Log In Register

The truth about dimmable ballasts....

  • Thread starter Thread starter TylerDurden119
  • Start date Start date
  • Tagged users Tagged users None

The truth about dimmable ballasts....

TylerDurden119 76 Replies 80,604 Views
Page 3 of 4 · Replies 41–60 of 77
OHM'S law only applies for DC voltage. Resistance in DC is static. In AC it is called Impedance and is dynamic based on frequency. Ohms law is Amps=Volts/Resistance, not Amps=Volts/Impedance. While ohms law is true for DC and AC it is impossible to factor in resistance in AC since it changes depending on frequency. And for the color change, that color difference would be seen when comparing a 600 to a 1000 every time. If you dim a 1000 to a 600, the minute color change would be the same if you replaced the 1000 with a 600. I have 2 Lumatek 1000w dimmables and both show a decrease of 7Nm when going from 1000 to 600. When i switch it with a 600W Lumatek digital on full blast its still 7Nm off the 1000 Watt. so while the 600 does show a slight drop, this is not due to the dimming. If i get around to grabbing a 400W lumatek, i'll test the 600 dimmed to a 400 and see the drop compared to a 400 at full blast.(note when i say decrease, Im taking about the location of the peak.)

What type of measurement device are you using? Do you mean nm/W? For data on the spd part of this question one would need a spectrocoloromiter. The aggregate kelvin values won't tell us what's missing re what the plants care about.

Northone why your plants always so tiny? Are you just vertical or any lights overhead?
 
to measure the point at which the intensity within a wavelength bandwidth peaks, one only needs to use a light-tight box and a prism. you break the light down to a spectrum then place colored film over the lens to filter out the colors section by section and then measure intensity. then you get a rough graph and then you fine tune your colored shades and you can get a pretty good graph of your light intensity. its fairly cheap and great for finding good flouros.
 
to measure the point at which the intensity within a wavelength bandwidth peaks, one only needs to use a light-tight box and a prism. you break the light down to a spectrum then place colored film over the lens to filter out the colors section by section and then measure intensity. then you get a rough graph and then you fine tune your colored shades and you can get a pretty good graph of your light intensity. its fairly cheap and great for finding good flouros.

Hmm, cool idea for sure, but finding a few peaks won't tell me what I want to know. Unless you had a ton more data points..
 
You can use color filter arrays. It uses a pattern in grid and runs it through a program that is basically just a sensor at every spot in the grid that measures intensity. When ran through the application it reads the data and based on the configuration of the grid and how the software is set to read it, you can measure intensity at all of the spectrum. Its used for digital camera processing, but it works just the same. To find what i need to know I just use the cheap box and prism. I only need to find my 2 peaks, one around 425nm for Chlorophyll A, and the other around 660nm for Chlorophyll B (those are the "ideal" peaks, for the best absorption of usable light) Anything out of that range isn't being used efficiently. Most HPS bulbs peak around 550 or so, which is a little too yellow, so I like to test out their intensity at the Max absorption level to see how intense the light is at that wavelength. I feel that is the only way to really measure usable light. Sure I could extrapolate many points and measure their intensity, but since 9/10 of those points would be out of the usable spectrum, its pointless.
 
You can use color filter arrays. It uses a pattern in grid and runs it through a program that is basically just a sensor at every spot in the grid that measures intensity. When ran through the application it reads the data and based on the configuration of the grid and how the software is set to read it, you can measure intensity at all of the spectrum. Its used for digital camera processing, but it works just the same. To find what i need to know I just use the cheap box and prism. I only need to find my 2 peaks, one around 425nm for Chlorophyll A, and the other around 660nm for Chlorophyll B (those are the "ideal" peaks, for the best absorption of usable light) Anything out of that range isn't being used efficiently. Most HPS bulbs peak around 550 or so, which is a little too yellow, so I like to test out their intensity at the Max absorption level to see how intense the light is at that wavelength. I feel that is the only way to really measure usable light. Sure I could extrapolate many points and measure their intensity, but since 9/10 of those points would be out of the usable spectrum, its pointless.
Can you fill us in more on that software? Sounds interesting. Knowing the luminous intensity at 425/660 might be a useful comparison, but I still want to know a lot more. With this method, an spd with narrow peaks might might appear just as one with much wider spectral power. I am also indeed looking to measure outside of visible spectrum, but not 9/10 (200-900nm is enough).
 
