Some DIY LED questions....

IggyP said:

just a quick update.....scrambling like mad but i did manage to get the first unit finished and turned on without issue....properly blinding ;P

regarding wiring....i wish i found this other one beforehand
https://www.amazon.com/dp/B00MOKO820

......this tinned wire i still dont know about
https://www.amazon.com/gp/product/B010T5Z1M2

...i saw some info here and there saying it should be fine, but i am still not convinced, some data still suggests otherwise.....this wire seems to me like 18-20ga thats "coated" in 50% tin lol...i did scratch into it and see some copper but i dont see how they gonna add the tin then still have it be 16awg and expect same conductivity when tin is much less conductive....seems like a total scam, unless ppl wanna use it as smaller gauge then perhaps the anti corrosive would actually be a benefit....did i miss something on that??

anyway yeah, idk why solid copper 16awg is so hard to find....but the magic word here seems to be "dog fence"....fwiw

Click to expand...

"tinned" wire is simply coated with solder - It does not affect the conductivity, nor does it make it significantly thicker - its only a few molecules thick. Tinned wire will not oxidize nearly as fast as pure copper wire will.

i know that is the sale point.....but if you look close at that listing, it isnt up front about the "tinning" which is the first suspect....true the tinning doesnt make it thicker, but that is actually my other gripe....i would prefer 16ga of copper, then tinning over top.....that would be "legit".....rather than a 16awg wire thats literally half tin....

if you look purely conductivity wise, say, on a chart like this one
https://www.bluesea.com/resources/108/Electrical_Conductivity_of_Materials

you can see an ocean between those stats....tin is also much cheaper, so i am pretty sure that is a "hoaxy chinese product".....

in this case, i nearly could have used 18 wire anyway so perhaps it is ok...house didnt burn down while i slept or anything, but i will need to dig more for better confidence...just waking up now, but im gonna grab some pics and do some more current tests....see if there is any unusual voltage drop or anything as a result of all that tin...

ok, ill get some pics up next post lol but quickly for now i did a current test and am def loosing .5 volts by the end of the 4' run which seems like it should be more like .05dcv ....even at the start of the run tho, i can only get it to 22.6v with the leds attached....for some reason??

....also noticed this driver is basically too hot to touch....wire isnt hot at all tho...that normal???

if the voltage drop was due to driver kicking to CC mode, amps should be higher....not sure exactly whats up hmm

You can measure voltage drop of the wire directly, just connect a volt meter to each end of the wire.

BTW, people often overlook that there are typically two wires, for a 4 ft. run you must calculate the loss of 8 ft. of wire.

Running 21 strips @ 23 Volts requires 483 volts. Not going to happen with a 320, even with the 700 mA model it's only rated @438 volts if I remember correctly.

IggyP said:

i know that is the sale point.....but if you look close at that listing, it isnt up front about the "tinning" which is the first suspect....true the tinning doesnt make it thicker, but that is actually my other gripe....i would prefer 16ga of copper, then tinning over top.....that would be "legit".....rather than a 16awg wire thats literally half tin....

if you look purely conductivity wise, say, on a chart like this one
https://www.bluesea.com/resources/108/Electrical_Conductivity_of_Materials

you can see an ocean between those stats....tin is also much cheaper, so i am pretty sure that is a "hoaxy chinese product".....

in this case, i nearly could have used 18 wire anyway so perhaps it is ok...house didnt burn down while i slept or anything, but i will need to dig more for better confidence...just waking up now, but im gonna grab some pics and do some more current tests....see if there is any unusual voltage drop or anything as a result of all that tin...

Click to expand...

You literally have NO CLUE what you are talking about. NONE.

ok....here i know we need to officially enter the photos, words only go so far...

some shots to show the spacing and some output/voltage testing....


here u can see the actual test ive done showing voltage drop by the end of the 4' run....

btw...i didnt claim to be an expert, i am willing to learn, but if u want me to trust that wire higgs then you will have to explain yourself not the other way around....burden of proof lies to you as ive linked and stated my logics....for this matter, if it is so well conductive why not they just make solid tin wire and get rid of the expensive copper altogether? ....it seems it would have to matter....its more resistive, less conductive...how can that not matter...

im willing to do any tests so long as we are here learning, but the next unit will be solid copper, so that will be the first comparison ;P
peace!

