some help with wiring plz:10w ledengin(green)cf4.1 +clash flash-shimmer!

[Eandori]

Ok, I read through one of the online specs for a buckpuck and I have a better understanding of the CTRL and REF pins.

1. The REF pin is a 5v output meant for powering a small uProcessor or digital logic. It can also be used to power a simple voltage divider circuit so you can vary the voltage into the CTRL pin from 0-5v.

2. The CTRL pin is a logic input meant to be set between 0 and 5v that will control the output driving % of the buckpuck. As the control voltage gets higher, the output current of the buckpuck drops. There is even a graph on the datasheet for the buckpuck showing that.

=========== Now having said that ===========

The last circuit diagram you showed has the REF output pin tied through a resistor to the CTRL input pin. That will permanently set the current output of your buckpuck to some value less then 100%. I seriously doubt that's what you were going for.

In short, how you have the buckpuck wired up in that diagram (ignoring the power extenders...) you have all 4 LED's set to be on all the time. 2 through CF, and the other 2 (meant for clash/shimmer on at some static less then 100% value)

=============================

Now I need to read up on Erv's spec for the power extender to see exactly how they are wired and meant to be used.

[8lade]

Thanks again Eandori,
I really appreciate your help!
As for the battery you are completely right, but nevertheless that is the least of my problems
I might try and get a higher voltage pack or figure out a solution for that later
cheers

PS:about the buckpuck being on all the time...yes that is my problem with wirring, for it is working fine when the cf is on clashing and shimmering but as I said before and you have noticed too,it stays fully lit when the cf is off and the ramp up/down works the opposite way it should work! :-/

[grayven]

I just got done with a clash flash using the CF4. I used an RBW RB in series for the main blade (purple) and W with an power extender for the flash. I set Xflash to 15. This worked extremely well, you shall see a video of this soon. I also used the Low batt LED but set the lbat at 650. Basically I got it on whenever the saber is running. It also has a nice fast blink effect to it. My power source it two Trustfires 14500's. Running the "Ancient saber" Sound and config from NovaStar I get a really nice dark purple, the rest give me a really nice bright purple with the LED setting at 45.

[Mad Hatter]

Actually, the two power extenders work great for this... Erv's design, not mine. I had to ask him how to vary the puck output with the shimmer voltage. (It's weird because the open circuit on the puck's REF CTL makes it turn ON instead of OFF.)

The resistor on the buckpuck ctrl line is to keep it from shorting to ground when both extenders are open.

The power extender on the shimmer uses the trimmer to tune the voltage used to open the fet. Resistance too high = the fet never opens. Too low = the fet opens all the way. Just right = the fet opens partially and allows the flash to shimmer.

The second power extender is triggered by the clash flash. So, without the clash signal, there is no path to ground and the CTL lead is getting (almost) the full 5v from the REF. i.e. the buckpuck is OFF. With the clash signal, the second extender opens, and the path from REF to ground is controlled by the voltage level on the shimmer output. When the blade is brightest, the first fet opens all the way, the CTL lead has ground potential, and the puck is all the way on. As the voltage drops, the resistance goes up on extender 1 and the puck dims.

As for voltage requirements, Erv measured the green at about 7.6v for 2 in series. (The one we buy isn't the one in the tech spec with the super high lumen and voltage, it's more towards the middle.) I have wired the 10W green and it is very bright, both normal and with the flash, but obviously the brightness starts to dip after 10-15 minutes.

I normally use the red, which averages < 6v and I get a long time before it starts to get dim. (maybe 45min to 1hr, but I've never timed it.)

Also, note my comment above ^^ that the main and led ground wires on the puck and the ground wire on the second extender have to share a common ground. It won't control right if they don't.

[Eandori]

Can you post your circuit diagram Mad Hatter? Because If it's working I'm sure it's different then the diagram 8lade has. That one can't work.