It's called demosaicing, and the software would be a custom tailored thing.
Its mainly used in digital camera's.
Adobe Photoshop's bicubic interpolation can simulate it, but it only reconstructs an image from a CFA.
Why is there a need to measure light outside of the spectrum plants can use?
 
It's called demosaicing, and the software would be a custom tailored thing.
Its mainly used in digital camera's.
Adobe Photoshop's bicubic interpolation can simulate it, but it only reconstructs an image from a CFA.
Why is there a need to measure light outside of the spectrum plants can use?
Sounds like I'm operating on the assumption plants use a lot more of the spectrum than you are. Roughly 200-315nm, 430-470, 490-530, 630-690, 700, 750ish, and maybe even ~850, all have specific effects I'd like to address. The breadth across these ranges is important too. Btw, I'm not referring to, nor interested in LEDs, except for maybe the 750-850nm phytochrome flashers (just to be clear and not contaminate the thread topic).
 
I've heard about using far red light at the first hour or so of darkness to help change Pfr into back into Pr but have yet to try it. Also I have heard about UV B light during the end of flowering increasing the THC count, but haven't tried that either. So with far red(~730nm),red(~660nm),blue(~425nm), and UV B (~300nm) a 2x2CFA with 4 color receptors would work.

The roll-off on the intensity once you venture from your "sweet spot"(which I'm sure is genetic specific) is drastic. Maybe not for the UV because of the exponential properties of light. The numbers are low enough where 10-20 nm wouldn't effect the end of the function much. but at higher wavelengths the farther you go the more the change.

For example: Fx=x^2.
if x=1 then Fx = 1
if x = 2 then Fx = 4
if x = 10 then Fx = 100
if x = 11 then Fx = 121
if x = 25 then Fx = 625
if x = 26 then Fx = 676
etc . . .
Notice how when x went from 1 to 2 Fx only went from 1 to 4. when x went from 10 to 11 Fx went from 100 to 121. from 25 it went from 625 to 676. That's just an example exponential function. Here lemme figure out this picture thing real quick.
 
AbsorptionSpectrum

See how sharp they are?
now here is a popular top of the line bulb's output
Shps

compare.
I have ADD super bad so this has nothing to do with anything.i venture off topic a lot, my apologies to everyone.
 
Ok, but that par graph doesn't tell the whole story. Check out this one: (although it's missing some things too.) Also think about Emerson effect, and how some green wavelengths interact with it (although certain greens also cause shade-avoidance behaviors). I'm just learning about this stuff myself..

Pigabs
 
Ok, but that par graph doesn't tell the whole story. Check out this one: (although it's missing some things too.) Also think about Emerson effect, and how some green wavelengths interact with it (although certain greens also cause shade-avoidance behaviors). I'm just learning about this stuff myself..

View attachment 211439

Wait a second here... the more closely I look at that graph, the less sense it makes; for instance, why would the absorption spectrum graph fall off at the red end BEFORE it gets to chlorophyl a's wavelength there?
 
that Beta-Carotene is exactly what i don't want my plants absorbing. its a survival tactic used to detect when fall is coming and lets the plant use other means to photosynthesize a different spectrum of light. It might help ripen buds during flower but like i said before i haven't tried any experimental lighting. I like my leafs greens, not orange.
 
stick with a nice 430-ish and 660-ish peak that rolls off to the peaks of each chlorophyll type.(all hypothetical, i doubt we can get a strong enough light to only produce those wavelengths) I wish there was a device that had an input and an output that could apply a function to the input's wavelength and output it using the same interface so the bulb knows no difference. We could then write a function(call it an EQ) and fine tune our HID lights.
 
