The far red thread

C

churchhaze

Well-Known Member
A Scmitt trigger would be useful if there was a noisy analog signal that often floated near the threshold. (like if i was trying to sense light outdoors during a sunrise, I would need a schmitt trigger, or it would turn on and off many times when it got toward the threshold)

Can you please be more specific of where you think I should use a Schmitt trigger? I really don't think I will need one just to see if the lights are on or off. It's not a signal that floats close to the center. It's hard on or off.
 
Last edited:
C

churchhaze

Well-Known Member
guod said:
the "It's hard on or off." Version

View attachment 3406028
Click to expand...
This schematic is a copy of my design, minus the differentiator (coupling capacitor and bypass resistor). This convert lights out signal into a negative trigger spike that starts the timer. (the derivative of a square wave is a spike wave). Without it, the timer would continue to close the relay throughout the night period, rather than stopping after 9 minutes.

The diode was used to clip the positive spike to improve stability of the timer. This differentiator fails if the rise or fall time is not lower than a few milliseconds (tested in simulation), so if the rise and fall of photocell and resistor input isn't fast enough, it will fail to produce a low enough spike to trigger the timer. However, from some testing, I think this will work reliably.

The potentiometer would be a good way to adjust the sensitivity, but I will just try a few resistor sizes until I get one that produces a reliable trigger spike on lights off.
 
C

churchhaze

Well-Known Member
I tested this design with a 1k pulldown resistor, a 4k pulldown, and a 10k pulldown, and they each had their own problems in terms of generating a reliable trigger spike.

I went ahead with @mc130p 's idea to use an NAND gate... I used an AND (with one input connected to 5V) .. 2 inverters would work fine too, or any gate really... This produces a way more clean output with sharp rise and fall, which generated a clean trigger spike every time... the problem is the timer frequently starts when the lights come on as well as turning off... This will require a schmitt trigger after all, even with sharp "on or off". I will need to order some schmit inverter chips, or hack one together with what I have.

As a side note, I think using a photocell to detect whether the lights are on is a bit backwards, and an opto-isolator to detect mains directly might be simpler and more foolproof.
 
Last edited:
C

churchhaze

Well-Known Member
Using a 100k pulldown resistor made a huge difference.. now instead of needing a very strong light to be considered ON, room with ambient brightness will cause it to be ON. I can go into any room now, turn the light off, and the output LED comes on reliably. This means I could put the photocell anywhere in the tent and not have to worry about how it's positioned.

But also when the light comes on... this could be thought of as a feature for some!! 730nm at lights out, and lights on! or the schmitt trigger should fix it. :dunce:
 
Last edited:
mc130p

mc130p

Well-Known Member
churchhaze said:
Using a 100k pulldown resistor made a huge difference.. now instead of needing a very strong light to be considered ON, room with ambient brightness will cause it to be ON. I can go into any room now, turn the light off, and the output LED comes on reliably, but also when the light comes on... this could be thought of as a feature for some!! 730nm at lights out, and lights on! or the schmitt trigger should fix it.
Click to expand...
I wanted to test my 730 at lights on, but I think it was you and some others that convinced me not to! Now I'm stuck in the comp and I have to wait..... haha
 
C

churchhaze

Well-Known Member
Because I do not have a 74HC14 (6 schmitt inverters), I do not have spare opamps, I made a schmit trigger using one of my 555 timers according to the following:

http://www.allaboutcircuits.com/vol_6/chpt_8/2.html

I used a 74HC04 to invert the photocell/resistor voltage divider, then the 555 based schmitt trigger inverts it a second time, only with hysteresis. Without the first inverter, the 730nm lights would come on only at lights on, not lights off. With the inverter, it comes on only at lights off.

Works perfectly so far. Thanks both @guod and @mc130p for your contributions. It really helped. I will post a schematic of my final design when I have the relays part worked out. (don't have 120v relays laying around)
 
Last edited:
C

churchhaze

Well-Known Member
Here's what I have so far. (without relay stuff)

730_timer_schmit.jpg

This could be made more simply, but it works pretty well. With R1=1M and C1=470u, output is on for about 8.61 minutes. You can adjust C1 and R1 to change the pulse time. So far it's been working very reliably.
 
