Gate Driver vs Opamp

Thread Starter

8dm7bz

Joined Jul 21, 2020
199
Hello,
I build an RLC circuit using the MCP1402 as a gate driver. At resonance the MCP1402 will drive a 20 ohms load (plus 3 ohms which is the DCR of the coil). Unfortunately the MCP1402 has around 12 ohms output impedance (according to this) and so it cannot drive the load effectively. A more detailed discussion about this can be found in this thread.

Now my question is wether to use a lower impedance gate driver, or a high slew rate opamp. As for the specifications, I want it to be able to source around 500mA. It will switch 10Vpp as denoted in the schematic (I would be fine with 5V aswell, since I can build a bipolar drive aswell)

If an opamp is the way to go, the following specifications would be needed. For the slew rate I calculated about 1.25V/us to switch a 20kHz signal from 0 to 10V. But the project requires a frequency up to 100kHz (about 6.3V/us). And the output impedance is probably very low for an opamp, although I couldn't find the parameter in the datasheets that would suggest this.

If on the other hand a gate driver is better, I think the key specifications would be the that the output current should be around 500mA and the output impedance is very low (I'm not sure what that is called in the datasheet, maybe "output pull-up resistance" ?).

If I forget any important specifications that are needed to drive the load under the stated frequency, please let me know.

-8
 

Attachments

Last edited:

Alec_t

Joined Sep 17, 2013
14,280
Most opamps (at least those at a reasonable price) are pretty wimpy. I'd go for a dedictaed MOSFET-driver IC with 1A or more output capability.
 

ronsimpson

Joined Oct 7, 2019
2,989
buffer
gate drive not in production
Motor driver Just thought of motor drivers. There are many types.
more motor driver
TI has many here is a little one
Here is another thought. 0.35 ohm 600 volt MOSFETs and a PWM. I think I can force it to 50% duty cycle. Onsemi
----------edited---------
More about OnSemi parts:
There are three different families of parts. Internal oscillator, output=50% duty cycle, not PWM but it changes frequency to control output power. It could be set to min or max frequency and left alone. A micro could pull down on "RT" and adjust the frequency. Internal MOSFETS (depending on what part you get the Resistance is in the 4 to 0.3 ohms range) These parts are often used in light bulbs and powered off the power line. I do not know how low the voltage can get and still work. Another part said 35V to 400V. I think the parts will work at a low 15V but do not know. There are more parts like this from other countries.
 
Last edited:

Thread Starter

8dm7bz

Joined Jul 21, 2020
199
The buffer has a great slew rate, but the output resistance is 12 ohms according to page 3. So not too much better.
For the rest I couldn't really find the output resistance in the datasheet. Although it may be the thing called "RDS" ? Is that the output resistance ?
 

Papabravo

Joined Feb 24, 2006
21,159
The buffer has a great slew rate, but the output resistance is 12 ohms according to page 3. So not too much better.
For the rest I couldn't really find the output resistance in the datasheet. Although it may be the thing called "RDS" ? Is that the output resistance ?
rds is the resistance of the channel in a MOSFET when the gate voltage is far enough above the source to turn the device on HARD.
 

ci139

Joined Jul 11, 2016
1,898
? what's your frequency and the output DC offset (?settability?) requirements . . . use any decent audio power Amp (or less likely a Coax driver) ?
 

Thread Starter

8dm7bz

Joined Jul 21, 2020
199
Thanks for explaining RDS

The frequency is fixed and not yet set (will be fixed with a counter IC and a crystal). Could be anything between 20kHz and 100kHz. And there is no DC offset. I can't seem to find any specifications in any opamp datasheet about the output resistance. Is that because there is so little of it ?

I found this gate driver (there is also a non-inverting one). It's typical output resistance is 1 ohm. Would that one be good for the prupose ?
 
Last edited:

ronsimpson

Joined Oct 7, 2019
2,989
I found this gate driver.
Looks better. Its output looks like about 1 ohm to ground or supply.
I can't seem to find any specifications in any opamp datasheet about the output resistance.
Opamps are not used in that mode and often not rated like that.
There is a type of amp that is rated. CMOS R-R OUTPUT they have FETs in the output and can pull to the supply rails.
Most opamps do not work well when the output is at the supply rails. They might only pull to 2 volts off the rail. Many opamps, when powered with 12V, only work well in the 2-10V range.
 

