More PWM questions

[Breezy]

Hi all. Thank you much for the help you have given to me about PWM’s and Mosfet’s. The 2 circuits I referred to in another thread were unanimously shot down as not very good designs.

I think I now have a better understanding of Mosfet’s. What I gather so far is summed up from the many replies I received. They work in switch mode when the circuit they are in turns the Mosfet fully on and fully off. In the linear mode, the Mosfet is operating in a region between full on/off which allows voltage and currents to be in them at the same time which makes them heat up and burn up. Is that the basic idea here?

Now I have gotten more curious about PWM’s, Mosfet’s, current limiting, SMPS and such. I have been looking for a circuit that will work to provide high currents (as high as 100 amps, more for curiosity now) and work in a switched mode, and the duty cycle is adjusted to provide output voltage to control the current level that is wanted to be kept constant at. I think that is what I am after now, is this right?

There seems to be a lot of hobby electronic sites with tons of circuits. I have not really found a circuit yet that can do what I want, or at least I have not recognized such a circuit if I did come across one. I did however find one that may be what I want and I attached the file to this post. Could this be a design I can use? Can any of you make a suggestion about other sites that may have what I want to find, or at least come close to what I want?

If I find a circuit that does the constant current properly (in switched mode) but say it is only for driving 2 or 3 amps, can such a circuit be modified to control higher currents? I think this may be possible by using the 2 to 3 amps to control a higher current Mosfet or multiple Mosfet’s. Is that correct?

You offered various ideas of circuits that could be models for what I want such as SMPS, LED drivers, and some other circuits. And you also suggested that I look at IC’s that are designed specifically as PWM’s that could be the basis of what I want.

This led me to search some of the IC makers and I have found some IC’s that could be what I want but they do not really tell me a whole lot more cause I just don’t understand a lot of what is said in tech-speak. If any of you know of any of these or take the time to look at the data sheets, or perhaps offer up some other options you know of, would be greatly appreciated. These are the ones I have found so far. (I would prefer to use an IC type that is DIP because the smaller things are scary to me, no idea how to solder with those.)

OB2354 Current Mode PWM Power Switch by On Bright Semiconductors
OB2354 datasheet, Pinout ,application circuits Current Mode PWM Power Switch
What I like about this one is 8 leg DIP and these words in the data sheet – “The PWM duty cycle is determined by the current sense input voltage and the FB input voltage.”

LM3424 National Semiconductor
**broken link removed**
Constant Current N-Channel Controller with Thermal Foldback for Driving LEDs
I tried to use the on line designer and chose LM3424 but it changed that to LM3409HVMY and gave a circuit design. This looks like a really good thing to use but I really don’t know much about what I find.

LM3477 National Semiconductor
**broken link removed**
High Efficiency High-Side N-Channel Controller for Switching Regulator


LM3485 National Semiconductor
**broken link removed**
…designed to drive high power, high brightness LEDs (HBLEDs) at a constant current of 1.4A

LM5025B National Semiconductor
**broken link removed**
Active Clamp Voltage Mode PWM Controller

In closing, I don’t remember which of you asked about that “boyfriend” I mentioned, well, I found out he’s a bad person. I won’t give you any details, but that was a good one to find out about before I got too far with him, what a jerk…

 

[Ubergeek63]

Supertex HV9910 is an LED driver that readily drives 3A continuous into a variable voltage load

 

[misterT]

Switching Regulators have a FB (feedback) input which is normally connected to the output voltage (to get feedback for constant voltage regulation). Perhaps you could use a hall sensor to measure the output current and connect the current sensors output to the FB pin. This way the Switching Regulator could become a constant current regulator. I have never tried that, but have thought about it. It could work. A big problem could be all the noise in switch mode devices.

I'm currently designing a switch mode servo amplifier that could be used for motor control or even as a digital audio amplifier. At the moment I'm waiting parts for the first prototype which will be a simple buck converter with voltage and current sensors. The controller will be an XMEGA microcontroller from Atmel. I could report here about the mistakes I made with the first prototype and all other things I learn with it. I will try to make a current/voltage regulator with maximum power, voltage and current limits.

 

[ronv]

It sounds like you are making a hydrogen generator. Is that correct? If so you probably don't need constant current, just an adjustable pwm, perhaps with over current protection. Most constant current circuits will have some inductance in the current path to slow down the current rise time to give the circuits time to respond to the current that is sensed. Small 100 amp inductors are hard to come by so share with us what the end product is and a little idea of the voltage and current being switched.