Putting strobes together

riri7707

Full time elf
Joined
May 9, 2010
Messages
222
Location
Paris / France
Hi,
Thanks to Tabor, i compiled 2 new variations of the strobes....
Try them. Works fine for me.
So actually i have three different led randoms files including the original one.
Let me know...

Cheers... Henri
 

Attachments

  • 3_15.hex
    873 bytes · Views: 29
  • 6_45.hex
    873 bytes · Views: 24

Tabor

Full time elf
Joined
May 16, 2010
Messages
249
Location
Perth, WA
Hello Henri

The strobes were designed to be left powered at all times during display. Then when required the trigger line can be pulled low.

You can tie multiple trigger lines together. so for all 160 strobe they can all be controlled by one channel, provided the channels can sink the required current, which would be very small anyway.
 

riri7707

Full time elf
Joined
May 9, 2010
Messages
222
Location
Paris / France
Thanks for this precious information.
I will do it like that... 3 wires and one channel.
Makes really sense..
i just need to cook the rest of PCBs now by soldering the PIC.

I'm planning to manage a variation of the original desing for economical reasons (i have270 modules to build).
I got the PICs yet but not the rest of components and the drive (additional $270 to add and the power leds $135 more) :-[
As the flash is not too bad with a white standard LED directly plugged on PIC with a resistor to manage 5 volts,
do you see any incovenient to work like this? I do not need power white led 500mw or 1w.
So means only the single PIC, white led 3v 20ma and and single resistor....

Cheers... Henri
 

rkhanso

New elf
Joined
Jun 25, 2010
Messages
28
Tabor said:
Hello Henri

The strobes were designed to be left powered at all times during display. Then when required the trigger line can be pulled low.

You can tie multiple trigger lines together. so for all 160 strobe they can all be controlled by one channel, provided the channels can sink the required current, which would be very small anyway.
I am also thinking this may end up working for me, with a little more thought and design.

I planned on using the strobes with a Renard 24LV, which we all know is a positive trigger, and later realized these strobes require a low trigger. Since I already have to drop the 24 volts that I'm running through the 24LV down to 5 volts to power the strobes, I'm sure it's not that much more to change the positive output of the 24LV board to a negative output. Shouldn't I be able to do this with just a logic IC, like a NAND gate or something - where I can take a high output from the 24LV and change it to a low output to connect to the pin #4 on the strobe PCB? If not a NAND gate, any other ways to do this simply? I would rather use through-hole parts since I don't want to take the time to etch boards. I only need to make 6 of these since I have just 6 free channels on the 24LV controller to operate these strobes, thinking that a simple prototype board would suffice.

If not a logic chip, like a NAND gate (thinking that a low output from the 24lv board to the NAND gate would give a high output, thus NOT triggering the strobe board. When the 24LV board goes high, the difference in the signal would drop the output on the NAND? I'm going off a rusty 25 year old electronics education here, so this may be completely insane!!) - if not the logic chip, can I futz with the high/low trigger difference between the 24LV and the strobe board in a different, simple manner?

Unless, I can just use the channels backwards in Vixen. Say, I DON'T want the strobes firing, so I keep the output high in vixen, which would not trigger the strobes. Then, when I want the strobes to start flashing, I can turn the channel OFF in Vixen. Could it be that simple? I could power the strobe boards directly off my 24v power supply with a 7805 regulator, keeping the load off the 24LV board. Just need to make 6 prototype boards for running groups of 25 or so strobes - keeping the current requirements down enough where an LM7805 may work...and then tying the low trigger outputs together for each group/channel of strobes so they are turned on when Vixen shuts the channel off.
 

David_AVD

Grandpa Elf
Community project designer
Generous elf
Joined
Jun 12, 2010
Messages
4,681
Location
Victoria Point (Brisbane)
You could use an opto-coupler (4N25, 4N33 or similar) to convert the incoming 24V trigger from the Renard to an "open collector" output.

Use a 2.2K resistor in series with the opto LED. The emitter (pin 4) of the opto will go to the strobe ground, while the collector (pin 5) of the opto will go to the strobe trigger input.
 

David_AVD

Grandpa Elf
Community project designer
Generous elf
Joined
Jun 12, 2010
Messages
4,681
Location
Victoria Point (Brisbane)
The base (pin 6) is left not connected.

MGalleryItem.php
 

rkhanso

New elf
Joined
Jun 25, 2010
Messages
28
David_AVD,
Thanks for that info. I was looking at the data sheet for the 4N33. It can handle the 24v input? I will output the 0v and 5v needed for the strobe board, right?