IggyP said:

ok....here i know we need to officially enter the photos, words only go so far...

some shots to show the spacing and some output/voltage testing....
View attachment 4141384 View attachment 4141386View attachment 4141392View attachment 4141387 View attachment 4141388 View attachment 4141390

here u can see the actual test ive done showing voltage drop by the end of the 4' run....

btw...i didnt claim to be an expert, i am willing to learn, but if u want me to trust that wire higgs then you will have to explain yourself not the other way around....burden of proof lies to you as ive linked and stated my logics....for this matter, if it is so well conductive why not they just make solid tin wire and get rid of the expensive copper altogether? ....it seems it would have to matter....its more resistive, less conductive...how can that not matter...

im willing to do any tests so long as we are here learning, but the next unit will be solid copper, so that will be the first comparison ;P
peace!

Click to expand...

Did you figure loss from all the wago connections daisy chained ?

1212ham said:

You can measure voltage drop of the wire directly, just connect a volt meter to each end of the wire.

BTW, people often overlook that there are typically two wires, for a 4 ft. run you must calculate the loss of 8 ft. of wire.

Click to expand...

He is losing .6V according to the pic. 2 meters of 16g wire will drop .35V at 13 Amps. Throw in a shit ton of Wago's and there's the other .25V

i will wait for this "dog fence" to get here sometime next week and do some raw wire comparisons before wiring up the next unit...

i am also curious what differences could be from using the 4way push connectors instead of the 3way like i did on this....i can still test the ends but with 4ways i could check all along it....but then again idk if that would drop more again for bulkier connectors...its not much i know, but im curious to test these things...

also getting a 20a multi meter on the way so i can test the amps properly on the dc side soon....

i will check some temperatures of driver and led later and share pics....by touch the lights dont seem hot, at all...the driver on the other hand is one hot potato being pushed to the limits....

i think with 15-20ft of 12-14awg i can mount them remotely outside the room....ON A FAT HEATSINK... without loosing much voltage. so it will be like that most likely 14awg to the 16 to the 18 lol but i didnt come all this way to loose efficiency by any easy avoided mistakes... i will test it all along the way next couple weeks before anything much serious in there...

will be 3 units.....i could put UVB between them....not gonna tho....i could put blue t5s between....not gonna do that either.....could put a purple sun pulse halide in an hid hood above lol....probly not gonna do that either tho sigh....i guess im having enough fun so 3 units and unless i start placing them vertically that will be all ;P

a close to true "1k watt replacement" in power use, not counting the AC i wont run....im def very curious friends...

IggyP said:

ok....here i know we need to officially enter the photos, words only go so far...

some shots to show the spacing and some output/voltage testing....
View attachment 4141384 View attachment 4141386View attachment 4141392View attachment 4141387 View attachment 4141388 View attachment 4141390

here u can see the actual test ive done showing voltage drop by the end of the 4' run....

btw...i didnt claim to be an expert, i am willing to learn, but if u want me to trust that wire higgs then you will have to explain yourself not the other way around....burden of proof lies to you as ive linked and stated my logics....for this matter, if it is so well conductive why not they just make solid tin wire and get rid of the expensive copper altogether? ....it seems it would have to matter....its more resistive, less conductive...how can that not matter...

im willing to do any tests so long as we are here learning, but the next unit will be solid copper, so that will be the first comparison ;P
peace!

Click to expand...

I'm telling you EXACTLY what "tinned" wire means. Its simply a solder coat - NOT TIN. Solder is a Tin/Lead alloy. It does NOT significantly effect conductivity. I've been working in this industry for 35 years, I do consider myself an expert. 0.6V is not a big enough loss to be too terribly concerned about, its less than 8 watts of lost power - but I would have used 14G anyway and cut the wire loss in half. You can use pure copper wire if you like, but in a humid environment, I can guarantee green corrosion all in those wagos in short order.