EDIT: Ok I spent some time looking over the Power Extender and I think I understand more how it works. I figured Erv' was using a high power transistor... but it does not seem to be the case. Rather he pipes the + rail straight through and is doing some kind of switching with the ground rail. So based on low level input, the power extended opens up a path to ground or not.

So... looking over that diagram, it seems that in the absense of clash/shimmer the REF pin is pulling the CTRL pin up to 5v and shutting off the 2 Clash/Shimmer LED's. But when a Clash happens, the bottom power extender is being used as a switch to connect ground to the CTRL pin, pulling it to 0v and turning on the buckpuck. The shimmer power extender will be pulsing the connection to ground as well. Unfortunately, it seems that both power extenders are hooked up in series so they both need to be activated for the buckpuck to fun at full power. Which is probably why you only see a solid value from them. I think they need to be connected in some form which is not in series to work right.

Also, it appears to me that the main reason those power extenders are needed is polarity. I think the clash and shimmer outputs are active high, while the CTRL pin on the buckpuck is active low. So, the polarity needs to be reversed. I think that's why they were even added in the first place. Not that you need something that can power a big circuit, but to change from active high to active low.

More to come as I keep learning about all the components in your setup.

[Mad Hatter]

He has the buckpuck/power extender part of the circuit right in the diagram here, other than the two - leads on the buckpuck aren't sharing a common ground... and he doesn't show where he's connected the Board - lead on the second power extender.

[8lade]

Madhatter can you download my diagram and connect the buckpuck - leads and the unconnected power extender lead to the right places?that would be a great help!
Thanks both of you for all your help!

[Eandori]

Hey Mad Hatter, your response above my last one is great and I'm reading it closely now. I must have missed it on my last reply. You seem to have a good handle on what Erv is doing with this circuit and I'm catching up to that now. Main issue I still see is that both power extenders need to be somewhat on to have a path to ground. As in, you won't see shimmer unless there's an active clash.

Yeah, even at 7.6v LED that's too high for the battery he's using. The rule of thumb for voltage/current regulators is.... AT LEAST 1v above the voltage of the unit you are driving is needed. So a 7.6v LED should have a battery rated at 8.4v or somewhere around there.

So... I fully believe you that it only stays bright for 10-15 min. And it's also possible that for those 10-15min it's still underdriving the LED!

[Mad Hatter]

Madhatter can you download my diagram and connect the buckpuck - leads and the unconnected power extender lead to the right places?that would be a great help!

Look at your diagram.. You have two black leads coming off the top of the buckpuck, and one black lead ("Board -") on the bottom power extender. All three of those need to have a connection to the Battery -

Main issue I still see is that both power extenders need to be somewhat on to have a path to ground. As in, you won't see shimmer unless there's an active clash.

That's correct. (Though the CF output shimmers normally) If you want the full 10W all the time, you drop the second power extender from the circuit and go straight to ground. ... and install a bigger battery. ....and a cooling fan.

Yeah, even at 7.6v LED that's too high for the battery he's using. The rule of thumb for voltage/current regulators is.... AT LEAST 1v above the voltage of the unit you are driving is needed. So a 7.6v LED should have a battery rated at 8.4v or somewhere around there.

So... I fully believe you that it only stays bright for 10-15 min. And it's also possible that for those 10-15min it's still underdriving the LED!

That's the main reason I stuck with the Red.

[Eandori]

Agreed, Red is MUCH easier for this kind of setup.

That's correct. (Though the CF output shimmers normally) If you want the full 10W all the time, you drop the second power extender from the circuit and go straight to ground. ... and install a bigger battery. ....and a cooling fan.

Actually I wasn't shooting for all 4 dyes running at full power all the time, I was shooting for only the main 2 dyes running on hum... on lockup the extra 2 dyes pump out some shimmer, and on clash you have a very quick pulse of full power on all dyes.

I think that can be achieved... but the two power extenders need to be placed side by side. Or... perhaps it's time to bring in NOR gate for the shimmer and clash outputs... both high for assertion with an output that is OR'd together and low for assertion... that might be a really simple fix...