Underperforming plants, various strains

[Rahz]

i doubt your feed ph of 6.8 is doing anything, if there is adequate lime in the media even if your feed ph is 6.8 within a few hours of watering the ph of your media will swing back to what it was at before you watered at 6.8. do you have a ph pen with the probe for inserting into the media to get its ph. run off ph is not an accurate measurement of the medias ph. although calmag cant hurt. however looking at your first picture if i were to guess i would say the ph of the media is more likely too low and what your seeing is a nitrogen deficiency due to low ph not for a lack of feeding i suspect.

I'll order one. Do you recommend the Bluelab or the Apera? The bluelab requires a pre-made hole and I assume the mode of action is for some amount of solution to enter the glass tube of the probe to get a reading? With the Apera the spear is the probe.

 

[Rahz]

Im still placing my bets on mediocre genetics for the plants that are underperforming. You can spend more time fine tuning everything and maybe they will perform slightly better, but I dont think they will meet your standards. You have plenty of growing experience and can clearly spot the difference between an excellent plant vs a mediocre one.

I'm not ruling that out. Purely conjecture but I wonder if all these new strains genetics are as stable as the old-school strains? At any rate, I'm going to get a soil probe and not rely on runoff. Probably seems silly but I hadn't checked for soil probes in years. All I found were the cheapo garden probes. Looks like the Bluelab probe was first available in 2019 and the Apera in 2020.

I'm also going to work in the cal-mag just because it seems like a reasonable thing to do.

 

[Thundercat]

While ph seems like it could definitely be some of the issue, I can’t believe no one mentioned temperature.

You said you run about 74degrees which is on the low side for LED lights and CO2. LED lights don’t put out the IR which warms the leaves and helps regulate transpiration. When LED first became popular there were tons of people having deficiencies that commonly just had cal/mag blindly thrown at it as a solution. After a few years it was discovered that raising temps into the low 80s actually would solve most of the issue without adding anything.

 

[Rahz]

While ph seems like it could definitely be some of the issue, I can’t believe no one mentioned temperature.

You said you run about 74degrees which is on the low side for LED lights and CO2. LED lights don’t put out the IR which warms the leaves and helps regulate transpiration. When LED first became popular there were tons of people having deficiencies that commonly just had cal/mag blindly thrown at it as a solution. After a few years it was discovered that raising temps into the low 80s actually would solve most of the issue without adding anything.

74 under the canopy. Thermostat set to 78. I'll look into that and probably bump the temps up. Thanks!

 

[thumper60]

74 under the canopy. Thermostat set to 78. I'll look into that and probably bump the temps up. Thanks!

Do you realize that the hp-cc should be run like Coir Ph 5.8-6.2 lots of watering never let dry out like soil never water with out nutes at least have cal-mag in the mix, An running your nutes have plenty of cal-mag handy week 4-6 of flower. I have been using promix for 4 decades I love it! I tried the hp-cc mix a few yrs ago hated it just could not seem to dial it in

 

[Blue brother]

Bad genetics all day long, while there's things you can obviously do to improve your game on your end, even badly treat decent genetics have frost of some level. Scrap those genes they aren't what you're looking for.

 

[Rahz]

Do you realize that the hp-cc should be run like Coir Ph 5.8-6.2 lots of watering never let dry out like soil never water with out nutes at least have cal-mag in the mix, An running your nutes have plenty of cal-mag handy week 4-6 of flower. I have been using promix for 4 decades I love it! I tried the hp-cc mix a few yrs ago hated it just could not seem to dial it in

I was under the mistaken impression that PH should climb a bit toward the end of flower, likely a combination of reading about soil and being aware of the high sphagnum peat moss content of HP-CC. I think I've got a better handle on how to treat it in flower now.

The reason I have used it, I've used it for mothers for over 10 years, light feed at PH6.8-7 and moms will veg strong and stay healthy for many months. I replace them only because they start to get really big and root bound. I'm not opposed to switching but I'm going to keep working with it in flower and see if I can improve.

Anyway, thanks to all for the observations and suggestions. Much appreciated.

 

[Rocket Soul]

I havent really got anything to add but posting to support a led oldtimers thread and to follow along. Happy to see you, Rahz

Are you going to add uva or was that just AI talking?

