PAR levels high enough?


I read on various sources in the internet that PAR levels should be around 200 for optimal growing conditions. The PFC that I build (following the original BOM) does not give more than 65 PAR at the leaves. Does anyone have better values (did I do something wrong?) or should we increase the amount of leds in the PFC ?

Is the lighting in PFC version 1.0 insufficient?
Is the lighting in PFC version 1.0 insufficient?

@Peperoni, what kind of PAR meter you are using?


The Grove digital light sensor that was described in the BOM…


So you must be converting lux to PAR manually? I’m not an expert, but I guess that it is not that straight forward conversion.
PFC is using blue leds, while:

Any light with a high amount of blue energy (wavelengths below 500 nm) is going to be severely underrepresented by a lux meter


Ah… so the ‘SLPA’ does not mean Sensor Light PAR’ ? sorry, I though it was. Now looking at the datasheet of this sensor it actually does not mention PAR metering… What is this value for ?


My guess is that necessary PAR is embedded into a design, while sensor helps with operations, i.e. even though it just measuring lux, it still helps to diagnose when your lights simply got broken :wink:


Hmmm… below a copy of a part of page 29 from the Motherboard Assembly Guide:
This is a message that contains a sequence of instruction codes:
● “GTYP”:“Stream” ­ This message type is a stream message.
● “SWPH 1”:6.9 ­ The 1 s t S ensor of W a ter P H is reporting a value of 6 .9
● “SWTM 1”:25.4 ­ The 1 s t S e nsor of W a ter T e mperature is reporting a value of
25.4 degrees C
● “SWEC 1”:2.6 ­ The 1 s t S ensor of W a ter E l ectrical C o nductivity is reporting a
value of 2 .6 mS/cm
● “SLIN 1”:33 ­ The 1 s t S e nsor of L i ght I Nt ensity is reporting a value of 33 lux
● “SLPA 1”:0.73 ­ The 1 s t S ensor of L i ght P Ar (Photosynthetically Active Radiation)
is reporting a value of 0 .73 umol/(s*m^2)
● “SATM 1”:24.7 ­ The 1 s t S e nsor of A i r T e M p erature is reporting a value of 2 4.7
degrees C
● “SAHU 1”:31.3 ­ The 1 s t Se nsor of A i r H Um idity is reporting a value of 3 1.3 %
relative humidity
● “SACO 1”:400 ­ The 1 s t S e nsor of A i r C O2 is reporting a value of 4 00 ppm (parts
per million)

It mentions actually that PAR is measured (as well as LUX at SLIN), but I don’t see how they can come to PAR with this Grove digital light sensor as the source.

Does anyone know how the value of SLPA relates to PAR (if at all) ?


@Peperoni, looking at the source code

lux_ = lux_average*calibrtion_to_vernier_lux_; par_ = lux_average*calibration_to_vernier_par_*measuring_indoor_par_correction_;

where constants for conversion are:

calibrtion_to_vernier_lux_ = 0.78; calibration_to_vernier_par_ = 0.02; measuring_indoor_par_correction_ = 0.86; //reduction by 14%

I don’t know how accurate this conversion is. One of the ways to find out is to buy PAR meter and compare readings. The problem, though, PAR meters are not cheap.

UPDATE: Giving it more thought, I think that calibration_to_vernier_par_ was actually calculated by MIT people based on Vernier PAR meter readings and TSL2561 lux reading. So if they did it for exact same LEDs which are on the BOM, than indeed PAR levels are quite low.


Serein, do you have a PFC up and running… ? if so, what are your SLPA (PAR) levels ?


No, I don’t yet. I decided to skip building V1 and wait for V2 design.

Currently, I only have RPI and Arduino running V2 software with a bunch of sensors connected. I also have some 60 watt LED lamp with mix of blue and red spectrum leds (dont have exact specs at hand as it is custom build). And my light intensity reading 70cm away from lamp is about 2800 lux, which based on conversion above gives only 48 PAR.

Is the lighting in PFC version 1.0 insufficient?

I’m confused about the calculations. I think the sensor is giving you a very bad approximation.
I may be wrong, but I think you can only do a PAR calculation if you have the complete spectral power density of your lighting setup. Datasheets only include a relative one.
I’m sure that the PAR levels on most PFCs are too low, the standard PPFD for lab tests is 120μmol·m−2·s−1
If you use the right green and red LEDs, almost all of your output light will be PAR
I’ve used Cree XB-D LEDs on my first prototype. I am planning to do a full SPD test with a spectrometer at my university when my 2nd prototype LED boards are done.
I’ll also write up a post on choosing the right LEDs and message here when I’m done

edit: corrected PPFD to 1200
edit: corrected PPFD to original 120


I’m not too familiar with the particulars of how the sensor works. I do know however that for converting regular lux into PPFD the calculations are rather complex:

That to me is the best guide on this topic.

Thank You,


What bothers me guys, is the way these sensors have been picked… There’s no way you can replace a real PAR CMOS sensor with this solution… sorry. It’s not even close to what the Aquarium hobbyists are doing.


It’s time to learn :slight_smile:


Thank you for sharing this @bertrand ! What I conclude from this video is, that when you know your light source (described in the BOM) you can easily find out your multiplication factor and, more or less, know what you PAR values are when you measure the LUX… I think that is exactly what is done in the software.
An acurate PAR meter is extremely expensive, and will undoubtedly be advised in a professional environment, but is hugely overpriced for our purposes. You should also consider that all other values have some kind of ‘noise’ which makes it less scientific than you might think. I am not sure if anyone has tested an environment with 2 CO2 meters by COzir. I am convinced that they will NOT have the same values all the time. Measurments should always be placed in perspective and are no substitute for common thinking…


Hi @Peperoni !

The thing is I doubt you know the wavelength of the LEDs you have in your LED strip… and also, nowadays, LEDs are not bare, they have phosphors to enhance some colours, or at least broaden their spectrum. I know a few manufacturers that tune their LED on purpose for grow lights, but we are not talking about cheap of the shelf components…
95% of the Chinese grow lights are using regular LEDS…


I understand, but in the BOM the Erligpowht LED panel is defined. I assume that the MIT-team has done some measurements on it and have come up with some kind of multplication factor that is in the software integrated. This way you can use a relative cheap LUX meter (it does IR as well if I understand correctly) and come up with a cheap, yet useable solution. The outcome might be very different if you use another LED source, I agree with that !


I got that… but:

  • The Atlas RGB sensor is good for pantone charts,
  • The TLS2561 is super noisy (I got a TLS2561 & BH1750 next to each other and the difference is huge). The good thing about the TLS2561 is the second sensor for IR. But there is no FarRed emitted by the Growlight…
  • I’ll be happy to see the spectroradiometer measurements… or a least the PPFD score.

Even good Chinese GrowLight mfg are publishing spectrum now…


I’m curious about their PAR sensor…


Looks to me like it is the same photodiode solution everyone else is using. Unless the light source is calibrated to the sensor it will be fairly useless.