Super easy DIY LED: Philips XF-3535L

[naoki]

I just posted a step-by-step assembly of super easy DIY LED, which I have done about a year ago. 1 year is a long time in the progress of LED technology, but Philips XF-3535L is still great in terms of efficiency. The price came down a little bit over the year, but COB (chip-on-board) LEDs (such as Cree CXB series and Bridgelux Vero) are still cheaper. But there are lots of positives about XF-3535L.

I mostly use it for orchids, but it can achieve fairly high light for CP if you use lots of them. So it may be useful for some of you.

Here is the link to the blog post:
Orchid Borealis: Easy DIY LED: Philips XF-3535L

 

[pmatil]

Very good info on that, thanks!

 

[Est]

Very cool! I appreciate the thorough writeup. Thanks for posting! Looking forward to your post about COB.

Edited addition: Also, great to see more people doing quantitative work! I've been working more with different types of sensors so I've been thinking of horticultural experiments. Interesting to see someone else using another Licor product -- I've used some of their gas analyzers for a few years now (on and off anyway).

 

[naoki]

Very cool! I appreciate the thorough writeup. Thanks for posting! Looking forward to your post about COB.

Edited addition: Also, great to see more people doing quantitative work! I've been working more with different types of sensors so I've been thinking of horticultural experiments. Interesting to see someone else using another Licor product -- I've used some of their gas analyzers for a few years now (on and off anyway).

Thanks. So you are a real plant physiologist! I also have an access to LI-6400, too, but I don't know well about it (it seems quite complicated). One of my former student did plant eco-physiology type research with it. My PAR sensor is old; I scored it on ebay for $40 because the seller didn't know if it works or what it is. So the sensitivity to blue may be a little weak according to this article. But I think that it is good enough for my hobby purpose.

Are you testing different types of PAR sensors? Like this study:
http://www.apogeeinstruments.com/content/Quantum Sensors-LEDs.pdf
I would like to learn what you find out. I wish that there were more affordable PAR sensors so hobbyists can use them.

 

[Est]

Ah, I'm more on the soil side of things really (biogeochemistry, specifically). Autotrophic respiration has been mostly something we calculate as a correction rather than as a primary measure of interests. I do work with people who do things like leaf respiration, but that's outside of my immediate skill set! Great that you were able to get one via ebay -- I've considered the same for hobby work. What's kind of work do you do?

While I'm not using any PAR sensors, I do use a few other Apogee sensors (O2 and barometric pressure) among others. Recently set up a pair of arrays with barometric pressure soil O2, moisture, temp, redox and will be adding rain gauges soon -- but no work with light measurement, unfortunately! What kind of data logger are you using for your Licor sensor?

Another sidenote: Currently working with some "soils" from Healy; pretty close to you!

 

[pmatil]

What kind of soil moisture sensor do you use?

 

[Est]

I currently use the Campbell Scientific CS655 shown here. They have a few different models for different soil types. Generally we use the shorter ones for soils which are more difficult to work in. Somehow I always seem to end up in soils that are difficult to work in.

Some of my collaborators have had good luck with Decagon soil moisture probes, and I've heard that the latest iterations have been significantly improved in terms of ruggedness. I haven't used them myself, but I may start in the next few months. Do you do related work?

 

[naoki]

Ah, I'm more on the soil side of things really (biogeochemistry, specifically). Autotrophic respiration has been mostly something we calculate as a correction rather than as a primary measure of interests. I do work with people who do things like leaf respiration, but that's outside of my immediate skill set! Great that you were able to get one via ebay -- I've considered the same for hobby work. What's kind of work do you do?

Est, that's cool. I'm an evolutionary biologist (studies population genetics and molecular evolution of plant mating systems as well as some ecological genetics of plants).

While I'm not using any PAR sensors, I do use a few other Apogee sensors (O2 and barometric pressure) among others. Recently set up a pair of arrays with barometric pressure soil O2, moisture, temp, redox and will be adding rain gauges soon -- but no work with light measurement, unfortunately! What kind of data logger are you using for your Licor sensor?

Another sidenote: Currently working with some "soils" from Healy; pretty close to you!

Apogee light sensor is pretty good, but it is a bit problematic for monochromatic LEDs, which may use deep red. But it is the most affordable one. There is a USB version of Apogee quantum sensor, which is kind of cool.

I'm not using a data logger with my Li-Cor. I have used HOBO for environment monitoring, but I don't think I can connect Li-Cor to it. I'm actually planning to make a temp/humidity logger with Rapsberry Pie (I have all components, but I haven't had a time yet!).

A small world. Yes, Healy is very close to Fairbanks. Have you collected soils there? It is a beautiful area!

Do you have any experience with CO2 sensor? Is there a cheap and decent sensor?

 

[Est]

I'm not using a data logger with my Li-Cor. I have used HOBO for environment monitoring, but I don't think I can connect Li-Cor to it. I'm actually planning to make a temp/humidity logger with Rapsberry Pie (I have all components, but I haven't had a time yet!).

Excellent! I've been thinking of trying something similar myself as a personal project. Realistically though I have very little background to go on. Do you have any good sources for documentation that you like?

A small world. Yes, Healy is very close to Fairbanks. Have you collected soils there? It is a beautiful area!

