2 Amp+ driver circuit needed.

[Hemlock_Mike]

Benm --

You touched on what I'm looking for. The Mosfet bank is OK, it's just one component on the sense/driver board which is bad. I'm looking for similar circuits to help identify my components.. Right now, it looks like a smt device marked 431P which may be toast and I don't know what it is. The 8 pin IC is likely OK because it is somewhat isolated in the circuit.

Mike

 

[GooeyGus]

[quote author=Hemlock Mike link=1240535024/0#12 date=1240712054]WOW -- Now I have to digest this info provided ......  The space for the driver is limited but I'm looking at that 350.  
My present supplied circuit is a small board driving 3 Mosfets on another board.  

I doubt the LM350 is the best approach here. It works exactly like the 317, but at these current levels you will burn a lot of power in both the lm350 and the resistor. To operate within spec, it needs 3 volts between input and output, and another 1.2 between output and adjust, resulting in a total of over 8 watts electrical use by the driver alone (!).

I would recommend using a battery/supply voltage closer to the LD voltage, and a circuit like this:
http://www.instructables.com/files/deriv/FTF/KKU6/Z2HEWIFNTP5/FTFKKU6Z2HEWIFNTP5.MEDIUM.jpg (R3 has ~0.7 volts over it, current calculation works like the LM317 circuit). The total drop for these doesnt need to be more than about 1 volt.

Depending on whats currently in the circuit, you might be able to use he existing mosfets as well. Also, if you need your diode case grounded, you can build a mirror image (using pnp instead of npn etc.) of the circuit. [/quote]

Any info as far as what transistors are used in that circuit? I'd like to build something like that. Simple and sweet.

 

[Benm]

Benm --
You touched on what I'm looking for.  The Mosfet bank is OK, it's just one component on the sense/driver board which is bad.  I'm looking for similar circuits to help identify my components..  Right now, it looks like a smt device marked 431P which may be toast and I don't know what it is.  The 8 pin IC is likely OK because it is somewhat isolated in the circuit.

The 431 marked device could be a zener diode (4.3 volt, philips in SOT346). Its unlikely that just failed on its own though, so simply replacing it would be an unlikely complete solution. Why do you think its toast?

 

[Benm]

Any info as far as what transistors are used in that circuit? I'd like to build something like that. Simple and sweet.

Q1 can be any small signal NPN transistor really, something like a standard BC337 or BC547 (or smt equivalent) works fine.

Q2 as drawn is a logic level mosfet, but can be substituted for another bipolar NPN transistor. This transistor should be able to handle the current and voltage across it, so for higher currents and voltage drops it needs to be a power transistor (BD139 works well).

I've built several of these current sources using BC547 and BD139 as the transistors, and they do work really well right down to just over 1 volt drop voltage. Also, it is possible to add a slow start to them using one more transistor, resistor and cap. I'll draw out the schematic when i have time to do so. They make very nice drivers for red diodes, but due to the lower voltage drop they can also be used to power blu ray diodes from 2 3.6V cells without any problem.

 

[Hemlock_Mike]

This driver circuit has a 1 watt 0.15 Ohm resistor for the sence feedback. I need it to be mor like 0.10 Ohm to adjust to 2 amps for the 808 2W pump. I was trying a piece of Nichrome wire and I saw smoke now it won't drive over 400 mA. If I drive the Mosfets direct, no problem. It's something in the control loop.
I'll have to do some testing here. I hoped someone might have a similar circuit to save me a little time. That's why this is a hobby !!!

Mike

 

[Benm]

I wonder what you blew up really... if you short out the shunt, chances are you damage the power fets, laser diode or both, but anything else seems somehwhat unlikely... unless you accidentily shorted out something else too (like a pull-up/bias resistor on the zener diode).

You might as well replace the driver with the transistor and use the existing fets to build the two-transistor approach, though you would need a larger (0.3 to 0.35) ohm shunt to regulate to 2 amps.

