Reversing String Connectors

aerouta

New elf
Joined
Aug 15, 2022
Messages
2
I am stringing a basic 12v RGB Pixels string to a custom prop. Once I was done stringing the model, in order to minimize wiring runs, I needed to swap the connectors to the opposite ends on the string. Now the string is not working. A few pixels blink momentarily while I am plunging in the string, but that is it. Other stings are working just fine. I double-checked the connection to ensure that I maintained the original wire paring. Am missing something? Are there special considerations when reversing connectors on a string?
 

merryoncherry

Senior elf
Generous elf
Joined
Apr 2, 2022
Messages
633
Location
Cherry St., Hudson MA USA
Each pixel chip takes a bit of data off its "data in" and passes the rest down its "data out". Feeding controller dat to "data out", which I think is what you did, doesn't work.

You want it reversed, you have to insert the string in reverse physically, and make the necessary adjustments in the layout software. If it is a custom prop, it's doable.
 

uncledan

Senior elf
Generous elf
Joined
Dec 27, 2014
Messages
545
Location
Buellton California
Swapping connectors isn't an issue as long as the data flow is correct. Data in is opposite the black IC on the pixel if using bullets. Most are labeled DI for data in. Square nodes generally have an arrow for data flow. Power flows both ways fine, but not data. It's a good practice to keep everything the same as far as connectors but not required. As others have mentioned it sounds like you have the data flow reversed which is why it's not working
 

Skymaster

Crazy elf
Global moderator
Generous elf
Joined
Dec 19, 2021
Messages
1,042
Location
Western Sydney
Shouldn't plug or unplug whilst powered either. That can kill chips.
I'd be curious to know the basis of this assertion. Other than the risk of using dodgy connections and potentially wiring reverse polarity, what would be the technical difference between plugging something in, and flicking a power switch?
 

TerryK

Retired Elf
Joined
Feb 9, 2020
Messages
655
Location
West Central Ohio
For pixels, little to none I would say. The data signal is so low power that even if a pixel saw a significant voltage on its DI pin it would be rare to cause any damage. And most datasheets call for a small resistance in the data line to help with signal reflection which would I think also lead to less damage if any.

This policy usually applies to devices with internal capacitors that need to be discharged or nearly so before power is applied, devices which have multiple power connections, or devices that draw significant amperage.

About the only thing that comes to mind is the form of contact bounce as the powered pins slide together. Both the V+ and V- would have quite a bit of electrical noise on them until the connector is seated. Could cause the IC to 'lock up' I guess but a power cycle should fix that.
 

Skymaster

Crazy elf
Global moderator
Generous elf
Joined
Dec 19, 2021
Messages
1,042
Location
Western Sydney
Thanks Terry. That does make sense. One thing that was discussed in chat earlier is the possibly of the ground connection being last, and the data line forming a "temporary ground" as the connection is joined in, which can cause issues, which sounds like a plausible explanation too. Obviously there's anecdotal evidence from folk who have had hot-plugging issues, it was just something that played on my mind as to "why" :)
 

TerryK

Retired Elf
Joined
Feb 9, 2020
Messages
655
Location
West Central Ohio
@Goliath : No offense meant. Hopefully non taken.
Over the years I've been hot plugging and unplugging quite a few things and to my knowledge have yet to destroy things. Have I inadvertently lessened a device's lifetime? Perhaps. About the only things I've avoided are those a manufacturer specifically says bad things happen if one does this. In any case, I pretty much routinely plug and unplug powered pixel strings and think little of doing it. Pixel power supplies, not so much.

@Skymaster : I thought about the V- last possibility when I wrote the comment but obviously did not include it. To upset the IC internals enough to damage the IC I would think the DI pin would need a low resistance path to GND. I'm not sure the upstream DO could provide that although it does generally spend more time held low than high. Goes without saying the IC internal design plays a part as well.

This is the typical situation where one can say it will never happen and somewhere a hot plugged string will go up in smoke. So, play it safe, take a chance? To each their own I guess.
 

Skymaster

Crazy elf
Global moderator
Generous elf
Joined
Dec 19, 2021
Messages
1,042
Location
Western Sydney
@Skymaster : I thought about the V- last possibility when I wrote the comment but obviously did not include it. To upset the IC internals enough to damage the IC I would think the DI pin would need a low resistance path to GND. I'm not sure the upstream DO could provide that although it does generally spend more time held low than high. Goes without saying the IC internal design plays a part as well.
I'm wondering if this happens more on the first pixel of the string, where a buffer chip on the output of the controller in the low-state would provide that low-resistance path. Usually they are only able source a few mA of current though, which I wouldn't have thought would be enough to cause a problem.
 

algerdes

Al Gerdes
Joined
Nov 29, 2012
Messages
289
Location
Lebanon, Illinois
After literally years of hot-plugging, tonight I lost the first pixel in a prop (singing ChromaBulb) that was connected via 18' of line. Not sure as to why, as this prop has been around for a long time and has been unplugged and plugged in several times a year. It could have been something with the length of the wire (18' of 18AWG SJEOOW with DIYLEDExpress connectors on each end). Or it could have been "just its time". We may never know.
 
Top