Diodes in Series misbehaving - need electrical help!

[dethlore]

Ok, so I am pretty sure I've nailed down a couple of reasons I was having issues... biggest one is the old and tired batteries I was using. The Power Supply is definitely what I should have been using to prototype this to get the right values. haha What was I thinking?

Now, I have a couple of questions regarding transistors and mosfets.
I am using a 2N3904 transistor as a switch that receives a 4.7V signal to turn on. I have the battery source positive (6-8.4V) go to the collector, then from the emitter to a mosfet gate (using a fairchild semi HUF76423P3). Battery positive again goes to diode pos, and diode neg goes 1 ohm resistor to mosfet's drain. mosfet's source goes to battery negative.

Problem here is that I would expect full battery capability going through the mosfet (minus a small forward voltage), but I'm getting a LOT less. I thought this could supply a couple amps worth of current, but I can only get up to 500mA draw.

Behavior of transistors and mosfets is really new to me - I've learned a ton, but I know there's much more to learn. Any ideas or light you could shed on this situation for me?

I appreciate it!

 

[The Lightning Stalker]

Is this for the laser driver or something else? I don't know what your'e trying to do with a diode on the gate
of a MOSFET. A schematic would help.

 

[dethlore]

Is this for the laser driver or something else? I don't know what your'e trying to do with a diode on the gate
of a MOSFET. A schematic would help.

this is for a high powered led driver. The diode will go to the mosfet drain. I attached a schem just now...

Now that I've worked out the details of my regulator for the LED, I haven't put in on this circuit yet since it doesn't seem like the fet is open enough to power the LT1185 (with even more voltage drop expected) when it can't even power one diode beyond 500mA.

PS Lightning Stalker, on the LT1185, I connected pin 2 to (+), and nothing to pin 4. That allows the regulator to work as a constant current variable voltage instead of being limited like it was before. Thanks for reminding me about that.

 

[The Lightning Stalker]

Well, if you don't connect anything to pin 4, you're not going to get anything since pin 4 is the output.

Now the FET needs a pull-down resistor between the gate and (-) or else it will come on and stay on.
Then the base of the bipolar needs another resistor to limit current from your 4.5V signal since a bipolar
transistor is like a diode and it needs a resistor. 10k should be about right.

 

[dethlore]

Well, if you don't connect anything to pin 4, you're not going to get anything since pin 4 is the output.

hahaha well... I meant I took off that resistor from pin 2 to 4.

Now the FET needs a pull-down resistor between the gate and (-) or else it will come on and stay on.
Then the base of the bipolar needs another resistor to limit current from your 4.5V signal since a bipolar
transistor is like a diode and it needs a resistor. 10k should be about right.

Yes, I forgot to draw in the pull down resistor on the fet. That IS there. I'm using a 10k ohm resistor.
Ok, so the transistor needs a 10k resistor in series to the base? And just so I understand... that resistor is to limit current as you say - and that is because using this as a switch, I need minimal current, else it will draw as much current as it can (being a diode) making it horribly inefficient. Correct?

EDIT: And I just read a bit more about the base resistors on transistors... They're to keep the magic smoke inside.

 

[The Lightning Stalker]

Yes, that is all correct. I hope it's up and running now.

 

[dethlore]

Added the resistors last night and immediately got a bump in power, not as much as I expected though.

Powering two diodes in parallel to drain of the same mosfet yields 450mA still, source to a 4.75 regulated power supply (measures 4V under load). If I remove the 1 ohm resistor from series with the diode, I get 500mA.

If I connect mosfet source to battery, I get 600mA each diode, or 800mA using only 1 diode, but there's no regulation or protection here...

I worry about relying solely on the voltage regulator, or no regulator at all. These need to be current regulated.

Connecting mosfet source to 700mA regulator with diodes in series, I get 450mA each. So I know the problem is there isn't enough voltage available for the driver to run 700mA to diodes with a 7V drop between the two of them.

The best result is practically a direct drive for 600mA each diode... I have seen some examples of direct driving, but I just don't feel right about that when it seems like the batteries should be supplying more power.

What do you think?

 

[dethlore]

It's not perfect, but I think it's satisfactory. End result, I've got 600mA to each diode.
I am running them in parallel with a current driver set to ~1400mA (I know I know, if one dies, the other one will be shortly after), but it's apparently still lacking the voltage I quite need. I contacted the LED supplier and asked if I could get a replacement since these are so far over their binned voltage. We'll see how that goes.