Sorry for the crude drawing - I'm not that familiar with the software to make this more legible. But, is this what you're thinking may work?

schematic.jpg
 

David_AVD

Grandpa Elf
Community project designer
Generous elf
Joined
Jun 12, 2010
Messages
4,681
Location
Victoria Point (Brisbane)
Yes, it will work with 24Vdc. The 2K2 resistor limits the LED current to a safe level.

Your diagram appears to be correct, assuming the "channel 1" wires are actually two wires.
 

budude

Way behind schedule - again...
Joined
May 13, 2010
Messages
127
Location
San Jose, CA
The Ren24LV is not positive trigger - it is a switched ground like most of the DIY DC controllers. You may be confused in that the FW settings are such that the PIC outputs *are* reversed for a "positive when ON", but that is because the ULN2803 driver inverts the output setting back to a normal switched ground. I don't think you would need the optos - just power the boards separately using the common power supply/regulators and run the output of the LV direct to the trigger.

Keep in mind that you won't be able to drive more than 6-7 strobes per 7805 regulator. At 24v and 600-700mA (if all on), the 7805 will be crying for it's life from the heat produced. I know the strobe effect will obviously help but I would definitely experiment a bit before going too far with this.
 

David_AVD

Grandpa Elf
Community project designer
Generous elf
Joined
Jun 12, 2010
Messages
4,681
Location
Victoria Point (Brisbane)
I did think all of the DC controllers were switched negative, but didn't bother correcting the O/P for fear of confusion.

The opto is still a good idea IMO, as it removes the possibility of leakage currents causing false triggering. It also helps keep the ground return currents in their place once longer cables and larger current draws come into play.
 

budude

Way behind schedule - again...
Joined
May 13, 2010
Messages
127
Location
San Jose, CA
riri7707 said:
Hi,
Thanks to Tabor, i compiled 2 new variations of the strobes....
Try them. Works fine for me.
So actually i have three different led randoms files including the original one.
Let me know...

Cheers... Henri

Maybe I missed it - what exactly do you change to get the multiple random outputs? I configured MPLAB to build the standard file OK - just need to know which bits to play with. I see the seed value but not sure what makes for good values.

Thanks!
 

riri7707

Full time elf
Joined
May 9, 2010
Messages
222
Location
Paris / France
Hi Brian,

As Tabor said :

There are two variables in the C file which can be manipulated in order to get a different strobe effects.

Code: [Select]
//the following two values are used to fine tune the effectsunsigned char longdelaytune = 4; //tune the random blink rateunsigned char ledontime = 30; //LED on time in milliseconds

So, you can change these values to customize your radom strobes and burn your pics with your own files.

For those who wants to make cheap strobes, makes sense : only pcb, 1xPIC and 1x resistor and 2 wires; that's it.
When you power the line, all strobes run in random mode immediately ( as you programmed with different hex files); i built 7 different effects shared over the modules frame ...
But the LED needs to be standard one here : 3v, 20ma.... ( if you go to basic cheap solution)

These is not exactly how the board was made for, but this is also another option too.

Personally, i use Blue LED and Star Covers, to balance colors between my falling snow lights who yet white.
I'll put a picture or a video asap.

Henri
 

budude

Way behind schedule - again...
Joined
May 13, 2010
Messages
127
Location
San Jose, CA
Ok - well - I just played with the seed value and got the results I needed. My setup will be 2-wire only as 3-wire is a non-starter for me. I really sensed no delay between hitting the trigger to ground or simply tying it solid to ground and turning power on/off. The PIC starts up faster than my eyes can tell so I will stick with 2-wire. I'll be bundling mine using Cat5 stripped down 2-3 feet from the end with a strobe on each end or 4 strobes per cable. Initially, each 2-wire interface will go to a single channel but if I complete enough, I will make some type of board/interface so that multiple strobes are on the same channel.

I used 105/115/125/135 for the seed values and created four different hex files and ended up with four unique strobes for each of the ends. I can always offset the start times on different strobes as well so I'm pretty sure I can get the effect I want. Anyway - I ran the group of four together and they are totally out of sync (which is good!).

It appears that 12mm heat shrink might work for these - it should be a very tight fit on the covers when done. I only had a short strip of it available so can only make four of them but if it works, I'll get more of it from eBay or something.
 

David_AVD

Grandpa Elf
Community project designer
Generous elf
Joined
Jun 12, 2010
Messages
4,681
Location
Victoria Point (Brisbane)
The reason for powering the strobes permanently and using the control line may be (I haven't checked the code carefully) to get the internal counter used in the random routine running so they are out of sync enough when you do trigger them.

If that's not how they work now, I sure the code could be modified to run the "random timer" constantly and use the trigger input to gate the output (enable it when required).
 
Top