Some DIY LED questions....

nfhiggs said:

Yeah, that's due to the long chain of connectors. You get the same effect if you connect strips end to end. Thats why I chose to use a distribution block.

Click to expand...

With so many strips in parralel, it's just a lot of damn wires! (Some series wiring next time. )

I had a look and found some buss bars with large diameter ports......
https://www.ebay.com/itm/5-Holes-Bridge-Design-Ground-Wire-Row-Copper-Tone-Terminal-Block-Connector/322256254574?ssPageName=STRK:MEBIDX:IT&_trksid=p2055119.m1438.l2649

I'd do this...
Mount one on each side of the frame, positive on one side and negative on the other. Run 18 ga. solid from the strips to the buss bar, putting several leads in each port. Done.

Bringing all the leads together at the buss should give low voltage drop and excellent voltage ballance on the strips.

ive got some "dual port" 12 gang terminal blocks with jumpers on the way....i figured i could try running 2 or 3 per junction and see if there is any ill effect, otherwise 2 blocks per unit i wont even mind, i found them cheap around $5 each....i am pretty wary of any series wiring on this build...seemed better to avoid it so far?....i dont know how many strips can be put in a series before any drop occurs, just seems less resistance with solid copper to each individually...i dont mind a medusa head on top if the #s check out lol...

i am still not sure which i prefer tho as to wires all on one side or on both sides....i sort of like the idea to keep them together, then fashion some kind of "wiring cover" for that one end.....well see, its good to have options ill ponder it while i wait on shipping

nfhiggs said:

Yeah, that's due to the long chain of connectors. You get the same effect if you connect strips end to end. Thats why I chose to use a distribution block.

Click to expand...

true i saw one similar build online and this long line of connectors is how they "overcame" the loss connecting direct thru strips....i am tending to think, they didnt look as close as i did to the output and/or perhaps figured it was good enough idk....they were also using a stronger driver so idk...

will be so nice to get this redone and check the output directly at each strip..

IggyP said:

i am pretty wary of any series wiring on this build...seemed better to avoid it so far?....i dont know how many strips can be put in a series before any drop occurs, just seems less resistance with solid copper to each individually...i dont mind a medusa head on top if the #s check out lol...

Click to expand...

I said series, not a parallel daisy chain.

I run one "lamp" that has 16x Gen1-560mm 3000K BXEB strips, wired in series with a HLG240C700, at full operating temp, it only loses 1 volt.....
357 cold start, 356 after 2 hours.

IggyP said:

true i saw one similar build online and this long line of connectors is how they "overcame" the loss connecting direct thru strips....i am tending to think, they didnt look as close as i did to the output and/or perhaps figured it was good enough idk....they were also using a stronger driver so idk...

will be so nice to get this redone and check the output directly at each strip..

Click to expand...

One method you can use to counteract the voltage drop when chaining the strips end to end (in parallel), is to connect the positive line from the driver to the first strip and connect the driver's negative line to the last strip in the chain. That will equalize the drop across all of the strips.

Chip Green said:

I run one "lamp" that has 16x Gen1-560mm 3000K BXEB strips, wired in series with a HLG240C700, at full operating temp, it only loses 1 volt.....
357 cold start, 356 after 2 hours.

Click to expand...

Yes, series connections do not generally have voltage loss issues. But you're working with much higher voltages - 350 VDC can be deadly.

nfhiggs said:

One method you can use to counteract the voltage drop when chaining the strips end to end (in parallel), is to connect the positive line from the driver to the first strip and connect the driver's negative line to the last strip in the chain. That will equalize the drop across all of the strips.

