Super Standard Strobe String. (SSSS)

mrpackethead

Full time elf
Joined
Jun 12, 2010
Messages
452
Location
Wellington
Folks, a new idea for some blinky kit got legs last night, made me spend a few too many hours awake..

The Super Standard Strobe String. ( ok, so it needs a better name ).

The Strobe String will be a string of high powered white ( or optionally blue or any other colour ) (500-1000mW) Leds, on a string. Spacing between "strobes" to be determined, but could be as much as 1meter, and the number of strobes on the string as many as 50-100. ( thats to be determined ). THe strobe string is a special use case of the WS2801 controller chip ( which we have all seen in use for RGB strings )..

All three outputs of the WS2801 are put in parrallel, tripling the current available to drive a LED. This on paper is 150mA. However the overriding factor in the WS2801 is more to do with how much heat it can tolerate before it cooks. Because strobes have a low duty cycle ( the percentage of time that they are actually on ) in the order of a maximum of 5-10%, we can safely "overdrive" momentarily.

Last night i hacked an RGB pixel, changed its operating parameters, connected a 1W pixel to it, and using our E16 controller, sent a specially constructed sequence to it to make it behave like a strobe. The good news is that some 7 hours later, the 1W LED is still flashing twice a second at 350mA without issue. Its nice and bright, and the chip is not getting hot. In fact for comparison i've set a WS2801 running an RGB led at FFFFFF ( white ), and its surface temperture is higher than the one running the strobe. Why? because its average power loss is significantly higher than the "bright strobe". I'm going to run this test for the rest of the week to see what happens, and see if it remains stable.. I'm optimistically confident it will however.

Now, thats as far as i've got. Before you all jump off and ask the "but" questions, theres some things you need to know.

(1) While this technically could be plugged into a "standard" SPI type controller ( such as the E16, TP3244 etc ), for anything other than very carefully constructed sequence material you would risk an accidental melt down of the pixels.. In the worst case situation, if you set all the strobes to be on 100% you'd cook everything pretty quickly.. the ws2801's would overheat, you'd need a huge amount of current to drive it all ( read large large power supply and BIG cable ).. This is not a situation that anyone could really tolerate..

(2) Becuase of (1) above, a Super Standard Strobe String will need to be connected to a controller that is configured to deal with the limits of the string, so it does'nt cook it. The nature of "strobes" makes it pretty much sensible for the controller to create the "random effect" of the strobe, so it looks like strobing rather than have to create this on litternally hundreds of channels in your sequencing.. A controller for the SSSS will therefor probalby just require 3 or 4 dmx channels to set the parameters of the string.. This is just a mind map right now, but i suggest that the channels will control;

rate: the average time between flashes on a given strobe
intensity: the intensity of the flash from the strobe ( yes these are dimmable! )
fill: the average number of strobes that can be on at a given tome.

Some good news; As far as our E16 controller goes, all that is required is a firmware change to cope with this. I can't speak for Tabor and Phils TP3244, but i'd suggest that they too could make the firmware changes to support the SSSS.

Next steps:

(a) I'm going to creat a few more "hack" strobes and wire them together, and run them for a week.
(b) shoudl (a) prove to be ok, i'll get some test strings manufactured for me in our china factory
(c) distribute a few test strings to the developers about the place, and get all the nessesary firmware changes made to kit as need be.
(d) work out costing, and do a "once only ACL special Co-op" for those than want so. After that you can buy my commerical ones.. :mad: or someone else can run a co-op

the stobes would come in an expoxy encapuslated body, like our RGB pixels, ready to be plugged in used... Comple.

Price.. Let me figure it out, but its going to be good.
 

fasteddy

I have C.L.A.P
Global moderator
Joined
Apr 26, 2010
Messages
6,648
Location
Albion Park NSW
Excellent work MPH, another great innovation for the community and another thing i have to budget for :-\
 

chrisl1976

Apprentice elf
Joined
Jun 4, 2010
Messages
98
Location
Kankakee, Illinois
Sounds like another HQ product.
thumbsup.gif


Look forward to checking it out.
 

SmartAlecLights

Im a SmartAlec what can i say!
Community project designer
Joined
May 4, 2010
Messages
1,533
Location
Murray Bridge, S.A.
Cant wait to get my hands on these babies
let us know on the price so we can all go nuts buying up
ps. great idea/job
 

TimW

Full time elf
Joined
Jun 15, 2010
Messages
168
Location
Melbourne
Nice innovation MPH. V clever as usual!

