Page 2 of 5 FirstFirst 1234 ... LastLast
Results 11 to 20 of 48

Thread: DIY meteor tubes/icicles without ray wu Tee's or other parts

  1. #11
    Join Date
    Jan 2017
    Location
    SF CA
    Posts
    524
    Post Thanks / Like

    Default Re: DIY meteor tubes/icicles without ray wu Tee's or other parts

    Quote Originally Posted by XmasinVancouver View Post
    Yup, you have to loop the data from the end of an LED strip to the start of the next one. Gotta keep that in mind and deal with that appropriately (as you could very easily make your cables longer than the data can travel reliably).
    Last edited by Bwinter; 09-13-2020 at 10:47 PM.

  2. #12
    Join Date
    Mar 2012
    Location
    Lebanon, Illinois, USA
    Posts
    2,907
    Post Thanks / Like

    Default Re: DIY meteor tubes/icicles without ray wu Tee's or other parts

    One thing that we experimented with was taking two strips, back to back, and feeding the two of them the same signal. In our case, from top to bottom. Worked great! At the bottom, we ran the data line off of one of the strips back up and to the next drip tube.

    The purpose of our testing was so we could hang these out in the middle of a hallway, and no matter which direction you were coming from, you would get a great view. BTW - our completed units were run at 15% and comments still came in as to how bright they were.
    Live, Laugh, Love.

  3. #13
    Join Date
    Jan 2017
    Location
    SF CA
    Posts
    524
    Post Thanks / Like

    Default Re: DIY meteor tubes/icicles without ray wu Tee's or other parts

    Yup, splitting the data-in, then bringing a data-return up for the bottom works (if you actually need to worry about the return data-line).

  4. #14
    Join Date
    Nov 2016
    Location
    North Vancouver, BC, Canada
    Posts
    912
    Post Thanks / Like

    Default Re: DIY meteor tubes/icicles without ray wu Tee's or other parts

    Quote Originally Posted by Bwinter View Post
    Anything can look good in a promo-pict.

    Strip LEDs simply donít benefit from seeing the individual pixels.
    Lol man don't worry, you don't have to defend that view. In fact, those pictures look kind of photoshopped honestly. I wasn't saying they look good. I was just using those close up pictures to show what I meant that I thought I had noticed with those kind of lights. It's still obviously diffused, but you could kind of make out a little more of the source than normal. Obviously not bare strip.

    And either way, I fully agree. The aesthetic should technically be the most important part because ultimately, that's what people are seeing. I haven't built or owned meteor tubes before, so I asked, and got the answer. Kind of the least of my worries honestly. Especially since I already have a length of white straight pex here without having to source some other material

  5. #15
    Join Date
    Jan 2017
    Location
    SF CA
    Posts
    524
    Post Thanks / Like

    Default Re: DIY meteor tubes/icicles without ray wu Tee's or other parts

    Quote Originally Posted by XmasinVancouver View Post
    Especially since I already have a length of white straight pex here without having to source some other material
    There are so many reason why that is the right choice...

  6. #16
    Join Date
    Nov 2016
    Location
    North Vancouver, BC, Canada
    Posts
    912
    Post Thanks / Like

    Default Re: DIY meteor tubes/icicles without ray wu Tee's or other parts

    Good point about how far the signal can travel... I've been lucky in that I honestly haven't ever had enough length for that to become a problem.. How far can data travel on led strip before its a problem? Not even considering the wire in between the icicles yet. Could I just add or splice in a sacrificial pixel in between a few icicles at a certain point to refresh the signal? I don't know if that is considered a null pixel or feasible

  7. #17
    Join Date
    Jan 2017
    Location
    SF CA
    Posts
    524
    Post Thanks / Like

    Default Re: DIY meteor tubes/icicles without ray wu Tee's or other parts

    Quote Originally Posted by XmasinVancouver View Post
    Good point about how far the signal can travel... I've been lucky in that I honestly haven't ever had enough length for that to become a problem.. How far can data travel on led strip before its a problem? Not even considering the wire in between the icicles yet. Could I just add or splice in a sacrificial pixel in between a few icicles at a certain point to refresh the signal? I don't know if that is considered a null pixel or feasible
    Given that you generally want the icicles on a long-ish tether (to spread them out)--this is a major concern. Unbuffered, data generally doesn't travel that far from the end of pixels.

    Even cable-type has an impact (twisted-pair for data would be a benefit). Using regular 18/3, I've seen troubles at 5'-10' off a regular pixel (but this can be highly variable). And remember, you're probably sending a lot of data--30 pixels/icicle (and that's if you split the data) or 60 pixels/icicle (if you're doing the up/down method). Daisy-chaining a few icicles together (which would be normal), can really add up to a lot of data. Null pixels/line-buffers would be a reasonable options (albeit a hassle)--I think a line-buffer would have an advantage over just a null-pixel.

    Oh, and remember that if one icicle goes down/has a problem--all your downstream ones go bad too (and now you have a problem high-up in a tree).

    So...beating the dead horse...that's why I specifically developed a control-module to take all of this into consideration and minimize the problems.

  8. #18
    Join Date
    Nov 2016
    Location
    North Vancouver, BC, Canada
    Posts
    912
    Post Thanks / Like

    Default Re: DIY meteor tubes/icicles without ray wu Tee's or other parts

    Quote Originally Posted by Bwinter View Post
    Yes...but that then means the LEDs (on the front and back) will have different addresses for the same position. Theoretically, it is possible to model this correctly. But it's much easier to just have a single line of thirty-LEDs that are simply mirrored on the front and back.



