Featured Article:
Growing Notes—The Hydroponic Adventure Continues: Creative Chaos and Catastrophe
Two ideas seem to be in dynamic tension when it comes to my hydroponic systems this summer. The first idea, kaizen, is a Japanese word that means continuing improvement. Commitment to perpetual improvement was imported to Japan after World War II by Americans trying to help the Japanese get back on their feet. Edward Deming taught the Japanese a set of principles that emphasized a never-ending commitment to quality. The second idea is due to G.K. Chesterton, the British journalist and sage, who once said: “Anything worth doing is worth doing badly.” That is, you must go through a learning process to do anything well.
As the tomato vines on my Winter II (home-office hydroponic garden; see The Growing Edge Vol. 9, No. 6) system started pressing against the roof, I wondered what I should do when spring arrived. I wanted to fully automate the system with injectors and tank filling that would be governed by sensors, as well as electronic nutrient and pH monitoring. But I just didn’t have the time. So, I decided to start with smaller improvements for this spring and summer.
My Summer II (see The Growing Edge Vol. 10, No. 4) system had worked adequately, but because I didn’t keep track of which seeds I planted, I ended up with lots of ‘Sun Gold’ cherry tomatoes and not so many of the other larger varieties. Still, the ‘New Jersey Giant Standards’ were good, and the ‘French Dona’ tomatoes were nearly perfect.
A commitment to kaizen means that you reflect on what didn’t go so well the previous year and think of improvements. The biggest problem I had wasn’t with plant growth, but plant support. I used an A-frame that worked when the plants were small, but as they grew larger, the vines were all tangled. However, the watering-trough float valve was a success. For this year, I decided to expand the NFT system to four tubes, which meant building another system but it wasn’t difficult because the design is simple.
My basic NFT hydroponic garden consists of two 10-ft pieces of 4-in. diameter tubes of PVC pipe. I had drilled seven holes in each pipe (which is probably too many for tomatoes because they grow too close to each other, causing the tangling). Then, I mounted the pipes next to each other across two sawhorses that are high enough so that the pipes with elbows will return the nutrient to the tanks by gravity-feed. The nutrient is fed from a holding tank into the system at one end using 1/2-in. plastic pipe attached to a splitter so that each PVC tube receives a feeder tube. The nutrient returns to the tank by gravity flow, so the sawhorse opposite the tanks has to be higher than the one near the tanks.
I use plastic trash cans for the nutrient-solution holding tanks. The ones I used last year seemed a little flimsy, so I bought thicker-sided soft-plastic cans as one of the small improvements. The previous summer, I also purchased some foldable steel sawhorses with the intention of expanding the system, so I was ready to double my production.
Toward the end of last season, I started collecting my own seeds. The procedure was simple: I removed the skin of the tomato that I wanted seeds from and blended the pulp. Then, I poured the pulp through a strainer to separate the seeds. I tapped the seeds onto an absorbent paper towel to let them dry. When they were dry, I transferred them to a fresh paper towel, wrapping them, and put them in a labeled envelope. I was curious to see what second-generation ‘NJ Giant Standards’ would be like.
Still dreaming about the new summer system, I prowled through the new tomato seed catalogs. I sent away for the latest ‘NJ Giant Standard’ seeds and received a packet of about six seeds, which I labeled “GS ‘99.” The seeds I saved myself were labeled “GS ‘98.” I ordered five varieties from Totally Tomatoes’ catalog: ‘French Dona,’ ‘Presto,’ ‘Bush Early Girls,’ ‘Tom Terrific 6816A,’ and ‘Tom 802C.’ A friend on the Internet, who noticed my e-mail signature “On the search for the PERFECT tomato,” sent me her “perfect tomato,” an Australian variety called the ‘Grosse Lisse.’ I was set with eight varieties of tomatoes!