 

[Rahz]

Hi Rocket Soul, good to see you too! Yes it was the AI talking but I'm probably going to add some UV. I've been in pots for about a year so it's time to go beyond the basics. The plant issues threw me for a loop but I'm mostly convinced at this point it's a genetics issue. I've got a handful of beans to work through but next time I order it will be from a different bank.

The PH tester arrived today. The readings were mostly 6.1 and 6.2 but I had a 6.4 in there. Same pot different spot was 6.2.

Adds/changes:
I started working in some Recharge once a week a couple weeks ago
80 degrees at the canopy now
3ml Cal-Mag per gallon since a couple days ago
CO2 raised to 1000 baseline last night
Will drop feed PH to 6.2 and monitor soil PH. This may lower PH slightly but with the added cal-mag I'm not sure what to expect
I have some 13w UVB bulbs, thinking about putting 2 (total) between the last 3 lamps where the late stage flowering happens, would be about 4.3% of the light in the late stage area.

Not planning so far:
bloom boosters
foliar spray
Silica (maybe at some point)

I'm also in the middle of building drain tables. As suggested, I don't plan to over water but I think it will be good to have the soil well drained at all times and it will make doing flushes much easier which I would like to do at least once per plant. Using 6063 square tubing and angle for the structure, 6063 square tubing for the table top, 5052 sheet for the drain pans, 316 stainless hardware with rubber washers in the drain pan. Will be 3 tables that wrap around the room with interconnected pans hooked to a drain line that leads outside. It's something I've put off doing for a year but I brought the metal home today, got my chop saw set up.

 

[Rocket Soul]

Hi Rocket Soul, good to see you too! Yes it was the AI talking but I'm probably going to add some UV. I've been in pots for about a year so it's time to go beyond the basics. The plant issues threw me for a loop but I'm mostly convinced at this point it's a genetics issue. I've got a handful of beans to work through but next time I order it will be from a different bank.

The PH tester arrived today. The readings were mostly 6.1 and 6.2 but I had a 6.4 in there. Same pot different spot was 6.2.

Adds/changes:
I started working in some Recharge once a week a couple weeks ago
80 degrees at the canopy now
3ml Cal-Mag per gallon since a couple days ago
CO2 raised to 1000 baseline last night
Will drop feed PH to 6.2 and monitor soil PH. This may lower PH slightly but with the added cal-mag I'm not sure what to expect
I have some 13w UVB bulbs, thinking about putting 2 (total) between the last 3 lamps where the late stage flowering happens, would be about 4.3% of the light in the late stage area.

Not planning so far:
bloom boosters
foliar spray
Silica (maybe at some point)

I'm also in the middle of building drain tables. As suggested, I don't plan to over water but I think it will be good to have the soil well drained at all times and it will make doing flushes much easier which I would like to do at least once per plant. Using 6063 square tubing and angle for the structure, 6063 square tubing for the table top, 5052 sheet for the drain pans, 316 stainless hardware with rubber washers in the drain pan. Will be 3 tables that wrap around the room with interconnected pans hooked to a drain line that leads outside. It's something I've put off doing for a year but I brought the metal home today, got my chop saw set up.

Sounds like you got quite a plan and project there. Ive done uva leds the last year, thru various different lights and i really like the results, except for where there was a lot of blue already where it seem to stunt a bit. Our best results was with lights that incorporated high and wide reds: low k, 90cri + 660 and uva in 2:1 400/365 proportions: gave it loud and bright smells and taste and good plant health. Also had some GLA strips and another HE type spectrum (4000k + 660 + uva) and it seems like with more green the uv had a more subtle but maybe complex smell. I would really recommend led uva, it seems like much progress has been made on that front; maybe pandemic had something to do with it. Even 1$ diodes from china in 3535 foot print seems to have decent efficiency.
Whats your lighting nowadays? Still rocking any high cris? Im still fairly sold on them but nowadays it seems like the 630nm diodes are getting so good that they may be a good replacement for that 90cri red bump.

 

[Rahz]

I've been using 3500K 90CRI (Vero29C G8 at 700mA) for a few years now. I think I'm going to add some reptile bulbs, easy drop in. I have repeat cycle timers but haven't decided how to use them. Old school recommendations in late flower is 5% UVA and 1-2% UVB which I could achieve with the "15.0" reptile bulbs running full time when lights are on. Some newer information suggests running no UVB and less UVA which I could get pretty close to with the "5.0" reptile bulbs and a repeat cycle timer. What would you suggest?