    Yellowness generally comes from UV damage--but the shrink-tube is inside the PEX tubing (thus protected). Haven't seen any problem/degradation yet (after two years). The shrink-tube isn't strong enough to deform the LED strips. But it does lock them in place and hold them together (can't rely on the adhesive on he back of the strips....that doesn't hold after awhile with the heat).



    Standard cable-glands. PG7 (standard size-chart) is the 12.5 mm thread, which screws perfectly into the 1/2 PEX.



    14awg--why?!? WAY too thick. You only need to power a single icicle--not a distribution-board.



    Yup, you have to manage your data-line (taking into account the weak data-signal off the LED strips), and you have to manage 5V power requirements. Those icicles do consume a reasonable amount of power AND 5V doesn't travel well. My system takes care of all that for you....
    Yes...but that then means the LEDs (on the front and back) will have different addresses for the same position. Theoretically, it is possible to model this correctly. But it's much easier to just have a single line of thirty-LEDs that are simply mirrored on the front and back.

    -I'll just go based on your feedback here. That was the intent after all, since I don't have any set positions or opinions yet. The way I had figured I would do this, or at least pictured, would be similar to the way people will rig one string of pixels for a megatree, folding it in half so that it becomes two "legs". With Vixen, it's really easy to mirror props to the same sequence item. By assigning the same 30 pixels in the preview for the back and the front, I would have identical mirrored signal on both sides....theoretically. So one item in the sequence, playing on two props in unison.


    Yellowness generally comes from UV damage--but the shrink-tube is inside the PEX tubing (thus protected). Haven't seen any problem/degradation yet (after two years). The shrink-tube isn't strong enough to deform the LED strips. But it does lock them in place and hold them together (can't rely on the adhesive on he back of the strips....that doesn't hold after awhile with the heat).

    -Absolutely correct... And also, stupid on my part, because as long as i'm not using fully clear tubes, it wouldn't make a lick of difference unless the color was really significantly or dense.


    14awg--why?!? WAY too thick. You only need to power a single icicle--not a distribution-board.

    -Don't actually need or want 14 AWG. That's what I was kinda saying, that it would be way too thick to work with in this context or at least a pain. The only reason that would ever be a consideration would be if voltage drop was an issue. And if I was doing it as one main distribution line with feeders for the icicles over a longer run. I brought it up because last year, I ended up having to try and use 14awg for power injection due to MAJOR problems working with the Ray WU "promo pixels" that are documented in many threads here. It was basically an overkill in desperation trying to resolve the power issues that didn't make sense.

  9. #19
    Join Date
    Jan 2017
    Location
    SF CA
    Posts
    524
    Post Thanks / Like

    Default Re: DIY meteor tubes/icicles without ray wu Tee's or other parts

    Quote Originally Posted by XmasinVancouver View Post
    [COLOR=#333333]The way I had figured I would do this, or at least pictured, would be similar to the way people will rig one string of pixels for a megatree, folding it in half so that it becomes two "legs". With Vixen, it's really easy to mirror props to the same sequence item. By assigning the same 30 pixels in the preview for the back and the front, I would have identical mirrored signal on both sides....theoretically. So one item in the sequence, playing on two props in unison.
    They can be modeled many different ways--I think it's just easiest to model a single-line of 30 nodes. Apply effects to that, and not worry if the effect is treating the both sides the same (xLights v Vixen possibly handle this differently). Also, this is effectively doubling your pixel-count (data-wise).


    Quote Originally Posted by XmasinVancouver View Post
    [COLOR=#333333]
    -Don't actually need or want 14 AWG. That's what I was kinda saying, that it would be way too thick to work with in this context or at least a pain. The only reason that would ever be a consideration would be if voltage drop was an issue. And if I was doing it as one main distribution line with feeders for the icicles over a longer run. I brought it up because last year, I ended up having to try and use 14awg for power injection due to MAJOR problems working with the Ray WU "promo pixels" that are documented in many threads here. It was basically an overkill in desperation trying to resolve the power issues that didn't make sense.
    Since you're operating 5V over a fair distance, and carrying a lot of current--thicker gauge makes sense. But since you need to propagate a weak data-signal over a long distance--thinner gauge is better. This really comes down to how you eventually plan to connect everything together.

  10. #20
    Join Date
    Nov 2016
    Location
    North Vancouver, BC, Canada
    Posts
    912
    Post Thanks / Like

    Default Re: DIY meteor tubes/icicles without ray wu Tee's or other parts

    Quote Originally Posted by Bwinter View Post
    They can be modeled many different ways--I think it's just easiest to model a single-line of 30 nodes. Apply effects to that, and not worry if the effect is treating the both sides the same (xLights v Vixen possibly handle this differently). Also, this is effectively doubling your pixel-count (data-wise).




    Since you're operating 5V over a fair distance, and carrying a lot of current--thicker gauge makes sense. But since you need to propagate a weak data-signal over a long distance--thinner gauge is better. This really comes down to how you eventually plan to connect everything together.
    Still have that BOM handy? Wouldn't mind at least pricing it out. I think my main hesitation is that I don't want to transition to fpp this season, and I'm not entirely sure how I would adapt the board to run off of let's say a espixelstick or a Wemos instead of trying to get the whole pi and gpio pins and fpp set up. I know many love it and its on my list, but I really really need and prefer the live control rather than playing back exported sequence. I tweak or adjust a lot based on what I'm seeing even through December. I change things, gamma, lighting curve etc. On the flip side, I keep coming back to the fact that I really really like the centralized distribution with individual strands that can allow randomized spacing to fit the tree. The thought of having just a straight spaced set just wouldn't be as great in that setup in a tree vs lining a roof trim. That's what initially caught my attention about your build

Page 2 of 5 FirstFirst 1234 ... LastLast

Bookmarks

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •