CD4047 output current gate driver?

eetech00

Joined Jun 8, 2013
3,859
Since the IC is driving MOSFETs the IC current output is not really a concern since MOSFETs are voltage controlled. The more gate voltage, the more current it will pass. Drive the mosfet with the most voltage its gate and the IC can Safely handle.
 

jpanhalt

Joined Jan 18, 2008
11,087
There are several calculators online for capacitor charge rates. This is one I use: http://hyperphysics.phy-astr.gsu.edu/hbase/electric/capchg.html

With some approximations, your RC time constant is about 10 us. One RC constant should turn the mosfet on (i.e., about 2/3 x15 V).

1592342055693.png

10 us = 100 kHz. However, the mosfet will be in its linear region for a good part of that time and may get warm. Why not use a simple multibvibrator and a purpose-designed driver that makes the complementary outputs? Simple one will output amps for a short time, not milliamps.
 

rsjsouza

Joined Apr 21, 2014
383
@jpanhalt shows a good way to calculate your way into it. You need to keep in mind the gate capacitance also changes with the output current (in the datasheet below it can reach up to 8000pF, for example).
https://www.vishay.com/docs/91215/91215.pdf

I would also plan to use a driver or another oscillator circuit to get higher currents. This image belongs to a blog that describes a high voltage generator to a Geiger-Müller vacuum tube, but these tubes usually consume very little power.
 

Thread Starter

Hamlet

Joined Jun 10, 2015
519
There are several calculators online for capacitor charge rates. This is one I use: http://hyperphysics.phy-astr.gsu.edu/hbase/electric/capchg.html

With some approximations, your RC time constant is about 10 us. One RC constant should turn the mosfet on (i.e., about 2/3 x15 V).

View attachment 209934

10 us = 100 kHz. However, the mosfet will be in its linear region for a good part of that time and may get warm. Why not use a simple multibvibrator and a purpose-designed driver that makes the complementary outputs? Simple one will output amps for a short time, not milliamps.
Yes, I suppose I could do that, but I thought maybe the CD4047 might do this better than I could.

Do you have an example of a "purpose-designed driver"? Maybe I could link that to the cd4047...
 

jpanhalt

Joined Jan 18, 2008
11,087
Yes, I suppose I could do that, but I thought maybe the CD4047 might do this better than I could.

Do you have an example of a "purpose-designed driver"? Maybe I could link that to the cd4047...
There are, or used to be, a lot of them. Here is one I have used: http://cds.linear.com/docs/en/datasheet/1158fb.pdf

It controlled paralleled mosfets to control the speed of a Ford long-shaft tractor starting motor being used as a winch (current > 200 A). That was 15 years ago. My guess is that more modern controllers must exist.
 

Papabravo

Joined Feb 24, 2006
21,159
The problem with driving MOSFETs with wimpy chips like 4000 series CMOS is that you need to rapidly move charge onto and off of the gate. To do this quickly, to avoid power dissipation problems, takes a push-pull driver that can source and sink large currents.
 

Audioguru again

Joined Oct 21, 2019
6,673
How does that get the upper mosfet gate to VCC+10 V (or whatever)?
The circuit does not have an "upper" Mosfet. Both Mosfets have their sources grounded so the gates need pulses that are 10V for ordinary Mosfets or 5V for logic level Mosfets. The transistors provide pulses that are about 0.7V less than the supply voltage but with a high current to quickly charge and discharge the gate capacitance of the Mosfets.
 

Thread Starter

Hamlet

Joined Jun 10, 2015
519
So the CD4047 supplies both negative an positive signals?

A push-pull circuit with discrete components amplifies these signals,
and helps hurry along the mosfet gate charge/discharge cycles,
keeping it out of the linear region, so it doesn't get hot or
stumble over itself at higher khz , right?

I keep reading mosfets are voltage controlled, which messes me up,
because in some applications, they need current too.


Add this in-between CMOS and MOSFET.
View attachment 209946
 

eetech00

Joined Jun 8, 2013
3,859
So the CD4047 supplies both negative an positive signals?