When I upgrade my lamps (could be a while) I'll probably look into CXM32 Hortilum. Just eyeballing it, looks like maybe 3.2 umol/j at 700ma Edit: doing the math the umol/j at 700ma seems to be about 2.5 rather than 3.2

Yes, very excited about the drain tables. I got the metal chopped up today. Hardware arrives Monday. I originally built a bunch of plant caddys with castor wheels which has worked out pretty good for moving plants around but the tables will be a big upgrade.

 

[Rahz]

After digging into the CFL reptile bulbs, it seems the difference between 5, 10 and 15 ratings are 5(5% UVB and 30% UVA), 10(10% UVB and 30% UVA) 15(15% UVB and 30% UVA). While this information was only available for one brand I suspect it's similar across manufacturers.

Trying to average UVA and UVB (morning+noon,+evening/3) at sea level I come up with about 6.15% UVA and .27% UVB. At 1000 meters the average goes up to 6.91% and .32% respectively.

A 13w 5.0 reptile bulb paired with one of my 200w lamps will provide UVA 1.83% and UVB .31%. If I wanted to replicate the UV intensity of Sunlight I couldn't do it with reptile bulbs alone.

If I used one 13w reptile 5.0 bulb and one 15w blacklight with each 200w lamp that works out to 7.63% UVA and .29% UVB. That would be pretty close to UV percentages in Sunlight.

But the suggested use rates of UVA are all over the board, from 1-10%. And UVB from 0-2%. I think maybe I will just stick with the 13w 5.0 reptile bulbs for now. Seems like it won't overpower the plants, especially considering the light spread with bulbs isn't exactly straight to the plants.

I read that aluminum is not a good UV reflector and I'm planning on using aluminum reflectors so I'll be painting them white. Some cursory digging suggests automotive paint may be better at resisting and reflecting UV than regular paint.

Any thoughts on any of this?

 

[Rocket Soul]

Ill get back with a detailed but but probably rambling response, i got copiloting to do (gps map reader on the danish country side) and some family stuff going on. But yes, i think your probably right in first and foremost try whta you can with what you got already. Happy to give my opinions, still remember the cri shootout you did many years ago which really guided some of my choices in the last few years

 

[Rocket Soul]

After digging into the CFL reptile bulbs, it seems the difference between 5, 10 and 15 ratings are 5(5% UVB and 30% UVA), 10(10% UVB and 30% UVA) 15(15% UVB and 30% UVA). While this information was only available for one brand I suspect it's similar across manufacturers.

Trying to average UVA and UVB (morning+noon,+evening/3) at sea level I come up with about 6.15% UVA and .27% UVB. At 1000 meters the average goes up to 6.91% and .32% respectively.

A 13w 5.0 reptile bulb paired with one of my 200w lamps will provide UVA 1.83% and UVB .31%. If I wanted to replicate the UV intensity of Sunlight I couldn't do it with reptile bulbs alone.

If I used one 13w reptile 5.0 bulb and one 15w blacklight with each 200w lamp that works out to 7.63% UVA and .29% UVB. That would be pretty close to UV percentages in Sunlight.

But the suggested use rates of UVA are all over the board, from 1-10%. And UVB from 0-2%. I think maybe I will just stick with the 13w 5.0 reptile bulbs for now. Seems like it won't overpower the plants, especially considering the light spread with bulbs isn't exactly straight to the plants.

I read that aluminum is not a good UV reflector and I'm planning on using aluminum reflectors so I'll be painting them white. Some cursory digging suggests automotive paint may be better at resisting and reflecting UV than regular paint.

Any thoughts on any of this?

Yes, youll get very varied numbers from trying to calculate and from people having tried things that either worked or not, and usual everybody seems to think that what they did is the only Truth. I think varied results depend much on varied applications, efficiencoes and actually on what spectrum they started with.
My own thinking: yes, by all means add uv, its good for the results but be careful not to add too much so that total blue+uv levels remain somewhere below 20%. We seen both no yield reduction with uv and smaller buds and lower yields depending on what we started with: adding uv to 4000k based blue levels didnt work well even if we added far red in similar proportions.
Adding 50/50 uv/wide reds worked well with 2700k 90cri though when adding too much some plants had weird reactions (blurple lile density loss and weird smells like celery). But our more sturdy genetics lapped it up even at 10% uva.

