Falcon Christmas

Other Controllers and Hardware => General Hardware => Topic started by: drlucas on January 24, 2015, 01:11:07 PM

Title: Ray's latest TM1809s - red LEDs failing big time
Post by: drlucas on January 24, 2015, 01:11:07 PM
I had Ray replace about 30 or so 3m strips (90 lights per strip) after most of the lights failed on me over two weeks. Well, the latest 1809s held up pretty good for me over December. I brought my matrix in on News Years day when the weather was half decent and just went to do some matrix tools testing this afternoon and I have a very large number of failed red lights. The temps were a yo-yo for us in December ranging from the single digits up to maybe 50 a few days....most around the freezing mark though. Well, anyways today on my 32x90 matrix I have what I'd guess is about 100 or so defective red lights. Has anyone else brought their new TM1809s inside after being out in the cold and done any testing? I did a real quick test about 2 weeks ago but didn't notice anything broken until I went to do the testing today. Will post a video so you can see what's happened.
Title: Re: Ray's latest TM1809s - red LEDs failing big time
Post by: drlucas on January 24, 2015, 01:13:27 PM
Here is the video

https://vimeo.com/117683643 (https://vimeo.com/117683643)
Title: Re: Ray's latest TM1809s - red LEDs failing big time
Post by: drlucas on January 24, 2015, 01:15:03 PM
And another with just red

https://vimeo.com/117683759 (https://vimeo.com/117683759)
Title: Re: Ray's latest TM1809s - red LEDs failing big time
Post by: drlucas on January 24, 2015, 04:49:04 PM
Up close pics

Tried to move the ports and strips around to where I had a good strip (not many) and no luck. Also tried the basic test mode of the f16 and same thing. So eliminated any xlights or fpd issues. Issue is these dang strips!!!

I asked ray if he wants me to ship him back at least one to troubleshoot.
Title: Re: Ray's latest TM1809s - red LEDs failing big time
Post by: JerryPlak on January 24, 2015, 06:29:43 PM
it hard to tell from the photo but do R1, R2, R3 of dead section in value R1, R2, R3 the same in the working section ?
Title: Re: Ray's latest TM1809s - red LEDs failing big time
Post by: drlucas on January 25, 2015, 06:16:27 AM
Not following you. I think each led now has a processor, a cap, a diode and 4 resistors (one for each color and one overall). As the blue and green work I believe that the RR resistor or LED  is flawed. Unless there is a different resistor    Type/value being used for red I am more leaning towards the led itself being the issue because it's hard to believe this problem would happen otherwise. Don't understand electronics enough to say for sure though.
Title: Re: Ray's latest TM1809s - red LEDs failing big time
Post by: JonB256 on January 25, 2015, 06:50:23 AM
Have you "massaged" the components in that area? If the problem is bad solder joints, then they will frequently come on. If they do relight, the fix is to slice open the sheath and reflow the joints.

My tm1809 strips are still outside.
Title: Re: Ray's latest TM1809s - red LEDs failing big time
Post by: drlucas on January 25, 2015, 07:08:06 AM
I did have that with a couple of the LEDs back in October when I first got the new strips....and I am noticing some of the lights that are off will come back on. I'm not sure with my multimeter if I can check any levels between resistors and see if it a component failed, or if it is just the solder.

Are your 1809s the newer model that you can splice in more frequently, or the older ones? The new ones also have slightly different spacing than the previous version. Of course, you being in Texas, might mean that the problem doesn't show up as much for you with the change in temperatures.
Title: Re: Ray's latest TM1809s - red LEDs failing big time
Post by: toozie21 on January 25, 2015, 08:34:05 AM
I know we've talked while we were both in the designing phase, but how did you affix your strips?

I used WS2811 strips from DIYLEDEXPRESS and didn't have a single failure. I ran 24x7 from Nov 1 until about Jan 4 or so. I am not tooting my horn, just trying to point out that if you want to stick with strips, do it. Because I don't see why it can't be done.

Short and badly typed since written from my phone...