On April 18, I planted 40 small rockwool cubes. (I wasn’t going to lose track of which varieties went where this year. Each seeded cube received a toothpick with a little flag that marked which kind of tomato it held.) I moved the cubes outside after the tendrils of the baby tomato plants appeared. And, because of the danger of frost in the Shenandoah Valley until about May 15, I faithfully brought the seedlings into the house each night.
Meanwhile, I ran over to the local home design store and bought four hose clamps and two more pieces of 4-in. diameter PVC pipe. I wandered around looking for ideas for a support system for the tomatoes. The supports would have to be high because the grow tubes are almost 3-ft above the ground. The store was selling bundles of furring strips that might be suitable. I wanted something that could be put up and taken down easily. I used flexible elastic straps (Bungie cords) to hold up the sodium vapor light in my indoor system, so attaching the support structure to the sawhorses with Bungie cords seemed natural.
This year, I set the tanks outside next to the house windows so that, if I added some instruments, the electrical wires could be snaked through the windows. This arrangement changed the system orientation from last year; now the NFT tubes were perpendicular to the back of the house. Last year, I had them aligned horizontal with the house to take advantage of the slight grade in the back yard so that I didn’t need to make one end of the system higher than the other.
But the yard is dead flat going away from the house. To compensate, I quickly stuck some boards under the legs of the sawhorses farthest from the house. Then, I filled the trash cans with water and strung the water hoses to the end of the channels, holding them in place with duct tape. The pond pumps worked nicely, sending the water to the back of the far end and flowing it back into the tanks. All the system needed now was nutrient solution and plants.
By late May, the plants in the rockwool cubes had grown to about 7-in. high; time to put them in the system. People often ask me how I do that: I just put them in. Sometimes, the nutrient flow is enough to cause the cubes to float down the channels, but they can’t get too far generally. After the plants get a little bigger and are tied to the support string, they stay in place.
I drew a color-coded map in my notebook that showed which tomato variety was in each hole. This year, I was going to know which tomatoes were where. (I didn’t attach the support strings right after placing the plants in the system. After they had grown some and the stems thickened up, I tied the bottom of a support string loosely around the stem; the top of the string is anchored to the support structure. As the plant grew, I twisted the stem around the string. This method worked fine until I got lazy and stopped twisting diligently, or until all the suckers that I was too lazy to trim sent shoots every which way. I’m a member of the “let ‘em grow any way they want” society. After this year, I’m thinking of joining the “one vine is all you’re allowed” group.)
Although I had put the seedlings in the system, I didn’t have the support structure up. After all, there’s no point in putting up the support structure when the plants are too small. But after they’re in the system and being fed constantly, they grow a lot faster, so it was time to get crackin’.
I figured we could build the supporting framework and just attach it to the sawhorses with Bungie cords; all that we needed was a couple of saw cuts, a few screws or nails, and we would have an 8-ft-tall support for the tomatoes. The “we” in this plan included my 14-year-old son Christopher who needs an outside interest. He lives on the Internet and plays computer games constantly. The rays from his CRT just aren’t enough to erase the indoor pallor so a little sunshine would do him good.
On Saturday, I got out the circular saw, a drill, some hammers and nails, and the furring strips. I roused Christopher who believes Saturdays are designed for sleeping late in the morning. After quickly sketching out a design on the back of an envelope, we were cutting furring strips and hammering them together. Christopher developed a unique reclining hammering style, lying on the grass with the framework pieces next to him. After we were finished with the supports, we outfitted the crosspieces with screw eyelets to attach the strings that hold the plants. To ensure that everything was just so, we used the Bungie cords to secure the ends in place before we drilled the holes that supported the crosspieces. Because the plants had already grown some, soon after the support frames were up, I tied onto the stems and started winding them around the strings.