When adding uv its a good idea to look at the spectrum and composition; i feel (from reasoning of my own and seeing stuff online aswell as our own tests) that you wanna try to recreate something similar to the sun: a downwards slope from blue to violet to uva to deep uva to uvb. So that each one of those is a bit lower the lower you travel. If you dont have any red sup in the spectrum you may wanna add this aswell at the same time or on a separate dimmer preferably.

Heres a look at one of our attempts maybe most relevant to you: adding uv to plain white. I couldnt get the funding to rebuild some other lights so i just decided on suplementing all my reds and uvs all on one channel, a sort of "horti centric blurple" so to speak. I got a china supplier to make me some 4up ledstars: in 4S config so i could run any diode on the same pcb. The total string (iirc) would be 4x400, 2x385, 2x365 6x680 6x660 1x630 1x730. Theres more reds than uvs but in voltage and output they would be fairly 50/50.
If i were to apply the same idea to your 3500k 90cri i would probably:
Use more reds (maybe 1:2 uvs/reds in output/voltage?) but add a bit more 630 and remove some +660 diodes. Its seems like pushing chloro b more than chloro a may not always be the best idea.
This horticentric blurple gave us good results for quality but some strains didnt react well to too high proportion of supplement: 240w over 900w of whites low K high cri gave some funny results and some density loss in some strains while others seem to do really really well. We got some cultivars to really shine with volatile/terpentine/solvent type smells, that type of smells we only really had from hps back in the day. Some of the funky results also gave volatiles; but not in a good way, dont like the celery smell even if it came with some.

But as a general advice to you: try uv. But be prepared to add some reds if you already have a full cycle type spectrum. Adding violets also seems to work wonders while still being photosynthetic. Seems like starting out with violets and then working yourself lower into the uv spectrum is a good way to procede. Main wavelengths id like to think as important are 400-420, 365 and 285.

Im going to try to track down that horti cmx you talked about, sounds interesting.

 

[Rocket Soul]

Had a look at the cxms, i wouldnt sneer at 2.5umols, pretty darn good spectrum. Low green and wide reds. A 2.5 light at 33w/foot gives the same levels as a 2.75 at 30w/foot and same as a 3.0 at but using 36w instead of 30w... I dont feel theres any point to be too shy with the watts as long as you get the results you want quality wise. In the end its just another 50ish watts per meter which is usually not the biggest of deals if you arent under hard amp limitations. Unless youre growing with absurd kwh prices and mega costs for cooling then 50w should be assumeable for better yields of the right quality weed. 80w extra per m2 is 30kwh per month; even with stupid high power rates thats just like 5euro/$ extra per month or 10 per cycle/m2. I feel its usually worth it as long as it doesnt mean you need to restrict the amount of trays you can do at the same time

 

[Rahz]

That's all pretty reasonable. The 3500/90 is already slightly over 20% blue so I'll have to keep an eye on the characteristics you mentioned and consider whether to modify the spectrum further. It may be that a rebuild using horticultural cobs or adding more red will be a good project next year. Either way it's potentially another bump in the wattage so figuring out which would be most efficient will be a rabbit hole of graph digitization. AI is generally helpful but it can't (yet) pull out the visual data in datasheets. It can digitize graphs if I snip them out so that might be something to play with.

My 200w Vero lamps output about 185w each. 6 CXM32s at 700ma would output about 214w so... that would be a slight bump in PPF, just slightly more than the extra wattage you suggested. I have a case of 133mm and 150mm pin fins left over from the Tasty days, would just need the cobs and drivers. Dang it, now you've got me thinking about building new lamps! thanks for the comments!

 

[Rocket Soul]

That's all pretty reasonable. The 3500/90 is already slightly over 20% blue so I'll have to keep an eye on the characteristics you mentioned and consider whether to modify the spectrum further. It may be that a rebuild using horticultural cobs or adding more red will be a good project next year. Either way it's potentially another bump in the wattage so figuring out which would be most efficient will be a rabbit hole of graph digitization. AI is generally helpful but it can't (yet) pull out the visual data in datasheets. It can digitize graphs if I snip them out so that might be something to play with.

