Many IR2117 past threads, yet none helped me resolve the problem ... thus this new thread )

Built the attached data-sheet schematic (I determined values).
Poor, intermittent operation ... sometimes nothing ... sometimes fan briefly starts then stops.
Using +12 Vdc for all inputs.

Sure seems like a straight forward circuit ... yet I am perplexed a bit )
Anyone that might have something to chime in is welcome ... and much appreciated.

The goal is to simply turn ON & OFF a power MOSFET (+12, 24, 48 Vdc), for a fan motor load.

 

okay ... so my post was premature ... I now believe that I simply need a P-channel MOSFET.
This N-channel MOSFET requires ~ 3-5 V + the supply (12V) to turn ON.
This is not realistic for my final application ... so I will change to P-channel and complete bench tests.

 

I have used the IR2117 many times and it works. The duty cycle cannot go to 100% on. It must be off for a time to recharge the capacitor.

P-FET is an option.

I would use a N-FET and switch the ground side of the fan. The Gate Driver is very simple this way.

 

If switching the Ground-side is not practical for your application,
use a Photo-Voltaic-Gate-Driver as shown below.
This works with High-Side, or Low-Side, Control equally well, no P-Channel required.
But, the Photo-Voltaic-type-Gate-Drivers can not be used for "High-Speed-Switching"
as they are somewhat "slow", ( several Milli-Seconds, depending upon the specs of the FET used ).
.
.
.
.

 

If switching the Ground-side is not practical for your application,
use a Photo-Voltaic-Gate-Driver as shown below.
This works with High-Side, or Low-Side, Control equally well, no P-Channel required.
But, the Photo-Voltaic-type-Gate-Drivers can not be used for "High-Speed-Switching"
as they are somewhat "slow", ( several Milli-Seconds, depending upon the specs of the FET used ).
.
.
.
View attachment 324158.

Hey there .. really appreciate the feedback ... thank you kindly )
So I do not anticipate a problem with this new circuit.
The HO value will always be 0 or +12 Vdc, with a power supply of +12, 24 or even 48 V.
The VGS will always be negative as required for the MOSFET to turn ON with a +12 Vdc HO, and turn OFF with a 0 Vdc HO.
My load will connect between the MOSFET drain and ground, which is ideal for my application.
I will bench test when parts arrive, my only real concern is the V drop of the MOSFET (0.1 Ohm x 20 A = 2 V).
I wish that I could eliminate this V drop.

 

No.
You are driving the Gate by Gate to Drain. No.
The 10k resistor should be G-S on the MOSFET.


This is simple if you switch the negative lead of the fan

 

Just select a more appropriately rated MOSFET, problem solved.

Don't get stuck on the one solution that You think You finally understand.
There's more than one way to skin a cat.

I think that there is nothing easier than driving a single LED, as I previously suggested.
.
.
.

 

I have used the IR2117 many times and it works. The duty cycle cannot go to 100% on. It must be off for a time to recharge the capacitor.

P-FET is an option.

I would use a N-FET and switch the ground side of the fan. The Gate Driver is very simple this way.

Thank you kindly ... I am not sure that I 100% agree though ... I'll bench test Tuesday and share the results. The 10k resistor to ground is simply to give a place for any in-rush currents to go after the MOSFET is switched OFF. Trying to protect the IR2117.

 

Just select a more appropriately rated MOSFET, problem solved.

Don't get stuck on the one solution that You think You finally understand.
There's more than one way to skin a cat.

I think that there is nothing easier than driving a single LED, as I previously suggested.
.
.
.

Thank you kindly LowQCab ... yeah .... as it turns out, there is a much lower Rds MOSFET 0.01 vs 0.1 Ohm. It is a $6 premium x4, but since my client keeps telling me that he wants all the premium features, well, easy decision )

 

~20-Amps DC for a Fan(s) !!!! that's serious Power.

Why 3-different Voltages ?
.
.
.

 

Analog Devices makes a high-side driver that can do 100% mark-space, because it has a built-in charge-pump.
https://www.analog.com/en/products/ltc7001.html

 

Thank you kindly ... I am not sure that I 100% agree though

The Gate Driver, when outputting a 0V signal is shorting Gate to Drain.
The Gate Driver, when outputting a 11V signal is putting a 11-volt battery from Gate to Drain.
Neither of these conditions are good.

 

~20-Amps DC for a Fan(s) !!!! that's serious Power.

Why 3-different Voltages ?
.
.
.

