site dedicated to scrubbers, managed by Santa Monica, who came up with the concept of the vertical-screen scrubber
http://www.algaescrubber.net/forums/ And he is developing a new device that is set for conceptual release in March or so, he's finishing up patent issues currently and I will be building those when he comes out with the design.
Also, technically the current 'modern' design isn't an Algal Turf Scrubber in the true sense, since "turf" generally means a tougher red or green type of algae. This is what is typically grown in a horizontal dump-style system like Inland Aquatics is famous for, and that is an Walter Adey invention (and is also patented). The vertical design is made to grown GHA which grows much faster and does a much better job of filtering. So 'we' call the current design and Algae Scrubber.
As far as me building the current version, I have been holding off on doing so until the new concept comes out. Bryan (Santa Monica) has touted to me that it will be a major improvement in many ways, and urged me to wait a little bit before going full bore with the current design and making my own line of scrubbers.
W/R to other filtration methods, the Algae Scrubber is fully capable of stand-alone filtration. There has been heated debate in the past regarding the skimmer issue, so on the RC thread you will note that the skimmer vs. scrubber debate is pretty much sidestepped. GFO is an easy one. Algae needs N and P to grow, so if you run GFO you limit P and N uptake will drop. Similarly, if you have a ton of established LR and your N is naturally kept in control, your P uptake will drop. This is actually what I have been experiencing in the Grove & Platt tank for some time(0.09-0.16), although after I reduced the screen size and let the growing cycle go for 10 days versus 7, it has started to drop (0.01 last time I checked).
The skimmer debate is hot, IMO, primarily because people are set in their ways; high-end SPS tank owners seem to be the ones that are most resiliently against scrubbers in this regard, and that is likely because the track record has shown that the most successful SPS tanks are running a monster skimmer which keeps the water really, really clean. Also people with a $5000 skimmer don't like the idea that the skimmer is not the best way to do things. But also because it does work to some extent. That extent has more recently been researched and IMO, the jury is still out. What I can tell you is that from a nutrient export perspective, scrubbers and skimmers do not remove the same things. A skimmer removes organics (food) and a scrubber removes inorganics (waste). This being said, a skimmer essentially strips the tank of all nutrients (food), while a scrubber 'waits' until the nutrients have been broken down, and then it removes the waste. This means that more food is available on a more continuous basis for a longer period of time.
Santa Monica runs a 90 gallon tank with 2 100 sq in scrubbers, has mostly SPS, and runs a continuous feed system for a DIY oyster feast, and he's been running the tank for over 3 years without PWCs. He has done many 'experiments' to see what it can handle. In one instance, he wanted to test the effect of Iron overdosing on the system. He incrementally increased the dosage until it got the point where he just dumped in a half gallon of it in one shot. The corals suffered but they eventually all came back (with no PWCs). He also more recently wanted to test the maximum filtering capacity of the dual-scrubber continuous feed system. He calculated the amount of food available on a natural reef and it was some insane amount of food needed - 1 pound of food particles per day per cubic meter, or on a 90 gallon system, 49 cubes of food per day. Right now, he feeds 72 mL/day which equates to 22 cubes per day. 22 cubes per DAY! At one point I think he upped it to about 26 cubes/day and that overwhelmed the scrubber.
I only know of a few people running NPS tanks on scrubbers, but if you think about it, this is where is makes the most sense. NPS corals (and SPS for that matter) need a continous food supply to thrive, just like they do in the wild, especially at night (for SPS too). Feeding this amount of food continuously would mean that you would have a massive upkeep need between changing media, monitoring levels, doing PWCs, you name it. Plus, all of that being done it doing nothing more that stripping the food out of the water before the corals can get to it. With a scrubber (done right) you literally need nothing else.
As far as lighting goes, LED is definitely the way to go. You want 660nm Red with a few 435nm blue (violet) or 455 blue works also. The blue doesn't grow anything, but rather just assists in strengthening the algae (roots and threads). I have a couple or fixtures I had custom made with different combinations of 630/660/435/455 LEDs and was (and still am) planning on running a set of experiments to determine the best spectrum. But for now, most people are getting the best results out of 660nm. Particularly, the e-Shine system 50W grow bar, which had 45 660nm red 1W chips and 5 455nm Blue chips is cheap, $200 for two of them shipped. When you consider that a similar footprint T5HO 2-lamp fixture costs about $65-$70 and you will need to replace the lamps every 3 months, the LED fixture pays for itself in material costs only in less than a year. So if you've got the itch, that's the fixture I suggest, and build around it. Also there is a 660nm Triple-Chip (Osram I believe) that you can use for a nano or smaller build. The problem with 3W chips though is spotty growth, which is why I recommend the 1W chip. It is more efficient (3 1W chips have more power than 1 3W chip) and you can get even spread. Also, experiments are revealing that not only does the LED fixture work equal to or better than T5HO, because the spectrum is essentially 'tuned' to the algae's preferred wavelength, LED powers through to green growth faster, whereas in an initially high-nutrient system, T5HO and CFL would have a 'black' growth phase. Also you can put a 1W array really close to the screen (like 2") and cut the photoperiod back (from 18 on / 6 off to 9 on / 15 off, or less) and get better growth (which means energy savings, which matters more in CA, etc, than here, but anyways).
I could go on and on...boring work day...but I'll stop for now. I tend to just puke out this stuff sometimes...
But anyways, for a 70g, with heavy feeding (lets say 7 cubes/day) then 7 x 12 = 84 sq in, get a couple e-shine fixtures (I'm working on a design to test these actually - I have 4) for $200 and save yourself the DIY hassle 'cause it's just not worth it for that price, and build a box around it. The e-Shine fixture has an effective width (light center) of about 16", so a 16 x 6 screen (plus height for into the slot and dipping into the water) would handle 8 cubes/day (or 26 mL of liquid food per day) no problem and that's a ton (from most people's perspective) of food.