The WAGO's themselves are another mess. Each of those introduces losses of its own. A much better method of wiring multiple strips is using barrier strip distribution blocks like in the picture I posted. You install jumper strips on those blocks, install a couple on the frame ends, then run 14G from the supply to the blocks, then 18g from the blocks to each individual strip.

Now to adjust your driver - turn the Io adjustment fully clockwise, then turn the Vo fully clockwise. Then monitor the voltage while adjusting the Vo counterclockwise, and as soon as you see the voltage drop by 0.1v then stop.

i will have the 14, so that can be an option...but im not sure i liked the distro blocks, as it ends up using longer/more wire... that can be negated some i know, by thicker wire and/or splitting it into 2 shorter parallel runs or something, but i would rather not scale up the wiring unless there is not another option....

i did a quick test for now, with a footcandle meter.....i can see very simply toward the driver is about 25% brighter than the middle, which could be part of the spacing i did, but then the back end drops down like 50% of the start side so something is def wrong i think....

for the tin wire....it will just have to be tested vs the pure solid copper i think....i dont mean to question your professional experience higgs, but there could be a few things going on here based on the info i gathered...i dont find it hard to believe at all that a chinese plant somewhere could putting out cheap wire....or at the very least, that whatever is making up 50% of this stuff is notably hurting current...

i believe i sorted out the dimming already, what i did, is set it at 23v "without" the leds attached.....but then soon as hooking em up the same connection point dives to the 22.5vdc shown in pic...wont go above 22.5v at all...not even a flicker..

ive also tested temps once maxed to 365w for a while....i put the driver on heat sinks for now, but still pretty hot....

clocked middle of the middle strip at 109degF
driver casing at 135degF

here is more data i cant make sense of idk about any of you.....

so the voltage max i was able to get of 22.5, i thought could be a result of driver going in CC mode....so i dimmed the AMPS, and wow some reason the voltage also dropped, a few digits...so guess its not that.....doesnt seem like that should even be possible regardless of CC or CV mode....curious how that works...

Is voltage drop dangerous or what? Strip just gets little less voltage (0.5V) so little less watts. Or...?

KonopCh said:

Is voltage drop dangerous or what? Strip just gets little less voltage (0.5V) so little less watts. Or...?

Click to expand...

hi,
im thinking of it for a few reasons....

1. one of my goals was a max efficiency possible type pursuit( i.e. the lower than average current and more strips per area type approach)

2. i made the plan for squeezing every lumen out of this driver possible, which kind of seems like it might need for blooming...but also given that it starts lower DCV than expected, and then drops DCV again by the end with a light dip even showing on a footcandle meter...i think that is where i start to defeat my own goals..

3. i dont trust this "tinned wire" so far...

if i have to monster it with 12 or 14awg and have wires going all over to max outputs i dont have a problem with that....wire isnt sooo expensive, but i am trying to approach it to solve each problem as it comes with minimal improves for desired effects...

this recent reveal, begs the question....how many of these strips can this driver run in one parallel line, with which gauge wire, without output dropping....i am tempted to break it into 3 jogs even but idk what the problem exactly even is atm lol......

OMG, way to many connectors!

Give up your obsession of over tinned wire, it's not an issue!

Look at the hundreds of builds on RIU for examples of good wiring!

Stop overthinking everything and listen to people that know what they are talking about.

18 strips and two wire nuts.
Three strips daisy chained isn't the best electrically, but it is clean and practical. Wire nuts aren't high tech, elegant or cool. They are dirt cheap and make an excellent connection of many wires at one point assuring all lines are at the same exact voltage.

hi,

to be clear, its not so much an obsession over tinned wire, as "this tinned wire" but i will see about getting more data asap...

i have looked around and listened but i dont just copy copy, or do whatever someone says unquestioning because they have "experience" ...lets try to not be so patronizing and have a bit of open minds to allow a little creativity and learning eh....

from my research, daisy chaining with the connections on the strips themselves, is even worse than using push in connectors....

is this true tho? wire nuts dont loose current like push in connectors? i dont see that making the total diff here, but im not exactly adverse to using them....push in connectors are dirt cheap too tho btw, 7.50$ for 100 of em...not exactly bleeding for those...assuming they perform the same tho idk

pretty sure im just going to have to break up the runs to be shorter, and possibly replace the wire....and possibly connectors i suppose lol...hell im open, but lets be specific and within the stated goals im after pls...

