March 2023. I was on the third step of an aluminum ladder in our propagation house, one arm wrapped around a yellowed plastic lens, the other reaching in to wipe dust off fluorescent tubes. The basil underneath looked ... okay. Not great. Okay was the exact word I used when people asked how the new expansion was doing.
If your search history includes how to remove plastic cover from fluorescent light fixture, you know the motion I was in the middle of: push the lens up on one side until it clears the metal lip, tilt it down, then slide it out before gravity does it for you. It is a two-minute job on a clean office ceiling. It is a different story after a year of greenhouse humidity, when the plastic has warped and an end clip crumbles in your fingers. I still have the scar on my forearm from catching that diffuser.
That afternoon, I added up what our free fluorescent lighting had cost over the previous year. Then I ran the same numbers for the Mars Hydro LED grow light system I had been postponing. The result made me feel stupid. This is what the math looked like, and the checklist I use now so I never repeat the mistake.
The free fixtures that weren't
Quick background: I handle operations and purchasing for a family-run greenhouse that grows baby leaf greens, basil, and microgreens for wholesale accounts. I have been in the role for six years, and I have personally made—and documented—three significant purchasing mistakes. This was the biggest one. Counting wasted electricity, replacement parts, crop losses, and labor, it came to roughly $4,700. (The other two: wrong spray nozzles, $38, and an undersized dehumidifier, $600. Small in comparison, but each one taught the same lesson about buying on price instead of total cost.)
In late 2022, we expanded into a new propagation bay. A friend whose retail store was closing offered me nine four-foot, four-lamp T8 troffers with the plastic covers intact. Free. A few tubes were dead, but I knew my way around fluorescent shop lights. I took all of them.
Free fixtures. Try arguing with that when you have just spent your entire budget on benching and HVAC.
Here is what free actually meant in practice.
The fixtures each drew about 145 watts including ballast losses. Nine fixtures at a 16-hour photoperiod came to roughly 21 kilowatt-hours per day. At our blended commercial rate of about $0.13 per kWh, that was $80–90 per month for one room. Not catastrophic on its own. Then summer hit, the AC ran longer to offset the heat radiating off the tubes, and that cost showed up in a different line of the utility bill.
The lamps were not free either. We ran 36 T8 tubes. After about a year of 16-hour days, their usable output drops off even if they still light up. Replacing 36 decent 6500K tubes cost about $150 per change-out. I remember thinking that is like another utility bill, and then not thinking much harder about it.
Ballasts started failing in month nine. We lost two in the first year—about $50 in parts plus my installation time. Not a big deal individually. But that was the pattern with those fixtures: never one large cost, always a steady drip of smaller ones. I was not tracking the drip because I was busy feeling good about the purchase price. Which was zero.
I don't have hard data on how much light those yellowed plastic covers were blocking by month fourteen. What I can say anecdotally: after I figured out the removal trick and actually cleaned them, the difference was obvious enough that our part-time greenhouse helper asked if I had installed new tubes. A dirty lens costs light every single day, and no one puts that on an invoice.
Cleaning, heat, and the failure that changed my mind
The real tax was maintenance. Every two weeks, I took all nine covers down, wiped the lamps and reflectors, and put everything back. That was about two hours of labor every two weeks, year-round. In July, it also meant standing under hot tubes while the AC ran behind me. I got very good at removing those diffusers. I also got very tired of it.
The most frustrating part is that I walked into this with my eyes open. I chose the free fixtures because I thought I was being smart with money. The free logic let me ignore labor, energy, lamp replacement, heat load, and the fact that the plants were growing slower than they should have been.
What finally ended the experiment came in February 2024. A mechanical timer relay welded shut over a weekend and left one set of lights running for 72 hours straight. The heat from the fluorescent tubes cooked a tray of basil starts and set another six trays back by about ten days. Ten days of propagation time across seven trays does not fit neatly into a spreadsheet, but it shows up at the end of the quarter. That was the moment I stopped comparing fixture prices and started comparing systems.
The comparison I should have run in late 2022
I will be honest: this was not a controlled horticultural trial. It was a purchasing decision made with a spreadsheet, a quantum sensor, and plenty of second-guessing. The area I needed to light was about 450 square feet of bench space plus a small germination shelf.
By 2024, LED lighting was no longer a frontier technology in commercial horticulture. The only real question was whether our budget could absorb the upfront cost.
On the fluorescent side: fixtures $0, about 630 kWh per month, roughly $150 per year in replacement lamps, four hours of maintenance per month, and mechanical timers with no dimming or alerts.
On the LED side, I priced three Mars Hydro TS 3000 LED grow lights for the main benches and one Mars Hydro 150W grow light for the germination shelf. The TS 3000 fixtures draw around 300 watts each at full power. But I rarely run them at full power. For leafy greens, running them at about 70% still gave us better, more uniform light than the old troffers, while cutting the room's energy use from about 630 kWh per month to roughly 300.
The fixtures work with Mars Hydro's Iconnect smart controller. The Zigbee smart integration was the feature I understood least when I ordered it—and the one I would least want to give up now. I can set photoperiods, adjust dimming in steps, and get alerts if a light goes offline. After the timer failure that cooked our basil in February, that capability alone justified the upgrade for me.
The phrase people use for this is total cost of ownership. My version: the price per fixture is only the first line of the story. The full calculation includes electricity, lamps, maintenance labor, heat load on cooling, crop timing, and the risk of a failure mode you have not planned for. Adding those up, the Mars Hydro system paid for itself in about 14 months by our rough accounting. Energy savings alone would have taken longer. The real return came from faster propagation, fewer losses, and all the cleaning hours I stopped spending.
What changed after the switch
We installed everything in April 2024. The first thing I noticed was not the energy bill—it was how much less time I spent on ladders. No diffuser removal. No tube changes. No ballast replacements. Cleaning is now a quick wipe of the fixture fins with a dry cloth every couple of weeks.
Plant response was less dramatic than the marketing photos suggest, but it was real. Our spring transplant cycle shortened by three to four days on average. Leggy seedlings became rare instead of normal. I cannot isolate how much of that came from spectrum, uniformity, or simply having a steady photoperiod. It does not matter much from a business perspective. The crop flow improved, and the electricity bill for that room dropped by more than half.
I still kick myself for not running this comparison sooner. If I had done it in late 2022, the upgrade would have been installed a year earlier, and this expensive lesson would have cost me a few hundred dollars instead of several thousand.
The lighting checklist I use now
Before I approve any lighting purchase, I write down the full cost over three years:
- Actual purchase price, including shipping and mounting hardware.
- Annual electricity at our real kWh rate and real photoperiod, not the number on a marketing page.
- Replacement lamp and driver costs over expected lifespan.
- Maintenance labor: cleaning, cover removal, tube changes.
- Heat load and extra cooling cost in warm months.
- Control capability: dimming, scheduling, alerts. What does a timer failure cost?
- Crop response, estimated conservatively rather than from best-case claims.
One note on timing: prices and specs move. The comparison I ran in spring 2024 used our local electricity rate and the specs Mars Hydro published at that time. As of January 2025, current TS 3000 and 150W product pages include updated PPFD maps and pricing. Verify those before building your own spreadsheet. The numbers will differ from mine. The framework will not.
We still have the old fluorescent troffers in storage. Every time I see them, I remember how easy it is to confuse free with cheap. The most expensive lighting decision I ever made was accepting fixtures that cost nothing to buy and everything to run.