Click to expand...

that would be a good test....if i still had it put together i would try that for a quick compare....i might anyway if i get bored waiting...

its been a couple hangups at the end sorting this last wiring bit, but i still have a good confidence to what i am after....

i like the idea to push the driver to the extreme for a case like this, (CV+parallel) i just want to be sure all the juice is making light tho for the same reason....no juice left to kick it up, only down...

once i get the strips better even i think it will be a good entry to led for me.....i want to thank again those who took the time to make very helpful responses...i hope the info can help others in turn....

for one, it is a bit less common led project than i usually see....most notably ive seen less strips ran harder, in smaller arrays, with CC drivers etc...

this takes an only slightly higher start expense to make a "more even and efficient" end result....once getting up to this ammt of strips tho i really didnt even see any feasible series wiring options....not without going against the start goals...ideally i can loose 1volt and thats all then im loosing light.....loosing amps in series seems even worse because there is no flex...(meaning, the driver is capable of extra voltage that the strips cant use, but not amps)

i also have hugely limited chances of thermal runaway by this approach...unless i suppose if i lost 3/4 of the array at once somehow....and yeah...not worried about it at all....

i feel it has at least stepped toward improving common ideas while also giving me a chance to learn and try some progression in my styles...the brightness, seems will pretty ideal for my target goal etc....not bad really, i will be very excited to put some clones under, which were just snipped last night....should time out well....

i gotta admit tho, if i can make them finish quicker with adding some far red or someth i probably should do that, im taking a little too long lol but still wanna run the later finishing lines...guess i have time for that tho so that should be next progress after wiring...seems less worrysome than UV at least....even if i can get perfect numbers for UV app rates im gonna leave it out first round but idk bout the far red is maybe diff..

nfhiggs said:

Yes, series connections do not generally have voltage loss issues. But you're working with much higher voltages - 350 VDC can be deadly.

Click to expand...

Indeed. Not a recommended configuration for most builds. On this particular lamp, all strips are soldered, all wiring connections are soldered and shrinktubed. This unit is permanently mounted in a secure area.
It is about as "safe" as one could wish for...

Chip Green said:

Indeed. Not a recommended configuration for most builds. On this particular lamp, all strips are soldered, all wiring connections are soldered and shrinktubed. This unit is permanently mounted in a secure area.
It is about as "safe" as one could wish for...

Click to expand...

Exactly. You can't use the connectors on a high voltage build. They are only rated for 300v. = D I'm sure you know that, just reminding people not to exceed voltage rating of any connectors, wires, or anything used.

mail truck came early
thought i would be waiting till tuesday for some of this stuff!

ive "quickly" slapped together a rewire job....did a quick voltage check etc..

gained about .5 volts basically, still about .5 less than id like to see(23v at driver, 22.5 where the terminal is, havent checked strips yet)
EDIT: this first test was with about 7ft of 14gauge from driver

i will have to come back later to try some fine tuning stuff, but i can see the output is much more even and yes also seems brighter so thats cool.....should be able to improve alot while working on the second one..

EDIT: hooked it up again with the actual length of wire i need from the driver to the light....dropped a bit more....22.6v at driver and 22.2v@ strips.....using 12gauge...maybe 10ft..

temperature control is important this round, so i might have to sacrifice a little output to keep the room cool.. i can put a shorter line on there in the winter for full output but i know its only a small diff anyway

Thats only a 5.3W loss in your wiring now. Thats actually a little lower than what is predicted (5.5W) for 12 AWG over 20 feet (6 meters) @ 6.3 A. You can't ask for much better than that.

No doubt, there is a point of diminishing returns.

Running the driver on 240v rather than 120v will give about 4% more efficiency.

Yeah, and not to forget to mention that each strips has a slightly different voltage/resistance. Luckily the differences in F-strips a fairly low and it is not visible with the naked eye, but it's still measurable.

Wherever possible one should use 230v with LED drivers because the most are designed for a wide input range and only a few are designed for 120v only.
Some people have changed all their LEDs for a 4% improvement.
Makes a lot more sense than trying to get less voltage droop over thicker wiring.
Optimal of course would be both!

1212ham said:

No doubt, there is a point of diminishing returns.

Running the driver on 240v rather than 120v will give about 4% more efficiency.

Click to expand...