that Beta-Carotene is exactly what i don't want my plants absorbing. its a survival tactic used to detect when fall is coming and lets the plant use other means to photosynthesize a different spectrum of light. It might help ripen buds during flower but like i said before i haven't tried any experimental lighting. I like my leafs greens, not orange.
There are other carotenoids at play too, and there's a lot more going on than what you're describing. Took me a lot of reading to even scratch the surface on this subject, but the data are there and I'm not pulling these ideas from thin air..
 
stick with a nice 430-ish and 660-ish peak that rolls off to the peaks of each chlorophyll type.(all hypothetical, i doubt we can get a strong enough light to only produce those wavelengths) I wish there was a device that had an input and an output that could apply a function to the input's wavelength and output it using the same interface so the bulb knows no difference. We could then write a function(call it an EQ) and fine tune our HID lights.
The function-driven spds you're suggesting are just not the way these light sources work. In fact, the graphs and numbers used for marketing by mfgs are mostly bs for several reasons, and they're hiding truth behind the famous veil of the hydro-industry (anybody want some watered down nutes at 1000x the price with cartoons on the jug?) The only reason those spds don't show sharper peaks is because the machinery is set to report rounded wavelength readings. Limiting to specific wavelengths definitely isn't my goal, and you'd be missing quite a lot with peaks at only 430/660 (and we're already missing so much using hps anyway).

Btw, there's already extensive work examining hid bulbs elsewhere in mj-forumland by knna, spur, whazzup, penguin, and others (would be silly for me to repost it all here). They even digitized an exhaustive list of mfg spds and worked them into a spreadsheet/calculator to figure out efficiency and spectral comparisons (pretty surprising results too).

Ttystikk,
 
Whoops, apparently I was replying to tystikk and got cut off. Something about the difference between making chlorophyll and then using it, sorry, lost it in a browser-freeze.
 
There are other carotenoids at play too, and there's a lot more going on than what you're describing. Took me a lot of reading to even scratch the surface on this subject, but the data are there and I'm not pulling these ideas from thin air..
Please elaborate.
The function-driven spds you're suggesting are just not the way these light sources work. In fact, the graphs and numbers used for marketing by mfgs are mostly bs for several reasons, and they're hiding truth behind the famous veil of the hydro-industry (anybody want some watered down nutes at 1000x the price with cartoons on the jug?) The only reason those spds don't show sharper peaks is because the machinery is set to report rounded wavelength readings. Limiting to specific wavelengths definitely isn't my goal, and you'd be missing quite a lot with peaks at only 430/660 (and we're already missing so much using hps anyway).

Btw, there's already extensive work examining hid bulbs elsewhere in mj-forumland by knna, spur, whazzup, penguin, and others (would be silly for me to repost it all here). They even digitized an exhaustive list of mfg spds and worked them into a spreadsheet/calculator to figure out efficiency and spectral comparisons (pretty surprising results too).

Ttystikk,
pm a link? You CAN tweak a ballast to give different color results. When you increase or decrease the amount of power(watts) supplied to a bulb, the gas being excited doesn't always act the same.(remember how my peak moved back 7nm?) this is because in AC electricity, you have constants, 1st degree variables, 2nd degree variables, 3rd degree variables, to infinity(this is because time is infinite and new "degree" is formed every time the wave starts going up or down). so with there being an infinite degree of variables, one change of a variable(watts) can effect more than just the intensity of the bulb. and back to the original topic, if the manufacture takes these variables into account, a dim-able ballast can be created this way. a 1000 watt bulb when put under a 1000 watt ballast gives us a peak color temp at x. under a 600 ballast of the same brand it gives us a peak of x-7. now if i were to create dimmable ballast i have to make sure that while I'm turning that knob, all variables are being tweaked(any differences between the 1000W and 600W have to be blended in a sense.) for instance if it turns out that the voltage drop across 2 points is different in the 600W and the 1000W , this will have to be accounted for while that knob is being turned. if all the manufactures did was put a variable resistor on the initial transformer, the that will not suffice and a huge difference would be apparent as you dimmed. also i also do not agree with any manufactures claims to spectral output, i was merely showing what they advertised as good intensity across the spectrum and was pointing out how that giant peak wasn't even in the sweet spot zone.
 
Page 3 of 4 · Replies 41–60 of 77
Back
Top Bottom