Deth's First PVC Saber Build - Custom Circuits

[Dethlore]

Why do my own circuits when they're available for purchase already? Because I'm outta money!
I'm building everything custom minus the 2014 Hasbro Mace Windu Electronic Lightsaber soundboard.

The circuitry is all a work in progress... as you can see in my subsequent posts. I welcome help where it is given! In fact... I need some! I can't figure out why the LM3914 isn't getting full voltage when the LEDs are turned on with the soundboard.

Any questions, comments, suggestions, PLEASE don't be shy! I'm new to sabers and welcome all input!

EDIT: I plan on using 3 accent LEDs from the soundboard, I do plan on adding a battery charge port, and the lm3914(not 1314) is currently hooked up to gnd instead of 7.2V (+). my bad on this picture...

Attachment 9904
Attachment 9905
Attachment 9906
Attachment 9907
SaberCircuitV1.2.jpg

EDIT: Made a simpler drawing as a general overview below. I will break out each component separately.

SaberCircuit-V1.3.jpg

Problems so far:
1. The pulsating 670nm (connected to R2) is not working when the other 670nm LEDs are connected. I suspected this was because as the 'R2' LED current was down, the other LEDs suck it all up, not allowing the 'R2' LED to conduct. I thought by adding a P-type mosfet and a 'dummy load' resistor (R1), I would be able to keep an amount of current separated from the other 2 LEDs. The diode and R5 resistor are for having a minimum on amount for the 'R2' LED, which works great without the other LEDs connected.

2. My volt meter drops quite a bit of voltage when I turn everything on. ie when everything's off, I have around 6 LEDs lit. When I turn on the circuit, it goes down to about 3 LEDs.

Does anyone see anything wrong with the overall circuit design that would cause these issues? Any suggestions?

Thanks!

[Dethlore]

Just an update for anyone interested:

I've been able to get all 3 LEDs working properly when I turn on the Hasbro Board. 1 pulses from 0~450mA? I can't measure with my cheap mmr because of the modulation from the 555 timer. It won't hold still! The other two LEDs are running at 560mA and 660mA (some tweaking to get them just right or leave it since I'd have to string up more resistors to get it just right).

Problems:
1. on freshly charged 18650s, I'm getting some leakage through the mosfets and the two constant LEDs are on at about 100mA.
2. When I turn on the circuit, the LED volt meter goes completely off.

To Do:
I am thinking of running transistors before each mosfet to give full power through the mosfets... something is being left out. I may also either run voltage dividers before the LEDs going to the mosfets, or I will try a voltage regulator so the mosfet gate voltage is well over the voltage of the LED going through the drain.

still not sure about how to keep the volt meter on when everything else is on... if anyone has ideas, please let me know.

Here's a teaser of the messy prototyping!

Attachment 10048
Attachment 10049

[Dethlore]

Using common transistors, I was able to feed the pulsing LED with more juice, and re-added the low-on resistor, so there's more light output without the pulse being too annoying. Constant LEDs are still running around 500 and 600mA.

Fixed:
Leakage issue. LEDs are all off if the soundboard is off, and the volt meter is on.

Problems:
*REDACTED*
EDIT: I'm retarded... the fact that batteries act very differently under load than a desktop supply does slipped my mind. Circuit is good.
I will be changing the circuit to have a P-transistor so the meter works when the saber is off, and an N-transistor with a different resistor in series to the LM3914 high sensor... thus giving it an approximate correct power level reading.

WOOT! I feel smart... and dumb at the same time.

Here's a short video of the circuit in action at the moment.
https://www.youtube.com/watch?v=IqsZ7Ho7qIA

[Dethlore]

As far as experience goes, I'm really just a beginner that loves playing with it. I've built a few laser drivers and a few basic circuits for nightlights and fan controllers and such... This is probably the most complex thing I've done. I'd LOVE to get my hands on a nano and play around with that! Here's the latest drawing of my circuit.. it has changed slightly, but for the most part... it gives the general idea...
and yes, I do know that I drew the mosfet symbols wrong. haha

[Dethlore]

...it looks like you are using some transistors to drive the gate of the FETs kind of like what's called a "Darlington pair". I have not read the spec sheet for your FETs that you are using, but this might not be necessary. You might be able to simplify your circuit and remove those and direct-drive your gate pins. Again, I've never used a 555 timer or some of the other components you are using, so I could be wrong, but typically drive my N-FET gates right from a microcontroller output logic pin.