The float valves were working like they should. I left them connected to the water source at the side of the house and checked the electrical conductivity (EC) every morning to see if I needed to adjust the EC and pH levels. I’m usually curious to see how much the plants are transpiring, however, and the easiest way to monitor the transpiration rate is to turn off the water to the float valves and observe the change in the water level over the course of a day. When the system is just starting out, the plants are small and they don’t consume or transpire much. So, theoretically, you can go days without doing much to adjust the system. But, I was in trouble right away.
When I drew up the system map, I creatively named the two systems “A” and “B,” and each tube was designated “a” or “b.” I numbered the holes one to six, starting from the end without the nutrient-holding tank. So, each hole had a unique designation, such as “Ab5.” The first hole contained the nutrient-feed hose.
Five days after starting the system and one day after installing the support frame, I recorded in my system log: “Checked the system this morning, and system A was down by two-thirds indicating a fairly significant leak.” (Talk about British understate-ment.) System A was losing about 20 gal, but system B was fine.
I thought that a leak that severe should be obvious. The elbows on the channels were dripping but it didn’t look like much, certainly not like 20 gal overnight. But, just in case that was the source of the problem, I reseated the elbows and increased the tilt of the system, which only slowed down the leaking. I added plumber’s putty to the elbow joints. Because it looked like some of the nutrient might be splattering out of the tank, I added extenders using duct tape and parts of milk cartons. System A still consistently ran out of water faster than system B. Finally, I replaced the thicker-walled, soft-plastic tank with the hard-plastic tank that I had used last year, and the problem went away.
The first tank looked fine and seemed to hold water, so I continue to be mystified. Meanwhile, the plants were growing like mad. Every afternoon when I came home from work, I walked around the channels, twisting stems around the strings. The plants grew rapidly, about 20% faster in height than the four tomato plants I planted in pots on the side deck as the experimental control. The roots grew fast and long, extending down the channels. The plants in system B grew long roots that reached into the nutrient tank. I’m fairly convinced that the reason that system B’s roots grew much faster than system A’s is that it had a more powerful pump that created a larger and faster volume of nutrient flow.
By mid-July, I started to have problems with system B. The nutrient tank was going down rapidly. So, I replaced system B’s tank with another hard-sided plastic tank but, this time, it didn’t work. I looked over the system for some indication of the problem. The sawhorse at the far end was wet. On closer inspection, I saw that the plastic tube going into the splitter was pressed into the frame in such a way that it leaked slowly. Freeing that alleviated the problem, but it still didn’t go away. Now, system B was losing about one-third of nutrient overnight.
Normally, plants in a hydroponic system use little nutrient in the evening. They need light to absorb nutrient. So, if your nutrient is going down at night, you most likely have some kind of a leak. The system structure was getting hard to see because it was covered with tomato plants and an increasing number of big green tomatoes. But I could see that system B sagged a little in the middle. When I looked closer, the nutrient level was up to the edge of the planting holes. I didn’t see any dripping so this was the only clue I had. Perhaps the liquid was wicking over the side or leaking from some place I didn’t see.
I decided to try to reduce the flow by adjusting the valves on the splitter tubes. The valves are designed to turn the flow on and off, not regulate it, so I had to be careful not to cut the flow off entirely. Well, reducing the flow stopped the leak.
Changing to the hard-plastic tanks introduced a complication that I hadn’t anticipated. If I left the water source open to the float valves on the tanks, the nutrient leaked slowly out of the holes that held the valve handles. This meant that I couldn’t leave the water on during those weekends we went on trips, which was annoying.
If leaks weren’t enough of an aggravation, we had more power failures this year than I can remember in the past five years. The system’s electrical components are attached to a ground-fault detector and, if the power goes out, the ground-fault switches the system off and doesn’t reset the power; it has to be turned on again manually. Once the flow stops, in only a few hours at most, especially on a hot day, the roots dry out and the plants are dead or dying.