The application is typically just +12 Vdc.
But sometimes, the same application requires +24 or +48 Vdc.
I doubt they will ever see the 20 Amps, just a required feature.

 

The Gate Driver, when outputting a 0V signal is shorting Gate to Drain.
The Gate Driver, when outputting a 11V signal is putting a 11-volt battery from Gate to Drain.
Neither of these conditions are good.

Thank you kindly for this input ... I am completing the bench test to verify solid functionality.
Gotta start with what I believe is correct, but I am by no means the MOSFET expert.

 

Gotta start with what I believe is correct, but I am by no means the MOSFET expert.

If you are going to do this:
Using a P-MOSFET, the Gate on voltage us below the +12/24/48V supply. So with a 48V supply, a Gate voltage of 40 will turn on the P=FET. A Gate voltage of 48 will turn it off. (note you must drive the Gate-to-Source not the Gate to Drain)
I connected the Vb supply to the (+12/24/48V) supply. Now when IN =high, HO=Vb and the Gate-Source voltage = 0V, MOSFET=off. When IN=low, HO=Vs which is (+12/24/48V supply)-11 volts.
That works BUT: C 10uF on Vb-Vs will never charge up. There will never be a charge on the cap so HO cannot function. In both circuits (using the P-mosfet) the IR2117 will never get a supply voltage stored on the capacitor. Will not work.

If you have to have a topside driver, then use post #11, using a P-FET.
or use post #4 if you are just turning on/off and are not using a PWM like 100khz. (slow turn on/off)
or use a bottom side driver.

 

You can lead a horse to water, but You can't make him drink.
.
.
.

 

If you are going to do this:
Using a P-MOSFET, the Gate on voltage us below the +12/24/48V supply. So with a 48V supply, a Gate voltage of 40 will turn on the P=FET. A Gate voltage of 48 will turn it off. (note you must drive the Gate-to-Source not the Gate to Drain)
I connected the Vb supply to the (+12/24/48V) supply. Now when IN =high, HO=Vb and the Gate-Source voltage = 0V, MOSFET=off. When IN=low, HO=Vs which is (+12/24/48V supply)-11 volts.
That works BUT: C 10uF on Vb-Vs will never charge up. There will never be a charge on the cap so HO cannot function. In both circuits (using the P-mosfet) the IR2117 will never get a supply voltage stored on the capacitor. Will not work.
View attachment 324328
If you have to have a topside driver, then use post #11, using a P-FET.
or use post #4 if you are just turning on/off and are not using a PWM like 100khz. (slow turn on/off)
or use a bottom side driver.

WOW, thank you kindly for all the details. I learn best by bench work ... so I will see all of this firsthand, very soon )

 

If you are going to do this:
Using a P-MOSFET, the Gate on voltage us below the +12/24/48V supply. So with a 48V supply, a Gate voltage of 40 will turn on the P=FET. A Gate voltage of 48 will turn it off. (note you must drive the Gate-to-Source not the Gate to Drain)
I connected the Vb supply to the (+12/24/48V) supply. Now when IN =high, HO=Vb and the Gate-Source voltage = 0V, MOSFET=off. When IN=low, HO=Vs which is (+12/24/48V supply)-11 volts.
That works BUT: C 10uF on Vb-Vs will never charge up. There will never be a charge on the cap so HO cannot function. In both circuits (using the P-mosfet) the IR2117 will never get a supply voltage stored on the capacitor. Will not work.
View attachment 324328
If you have to have a topside driver, then use post #11, using a P-FET.
or use post #4 if you are just turning on/off and are not using a PWM like 100khz. (slow turn on/off)
or use a bottom side driver.

Disappointed in the bench tests that confirmed your suggestions. Well I guess I have to wait another day to solve this with the new parts that are on order ) I'll try both of these suggestions .. FDA215 & ITC7001. Many thanks for the extremely helpful contributions )

 

Analog Devices makes a high-side driver that can do 100% mark-space, because it has a built-in charge-pump.
https://www.analog.com/en/products/ltc7001.html

My approach failed ... so I'm trying your suggestions tomorrow ) Many thanks )

 

If switching the Ground-side is not practical for your application,
use a Photo-Voltaic-Gate-Driver as shown below.
This works with High-Side, or Low-Side, Control equally well, no P-Channel required.
But, the Photo-Voltaic-type-Gate-Drivers can not be used for "High-Speed-Switching"
as they are somewhat "slow", ( several Milli-Seconds, depending upon the specs of the FET used ).
.
.
.
View attachment 324158.

My approach failed ... so I'm trying your suggestions tomorrow ) Many thanks )