Very nice Larry, you are truly an inspiration to growers who like to tinker. 
Out of curiosity, what kind of budget are you at build wise (so far)?
1 Like
Thanks!
I havent added it all up recently, and Ive lost track somewhat with all the returns and re-orders, but I think Im somewhere between $350-$375. Maybe as much as $20 - $30 of that total has been spent on adapters to go from one size pipe or fitting to another. If you are going to do this, its well worth spending a little extra up front to try to get everything the same size and type of fitting. The two main things I screwed up on were the filter, and the check valve.
Keep in mind that total does not include the grow tent, lights, 12 volt power supply, temp controllers, or rez. Its just for the HPA parts, and root chamber stuff.
1 Like
Some more tank volume math.
This new tank is listed as 4.5 gallons.
I just did a series of tests with the pressure switch set to turn ON at 85 PSI and OFF at 115 PSI. Im able to get .86 gallons or 3.25 liters of water out of the system with that pressure differential. The bladder is pumped up to aprox 82 PSI. That works out to a true volume for the tank of 3.7 gallons. Thats higher than I was expecting, so Im happy with that.
It takes the Ukla solenoid pump right at 7 minutes to replace that water, and get back up to 115 PSI. One unanticipated consequence of that 7 minutes of run time, is that the pump heats up from 72F to 155F. Thats well within the spec for the pump, but I prefer it run cooler than that, so I rigged up a small 12 volt .9 watt CPU cooling fan to blow on the pump. With the CPU fan running, the temp maxes out at 115F. Thats a very comfortable temp. I would prefer a 110 volt fan that was only on when the pump is on, but Im not going to sweat an extra .7 watts.
For now, Im going to leave the cut off pressure set to 115. I’ll make adjustments depending on what my final flow rates actually are. Im ok with the pump cycling a little more often to reduce the length of the run time.
At the max flow rate Im anticipating, and assuming 4 nozzles, the pump should only have to run that 7 minutes once every 10-12 hours or so.
So, I can cross off items 1.2 and 3. Now to work on a fixture to hold the solenoids and nozzles.
2 Likes
After re-reading several of Atomizers old posts, and doing some more math, Ive changed my mind. Im going to start with three nozzles, and see how it looks.
One of the things he said at one point had to do with over misting and getting the roots too wet.
If you have too many nozzles, you might not be able to reduce the ON times enough, to keep the roots from getting too wet, without loosing mist quality due to the lag in the system. With fewer nozzles, you can always increase the mist time.
I happen to have a bunch of left over PVC fittings and pipe from various projects, so I made two fixtures to hold the nozzles and solenoids. One will hold 4 nozzles and one will hold three. It will be easy to switch between them if needed.
The construction is simple. I have a heat gun that I use to heat up the PVC pipe. To make the flat, angled mounts for the solenoids, I just heat the ends until they get soft, then squeeze them between two pieces of wood in a vice. Bend the pipe to the angle you want and allow to cool. Same deal for the two angled legs on the three nozzle fixture. Heat until soft, then bend to the angle you want and allow to cool. If you dont have a heat gun, a propane torch will work if you are careful. You can even do it over your kitchen stove - as long as the wife doesnt catch you 
These are just sitting in the tent to make it easy to take some pics. They will end up sitting on the floor, inside a water heater drain pan. The pan will catch any runoff that drips out of the chamber. The solenoids will be outside the chamber and the nozzles will just stick inside. The root chamber is about 23" deep, and these are tall enough to place the nozzles aprox 13" down from the top of the chamber. They will be firing up at about 45 deg or so.
3 Likes
That’s pretty cool. The PVC trick will come in handy.
Very “pro” looking.
1 Like
Ive used that trick a bunch over the years making racks to hold airplanes, rod holders for the boat, or just odd shaped brackets and holders. I even build a drone out of PVC pipe. It wasnt a very good drone, but it did fly
I finished roughing in the fixture with three nozzles so I could run some tests. Thanks to Atomizers suggestion about how to light the chamber, the mist shows up much better. Im getting a solid 3 to maybe as much as 4 seconds of decent looking hang time. They dont show up in the video, but there are still some small droplets floating around after 5 seconds, when the next shot starts. Im assuming those are the smaller ones. They seem to just kind of float aimlessly rather than dropping.
