MOSFET Gate Driver

I have had good luck with the IR2121.

emc_c

for the power MOSFET i want to be able to vary Vds as well as gate voltage. does the this chip do that?

Sorry, I can't take the time to research the data sheets for you, but:

1) Gate driver chips are made to be FAST (typically faster than a "roll your own" circuit)
2) Linear Tech is a very good company

Therefore, you might want to double check and triple check the data sheet, schematic and circuit that you currently have. Double check resistor values, etc. You might only need to fix a little "oops".

http://www.chipdocs.com/datasheets/datasheet-pdf/Linear-Technology/LTC1255.html Use this link to download the Datasheet of LTC1255. i will u help u.

I bought this chip based on the the section on resistive load. when i set it up I realized that there are essential parts that I need missing. for the power MOSFET i want to be able to vary Vds as well as gate voltage. this IC wasn't the correct choice right?

To much ringing of a to fast signal. That sounds exactly like an unanticipated transmission line problem. Review your wiring and think in RF.

I realize I have made a huge mistake in component choice.

for the power MOSFET i want to be able to vary Vds as well as gate voltage.

Is there a chip out there that gives me the ability to test Vds of 5-20 volts . Also I need to vary my gate input anywhere from 10-20 V.

thanks

I cannot give you a simple answer for you there. A MOSFET is a three lead device. Like all three lead devices one can easily directly control the voltage across or the current through the two leads the driver circuit and transistor have in common (V_gs). To know how this effects a different pair of leads (V_ds) one must both know how the three lead part acts and what is rest of the circuit. Since you do not provide at all a circuit description, or better yet a schematic, the only help I can give you is a lesson on how to read a MOSFET data-sheet.

Now for how the part you've chosen, IRF520N, you should have the data-sheet. If one looks at the three dimensional chart for figure 1 on the data-sheet, one will see that for each V_gs value many different values for V_ds can exist. (A common mistake people have when they view these curves is that they think that the horizontal axis is a time axis. Read the chart again it is not.) Now you might think that you wish to keep the circuit in the right hand part of this series of curves where the values of V_gs has the largest separation. But the separation is along the Y axis, the current through I_ds and not the voltage V_ds is the separation. This mode of operation called the saturation or active region, V_gs modulates the current through I_ds. (If you notice the separation of the different gate voltages is not a linear relationship. This too is explained in the Wiki link.) Now you might think the that you want to be in the linear region on the left hand side of the charts. In this region V_gs effectively modulates the effective resistance between the drain and source (R_ds). In this region there is a nearly linear relationship between gate voltage and R_ds. This jumping between linear and non-linear control is why a power MOSFET is normally used as a controlled switch and not linearly.

Now Pulse Width Modulation (PWM) does approach a linear like response while using a switch that can be filtered to produce a linear waveform. But this will involve a lot more theory.

Lastly, my earlier comment about transmission lines comes from the fact that a MOSFET does this all in the realm less than 0.1 microseconds. Thus megahertz frequencies exist to make reflections off of unanticipated transmission line mismatches.

The drive circuit and the gate input are forming an under-damped RLC circuit which provides the excessive ringing. You need to find a series R value which will dampen the ringing. Ballpark R= squareroot(L/C). There is usually a compromise between rise/fall time and damping factor so you will need to adjust the R value to get acceptable results. Adding the correct R works 9/10 times for me, but there are always exceptions.

Thank you for all the advice but I have attempted all the suggestions give here and that is why I resorted to buying an IC gate driver.

Now I do not know how to look for an IC that gives me the abiliy to:

Vg= 1kHz, 40% duty cycle, and 10-20 Vpp

Vds= 5-20 Volts

Each IC I find focuses on the supply voltage (which i believe is the Vds that I need)

but which IC allows me the Vg input I need?

look out for datasheets for ur requirements because it will give additional details regarding your requirements. No one can't advice in this. Because you are in the situation. Better checkout and simulate results... it will be better. Lots of Student version of circuit simulators are available

Maybe it's not necessary to put a mosfet driver. You could try to put a resistor on the gate of the mosfet. The combination of this resistor and the gate (parasitic) capacitance of the gate (i.e. RC) will probably make the pulse more "smooth" avoiding the "ringing". Don't put a high value resistor because a large RC will cause a large "rising time" of the pulse (this is not good as the mosfet will be in its linear region for longer time, increasing its power dissipation). Probably a 100R value (or even smaller) will be sufficient. You could experiment on this.

Again thanks for the suggestions...

I have experimented with the 100Ohm resistor and capacitors but the ringing is still there. also, the simulators do not incorporate the voltage depending components within a power MOSFET, therefore it comes out perfect.

That is why I was desperate for a driver. But I dont know how to search for a pulse input driver

Most simulators do incorporate the voltage dependent components of a MOSFET very well. What is not captured well is the parasitic capacitances and inductances from wiring. You've entered into your simulator precisely what you wished you could create and the simulator gave you a usable response. Your problem appears to be that you cannot build precisely the ideal circuit you think you've built.

There is also another possibility, keep in mind your oscilloscope probe adds capacitance and changes the loading on the gate of the transistor. Also where and how your oscilloscope probe gets grounded will dramatically change how a high frequency component looks.