Growdown! Mars Hydro vs Grow Lights Australia

[Prawn Connery]

This comparison has been almost six months in the making, but I think we can finally start. The side-by-side grow has just finished and we don't have the yield or cannabinoid levels yet, but we will be adding these to this thread once the final product has been fully dried, weighed and tested.

So you will all have to wait for the results! Even we don't know what the results are at this time . . . but we will know in a couple of weeks.

To begin, Mars Hydro kindly sent me one of their current FC-4800 LED fixtures to review earlier this year, but I don't think they reaslised that I design LED grow lights for Grow Lights Australia, whom I am affiliated with. We (@Grow Lights Australia and I) thought it would be a great opportunity to pit the Mars light against one of our own, so I will detail the full specs of each light below, but first . . .

FULL DISCLOSURE
We tried very hard to make this the fairest possible comparison we could, and went to some lengths to ensure the Mars fixture was not handicapped in any way (which I will detail throughout this thread). This entailed entrusting a third party with the actual grow. I did not take any part in it.

The "grow-off" was conducted by an Australian commercial grower in an 8'x4' (2.4m x 1.2m) tent that was divided down the middle with one fixture on each side using the same clones, nutrient, and all other environmental conditions. I checked only sporadically on this grow to see how it was going and to take photos. The grower was simply told to do his best, and being a commercial grower he aimed to get the best yield and quality he could. I helped set up the lights with a light meter to ensure the light spread was pretty even – which it was throughout the grow.

Here we are at the start. The clones are Laughing Buddha from Barney's Farm and the biggest and best clone was put under the Mars FC-4800 while the smaller clone was put under the GLA Quad. The clones were grown in run-to-waste perlite, and a divider was placed down the middle of the tent after these photos were taken. Both lights were set at a height to achieve a uniform 500-600 PPFD across the canopy during veg, and 800-900 PPFD during flower. This meant the Mars light was hung lower throughout the grow.

The Buddies side-lighting was not turned on at all during this grow to ensure there were no external factors affecting the lights.


Grow Lights Australia Quad


Mars Hydro FC-4800

 

[Rurumo]

Very cool Prawn! Looking forward to it.

 

[Prawn Connery]

The Grow Lights Australia 500W Quad features the Gen2 LED panels with 1% UVA and 10% Far Red which is 3100K. Full specs are on GLA's website, but the fixture is water-resistant and has 840 Nichia and LED Teknik custom LEDs with a Mean Well HLG-480H-54A driver that was measured at around 500W from the wall.

Photos are courtesy of @Grow Lights Australia who did the tear-downs and initial testing. The driver in the photos was not the one used for the test, but the wattage reading at the end was a test of the actual driver used










Here is the spectrum and efficiency test. For whatever reason LED Teknik did not format the results so there are a couple of anomolies I'm hoping will be addressed soon. But here's what we have in the mean time. The 2.54 umol/j result makes up 98.1% of the total output, so the actual efficiency is 2.59 umol/j. The test was done at the board level, so system efficiency is 96% of this figure (4% driver losses) or 2.49 umol/j. The water-resistant conformal coating was not added to this light (our confomal coating results in a slight red shift which drops the board to 3100K), which is another loss of 1.5% – so the total figure is around 2.45 umol/j and that is at FULL POWER. Remember these figures when we look at the Mars Hydro FC-4800 . . .

Now, 2.45 umol/j system efficiency might sound low, but remember this light includes 1% UVA and almost 10% Far Red and Deep Far Red, which drag the overall efficiency down. But the High Light 420 boards are still one of the most efficient lights with this spectrum on the market. Once you add supplementary UVA and Far Red to a typical "2.8 -2.9 umol/j" LED fixture, the results are substantially lower (because UVA and Far Red are not very efficient to produce). We'll prove this in a minute.

 

[Prawn Connery]

The FC series is the top-of-the range bar light produced by Mars with Samsung diodes (as opposed to the FC-E series that uses Bridgelux LEDs). The FC-4800 is rated at 480W and claims to have UVA and Far Red diodes. So what better fixture to compare to our own 500W High Light 420 (Gen2) Quad that also features UVA and Far Red light, as well as a 480W Mean Well driver. Both fixtures are designed to light up a 4'x4' (1.2m x 1.2m) area.