With three nozzles, the coverage seems to be ok, but not perfect. There is an area all the way around the perimeter of the chamber at the top that barely gets any mist. I can feel it on my hands, but its not much compared to the center area. The ‘dead zone’ is maybe 3" to 4" wide. Everything inside that radius seems to be getting good coverage.
Tomorrow I’ll play around with pointing them more up than they are now to see how that does. They are angled up at roughly a 45 deg angle now, and the nozzles ended up 11" down from the top.
I have lots of little details still to do to get things neater, seal up the root chamber, fabricate a catch basin for the runoff so the PPM meter will work, etc etc, but the main work is complete. A few more days and I will be ready to start some clones!
Here is a pic of one of the solenoids under the work bench.
Here is a shot down into the chamber.
Here is a quick video. This is 1 second ON and 5 seconds OFF @ 80PSI.
3 Likes
I re-ran the flow tests now that the nozzles are in place, and I have the entire system up and running. I did runs of 100 cycles at .5 seconds, 1 second, and 1.5 seconds of nozzle on time.
At .5 seconds, the flow rate for all three works out to 3.92 ml/second.
At 1 second its 3.69 ml/sec
At 1.5 seconds its 3.60 ml/sec.
Not surprisingly, it looks like the extra drips have a larger influence the shorter the ON time is, with a significant increase as it gets under 1 second.
It looks like these final flow rates are slightly higher than the original test. The individual nozzle flow rates went from around 4.3 l/hr to an average of 4.4 l/hr. There may still be some air in the system, so maybe that will decrease a bit over time.
Three nozzles looks, to my uneducated eye, like it is just barely good enough to provide good coverage in the chamber. Things improved a lot as far as the dead zone around the edges just by pointing the nozzles slight more UP than before. Now, most of the chamber walls get fairly even misting from top to bottom all the way around. There are a still spots that get more or less mist, and a few small spots that get almost none, but it looks ok to me for a starting point.
Timing wise, it looks like I need at least .5 seconds, with three nozzles, to fully fill the chamber. The center, upper portion gets the most as that’s where the nozzles are pointed.
Given all of that, I did some more math based on Atomizers recommendations.
.02 - .04 liters/gallon of chamber for each mist cycle for early growth, up to .06 liters/gallon for later growth.
For daily through put, the targets are 1.3 liters /day early on and 2.65 liters/day in later growth.
A .5 second On cycle gives me .043 liters/gallon in my 45 gallon chamber. That’s a little hi, but I don’t think I can go much below .5 seconds and still fill the chamber. I also doubt two nozzles will give decent coverage, so I think Im stuck with that as a minimum.
An on time of 1.0 seconds gives me .082 liters/sec. So it looks like Im going to need an on time of around .7 seconds to hit the later growth target of .06 liter/gallon.
So to hot the 1.3 liters/day target for daily through put in early growth, that works out to an on time of .5 seconds, and a pause time of 136 seconds. Later growth is .7 seconds on and 89 seconds off to get to 2.65 liters/day.
So, if I did all that correctly, I now have starting and ending cycle times to work from. I’ll have to let the roots, and plants tell me if Im close or not.
Here is a pic of the flow testing.
Ive been moving slowly the last several days, but I have made some progress.
Ive been worried about the brass pressure relief valve I installed contaminating the nute mixture. I ran across another of Atomizers posts suggesting using a check valve between the brass T&P valve and the tank. That should isolate any dissolved copper from the nute solution. I will probably need to burp the valve every so often to eject any contaminated water and make sure its not corroding shut, but this will be better than before and a lot cheaper than a stainless valve.
I also managed to ‘fix’ my pressure reducer. It was only allowing me to set the nozzle pressure up to about 85 PSI. Its supposed to go up to 150 PSI. I took it apart and added a nylon washer under the spring. It now goes up to at least 110 PSI.
With the higher pressures, Im now getting much better looking mist with much longer hang times - close to 10 seconds @ 100PSI. The flow rates will increase because of this and there are other consequences. It takes the pump longer to top off the accumulator tank at the higher pressures, and the flow rate in crease will mean shorter times between the pump cycling. I need to re-do the flow tests at higher pressures to see how much more it will be.