C

churchhaze

Well-Known Member
Moved it to something more permanent.. Still didn't handle the relay part. I might use that one (had it laying around). I'm not too familiar designing with relays, would there be any reason not to use that relay to power an APC-25-700 (driver for 730nm leds). I would obviously need a second 9-12V source to power the relay coil.

Right now, if it's "light" in the room, and you put your finger on the photocell (or turn the lights out and make it mostly dark) , the led next to the relay will come on for about 8-9 minutes.

IMAG0402.jpg

IMAG0408.jpg
 
AquariusPanta

AquariusPanta

Well-Known Member
EfficientWatt said:
Hi AquariusPanta,

Did you get the new Semileds Far red, or the previous model ? http://shop.stevesleds.com/SemiLeds-Far-Red-3W-Brand-NEW-8794102596.htm
Any chance you could give your impressions +tell us what parts you went for ?
Click to expand...
Hey there, EfficientWatt!

I just noticed the new inventory. It would appear that they removed the 730nm Epistars that I had bought last year during the fall. I'm sure the new models, or current selection, will do just as good if not better.

I was growing in a 3' x 1.7' tent and implemented two of the Epistars. As for the driver, I used this 3W driver and placed all of this on a 10" long piece of aluminum tubing, which they sell at SL.

My impressions were that the sucker worked quite well and did what it was meant to do, which was induce flowering and accelerate speed of pR/pFR conversion, thus allowing for more time with lights on during budding (an extra hour or two max). After a few days of using the far red Epistars, the results were blatant (vertical growth was staggering). I did manage to harvest after ~53 days of flower, with a nice combination of milky and caramel trics.
 
coolbreez1

coolbreez1

Well-Known Member
Does the far red have to turn on after the main LEDs turn off?

I know this might sound like a stupid question... But if you can just have the far red LEDs turn on for the last 10-15 min of the main lights then it would be a simple matter of plugging in an analog timer in the main timer and setting it to turn 15 min before lights out.

On this same train of thought some one mentioned a battery. One might be able to do something similar by having an analog timer that is set to turn on before lights out long enough to charge a battery, which then has enough power to run the far red LEDs for 10-15 min after it gets disconnected when the main power times out.
 
AquariusPanta

AquariusPanta

Well-Known Member
coolbreez1 said:
Does the far red have to turn on after the main LEDs turn off?

I know this might sound like a stupid question... But if you can just have the far red LEDs turn on for the last 10-15 min of the main lights then it would be a simple matter of plugging in an analog timer in the main timer and setting it to turn 15 min before lights out.

On this same train of thought some one mentioned a battery. One might be able to do something similar by having an analog timer that is set to turn on before lights out long enough to charge a battery, which then has enough power to run the far red LEDs for 10-15 min after it gets disconnected when the main power times out.
Click to expand...
The biggest problem that I initially experienced was properly syncing my timers with one another; one for the main lights and another for the far red. Because the mechanical timers run on 15 minute relays/clicks, I'd always have the far red either turn on too early or late. I eventually settled with the far red lights coming on a few minutes before sun down, with no ill-effects. I think the solution next time I implement far red into my garden will be to figure out how to use a digital timer and just have the far red hooked up to it, to keep it simple.
 
mc130p

mc130p

Well-Known Member
AquariusPanta said:
The biggest problem that I initially experienced was properly syncing my timers with one another; one for the main lights and another for the far red. Because the mechanical timers run on 15 minute relays/clicks, I'd always have the far red either turn on too early or late. I eventually settled with the far red lights coming on a few minutes before sun down, with no ill-effects. I think the solution next time I implement far red into my garden will be to figure out how to use a digital timer and just have the far red hooked up to it, to keep it simple.
Click to expand...
The 555 timer-based optical switch that @churchhaze designed on this very thread is simple to implement, reliable, and wholly suitable for this purpose:) Turns on whenever the lights turn off...simple and effective.
 
You must log in or register to reply here.
Top