@IggyP
For efficient parallel wiring I like these quick connect "current/power thieves". You can use a thicker main wire(up-to 4mm² with the blue connectors) and a short piece thinner awg18 wire to connect the strip. They are available in red(0,5-1,5mm²), blue(1,5-4,0mm²) and yellow(2,5-6mm²) and a set of 10 pairs(power thieve + male connector) costs only 1,10$.
They cause certainly a similar voltage droop like the Wago's, but you can keep the thinner wires extremely short.

https://www.ebay.com/itm/20X-Blue-Quick-Splice-Scotch-Lock-Wire-Terminals-Crimp-Connector-Cable-Snap/372257981268?hash=item56ac4e0354:g:L0gAAOSwElpasL3M

nfhiggs said:

Thats only a 5.3W loss in your wiring now. Thats actually a little lower than what is predicted (5.5W) for 12 AWG over 20 feet (6 meters) @ 6.3 A. You can't ask for much better than that.

Click to expand...

hmm is it really just 5watt tho? by dropping effectively 1.8v? its only .8v off the actual output i know just seems that would add up higher....

1212ham said:

No doubt, there is a point of diminishing returns.

Running the driver on 240v rather than 120v will give about 4% more efficiency.

Click to expand...

wow....thank you so much to say that....i hadnt realized that was even an option....i have 220 i already ran all down the hall and its all i use for HIDs...it was in the back of my head....i thought i bought the wrong ones tho and would have to search 220drivers or something...def on the list, just need some 220 plugs is all for that....some kinda junction area too i guess, simple enough tho...

Randomblame said:

For efficient parallel wiring I like these quick connect "current/power thieves". You can use a thicker main wire(up-to 4mm² with the blue connectors) and a short piece thinner awg18 wire to connect the strip. They are available in red(0,5-1,5mm²), blue(1,5-4,0mm²) and yellow(2,5-6mm²) and a set of 10 pairs(power thieve + male connector) costs only 1,10$.
They cause certainly a similar voltage droop like the Wago's, but you can keep the thinner wires extremely short.

Click to expand...

interesting....i may just try these for a comparison after getting a couple more built....i actually am using 23" heat sinks on the strips, so i have a little channel left at the end where they attach to the L metal, to put all the wires....dang its alot of wire tho haha

ill drop a few more pics before long, just going to get this second one finished and hanged first at least(half done )




right when i thought i was nearly done buying stuff lol, but yeah thats some good points thanks everyone...def need to do some of this asap

IggyP said:

hmm is it really just 5watt tho? by dropping effectively 1.8v? its only .8v off the actual output i know just seems that would add up higher....

Click to expand...

Its NOT dropping 1.8 according to what you said below:

IggyP said:

22.6v at driver and 22.2v@ strips

Click to expand...

Its dropping 0.4V - no more, no less. Thats 5.3W at 13.3 amps (driver max current)
22.6 V at the driver is 22.6 volts at the driver - there is no "voltage loss" from 24V. Stop expecting to see 24V, because that's impossible with the number of strips you have and the current they are running at (13.3/23 = .578 mA per strip). It simply can't go any higher because it's at the current limit of the driver with that load. If you were to reduce the number of strips it would go higher, because the strips would need a higher voltage to get a higher current in order to hit the drivers max current.

i just mean that the driver puts out 24v....so im trying to keep wiring and connectors only eating up that extra volt the strips cannot use....which of course is impossible if i want to mount the drivers remotely, but otherwise no problemo (plan for wintertime)

i did have this set as a 22 strip build to pull the mA up a little but supping in the 4k strips they fit best with an odd one exactly in the middle...overall i like the spacing so far tho....with reflection, it is pretty even at both < and > 12" ....

i wanted to keep it "low current" too, so it seems this is my sweet spot im stuck in....i am trying to be careful tho, in order to not loose the efficacy i gained by this approach by bad wiring plans or such...it is hard to not try to maximize everything but i can see so far this easily compares to an hps type output (by basic footcandles at least)


as to randomblames suggestion, i do not think it can work so i abandoned the idea for now....i think all these type of "daisy chain" parallel approach will ONLY work for drivers that do not need to be maxed like this build....and even then, smaller units only....unless consistant output is not considered important....

even if i use 10wire as a base to splice from, every node down the line pushes just a little more current until the end of the line...where, with a build like this the voltage drop would be unavoidable and very noticeable...

the only possible scenario i am left considering is making a spider of lead wires(the 12gauge atm from driver to light), then using more smaller terminals(4+) with shorter 18g wires.....but that seems like maybe over ridic lol....idk, im curious what diff it would make i gotta admit.....also got some 7port terminals on the way so might do a test idk.......i have the 12 wire too, it just seems like the thing u look at and say wtf lol.....