I can get a second pump for $12. I may do that and plumb it in parallel to the existing one. That will halve the fill times and give me a back up if one fails. Still thinking about that.
Unfortunately, even 10 seconds is way too short a hang time according to Atomizer. He has been very helpful in many ways, like that tip above, but Im getting annoyed at him for his attitude on nozzle information. He absolutely refuses to give out any details on nozzles or answer any questions about them. That seems really strange to me. Why share any information at all if your going to withhold the one critical part that everything else hinges on?
Its become clear that the “math” he proposes using for cycle timing, and flow rate numbers just dont work on these Hypro nozzles, or the first ones I bought. The hang time Im getting is at least an order of magnitude less than what he is talking about, but he wont answer any questions about that, or anything else on, or near the subject of nozzles - other than saying that my hang time isnt nearly good enough. I need minutes of hang time, not seconds.
The problem is, if you dont have the correct size droplets, with the correct hang time, at the correct flow rates, his numbers, and most of his other advice, just wont work. You will end up with over wet or over dry roots, and drain to waste will be too $$ - and of course, no fuzzy roots. All of this adds up to making it virtually impossibly for a newbie to get the same results he is getting - unless they have the time, and $$$ to keep trying different nozzles at random.
So, just in case anyone has been following along and was thinking about trying this with these nozzles - forget everything Ive said up to this point. NONE of the math Ive done so far is worth crap with these nozzles - and I have no idea what nozzles might possibly work better.
Anyway, sorry for the whining rant. I know it sounds pathetic, but Im feeling pretty frustrated. Like Ive been lied to or deliberately misled into buying the wrong nozzles - twice. I thought these forums were for sharing information and helping others get over the hard parts, not make them worse.
(grumblebitchmoangrumblegripewhinegrumble…)
1 Like
Hmmmm. Just re-read the post above and it is kind of whiney. Just to be clear, Im not giving up.
Even if I cant get fuzzy roots with these nozzles, I should still be able to get some nice roots that are at least as good as with LPA, and grow nice plants. Of course, LPA is a lot cheaper and easier, but oh well. Im still learning new fun stuff
At first blush, I’d think that 10 seconds of mist floating around is pretty darn impressive. Sounds like you are trying to manage quite a balancing act. How critical is it? Atomizer is saying that his mist is aerosolized for over 100 seconds?
1 Like
I didn’t get that impression…
1 Like
Yes. After saying that my original 1.25 seconds of hand time was ‘not bad for HPA’. his latest comment was that since I didnt have the required hang time, I would have to make do with alternate timing rather than follow his original recommendations - which we had been talking about from day one.
The gist of it is, you need a hang time at least as long as the pause time between mistings - preferably significantly longer. The ideal situation is to have the mist environment remain as constant as possible in the root chamber. In other words, you dont want a wet cycle followed by a dry cycle. You want a constant level of mist of the correctly sized droplets - or as close as you can get. The goal is that Goldilocks level of just right wetness with no big variations either way.
His Goldilocks numbers for my size root chamber work out to roughly .5 second ON and pause times between 89 and 136 seconds. Those numbers are on the hi side for flow rates, so ideally, the pause times would be even longer - more like 100 seconds to 180 seconds or so.
So, even my best hang time so far of not quite 10 seconds is not even close - by 10 to 20 times. There is no way these nozzles could ever work to give the numbers he says they need to produce.
The really irritating thing is, I asked him specifically about the first Tefen nozzles I ordered BEFORE I ordered them, but he did not answer any of my questions. Only AFTER I bought them, and did some tests, and found they were not going to work, THEN he said, “those are junk, you need to toss them in the trash”.
Same thing with these new Hypro nozzles. I asked about them before I bought them, and he said he had suggested them to someone else once. He also said he uses hypro’s in some of his HPA setups. So, I assumed that meant they were a good choice, but it turns out thats not true at all.
Now, he wont reply to any questions about nozzles at all. I gave up on asking for a recommendation for a specific nozzle that would work as described, and just wanted to know what kind of hang time he is getting from his Hypros, and if he is using the same ones I’m using - no answer to that either. S I have no clue if he is using these same nozzles or some other one that he wont name.