Ian0

Joined Aug 7, 2020
9,677
Your 5V supply is becoming your limitation. Before long, you might simply need more power, and in that case more voltage would be a better choice than same voltage more current.
With a little more voltage, a complementary emitter follower would do the job nicely, and the 0.6V drop wouldn't be a problem.
You may also find that the gate driver ICs have lower output resistances working at larger supply voltages.
If you want to stick with 5V, then get the one with the highest output current that you can find.
You won't find "output resistance" in op-amp datasheets, as negative feedback can reduce it to a negligible amount so it becomes irrelevant, but a negligible output resistance doesn't mean you can get a huge output current.
You will find maximum output current specified, though.
 

Thread Starter

8dm7bz

Joined Jul 21, 2020
199
Alright thanks @ronsimpson. I think I won't use an opamp then. I found an even better gate driver (MCP14A1201) with even lower output impedance.

@Ian0 I think I try to get the best out of the above mentioned gate driver. But in case this won't work well, could you explain this thing about the "complementary emitter follower" a bit more please. I don't quite understand how this thing works.

EDIT: I just read a little about the complementary emitter follower. Is that the same as the Class AB thing that @Papabravo was talking about ?
 
Last edited:

Ian0

Joined Aug 7, 2020
9,677
EDIT: I just read a little about the complementary emitter follower. Is that the same as the Class AB thing that @Papabravo was talking about ?
Almost - a proper class-AB amplifier needs biassing correctly. If you just need a square-wave that doesn't matter.
Take two transistors, one NPN, one PNP. Connect their bases together. Connect their emitters together. Take the NPN collector to the positive supply and the PNP collector to the negative supply. Input on the bases. Output on the emitters.
Only snag is that the output signal is 0.6V smaller than the input signal.
On a supply with plenty of volts you'll not miss 0.6V.
 

Thread Starter

8dm7bz

Joined Jul 21, 2020
199
yes, I probably won't miss the 0.6V. Do the transistors need any special specifications ? In other words, do you have some in mind ?
 

Ian0

Joined Aug 7, 2020
9,677
You will on a 5V supply!
There are a lot of transistors to choose from.
Obvious things: enough collector current, and enough Collector-emitter voltage.
Then plenty of Hfe. Your input has to provide current equal to Output current / hfe
Finally don't forget heat dissipation. It will dissipate 0.6V x your output current.
 

Thread Starter

8dm7bz

Joined Jul 21, 2020
199
I picked two transistors: 1, 2.
One is a throughhole, just so I can test it easier. The other one is a smd component. Would they both be suitable ?

throughhole, smd

both are npn, but I think they have similar pnp aswell

EDIT: changed the smd component
 
Last edited:

BobTPH

Joined Jun 5, 2013
8,813
Did I miss something here?

Why are you trying to use a gate driver to drive your load instead of using it for its intended purpose - driving the gate of a MOSFET which then drives the load with an output impedance in the mOhm region?

Bob
 

Thread Starter

8dm7bz

Joined Jul 21, 2020
199
We had an earlier discussion here about how to drive the coil.

Firstly, there is no diode to clamp the back EMF when the MOSFET switches off, and it will blow up the MOSFET. And secondly, if you do clamp it with a diode, the ON Volt-time constant is much longer than the OFF Volt-time constant so the current will increase indefinitely until your battery goes flat or something overheats. You might make it work by clamping it with a diode and a zener in series, but you won't get a nice waveform.
That is why I changed to a MOSFET driver. But since then a lot changed as far as the setup goes (I'm now trying to realise an RLC circuit). So maybe it's right to change.

How would the circuit look like with a MOSFET ? @BobTPH

At the moment I'm just trying to maximise the current going through the RLC circuit at a specific frequency (with a square wave). So whatever does the job best I will adapt.
 

Thread Starter

8dm7bz

Joined Jul 21, 2020
199
@BobTPH What specifications would the MOSFET need to have ? I read a little, and it seems to be that the gate driver is used to actively push and pull the gate high and low. Is that right ?
 

Ian0

Joined Aug 7, 2020
9,677
I picked two transistors: 1, 2.
One is a throughhole, just so I can test it easier. The other one is a smd component. Would they both be suitable ?

throughhole, smd

both are npn, but I think they have similar pnp aswell

EDIT: changed the smd component
Make sure that the BC816 doesn’t run too warm.
 

Ian0

Joined Aug 7, 2020
9,677
Did I miss something here?

Why are you trying to use a gate driver to drive your load instead of using it for its intended purpose - driving the gate of a MOSFET which then drives the load with an output impedance in the mOhm region?

Bob
Since when did anyone use a device only for its intended purpose?
MOSFET gate drivers make excellent push pull drive for all sorts of purposes. Did no-one ever put a whole load of logic gates in parallel to drive something other than logic?
 
Top