just finished wiring the second unit...ill get some shots...skeptically proud so far lol....wiring is in tight, the whole lot of it, but im sticking with it as im most feeling this logic that they dont loose much to each one and if it ammts to 5 watt overall and not per strip or anything i probably will just convert to 220v and leave it to that

IggyP said:

i just mean that the driver puts out 24v....so im trying to keep wiring and connectors only eating up that extra volt the strips cannot use....which of course is impossible if i want to mount the drivers remotely, but otherwise no problemo (plan for wintertime)

Click to expand...

There is no extra volt to eat up, because the strips do not NEED it. Its only going to output the voltage the strips need (to max out the current load), no more, no less. The only thing being "eaten up" is the difference between the voltage at the driver (22.6V) and the voltage at the strips (22.2V), which is 0.4V. What you have is 1.4V of 'headroom" - voltage that is not being generated, but potentially *could be* generated if needed. For example - 12G wire drops approximately 0.5V per 6 meters (20 feet) of length at 13.3 amps. That means you could extend the 12g leads another 10 feet, and still have about 0.9V of headroom. You would still see the same voltage at the strips, but at the driver you would see around 23.1 volts.

IggyP said:

i just mean that the driver puts out 24v....so im trying to keep wiring and connectors only eating up that extra volt the strips cannot use....which of course is impossible if i want to mount the drivers remotely, but otherwise no problemo (plan for wintertime)

i did have this set as a 22 strip build to pull the mA up a little but supping in the 4k strips they fit best with an odd one exactly in the middle...overall i like the spacing so far tho....with reflection, it is pretty even at both < and > 12" ....

i wanted to keep it "low current" too, so it seems this is my sweet spot im stuck in....i am trying to be careful tho, in order to not loose the efficacy i gained by this approach by bad wiring plans or such...it is hard to not try to maximize everything but i can see so far this easily compares to an hps type output (by basic footcandles at least)


as to randomblames suggestion, i do not think it can work so i abandoned the idea for now....i think all these type of "daisy chain" parallel approach will ONLY work for drivers that do not need to be maxed like this build....and even then, smaller units only....unless consistant output is not considered important....

even if i use 10wire as a base to splice from, every node down the line pushes just a little more current until the end of the line...where, with a build like this the voltage drop would be unavoidable and very noticeable...

the only possible scenario i am left considering is making a spider of lead wires(the 12gauge atm from driver to light), then using more smaller terminals(4+) with shorter 18g wires.....but that seems like maybe over ridic lol....idk, im curious what diff it would make i gotta admit.....also got some 7port terminals on the way so might do a test idk.......i have the 12 wire too, it just seems like the thing u look at and say wtf lol.....

just finished wiring the second unit...ill get some shots...skeptically proud so far lol....wiring is in tight, the whole lot of it, but im sticking with it as im most feeling this logic that they dont loose much to each one and if it ammts to 5 watt overall and not per strip or anything i probably will just convert to 220v and leave it to that

Click to expand...

You need to understand how LED's and LED driver works together. If you drive an F-strip (single row) for instance @1,4A it would take ~24v, @525mA it's only 22,5v. Take only 9 strips in parallel and you measure 24v at the strip and 24,4 or so at the driver, take only 6 and voltage would be 25v.
There is a term in the drivers datasheet "the highest voltage at the output of the driver depends on the configuration of the end systems". This means the LED circuit design dictates the voltage output. With a 24v CV/CC driver it could be between 22 and 27v according to the datasheet.

BTW,
the driver will only switch to constant current mode if the design voltage is below 22V. In order to deliver his amperage, he switches to CC mode and reduce the voltage.
That would for example happen if you use EBgen2. strips (19.5v) with an 24v HLG driver of when you use so many parallel F-stripes that the required voltage drops below 22v.
As long as the required voltage is above the drivers minimum voltage(see "voltage adjust range") it will work in CV mode.
The "constant current area" is only used in CC mode and the voltage poti will be disabled in this mode, which is a good indicator to figure out which mode is used.