 

[mauswiesel]

Dunno if im telling shit, but im planning to build a 40 Amp regulated power supply with 50 lm317 (yes, i didn't studied the power of mosfet befor buying my 50 regulator).

So 2 lm317 can supply 2x current if added in parralel :

An even simpler way to parallel multiple voltage regulators together is to add very low (below 1 Ohm) ballast/equalising resistors in series with the regulator outputs. Doing so should make the regulators share current equally and therefore operate together without problems. It is worth noting however that this approach to load balancing makes regulation a bit less accurate.

 

[positron]

Dunno if im telling shit, but im planning to build a 40 Amp regulated power supply with 50 lm317 (yes, i didn't studied the power of mosfet befor buying my 50 regulator).

So 2 lm317 can supply 2x current if added in parralel :

An even simpler way to parallel multiple voltage regulators together is to add very low (below 1 Ohm) ballast/equalising resistors in series with the regulator outputs. Doing so should make the regulators share current equally and therefore operate together without problems. It is worth noting however that this approach to load balancing makes regulation a bit less accurate.

OMG

Do you light your living room with 50 candles or buy a light bulb? ;D

 

[Hemlock_Mike]

Benm --- I was using a regulated PS at the time limited to 2.4 Amps. The load was a dead (LED) c-mount which still lights LED style. I believe that it is the three lead SMT device that's bad but I have to figure out what it is.

Mike

 

[Benm]

On the dual lm317 approach:

Doing so should make the regulators share current equally and therefore operate together without problems. It is worth noting however that this approach to load balancing makes regulation a bit less accurate.

Both circuits are current sources, there is no need whatsoever for load balancing to achieve exactly the combined current through the load. This is very different from using voltage sources in paralel where slight differences give big problems.

Benm -- I was using a regulated PS at the time limited to 2.4 Amps. The load was a dead (LED) c-mount which still lights LED style. I believe that it is the three lead SMT device that's bad but I have to figure out what it is.

My guess is still a zener diode, that matches the markings.. but what kind of package is in?

 

[billg519]

There is a TL431P three terminal programmable precision reference. Perhaps this is what you are looking at?
Is it an SOT-89 package?

 

[GooeyGus]

Benm ---  I was using a regulated PS at the time limited to 2.4 Amps.  The load was a dead (LED) c-mount which still lights LED style.  I believe that it is the three lead SMT device that's bad but I have to figure out what it is.

Mike

You have a picture?

 

[GooeyGus]

Dunno if im telling shit, but im planning to build a 40 Amp regulated power supply with 50 lm317 (yes, i didn't studied the power of mosfet befor buying my 50 regulator).

So 2 lm317 can supply 2x current if added in parralel :

An even simpler way to parallel multiple voltage regulators together is to add very low (below 1 Ohm) ballast/equalising resistors in series with the regulator outputs. Doing so should make the regulators share current equally and therefore operate together without problems. It is worth noting however that this approach to load balancing makes regulation a bit less accurate.

the heat coming off of that thing is going to be ridiculous.... please sell those on ebay and build yourself a beefy passbank

 

[rkcstr]

Dunno if im telling shit, but im planning to build a 40 Amp regulated power supply with 50 lm317 (yes, i didn't studied the power of mosfet befor buying my 50 regulator).

How the heck do you power it?? That's a lot of amps

 

[GooeyGus]

[quote author=mauswiesel link=1240535024/0#22 date=1240924485]Dunno if im telling shit, but im planning to build a 40 Amp regulated power supply with 50 lm317 (yes, i didn't studied the power of mosfet befor buying my 50 regulator).

How the heck do you power it??  That's a lot of amps  
[/quote]


Yeah.. I think the only way to do that would be a bridge recitfier, some filter caps and a big ass transformer to get down to the voltage you want (and still be able to provide more than 40A current....) and power it on mains...

 

[Hemlock_Mike]

I don't need 40 amps for this.  WTH ??  I will study my circuit some more.

Mike

Billg519 -- It looks like a basic SOT23 or SOT323.