My 200w Vero lamps output about 185w each. 6 CXM32s at 700ma would output about 214w so... that would be a slight bump in PPF, just slightly more than the extra wattage you suggested. I have a case of 133mm and 150mm pin fins left over from the Tasty days, would just need the cobs and drivers. Dang it, now you've got me thinking about building new lamps! thanks for the comments!

I can honestly say its hard for me to feel more complimented than by hearing you say that

That's all pretty reasonable. The 3500/90 is already slightly over 20% blue so I'll have to keep an eye on the characteristics you mentioned and consider whether to modify the spectrum further. It may be that a rebuild using horticultural cobs or adding more red will be a good project next year. Either way it's potentially another bump in the wattage so figuring out which would be most efficient will be a rabbit hole of graph digitization. AI is generally helpful but it can't (yet) pull out the visual data in datasheets. It can digitize graphs if I snip them out so that might be something to play with.

My 200w Vero lamps output about 185w each. 6 CXM32s at 700ma would output about 214w so... that would be a slight bump in PPF, just slightly more than the extra wattage you suggested. I have a case of 133mm and 150mm pin fins left over from the Tasty days, would just need the cobs and drivers. Dang it, now you've got me thinking about building new lamps! thanks for the comments!

- You make me wanna build a better light -

About the best compliment you could give me.

The 3500k 90cri is +20 blue? Though it was around 17?? If you have solid dets on rgb on some spectrums id love to see i know there was a resource somewhere but i cannot recall where on the led section

 

[Rahz]

I can honestly say its hard for me to feel more complimented than by hearing you say that

View attachment 5443981

- You make me wanna build a better light -

About the best compliment you could give me.

The 3500k 90cri is +20 blue? Though it was around 17?? If you have solid dets on rgb on some spectrums id love to see i know there was a resource somewhere but i cannot recall where on the led section

Just running off memory. I just digitized a Citizen 3500K 90CRI spectrum -> WebPlotDigitizer -> Excel and come up with 29% between 400-500mn (of 380-780). 420-480 is 25.7%. I don't suspect the Bridgelux SPD is much different than Citizen.

I have cvs files for the Citizen and also Hortilum if you are curious but they're not valid file extensions here.

 

[Rahz]

I've tried using AI to do the whole process. It can in theory complete all the steps, open the datasheet, pinpoint the SPD, crop, digitize and analyze, but the numbers are always messed up. I thought I could give a cropped image and skip some steps but it somehow screws that up too. Eventually it will be able to analyze datasheets effectively but it's not there yet.

 

[Rocket Soul]

Just running off memory. I just digitized a Citizen 3500K 90CRI spectrum -> WebPlotDigitizer -> Excel and come up with 29% between 400-500mn (of 380-780). 420-480 is 25.7%. I don't suspect the Bridgelux SPD is much different than Citizen.

I have cvs files for the Citizen and also Hortilum if you are curious but they're not valid file extensions here.

To my recollection i found some small differences between crees/samsungs 2700k 90s and bridgelux of same spectrum. Iirc correct from talk around threads and those old threads of Supra and Cobkits where they looked at cobs with licor: bridgelux tends to have slightly redder spectrums with less green but a little more cyan. Ive done both crees and blux in that spectrum but not side by side so hard to say but my guess is that they push a bit more red. All cct arent created equal but differences are still minimal.

Be careful about ai, they do not reveal truth, they reveal truthy. Job wel done is you giving them thumbs up, happy with the answer. My dad does family research going back long time. I once asked ai about a name that surfaced and got the response that he was the foreign minister of my country over several governments including over ww2! I was ecstatic and proud but double checked and it was all a lie; i guess ai thought i could use a little boost and made something up.
Ive also had very dubious behaviour from ai when i started to ask it questions it didnt like, almost passive aggressive and going from responding in a way i "trained" it back to responding in the most impossible to understand way it could find.
I think its just good for creating some blurby content and to use as a sort of google for complicated questions; when standard search terms wont do it. Dont try to get language models to do math for you. You have to check every single step of what it does, and it will never give any indication on when its actually doing things right or wrong.