I've never been up to Alaska myself, but we do have some collaborators that work up there. I just had the soil (okay, ball of moss) sent to me. I'd love to go to visit some day, but something tells me that a field season out there would be pretty miserable -- ~24 hours of sun sounds like a good excuse to never be allowed to stop working.

Do you have any experience with CO2 sensor? Is there a cheap and decent sensor?

You know, I'm not sure. I'm at least vaguely familiar with eddy flux towers for landscape-scale measurement, Li-Cor for realtime and static flux chambers (with subsequent gas chromatography analysis), but I've never used a CO2 sensor. Taking a look, I see that some certainly exist. Huh. That's actually super interesting... Wish I had some out with the rest of my probes in Springfield. I imagine that CO2 data in conjunction with O2 and redox data could show some interesting relations.

But I suspect that it's a case of "if you give a mouse a cookie." Typically, any time I look at CO2 I'm also analyzing for CH4 and N2O (potent greenhouse gases) since I can do them simultaneously, so if I'd probably want to figure something out for methane at least, too. Oh, in an optimal world with decent funding...

I actually just wrapped up another round of a related experiment using flooded/slurried soils in an anaerobic headspace looking at iron reduction in microbial respiration and resulting CO2. It would be interested to have some field data to complement that.

 

[naoki]

Excellent! I've been thinking of trying something similar myself as a personal project. Realistically though I have very little background to go on. Do you have any good sources for documentation that you like?

There are several tutorials, and I haven't decided which method to go with the software side. I kind of like the simplicity of google doc based method here:
https://learn.adafruit.com/dht-humidity-sensing-on-raspberry-pi-with-gdocs-logging
https://learn.adafruit.com/dht-humi...cs-logging/connecting-to-googles-docs-updated

Also, this vivarium controller seems interesting:
Raspberry Pi Vivarium Controller DHT22 Temperature & Humidity Sensors
I've been controlling exhaust fan (to prevent overheat) and humidifier with ZooMed Hygrotherm. But the sensors seem to shift over a long term, and this RapsberryPi based method might be nicer.

I've never been up to Alaska myself, but we do have some collaborators that work up there. I just had the soil (okay, ball of moss) sent to me. I'd love to go to visit some day, but something tells me that a field season out there would be pretty miserable -- ~24 hours of sun sounds like a good excuse to never be allowed to stop working.

Well, we could get lots of fieldwork done in the summer time. It's not too hot, no poisonous animals/plants to worry about, so it is great for field work!

 

[pmatil]

Do you do related work?

I have been eperimenting a bit but haven't found a sensor that could realiably tell whether the soil is wet enough or not. The problem is with CPs that the soil is often mostly LFS and perlite and so very loose. So it's hard to measure conductivity in that type of soil.

 

[SubRosa]

I have been eperimenting a bit but haven't found a sensor that could realiably tell whether the soil is wet enough or not. The problem is with CPs that the soil is often mostly LFS and perlite and so very loose. So it's hard to measure conductivity in that type of soil.

Perhaps a humidity probe designed to measure atmospheric humidity in a sleeve of a breathable membrane which would exclude liquid while allowing water vapor to contact the probe ala' Goretex could do the job?

 

[pmatil]

Yes, that could be and there are some sensors that use a humidity sensor. I'm just very suspicious because in the terrarium the humidity is already 80 % and then under the soil probably at least that if not close to 100 % so measuring the difference in soil moisture could be difficult.

 

[SubRosa]

Yes, that could be and there are some sensors that use a humidity sensor. I'm just very suspicious because in the terrarium the humidity is already 80 % and then under the soil probably at least that if not close to 100 % so measuring the difference in soil moisture could be difficult.

The sensor values which equate to acceptable soil moisture might well be something you'd have to figure out by experimentation, but I don't see how the humidity of the air above the media would affect anything other than of course the rate at which the media dries out.

 

[Est]

Hmm, yeah, in general soil moisture probes are calibrated for (and function best in) mineral soil. If your bulk density is really that low (loosely packed sphagnum and perlite) something like a a relative humidity measurement as described above may be a good bet (though it'll cease to function properly as soon as that membrane gets wet, so you'd likely need to make additional modification for consistency).

But realistically you're not trying to quantify soil moisture, rather keep things acceptably moist, right? So long as you can get something which gives internally consistent results, you can do a pretty simple calibration for your needs. If you know that you typically need to water every ~3 days, for example, take a measurement when you water and then again when you'd typically water again -- you now have your desired range. You don't have absolute values, but you have a decent index, which is what you're looking for, right?

How variable are the measurements you're getting with the probe you've been using? I'm assuming that you're looking at using a probe that you stick into media for a measurement in multiple pots, not something you plan on leaving in the media, correct? I can think of a few options depending on those answers.

 

[pmatil]

I was planning to use a sensor that is always in the soil so I can immediately tell which plants need watering. As I haven't got good results with the in-soil probes I have been planning to just sense the water level in the saucer below the pot. Water when there is no water in the saucer. Simple.

 

[Est]

Sure, checking saucer water level is generally a good way of keeping track of things. If you want to use a sensor that you leave in place:

1). You can find a spot that you like, realize that it won't be absolutely accurate, and use the values as an index.

or

2). you can probably just put in a "plug" of peat to encompass the probe electrode(s). That'd give you a relatively reliable, consistent number. Granted, that'd probably work much better for larger pots.

Honestly though, sticking to a routine and visual moisture cues is just fine.