One day, I was at work and my monitor went whoosh, the picture collapsed, and I was sitting in pitch darkness. In a moment, the power was back on, and the computer was starting to reboot. I called my wife Jessica. No answer. I only live three and a half miles from work, so just to be safe, I jumped in the car and drove home. Sure enough. The pumps were stopped, and no nutrient was flowing. I reset the power and went back to work. And, before August was over, we had two more power-outages.
When I harvested those first red tomatoes from the back-yard hydroponic NFT garden on July 22, I felt like I had really earned them. Also, they were unusual: a triumph of inadvertent genetic inheritance. I picked five tomatoes, each a perfect little orb weighing about 1 oz. They were too big to be cherry tomatoes, and two of them were a bright golden yellow. So, I checked the identity of the plants. They were GS ‘98s - but they certainly weren’t giants! I brought them in to Jessica laughing. My 1998 ‘NJ Giant Standards’ had been messing around with the cherry ‘Sun Golds.’ Instead of producing big, fat, red luscious tomatoes, I was growing some sort of hybrid giant cherry tomato.
Because I like to know exactly how many tomatoes I can grow using the hydroponic system, I started my running count and weighing in of the tomatoes, sometimes individually and sometimes in groups, as they were harvested. The first entry in my log recorded the hybrid giant cherries: five tomatoes totaling 6 oz. In the next few days, I harvested 10 lb of tomatoes. Then, the flood really began.
Each day that I picked fruit from the system, I recorded the number and the weight. In some cases, especially if the tomatoes had any interesting characteristics, I recorded which hole the plant came from using the system/tube/hole notation. For example, a tomato from Ab3 would be a ‘French Dona.’ The ‘Tom Terrific 802Cs’ were early producers and so were the ‘Bush Early Girls.’ I was particularly impressed by the ‘Prestos.’
Soon, I was harvesting so many tomatoes that they were overflowing the kitchen counters. Jessica pointed out, reasonably enough, that three people couldn’t eat this many tomatoes. Luckily, tomatoes, unlike kittens, are amazingly easy to give away, especially if they’re perfect, seamlessly smooth, hydroponic tomatoes. I took 20 to work and put them on a table in the company cafeteria with the tag “Free Hydroponic Tomatoes.” About 15 minutes later, they were all gone. By the end of August, the system has produced 469 tomatoes weighing a total of 2,135 oz (133 lb). I had to give more than half away. Fortunately, people seemed disappointed when I didn’t bring tomatoes to work.
The tomato glut was finally stopped by a catastrophe. The load of tomato plants caused the support frame to sag, and it had begun to progressively tilt, a little at a time. I wasn’t too worried: the system was heavy-duty, and the tilt didn’t look severe. Then, one Saturday afternoon (September 4 to be exact), as I was adjusting the nutrient level in system A, the channels started to lift up. If I had reacted quickly, I might have had a moment when I could have saved the system. But I didn’t. The whole system tilted over and collapsed, with a loud whoomp, into an impressive heap. Looking over the rubble, I discovered a large cluster of ripe red tomatoes that I hadn’t noticed before, about 2 lb worth.
Later, as I tried to salvage what was left of the downed system, I collected the unripe fruit still on the vines: 256 tomatoes, ranging from large mature tomatoes just starting to turn red to little babies the size of gum drops. After bringing the large ones into the house, Jessica and I wrapped them in newspaper to finish their ripening.
To console myself after the catastrophe, I sat down at the computer and entered the tomato data into a spreadsheet. I plotted the total weight and the number of tomatoes taken off the system against the dates. The result was a steady production curve. If the 256 tomatoes salvaged from system A had reached maturity, they would have added another 72 lb to the total.
Well, my experiment in kaizen wasn’t entirely successful. The tomatoes are still coming fast and furious from system B. We certainly won’t be missing out on tomatoes at home. The tomato fans at work will just have to go to the supermarket. On the bright side, I reflected, “Chesterton was right. Anything worth doing, is worth doing badly.” Now, of course, I have a whole list of improvements to add to next year’s hydroponic garden.