Title: Re: Ray's latest TM1809s - red LEDs failing big time
Post by: drlucas on January 25, 2015, 11:14:25 AM
If you zoom in on the pic I posted in reply 3 above you'll see all those red failed LEDs, well, in that picture it shows how I attached them the best. I used 1/2" PVC screwed into a 4x8 piece of plywood. From there I drilled holes on the top and bottom of the PVC about every 12" or so along the plywood. I then left about 1/8" inch gap and put the next piece of PVC on the plywood. Then the strip goes on the PVC and I used zip ties through the holes I drilled to keep them in place. I pulled the zip tie pretty snug, but not too tight - enough so I could slide the strip down or rotate it a bit so that all the LEDs would line up nicely. I thought about gluing them onto the PVC and also though about using a sheet of plastic instead of the ply and PVC...would of saved a TON of weight, but didn't find a guy who could make the sheet for me until after i bought all my PVC. At 30 cents a LED and the fact it was Ray, I had a higher expectation with the product...as for a US store vs a Chinese store, I figured most of the US stores were just reselling the stuff from China anyways. If that's not the case, then I might consider buying state side.

I know we've talked while we were both in the designing phase, but how did you affix your strips?

I used WS2811 strips from DIYLEDEXPRESS and didn't have a single failure. I ran 24x7 from Nov 1 until about Jan 4 or so. I am not tooting my horn, just trying to point out that if you want to stick with strips, do it. Because I don't see why it can't be done.

Short and badly typed since written from my phone...
Title: Re: Ray's latest TM1809s - red LEDs failing big time
Post by: drlucas on January 28, 2015, 05:17:53 PM
So...seems that Ray is not overly convinced he has any quality issues with his strips. Given this matrix worked from November - December with no issues, having so many lights fail doesn't seem like a PSU related issue...especially since I have 4 power supplies (some from DL380s and some from a IBM server) and strips on each of the different supplies all failed. It's not clean power from house to the power supplies through a UPS, so there is a chance of a spike, but before I send some things back to china, is there any testing on my own I can do?

Quote
Dear Mr.Ryan,
i have explained this to my engineer, we could not find where is the problem.
i am sure the resistor is no problem,please check if your power supply has a higher voltage(more than 12v) output,
 
because if your power supply is not stable, it would burn the red led first(as the red chip need only 2.0-2.4V voltage, the green and blue can work more than 3.5V)
that is fine to send me some pieces of the defective strip for research
 
thanks and best regards
Title: Re: Ray's latest TM1809s - red LEDs failing big time
Post by: JonB256 on January 28, 2015, 06:35:17 PM
That does bring up something I considered back in December. Since I knew from experimentation that my TM1804 bullet pixels were stable with an 11vdc input (120 pixels per strand), I did dial down the power supply output to 11.5vdc. 

I did not do the same for my TM1809 strips like yours, though I could have. For them, I just made sure the voltage was 12.0vdc with no load and verified that it didn't rise or drop when my strip matrix was full on (also 120 pixels).
Title: Re: Ray's latest TM1809s - red LEDs failing big time
Post by: drlucas on January 28, 2015, 08:01:27 PM
I'm using the supplies in this post here...

http://www.tjinguytech.com/my-projects/HP47A

It does say 12.33v, so there is a chance I'm completely responsible for this power issue on my strips. I did measure the voltage at one point at the end of the strip and it was always consistent and never over 12v.

How does one reduce the voltage easily?

I've heard of buck converters before, but never purchased one.
Title: Re: Ray's latest TM1809s - red LEDs failing big time
Post by: JonB256 on January 28, 2015, 09:23:00 PM
Well, there's the big difference. I am using the Ray Wu power supplies that are clones of the Meanwell supplies. They have an adjustment pot on the left side, making it very easy.

Computer power supplies may not have such an adjustment. If you see an adjustable resistor anywhere, that is probably what its for.

My TM1804 square pixels are powered by a PC supply and I could do nothing. Next year, I'm not using any PC supplies, just the Meanwell style.

JonB
Title: Re: Ray's latest TM1809s - red LEDs failing big time
Post by: zwiller on January 29, 2015, 11:04:02 AM
Just caught a glimpse of this in tapatalk...  I found the same thing as Jon, I found most of my stuff preferred less than 12v.  In the end, I dialed them in visually and not with meter at is was around 11v.