If I knew what his hang times were with the same exact nozzle, I would at least know if mine are defective or not. It would be a start. As it is, Im still 100% clueless.
Im not going to let this stop me. I happen to agree with you that 10 seconds of hang time is better than a poke in the eye - especially when compared to any type of LPA. Plus, I know I have smaller droplets than with any type of LPA, so at least I have a start in the direction I wanted to go.
It still remains to be seen if I can make drain to waste work at these flow rates, and just how good or bad the roots will turn out. Thats still worth pursuing even if I dont get fuzzy roots.
Its looking a lot less likely, but fuzzy roots is still the goal.
1 Like
I wonder about this. It’s going be pretty humid in there. How dry could the roots possibly get after two minutes minus the hang time? p.s. I know nothing about this technique.
You are blazing paths. I have a feeling your going to be successful one way or another…
1 Like
Its kind of it like a ‘Just-in-time’ supply system for a retail store. You dont want to supply more ‘stuff’ than will fit on the shelves, because it just piles up and gets in the way. When that happens, the customers cant get what they want because all the excess ‘stuff’ is in the way. With the roots, too much water gets in the way of O2 delivery, and clogs things up.
At the same time, you dont want any empty shelf space with customers having to wait, and not getting what they want in a timely manner. With the roots, that means they starve for nutes and water even though they may be getting plenty of O2.
According to NASA, if you have a steady supply of 50 micron drops, the roots will have the perfect environment to up-take nutes, and O2 at the perfect rate for optimum growth. Any time spent too dry or too wet, is less than optimum for best growth.
The trick is finding that balance.
With the ideal nozzle, the droplets will be in the 20-80 micron size, with the bulk of them around 50 microns. At that size, you do get long enough hang times to be able to do Atomizers schedule, and maintain a reasonably constant supply of mist at a reasonably good level of ‘wetness’. The math works in that case.
My nozzles dont come close to having that kind of hang time, which tells me the droplets are way too big on average.
You dont want them too small either - like with the ultrasonic foggers. They tend to produce 5 micron or so droplets, which float a loooong time. The problem is that 5 microns isnt “wet” enough. Its a dry fog. Plus the roots dont absorb that small a droplet as efficiently.
The problem is, if you cut the size of a droplet in half, the volume goes down by a factor of 8. The volume changes as the cube of the diameter. So a 50 micron droplet has 1000 times as much water as a 5 micron droplet.
There are a number of people who have been doing HPA for a good long while, so I am far from blazing any trails 
However, there are a lot fewer of them than any other type of hydro. From what I have seen reading threads on several boards, most of them dont get even close to Atomizers goals. They may end up with very nice roots, and plants, but not fuzzy roots and drain to waste with out tooo much waste
The few who do seem to have it all together, seem to be a fairly secretive bunch, and dont like to share. Thats especially true of those doing Air Atomized Hydro. Thats something I just dont understand or agree with.
Atomizer does share a lot of info on the theory - as long as its not about specific equipment questions - especially nozzle questions. His posts are well written and dense with information, but often short on misc details, and context, which can be a challenge for a newbie, because he usually wont answer questions once he has posted something. Trying to push him beyond what ever he is in the mood to share just doesnt work. He only answers maybe one question in 5, so you really have to poor over his comments carefully.
I really like his theories and the ideas he presents. He is a wealth of information if you dig deep enough. I mainly wish he would share more equipment details, and not wait to tell someone they bought the wrong stuff until after its too damn late.
1 Like
I would highly recommend, that any brass fittings are changed to 316L Stainless Steel before you start growing.
If plastic isn’t an option
1 Like
I wish I could. The only brass piece in the entire system is that pressure release valve. Every other piece of metal is stainless.
Unfortunately, I have not been able to find any plastic valves that are true safety release valves. The stainless versions are way outside my budget.
Thats why I added the check valve between the brass part and the rest of the system. Water/nutes can only flow one way. Water can flow toward the valve, but not back to the system. That should prevent any dissolved copper from getting into the nute water. That still leaves the risk of a corroded safety valve after some time, so I will have to keep a close eye on that.