"SparkleStrings"?... or "SparkStrings"? given the effects. (After all, 'Strobes' are so 2010! :) )

What about another channel for 'direction' - given the strobes are on a string it would be posible to get some directional effects such as individual runs or flowing 'flashy areas' down a string.

A one off coop is really interesting in terms of balancing community development & input (to the extent it occurs) with commercial objectives. Sounds interesting. Has anyone considered this as a develoment model before?
 

mrpackethead

Full time elf
Joined
Jun 12, 2010
Messages
452
Location
Wellington
SSSS becomes SS

TimW said:
Nice innovation MPH. V clever as usual!

"SparkleStrings"?... or "SparkStrings"? given the effects. (After all, 'Strobes' are so 2010! :) )

What about another channel for 'direction' - given the strobes are on a string it would be posible to get some directional effects such as individual runs or flowing 'flashy areas' down a string.

A one off coop is really interesting in terms of balancing community development & input (to the extent it occurs) with commercial objectives. Sounds interesting. Has anyone considered this as a develoment model before?

Thanks Tim, I liked "SparkString", and with a tiny mod I'm going to get call this the "Sparky String" So, we are back to the SS..
 

David_AVD

Grandpa Elf
Community project designer
Generous elf
Joined
Jun 12, 2010
Messages
4,681
Location
Victoria Point (Brisbane)
What happens if the data flow stops to the pixels and some are left on? Won't they heat up and die?

Maybe you can incorporate some sort of watchdog into the controller to detect this and prevent damage.
 

mrpackethead

Full time elf
Joined
Jun 12, 2010
Messages
452
Location
Wellington
Sparky String Schematic ( SSS )

Just for the technically inclined, here is a schematic of the guts of what i'm testing..

Current in each channel is determined by the Current sense resistor..

So far is working just fine, and no magic smoke.
 

Attachments

  • spark_string_schematic.png
    spark_string_schematic.png
    6.8 KB · Views: 115

David_AVD

Grandpa Elf
Community project designer
Generous elf
Joined
Jun 12, 2010
Messages
4,681
Location
Victoria Point (Brisbane)
Would be worth testing one on full power for 100% duty cycle. Hopefully the chip will just shut down the outputs if it gets too hot.
 

TimW

Full time elf
Joined
Jun 15, 2010
Messages
168
Location
Melbourne
For strobe operation, I wonder if you could do away with the current limiting resistors all together... just setup the 2801 in 'constant voltage' mode and rely on the duty cycle to protect everything? The led doesn't have to be on for very long to strobe... so if the resistors don't save the chip in case of failure are they necessary? Might be a risk of overdriving the Led .. but wouldn't the same duty cycle argument apply?
 

mrpackethead

Full time elf
Joined
Jun 12, 2010
Messages
452
Location
Wellington
TimW said:
For strobe operation, I wonder if you could do away with the current limiting resistors all together... just setup the 2801 in 'constant voltage' mode and rely on the duty cycle to protect everything? The led doesn't have to be on for very long to strobe... so if the resistors don't save the chip in case of failure are they necessary? Might be a risk of overdriving the Led .. but wouldn't the same duty cycle argument apply?

I'd had that thought. however i dont' think we can turn on/off the 2801 quick enough to get the control that we would need to pull it off.... its one control to turn on, one to turn off.. also, then we woudl have the same old problem with all constant voltage circuits on strings, differnet nodes on it will be at varying intensity..
 

TimW

Full time elf
Joined
Jun 15, 2010
Messages
168
Location
Melbourne
mrpackethead said:
also, then we woudl have the same old problem with all constant voltage circuits on strings, differnet nodes on it will be at varying intensity..

Yes.. the irony of my question was not lost on me in that regard given recent discussions!

And I guess the inclusion of 3 smd resistors is hardly cost prohibitive...
 

ryanschristmaslights

Senior elf
Administrator
Joined
Jun 30, 2010
Messages
951
Location
Adelaide, Australia
This sounds great - a completed strobe string is just what we need.

So from somebody who has no DMX gear other than a single USB dongle and an XLR to RJ45 adapter, what other elements do I need to get these up and running?

Currently running only LOR boards using the LOR protocol however in 2012 I'd have to start using DMX - either LOR adds DMX output from the software by the time I need to pay for a licence upgrade or I'll have to switch software.

Any ETA on this.. 2011? 2012?
 
Top