I cannot help but draw the same conclusion as Ray and his guy with all the failures being red.  I would think there would be a adjustment pot inside the HP PS?  I am with Jon that I am sticking with these PS.  On Amazon there are plenty of the Meanwell clones for like $20 shipped.  The few I bought did not output more than 11V but in the end, they worked better and I had to dial the strong ones down...  weird.  That said, I would think a little more voltage would not do damage like that.   
Title: Re: Ray's latest TM1809s - red LEDs failing big time
Post by: twooly on January 29, 2015, 12:02:25 PM
I'm doing away with the computer power supplies because they are a pain in the end trying to "mount" in some fashion.  This just sells that for me because honestly I never checked the voltage on them.  :-[
Title: Re: Ray's latest TM1809s - red LEDs failing big time
Post by: zwiller on January 29, 2015, 12:38:29 PM
I will tell you to have a few extra of these Meanwell clones on hand.  The fan in everyone of mine even Ray's is wonky.  Stalls/stutters/loud/what have you and it affects output.  Most times a knock or power cycle and it will be fine until you stop it again.  I leave mine all constantly...  If you like to power things down every night, you might have to get the real deal.   

http://www.amazon.com/Generic-Universal-Regulated-Switching-Transformer/dp/B00NTXRR3I/ref=sr_1_11?s=electronics&ie=UTF8&qid=1422560215&sr=1-11&keywords=12v+360w
Title: Re: Ray's latest TM1809s - red LEDs failing big time
Post by: twooly on January 29, 2015, 12:57:11 PM
I've got 3 power supplies from DIYLED and I like/have no issues with them.  I also have 2 from HolidayCoro and the fan on the HolidayCoro ones are constant/loud/annoying.  Not sure the real difference because you really wouldn't know it looking at them but something is sure different :)
Title: Re: Ray's latest TM1809s - red LEDs failing big time
Post by: zwiller on January 29, 2015, 02:14:18 PM
DIYLED just probably figured out where to lube them. ;D  I need to just crack one open and give it chance. 

Getting back on topic.  I had a bunch of issues with 2811 strips.  I got them from another member so not sure where from.  Best I could tell was that some components were not solderered well BUT "massaging" made them work.  Of course, they never lit once I got them where I wanted and had to splice  ;D  In the end, I got really good at splicing them.  That said, maybe everyone figured this out already, but I think the best way to fix is not to splice but to remove the entire strip from the sheath and fix/test/etc and then install into new sheathing and connectors.  Just buy a roll of extra sheathing. 
Title: Re: Ray's latest TM1809s - red LEDs failing big time
Post by: drlucas on January 29, 2015, 06:43:04 PM
Yes...i ran into the problem, pull the entire strip out, splice, pull back again. just it's tricky to pull through that plastic. I need to figure out if a pull string works as my strip always got stuck 1/2 way. next year matrix will be replaced with p10s. I'll see if Ray will do something for me with the failed lights, maybe toss in a few extra rolls and put a warning sign when people buy these to use his or adjustable PSU. This will save many people a lot of grief in the future. I think I'm somewhat convinced of too much good power pushing the resistors to their max. Now to see how much I can save after splicing. UGH!!!

I'm still disappointed to hear about this now, and a bit concerned that I never had a failure during the season. Maybe its a threshold thing and I was just pushing the limits.
Title: Re: Ray's latest TM1809s - red LEDs failing big time
Post by: zwiller on January 30, 2015, 05:52:44 AM
I remember reading of your frustration but couldn't think of anything at the time, I just thought you got a bad batch...

I would go back to Jerry's post.  I am getting more familiar (dangerous) with led circuitry and I would compare resistance and continuity of RR1/2/3 of a good section to a bad one unpowered.  Maybe the resistor failed and the led is fine.  But is likely the resistor AND led.  If RR is good I would be tempted to try swapping good for bad 5050s to fix for giggles.  That said, having spent an inordinate amount of time swapping rectangular modules myself, I got a bit aggressive and looked at the problem differently.  What is the fastest way to fix?  You'd be tempted to swap out each bad section for good.  Instead, try to join 2 good sections together cutting out bad sections to get one good strip.  There is some waste but a huge time saver. 
Title: Re: Ray's latest TM1809s - red LEDs failing big time
Post by: drlucas on January 30, 2015, 06:31:21 PM
Thanks -  I'll do some testing this weekend for sure....I am sending two strips back to Ray. Anyone know any tricks on Fedex or DHL to get a better price than $120 for 1/2lb package?

Title: Re: Ray's latest TM1809s - red LEDs failing big time
Post by: bcstuff on January 31, 2015, 06:56:15 AM
$120, WHAT?

