co2 stability

Hello everyone, I recently setup a pressurized co2 system and am having some problems with it. My tank is a 75g pretty heavily planted tank with 4 T8 bulbs running at 2xODNO each. This brings me to about 3.4wpg T12 equivalent. I'm running 2 x 5000k Philips ADV850 and 2x 8000k AGA bulbs. I'm dosing EI 5ppm (pretty decent fish load) nitrate 1.5ppm phosphate and 6ppm total potassium 3x a week, also 15ml of flourish comprehensive 3x a week. I've been having some problems with thread algae which I'm pretty sure is due to the unstable levels of co2 throughout each 24 hour period. I'm running my lights 8 hours a day 12pm-8pm my co2 comes on at 10am and goes off at 6pm. The only surface agitation is the spraybar from my XP3 which lightly skims the surface and creates surface movement along with very light surface agitation. Now what's happening with the co2 is at 10am when the co2 turns on my drop checker is a dark lime green, by 12pm when the lights go on it's a nice pea green which is where I'm trying to keep it. The problem arises about an hour or 2 after the lights come on, around 2pm when the drop checker goes back to being a dark lime green again. I've tried injecting copious amounts of co2 in the tank both before and during the first 2 hours of the lights coming on and the results are always the same no matter what I do. I've even had the co2 going so strong that the bubble counter was just a stream of air and my internal reactor could not keep up with dissolving the co2 fast enough and the top 1/3-1/2 of the reactor is just completely filled with air. I can actually see the water shooting in from the powerhead. The next thing that happens is around 4pm the co2 starts to dramatically climb, even if I have turned off the co2 and reactor. By 6pm the drop checker is a very yellowish green and my fish start showing signs of stress and labored breathing. At this point I have no choice but to turn an airstone on full blast in an attempt to dissipate some of the co2 from the tank. The drop checker stays the yellowish green color till about 1-3am at which point it's a nice pea green and I turn the airstone off. Then it's back to the beginning with a dark lime green at 10am before co2 comes on. I figure the problem stems in my reactor and it's not able to dissolve co2 fast enough for the plants to consume. The reactor is DIY and powered by a rio 600 200gph pump. It's just like all the other gravel vac reactors out there. The water comes in from the top the co2 from near the bottom and is contained in a 9" clear tube, and a sponge at the bottom the keep the co2 bubbles from escaping. I've tried making multiple modifications to it including using a wooden air stone for finer bubble injection into the reactor, changing the output of the water from just the round fitting to a flattened jet for more pressure, and even added dual venturi to it. These modifications seemed to help dissolve co2 a little faster but to no avail as I'm still stuck in the same cycle. My only thoughts at this point is to toss out the reactor and just have the co2 bubble up from wooden air stone and into the powerhead to be chopped up even more than shot across the tank. Any ideas or insight into my problem is greatly appreciated. Also I apologize for the very winded post, just trying to provide as much detail into my situation as possible so you all can get a good idea of what's going on. Thanks in advance for any help you can provide.

 

Update: I have now removed the reactor and have bubbles feeding strait into the powerhead which chops them up into micro bubbles and shoots them across the tank on the opposite end of the spraybar about 2/3 of the way down in the tank. So the micro bubbles are shot across the tank and slowly rise then are hit by the spraybar current when they get near the top and shot back across. I'm hoping that this type of co2 delivery coupled with the extra current will help get co2 levels equalized throughout the tank and thus eliminate the algae and increase co2 levels. The bubble rate I would guess somewhere between 4-5 bps as it is almost too fast to count. It's ugly, low tech, slightly noisy, and fills my tank full of microbubbles but if it gets the job done better than the reactor I don't really care.

 

Second Update: So I took Vaughn's advice and cut my impeller blades in half and bent them. This worked to an extent but dramatically reduced the flow from the powerhead and the bubbles weren't that much smaller than with the stock impeller blades. I did some googling and stumbled on this post. So I got some bio balls and made myself a pin wheel. Luckily if fit very snugly onto my impeller after I cut the blades off and locked onto it so I didn't have to use any glue. It works amazingly! The bubbles that come out of the output are so fine now you can barely see them and most of them dissolve immediately. Also the flow is as good if not better than the stock impeller was. I'll keep this post updated on the effect this method of co2 injection has on stability of co2 levels vs the reactor. If anything at least I have more flow in the tank than I did before, although my discus aren't too happy about it lol.