Dual MOSFETs - Gate-Source Capcitor or Resistor?

Thread Starter

mriksman

Joined Aug 31, 2010
113
Hi,

I've followed the datasheet https://www.mouser.com/datasheet/2/149/FL5150-964157.pdf to create a dimming circuit using 2 MOSFETs. It works (still ironing out some bugs though...). My question is the capcacitor they have between the Gate and Source. While looking around online, and talking to people, it seems this is normally a resistor.

Additionally, I don't get good simulations when using a capacitor. This is wrong
1599892664623.png

This is right;
1599892744968.png

Any ideas why?
 

ericgibbs

Joined Jan 29, 2010
18,766
hi,
Checking your LTS asc file and FL5150 datasheet, you are trying to create a FL5150 circuit using LTS models.?

Do you have an actual internal circuit diagram for the FL5150, as reference.?
E
 

ericgibbs

Joined Jan 29, 2010
18,766
hi Dick.
I have the PDF , ref page #2, I see a block diagram of the FL5150, but no internal circuitry.?

Attached a clip, is this the page you are referring too.?

Eric
 

Attachments

DickCappels

Joined Aug 21, 2008
10,152
yes

I thought that we might be able to find a datasheet for the controller but that is the datasheet.
The other document that I found is product overview (attached)
 

Attachments

Ian0

Joined Aug 7, 2020
9,671
I'm assuming that the good people at Fairchild (previously National Semiconductor) now OnSemi, (previously Motorola) know what they are talking about, they've been making MOSFETs for long enough.
Normally an RC filter on the gate drive is a BAD THING because it slows the turn on/off time and increases switching losses. However, one of the switching transients corresponds to mains zero crossing, so that doesn't have to be quick; and the other, if it were a triac dimmer would require an LC filter to slow down the current risetime to suppress the interference, so my guess is that the slow turn on (or off) is to sacrifice a bit of efficiency to reduce the interference, by slowing the risetime of the pulse.
The gate drive is almost certainly push-pull, so simulating it with a 4N25 isn't going to work.
 

Thread Starter

mriksman

Joined Aug 31, 2010
113
This is as good as you get.
1599916357956.png


So it seems people are alluding to
* The use of a Resistor or Capacitor between the G and S of the MOSFETS and whether it works or not is based on how the 'DRV Gate' internally functions...?
....?
 

Thread Starter

mriksman

Joined Aug 31, 2010
113
Hmmm... That works with the 22nF capacitor now...

But I don't understand why. I'll have to research MOSFETs a bit more and Gate Drivers....

Any idea why I have those... bumps in the first 100ms? They disappear when I remove the LTC1693 gate driver...
1599918320065.png
 
Last edited:

Thread Starter

mriksman

Joined Aug 31, 2010
113
Now I have 2 questions in here. Although they may be related (requiring the use of a Gate Driver?)
1. How to simulate the Low Power Mode. I assume I'll need to try a PMOS Gate Driver. I'm not sure what the PMOS threshold should be set to, or if the connections are right.
2. Now I have the simulated Gate Driver for 'DRV Gate', I can use the 22nF between Gate and Source. I'll research more on this to try and understand why. But what are those 'bumps' in the first 100ms??
 

Ian0

Joined Aug 7, 2020
9,671
I couldn't see the voltage source that defines Vs. Is the other end of it connected to gnd_a or to spice's 0V which is represented by the triangle-ground symbol?
 

Thread Starter

mriksman

Joined Aug 31, 2010
113
Vs is circled. It is created by the 17V Zener, C2_1, R2_1 and D1_1 - creating a transformerless power supply.

1599927608820.png


You can see the bumps/noise just by placing the LTC1693 component in
1599927687529.png
 
Top