Mars makes some pretty ambitious claims for their light, including "3 grams-per watt" (if anyone here has ever harvested 3gpw, please let me know!!!), IP65, 2.85-2.9 umol/j, "quality guaranteed", and the above-mentioned UVA and Far Red.

So how do these claims fare in the real world? And how does this light compare to other lights that are similarly powered and (in some parts of Australia at least), similarly priced (the GLA Quad is generally more expensive than the FC-4800, but not at every online store we checked).

Indeed, how does a Mars bar light compare to a GLA panel light? We'll start with a couple of screen shots from Mars' official site:

Note the "Amazing Yield" claim. I'm pretty sure they are talking about tomatoes and not cannabis, as I don't know anyone who gets 3gpw.


This is the advertised spectrum. We tested this – as well as the efficiency – on the same LED Teknik US$50,000 goniometer.


If you didn't see any UVA or Far Red in the above spectrum, then you're not the only one. Here it is again, according to Mars


Note the "IP65" rating – we are going to test this later.


And here is the actual spectrum test with efficiency. 2.57 umol/j at full power. So there actually is about 2.5% far Red and 0.2% UVA

 

[Prawn Connery]

Time to put the Mars Hydro FC-4800 on the bench (thanks to @Grow Lights Australia for doing the tear-down).



Unfortunately, our Mars FC-4800 came with a dead 660nm diode. It didn't affect the rest of the light, and interestingly – even though you can't see it in the goniometer test – there do appear to be a few UVA (actually 410nm) and Far Red diodes on the light. But they are so weak, they don't really show up in the spectograph.

Dead diode top left




The IP65 rating is going to be interesting considering this fixture is not in any way waterproof. There is a silicon conformal coating on the LED strips, but some of the solder joins are exposed. There are dabs of glue on some of the others, but they are not sealed against water.

 

[bk78]

Subbed up

 

[FirstCavApache64]

Love the idea of a real world comparison. Wish this was being done with all of your competition. Can't wait to see the plants. Thanks for the science and not opinion. I'm sure you're going to get blowback from quite a few but it seems like you're being very transparent about everything and that's what counts to me.

 

[MidnightSun72]

Another quality post. Will be following this with much curiosity.

 

[Prawn Connery]

Mars side cover removed.


The FC-4800 comes with a Moso driver and sealed DC cable, but the AC cable is not waterproof and uses a typical C14 joining plug.


The Mars ight has a dimming dial. The GLA Quad does not, but it does have onboard dimming on the Mean Well A driver and there is an optional AB driver that has an external dimmer. The GLA dimmer is not water-proof, but the Mars dimmer is not waterproofed either.


The Mars Hydro PCBs are 1mm thick. GLA uses 2mm PCBs.


The FC-4800 pulled 470W from the wall after it warmed up a little.

 

[FirstCavApache64]

Is the difference in thickness for stiffness, heat transferrence or a little bit of both? I'm clueless with the technical side of lights but would like to know why it's better to use one over the other. I'd pay more for quality. It's why I bought my Timber Grow Light 3 years ago over the other brands. The build quality was overkill compared to what others were doing. Thanks

 

[Prawn Connery]

Love the idea of a real world comparison. Wish this was being done with all of your competition. Can't wait to see the plants. Thanks for the science and not opinion. I'm sure you're going to get blowback from quite a few but it seems like you're being very transparent about everything and that's what counts to me.

Thanks. Whatever the results, we cannot deny our obvious bias. But if we don't do true blind tests, then we are only fooling ourselves. If you build a grow light and it doesn't perform, you need to go back to the drawing board.

Mea culpas aside, I could not conduct this test myself so I asked someone else to do it. As a reward I told them they could keep the Mars light, so it was in their interest to try to get the best out of it. They also run our lights, so of course there might be an asumed bias there. But they sell what they produce, so they don't like losing yield!

While @Grow Lights Australia did the tear-down and power tests at his workshop, he did not want to be part of this thread so I don't think he will be posting. He seems to think it's a bit of a no-win thread (and I kinda agree), but I don't mind if people read this all with a pinch of salt and I will try to answer questions because at least the photos don't lie.

 

[FirstCavApache64]

While I agree the results will be looked at through rose colored glasses by many, the facts provided by the data are very valuable I'm sure, as are the tear down photos and side by side grows which there are never enough of around here. It will be a benefit to a lot of folks that don't understand lighting as well hopefully, if you guys can explain why certain elements are so important in a light.