Hmmmmm. It just occurred to me I can treat the brass valve with a marine corrosion blocker. The check valve will also prevent any chemicals in the blocker from getting into the system. Im going to do that today!
Try google PTFE Pressure Relief Valves, im sure you can find a unit that fit’s your needs.
This is rated to 90 psi
1 Like
Thanks! It never occurred to me to search under that category. Unfortunately, those appear to work like the other plastic one I tried. They seem to be more of a pressure reducer or flow regulator and not a true safety release valve.
I found some cheap stainless valves, but none over 100 PSI. I need 150 PSI for this setup.
I’ll keep looking though.
Thanks again for the suggestion. I did some more searching with those terms and found several plastic valves that would work just fine. Unfortunately, they start at about $160 for the cheapest one, up to $2000+ for ones that will handle 150 PSI. Im stuck with this option for now at least.
I made some more progress on the small details. I think Im close to being ready to do some trial runs with clones and seeds.
For the seeds, this is another good idea from Atomizer. he may be a jerk some times, but he is one smart jerk
He used coco mat of some type, but all I have is some fabric batting. We will see if this works. Basically, you put something - like coco mat or fabric - under a foam net pot puck, then cut a small hole in the puck, and drop a seed in and let it grow. You also need to place something over the top of the seed. Im going to use either some perlite or another small piece of fabric. I will probably just fill the hole with perlite.
I sealed up all the holes and open places in the root chamber and installed a catch basin in the bottom to collect runoff water so I can monitor the outgoing PPM with my dual meter. The other chanell is plumbed into the tubing near the pressure reducer. That way I can keep track of the difference in PPM as the roots drink. Maybe.
Atomizer recommends using a fabric that is water proof. My fabric pots are not water proof at all. I have been worried they would just soak up all the excess water that drains down, and I would not be able to collect enough to measure. So I made a drain piece for the bottom out of some PCV parts that traps a piece of black plastic from a garbage bag. I have that plastic sheet covering a large portion of the bottom of the pot. Now, any drips from the roots hit the plastic sheet and run down into the plumbing pieces under the pot so they can be measured. The excess will run out and into the drain pan under the whole mess. It seems to work ok.
I deliberately didnt cover the entire bottom of the chamber. Im only interested in the water that drips from the roots.
That looks like metal under the bottom fo the pot, and it is. I hot glued an old socket to the pvc cap that Im using as a drain. I needed some weight to dimple the bottom of the pot enough to form a bowl shape. The probe connection is on the right side.
I also made some mods to the net pots to open up the spaces so mist can get through better. I almost cut the entire bottoms off, but I think the roots will get some support if they are wrapped into the net pot. It should also help to keep the plant stems from sinking down into the foam puck as they get bigger.
LOL! I just found this poor baby. She is left over from my earlier clone experiments. She has been sitting totally neglected for the last few weeks, so is not looking very good. She has roots though, so I think I will toss her in there just to see what happens
1 Like
Im finally ready to put “something” in the tent and start playing!!
I didnt think this through very well though. I have one single seed germinating in some paper towels. The seed has cracked and I think a root will be showing soon. As soon as it does, I will toss it in the puck and see what happens. Unfortunately, due to some clerical errors on my part, and being stoned when I stored them, and not getting the lid on well enough, a bunch of my seeds got mixed up. So, I have no idea what this is or if its one of my fem seeds or regular.
I also have that one old nasty clone I want to stick in.
The problem is, Im afraid the seed will want different wet/dry cycles than the clone.
Oh well. This is an experiment, so if someone has to die for science, so be it 
LOL. I didnt realize what I had done when I made my access hatch to get into the roots until just now as I was editing the pics. The joys of unintended consequences!
I covered the foam with some white mylar tent material as an air seal, and vapor barrier. The access panel is also cut out , and covered.
Im pretty sure this is going to be a happy grow! This will put a smile on my face every time I look in the tent
Trying to get good sharp pics may be a challenge. I may end up cutting an access hatch in the side of the fabric pot. I dont want to do that unless I have to. Im worried about air leaks drying out the roots.
2 Likes