IggyP said:

have the 14, so that can be an option...but im not sure i liked the distro blocks, as it ends up using longer/more wire... that can be negated some i know, by thicker wire and/or splitting it into 2 shorter parallel runs or something, but i would rather not scale up the wiring unless there is not another option....

Click to expand...

What you are missing is that only the wire going TO the block carries the whole 13+ amps of current. The wire going from the blocks to the strips only carries 525ma (in your build), and 18g is plenty large enough for that amount of current - even over several feet. You are simply worrying about what doesn't matter here. Voltage drop over the wire increases with the amount of current in the wire. Reduce the current and you reduce the need for larger wire. Thats the whole point of the distribution block.

IggyP said:

daisy chaining with the connections on the strips themselves, is even worse than using push in connectors....

Click to expand...

Yes, I do not recommend doing that. You end up with the last in line getting lower voltage than the first. Its OK to chain pairs together (to simplify wiring), but I would not go more than that.

IggyP said:

wire nuts dont loose current like push in connectors?

Click to expand...

Yes, wirenuts give a better overall connection than push-in connectors, without a doubt. Thats why they are used for household wiring.

hey i just came across this, if anyone else is interested.....comparing connectors...


spoiler alert:
in this test,
plain(tinned) wire resist by .0049 ohm (bit longer)
wirenuts resist by .0045 ohm
push-in/wagos by .0055 ohm

seems like a pretty small number but yeah i need to get this into shorter runs somehow....

i get that 18 is fine for the strips....i have solid 18 copper in hand too so im not even wondering about that part, but i will have to test to see the optimal gauge and layout for the part that parallels the current to each strip....

i have 16awg solid copper and 14 solid on the way.....here atm i have 18awg soild, 12 solid, then the 16 tinned.....need more wire to test lol...if i was in more of a rush i would just over build the wire, but there is at least a few weeks here before go time so maybe we can uncover something helpful...

IggyP said:

plain(tinned) wire resist by .0049 ohm (bit longer)
wirenuts resist by .0045 ohm
push-in/wagos by .0055 ohm

Click to expand...

No, the resistance of the wire nut is not .0045 ohms. The resistance of the wire is .0049. The wire, with a wire nut inserted in the middle is .0045, so the the wire nut reduced resistance by .0004 ohms. Obviously that's not possible, the discrepancy is a minor difference in wire length.

Dumb test, it's far better to test a connector under load. To measure the voltage drop across a connection (or wire), put a volt meter across it and power up the circuit!

IggyP said:

hey i just came across this, if anyone else is interested.....comparing connectors...

spoiler alert:
in this test,
plain(tinned) wire resist by .0049 ohm (bit longer)
wirenuts resist by .0045 ohm
push-in/wagos by .0055 ohm

seems like a pretty small number but yeah i need to get this into shorter runs somehow....

i get that 18 is fine for the strips....i have solid 18 copper in hand too so im not even wondering about that part, but i will have to test to see the optimal gauge and layout for the part that parallels the current to each strip....

i have 16awg solid copper and 14 solid on the way.....here atm i have 18awg soild, 12 solid, then the 16 tinned.....need more wire to test lol...if i was in more of a rush i would just over build the wire, but there is at least a few weeks here before go time so maybe we can uncover something helpful...

Click to expand...

What good is 12 awg going to do for you when the leads from the driver is 14 awg on the 320. My 240 leads are only 17 awg. Every connection you make is another possible electrical loss and failure point in your build.
And I just have to ask this, What the heck is all that plant material laying everywhere with a half dead plant sitting under your light ?