 

[Prawn Connery]

Is the difference in thickness for stiffness, heat transferrence or a little bit of both? I'm clueless with the technical side of lights but would like to know why it's better to use one over the other. I'd pay more for quality. It's why I bought my Timber Grow Light 3 years ago over the other brands. The build quality was overkill compared to what others were doing. Thanks

A few reasons. Heat dissipation is influenced by surface area exposed to (preferably moving) air. However, more mass can absorb more energy before it heats up to the same level. The High Light boards are designed to be run with or without heastinks, so they need good sinking ability as well as rigidity, which the 2mm thick copper-alloy does. The extra rigidity also keeps the PCB flat when it is being assembled, so it helps with LED placement. Reduced flex causes less stress on the solder joins and LEDs themselves. We also use 2oz copper and wide traces on the circuit which helps reduce voltage drop and heat build-up from current flow. Copper and aluminium are pretty expensive these days.

 

[FirstCavApache64]

A few reasons. Heat dissipation is influenced by surface area exposed to (preferably moving) air. However, more mass can absorb more energy before it heats up to the same level. The High Light boards are designed to be run with or without heastinks, so they need good sinking ability as well as rigidity, which the 2mm thick copper-alloy does. The extra rigidity also keeps the PCB flat when it is being assembled, so it helps with LED placement. Reduced flex causes less stress on the solder joins and LEDs themselves. We also use 2oz copper and wide traces on the circuit which helps reduce voltage drop and heat build-up from current flow. Copper and aluminium are pretty expensive these days.

Thanks for the really easy to understand explanation. That really helps.

 

[Prawn Connery]

Here we are five weeks later. I will leave you with this photo and will post up some more grow shots tomorrow.

GLA Quad on the left and Mas Hydro on the right. Here's a hint: the Mars hydro is stretching more than the GLA even though the Mars is 3700K and the GLA 3100K. How is that possible? Far Red (GLA) stretches and Blue (Mars) shrinks, right? Maybe not . . .

 

[Prawn Connery]

So by the first week of flowering the smaller Fighting Buddha clone under the GLA light had caught up to the larger Mars Hydro plant.

This is the GLA plant with the divider curtain raised (you can see the dividing line to the right where the line of red netting is). A few things you start to notice are that te GLA plant is shorter but wider and fatter. The leaves are green, but a slightly lighter shade of green than the Mars plant. There are reasons for these differerences in the same clones grown under the same conditions (see below).



The Mars plant is taller and a bit narrower. I thought at first this was down to the way the two plants had been scrogged and trained, but the grower said they were growing like that, with the Mars plants stretching a little more under the same light intensity (by now around 800 PPFD each side). The greener leaves are a function of the light: the camera sees the leaves as greener under the Mars light (which is 3700K and CRI 89 vs the High Light's 3100K and CRI95), but in real life the leaves are still a little bit greener because the pigments have developed to reflect a little more blue light than the GLA, which had more red in it.

The reason I suspect the Mars light was stretching more than the GLA comes down to the lack of UV. Even though the GLA has 10% Far Red and less blue light than the Mars, UVA is known to oppose cell expansion and shade avoidance.

 

[Prawn Connery]

Mars Hydro at one week of flower



Grow Lights Australia at one week

 

[Prawn Connery]

At week 5-6 of flower things were really starting to stack up. Unfortunately I couldn't get over to see the grow as often as I'd have liked, so these were the last photos before harvest.

The main things we noticed were the GLA buds where a bit thicker and chunkier, and there appeared to be more bud sites. This was despite the original clone starting off smaller than the clone under the Mars Hydro FC-4800.

Grow Lights Australia


Natural light



With flash



Mars Hydro


Mars Hydro right, GLA left. The Mars plant is taller, but the Mars light was hung lower throughout the grow with equal PPFD levels each side. While the GLA's Mean Well driver pulled 490-500W from the wall versus the Mars' Moso driver's 470-480W, the difference in system efficiencies (2.45 umol/j vs 2.57 umol/j) means both fixtures were putting out almost the same PPF – the main difference was in the spectra.

 

[Prawn Connery]

Both plants were cut down this week and are now hanging up to dry. The grower thinks he know swho the yield winner is, but we won't know the actual weights for another week or so. I'm also planning to get samples from each light tested for cannabinoids.

 

[TheHouseOfGreen420]

love your work mate!