I can draw a better diagram if you want... but first let me explain the reasoning on this last diagram since it is mostly the same layout.
The "Switch" on the diagram is actually the soundboard. My logic coming out of that is the power to the speaker (I don't want the flashing lights on power up/down from the LEDs on the soundboard). That logic is at about 4.3V according to my MMR.
Originally, I used IRF610 N Channel Mosfets because that's what I had on hand. Those required the Vds to be 1.5V lower than Vgs, so I direct the logic signal to the transistor to allow 7.2V into the FET Gate. The 2 LEDs each go to their own FET drain, and source goes to 4.75V regulated ground. They used to run through a diode to lower voltage to 4V so I could use correct resistor values I had on hand.
The 555 timer (seems to like more voltage rather than less, so it is connected directly to full voltage) has a PWM signal out for LEDs, but has a max of 200mA, so I run that to a transistor to allow full voltage to FET which has LED to drain and source to a second FET drain which has source to 4.75V regulated ground. This second FET has the soundboard connected to a transistor and then to the FET's gate for full saturation.

I removed the diodes from the LEDs which has given me about 400mA each instead of the previous 300mA. If I connect one to full voltage ground, instead of 4.75, it gets about 600mA, but I will need to rewire the circuit slightly to make sure it only turns on when I want it on...

I would LIKE to have every LED run at 700mA.

What are your thoughts?

EDIT: Added new diagram
circuit2014June10-02.jpg

[JakeSoft]

That's very interesting using the speaker output to control the FETs. I've never seen that done before. I'm surprised it works. I had always thought that the speaker signal would be oscillating rapidly.

You might find faster help at the Arduino forums. I go there a lot when I'm deep into a project and frequently get fast answers. They have an entire subsection for general electronics questions and circuit analysis stuff. While there are plenty of smart folks who frequent these TCSS forums, the Arduino forums are way bigger and more geared toward what you are trying to accomplish. Experience has taught me that you don't actually have to have an Arduino in your project to post questions about it there; they are cool like that.

[Dethlore]

That's very interesting using the speaker output to control the FETs. I've never seen that done before. I'm surprised it works. I had always thought that the speaker signal would be oscillating rapidly.

You might find faster help at the Arduino forums. I go there a lot when I'm deep into a project and frequently get fast answers. They have an entire subsection for general electronics questions and circuit analysis stuff. While there are plenty of smart folks who frequent these TCSS forums, the Arduino forums are way bigger and more geared toward what you are trying to accomplish. Experience has taught me that you don't actually have to have an Arduino in your project to post questions about it there; they are cool like that.

I thought maybe the oscillation would be the reason I can only pull 450mA max per LED through the FET, so I changes it so the LED outputs go to a P FET gate and the leds going through there... nope, still 450 ish mA.

I am going to try setting up a second volt/current regulator for 6V/700mA and see if I can run the two static LEDs in series at that power.

Just a couple other things to try and then I'll check out the Arduino forums. Thanks for the advice!

[Dethlore]

Just a quick update... I'm am nearing completion on this circuit. I now have the LM3914 volt meter showing proper levels all the time, 2 LEDs on near full power at 600mA and 1 LED flickering on the 555 timer. I MIGHT add a pot on there so I can manually control how fast it pulses... we'll see.
I'll add some circuit shots soon, but here's a couple shots of the blade I just got THANKS TIM! with a 1000lm cree XML2 in it (no cellophane inside).

bladeholdermod01.jpg

blade-test.jpg

[Dethlore]

Updates:
I didn't like the thinner section in the middle, so I've added some pieces around it for character and width. I also cut a slit between two of those pieces where I will show some wiring going up the hilt... not sure if that will look good or not. worst case scenario is I can put some putt over it and cover it up.

I also added in the main power switch and momentary soundboard switch. Thanks again for the yellow ring switch Tim, you rock!

I also took out the 3mm LEDs from my prototype as I work through the prototype and solder things together. I've replaced them with 0803 SMD LEDs. much smaller, but looks fantastic.

meter3mm.jpg
meterSMD.jpg
PVC-LS_19.jpg
PVC-LS_20.jpg
redblade.jpg

[Dethlore]

Put the blade, heatsink, and LEDs IN the hilt for the first time together! circuitry is still outside... but this is my first saber... THIS is EXCITING! I had to share some shots.

Blade: 1" x 34" thick wall trans white with round trans white end. Cellophane wrap from Walmart ($3) wrapped around maybe 7 or 8 times?
LEDs: 3-up rebel Deep Red (670nm). 2 at 650mA, 1 flicker-fading from 450mA to 600mA.

redblade00.jpg
redblade01.jpg
redblade02.jpg

Pardon the mess on the desk - my projects usually fill it up with drawings and diagrams and bits of wire and other components and tools and etc etc