I did a quick Google, I know you are in Canda, and saw rates for the US $12-24 via USPS.
http://www.quora.com/What-is-the-least-expensive-way-to-ship-to-China-from-U-S (http://www.quora.com/What-is-the-least-expensive-way-to-ship-to-China-from-U-S)
Title: Re: Ray's latest TM1809s - red LEDs failing big time
Post by: drlucas on January 31, 2015, 07:34:17 AM
Attached  is the quote for the  Canadian version of USPS.....Canada Post. UPS/Fedex/DHL/Purolator all the same ballpark.

Here is a slightly modified email from Ray......he is having some delivery issues with USPS....
Quote
could you help to send me several pieces of strip, so we can check.
my address is:
Post code:518111
company:shenzhen rita lighting limited
please do not use USPS,i have lost more than 10packages,my local post office is very bad, they never make delivery to me.
please use DHl or fedex
Title: Re: Ray's latest TM1809s - red LEDs failing big time
Post by: drlucas on February 01, 2015, 06:58:47 AM
it hard to tell from the photo but do R1, R2, R3 of dead section in value R1, R2, R3 the same in the working section ?

Well, without the strip powered up, I tested the resistance across the three resistors and they all gave readings that were extremely close to what the resistor had printed on it +/- 1 on the dial, so I think those are good. I'll power it on and do a comparison though against good and bad to make sure. I'lll look at the schematic of the LED (http://www.luckylightled.com/UploadFiles/pdf_2013110513836348311.pdf)  to see if I can quickly figure out anything else. I'm still gonna ship to Ray, but my guess is what the pdf says on page 11 happened - slight voltage change, burnout of led, but dunno really.
Title: Re: Ray's latest TM1809s - red LEDs failing big time
Post by: drlucas on February 01, 2015, 12:12:00 PM
OK - here is what I have found.....

With no power on the strip, the resistance on all 6 resistors looks to be showing correct readings.

RR1, RG1, RB1 (working red LED)
RR2, RG2, RB2 (not working red LED)

When I fire up simple test mode (on the F16)  , the resistance across all three of the Rx1 resistors change values and voltage changes values, however on the Rx2 group the voltage reading across RR2 is 0 and the resistance stays constant (the same as it was with no test sequence going).

.....so basically, there is no power going to the red led.
Title: Re: Ray's latest TM1809s - red LEDs failing big time
Post by: CaptainMurdoch on February 01, 2015, 10:17:14 PM
.....so basically, there is no power going to the red led.

Based on my understanding, I would change the word 'to' to 'through'.  I believe the chip is sinking current through the LED, so the LED should have positive voltage fed to it and then power flows through the LED, through the resistor and through the chip to ground.  Can you measure resistance through the red LED to verify conductivity?  I also think you should be able to measure positive voltage between the red resistor and the ground if the LED is good.  If the red portion is burnt out there would be no voltage at the red resistor.  you can double check one of the blue/green resistors to confirm if you see power there but not on the red resistor.
Title: Re: Ray's latest TM1809s - red LEDs failing big time
Post by: zwiller on February 02, 2015, 08:02:17 AM
That's kinda what I was thinking.  That said, thinking more about this, the led might likely be common anode and the resistors on the cathode side.  I would check for continuity across the led.  I would wager red is not.  Then I wonder if swapping the led might fix.  Doesn't look impossible but "fun"  ;D  Try this crudely for giggles and if it works (test burn in a few hours with new PS) research and find a better tool and technique.  I think the leds are ridiculously cheap. 
Title: Re: Ray's latest TM1809s - red LEDs failing big time
Post by: Jeffl on February 11, 2015, 07:57:22 PM
Curiosity got me so I bought a few of these supplies of eBay (HP DPS-600PB) to experiment with.  I'm able to mod them to turn the fan down as well as the voltage.  Mine are running about 12.45 unmodified and no load.  With a resistor, I can bring it down to 11.75 and possibly lower with more experimenting.  The question I have is; is this enough?  I can see where over 12v could be an issue, but having to run them way lower doesn't seam reasonable either.

I also have some DIYLEDExpress supplies on the way as well.  If nothing else I will be well supplied.

Title: Re: Ray's latest TM1809s - red LEDs failing big time
Post by: zwiller on February 12, 2015, 10:24:26 AM
I would burn a strip in at 11.75vdc and see how it goes.  I would think it would be fine. 

Not sure why the love for those PS.  They're used, funky sized/can't fit into a CG1500, and don't seem like bargain to me.   