A push-pull circuit with discrete components amplifies these signals,
and helps hurry along the mosfet gate charge/discharge cycles,
keeping it out of the linear region, so it doesn't get hot or
stumble over itself at higher khz , right?

I keep reading mosfets are voltage controlled, which messes me up,
because in some applications, they need current too.
Mosfets are voltage controlled and usually only require a tiny current to function. That makes them nice to use with CMOS logic devices when you need to drive large current loads. The more voltage applied to the gate, the smaller the "RDS_on" resistance, so more current will pass thru the drain and source. Think of it as a voltage controlled resistor.

What switching frequency do you intend to use?
How much load current?
 

Thread Starter

Hamlet

Joined Jun 10, 2015
519
Mosfets are voltage controlled and usually only require a tiny current to function. That makes them nice to use with CMOS logic devices when you need to drive large current loads. The more voltage applied to the gate, the smaller the "RDS_on" resistance, so more current will pass thru the drain and source. Think of it as a voltage controlled resistor.

What switching frequency do you intend to use?
How much load current?
I don't even know where to begin, that's why I'm asking questions.

I've lost some hearing, so 10khz sounds good to me! Yet, I would be happy with
1khz, as that seems the limit for iron transformers, which I have in abundance.

As for mosfets, anything goes, 50A? 110A? Maybe try 70A igbt? I'm thinking it might be
fun to try building a wild, non-sine wave, non-60Hz inverter, a big one, and push some electrons thru a big dc motor
or something, and most importantly, learn something along the way.

To me, all this boost converter stuff is new, and wonderful, and I'd like to know what I can and can't
get away with.
 

rsjsouza

Joined Apr 21, 2014
383
I keep reading mosfets are voltage controlled, which messes me up, because in some applications, they need current too.
They are controlled by both, but in different ways.

The voltage controls the resistance value between Source and Drain, but the current controls how fast this resistance can be controlled.

In this case, given the CMOS devices can reach higher voltages (typically 18 or 20V), they are not a problem to do the direct drive for the MOSFET you mentioned.

The "tiny current" is the core of what was being discussed earlier, which is significant depending on the driver that is trying to control it. The low power fanout of CMOS little logic devices such as the CD4047 will take a while to charge the gate capacitance of the MOSFET to its desired voltage, thus limiting the maximum number of charges/discharges per second (the switching frequency). The larger the MOSFET, the higher the gate capacitance.
 

Papabravo

Joined Feb 24, 2006
21,159
I don't even know where to begin, that's why I'm asking questions.

I've lost some hearing, so 10khz sounds good to me! Yet, I would be happy with
1khz, as that seems the limit for iron transformers, which I have in abundance.

As for mosfets, anything goes, 50A? 110A? Maybe try 70A igbt? I'm thinking it might be
fun to try building a wild, non-sine wave, non-60Hz inverter, a big one, and push some electrons thru a big dc motor
or something, and most importantly, learn something along the way.

To me, all this boost converter stuff is new, and wonderful, and I'd like to know what I can and can't
get away with.
You should probably learn to crawl and walk before you try to run a marathon. This stuff is not really a good place for dabblers -- it can be hazardous to your well being.
 

Thread Starter

Hamlet

Joined Jun 10, 2015
519
You should probably learn to crawl and walk before you try to run a marathon. This stuff is not really a good place for dabblers -- it can be hazardous to your well being.
Thank you for your concern.

I regret that my previous post may have seemed glib or naive to your reception. Online text communication often
fails at important nuances of information transfer. You may assume that I lack experience with TTL/CMOS/Logic families,
but let me assure everyone, I have handled/designed/installed electrical equipment & systems, and understand the need to apply good working practices and caution.

I will explore "pull-pull" drivers that you offered as a possible solution in your previous posting #10.
 

Thread Starter

Hamlet

Joined Jun 10, 2015
519
In this case, given the CMOS devices can reach higher voltages (typically 18 or 20V), they are not a problem to do the direct drive for the MOSFET you mentioned.
Thanks! I am going to read up on the differences in "logic families" (ttl,cmos, etc.). Then I'll grab the iron and see how much smoke I can make.
 
Top