You give up Ryan? 
Title: Re: Ray's latest TM1809s - red LEDs failing big time
Post by: MyKroFt on February 12, 2015, 11:08:21 AM
Not sure why the love for those PS.  They're used, funky sized/can't fit into a CG1500, and don't seem like bargain to me.   

+1

Sent from my SCH-I545 using Tapatalk

Title: Re: Ray's latest TM1809s - red LEDs failing big time
Post by: Jeffl on February 12, 2015, 12:00:39 PM
I picked up 2 for $23 shipped.  At 47 amps each they seamed like a worthy experiment until my others arrive.  I'm not saying to go out and buy any, but it has been an educational experiment for me.  I have a 4'x4' case that I plan to install them in as well as some hubs in.  You are correct that they are too large to fit into a CG1500.
Title: Re: Ray's latest TM1809s - red LEDs failing big time
Post by: zwiller on February 12, 2015, 02:44:55 PM
For fooling around that's not too bad.  Now if I was an IT tech with a bunch lying around I would put them to use for sure.  I think the jury is still out whether the PS caused Ryan's issue or not but I think lowering the voltage is a good idea. 
Title: Re: Ray's latest TM1809s - red LEDs failing big time
Post by: Jeffl on February 12, 2015, 03:50:55 PM
How long is a safe time to burn in strips or nodes before you feel like they are stable?
Title: Re: Ray's latest TM1809s - red LEDs failing big time
Post by: zwiller on February 12, 2015, 04:13:25 PM
Truthfully I've never had anything fail during burn in.  The duds present themselves immediately. That said I run them several hours a few times and that is good enough for me.
Title: Re: Ray's latest TM1809s - red LEDs failing big time
Post by: drlucas on February 12, 2015, 05:47:46 PM
I was of the same mindset....immediately.....

I ran mine inside for 2 days, then outside from October - end of December. Brought them inside, and in 2 days had the 100 or so damaged red LEDs.

As for the PSU....i had a bunch of DL380s collecting dust, looked at what the RC guys were doing with those supplies and it made sense to use them vs buying a few other supplies...of course this was also before the ATX power kit was brought to my attention. 

As for fitting them in a cg1500...my cg1500s are for the LEs only....in Canada it's hard to find people to ship them, so I only have 4 here and used marine battery cases for the DL380 supplies.
Title: Re: Ray's latest TM1809s - red LEDs failing big time
Post by: zwiller on February 13, 2015, 07:48:45 AM
That's tough...  Any leds go during the show?  Same PS you used outside was used inside?

I wouldn't think they would fit in a marine case.  I did notice there are some 1,000 watters for $20 shipped...  Until we find out what's going on I am staying away.  I'll let Jeffl be the guinea pig  ;D  So are you putting a resistor across pins 3 and 9 instead of a pot as per the webpage linked earlier? 
Title: Re: Ray's latest TM1809s - red LEDs failing big time
Post by: Jeffl on February 13, 2015, 08:36:49 AM
I bought some test leads and breadboard along with a sample package of many different resistors values.  I have been experimenting with them but so far this is what I'm doing with them.

Power On

Voltage Set

Low Fan Speed

The breadboard works good to test with.  I have been  using resistors instead of a trim pot.  The reason for this is there is a point under 9.5-10v where the power supply leaves the fan on, but shuts off power.  My trim pots I have are 10k 25 turn so I just get confused where I'm at.  Some web links say it's adjustable lower but this has not been what I have seen.  I haven't tested it at lower voltages to see how stable it is.  I have only tested it in the 11.5v-11.9v range.

As for the fan.  This can be set lower as well but from my testing, the fan does not rise when it heats up.  Some web videos show it does but not mine.  It's either high (un jumped) or low (jumped).  I tested blowing the heat gun in the fan and the only thing that happen is the fan stays running for a little while and but the power output is killed.  I think the fan then shuts off as well until the power cord is pulled and cooled..

I'll take some pictures this weekend for those that are interested.  As always,  check what I have against other resources on the web.

Title: Re: Ray's latest TM1809s - red LEDs failing big time
Post by: Jeffl on February 14, 2015, 02:21:34 PM
Here is a video of my experiment with the HP DPS-600PB B power supply running with 100 WS2811 nodes,  Falcon F16 Hub, uSC and 100' of  cat5 .  It's my first attempt at a video, so don't chew me up too bad.  Or the fact that I'm using speaker wire for power because it's all I currently have. :)

https://www.youtube.com/watch?v=MgyBsI4uq1Y