Continue to Site

Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

High side FET drive circuit for Offline Buckboost LED driver

Status
Not open for further replies.
T

treez

Guest
Hello,
We are doing a 150W Offline Buckboost LED driver, as in the attached schematic.
The high-side FET is to be driven via a digital isolator which is transferring the gate pulses up to a gate drive IC which is in the high-side circuit.

The high side bias supply is also shown attached.

Can you foresee problems with locating IC’s in this high-side circuit?..after all, it is referenced to the switching node. (eg noise problems and transients etc).
 

Attachments

  • High side Gate drive supply.pdf
    49.9 KB · Views: 160
  • Schematic _Offline Buckboost.pdf
    13 KB · Views: 150

Quite expensive. Smells like over-engineering...
 
  • Like
Reactions: treez

    T

    Points: 2
    Helpful Answer Positive Rating
Thanks, the thing is this is a constant off time controller, and the on-time can be very long at the low parts of the mains cycle, and so a gate drive transformer could saturate, so we need to do it like this.
 

Small isolated supplies for gate drivers don't usually need feedback like this (so long as the primary supply voltage is fairly well known). A simple push pull transformer driven at a fixed duty cycle is normally sufficient. Maxim and other manufacturers make ICs specifically for implementing this.

Or you could just use a flyback instead of a nonisolated buckboost, and use a simple low side switch...
 
  • Like
Reactions: treez

    T

    Points: 2
    Helpful Answer Positive Rating
Small isolated supplies for gate drivers don't usually need feedback like this (so long as the primary supply voltage is fairly well known). A simple push pull transformer driven at a fixed duty cycle is normally sufficient.
Thanks, but as you know, a simple forward mode transformer cannot be applied to this application with just two coils and simple surrounding circuitry....As you know, there is the resetting of the core to think about etc...and in the end , the bias coil regulated flyback that i have shown in the top post ends up being the simplest way of providing the high side bias supply..
 

Thanks yes, as you know, a push pull has a two coil primary. This is what i mean about the complexity being no better if topologies other than the one i demonstrate above are engaged with.
 

You can have a look at simple DCX for isolated supplies for your high side driver and see if there is some that meets your needs. Otherwise, you can design your own push-pull isolated and unregulated supply, like mtwieg said.

Why an isolated gate driver is not a solution for you ?
 

Why an isolated gate driver is not a solution for you ?
Thanks, it is isolated...do you mean why not use a specific gate drive transformer? (ie a pulse transformer)
 

Thanks, i see, you mean a bootstrap high side driver.
Would have to investigate that for long on times....also, the situation with high voltage mains transients.......one of the things with this application is that it must have above average survivability of high mains transient spikes.
 

Thanks, i see, you mean a bootstrap high side driver.
Would have to investigate that for long on times....also, the situation with high voltage mains transients.......one of the things with this application is that it must have above average survivability of high mains transient spikes.
No. I have said an ISOLATED GATE DRIVER IC, like the ucc21520. The high side supply can be made with DCX if there is one that fits your needs, or design your own.
 

Thanks yes, as you know, a push pull has a two coil primary. This is what i mean about the complexity being no better if topologies other than the one i demonstrate above are engaged with.
Cheap center tapped transformers are available off the shelf for this exact application: https://www.coilcraft.com/mintrans.cfm

It's less than ten components...
 

Thanks Mtwieg for the Coilcraft link.
Also Thanks CataM, the "DCX" is a term i am not familiar with....i presume it means an isolaed DCDC module?
 

the "DCX" is a term i am not familiar with....i presume it means an isolaed DCDC module?
DCX stands for DC Transformer and yes, it is an isolated and unregulated DC DC converter.
 

Thanks Mtwieg for the Coilcraft link.
FYI this is an extremely common way to implement isolated gate drivers:
https://www.digikey.com/en/product-...grated/transformer-drivers-for-isolated-power
https://www.ti.com/isolation/transformer-driver/products.html
https://www.linear.com/product/LT3999

Analog also makes modules which contain both isolated DC/DC as well as data isolators:
https://www.analog.com/en/products/interface-isolation/isolation/isopower.html

Also Thanks CataM, the "DCX" is a term i am not familiar with....i presume it means an isolaed DCDC module?
Assuming it refers to something like what is implemented with the MAX258, i.e. a DC-DC without feedback operating at a fixed duty cycle, thus giving a fixed Vout/Vin ratio. Effectively a "transformer" which operates down to DC.
 

you could possibly use a couple of xtors (emitter follower) instead of your high side gate drive IC. We typically use a local low side 15V supply and run a simple xtor H bridge to drive a small isolating Tx 48% duty cycle (100kHz) to provide the high side 12V for the high side drivers (full bridge of schottkies to rectify) [in fact our circuit gets +15/-5 - but that is another story] this gives an easy & small step up stage given that the GD power required on the high side is likely <1W
 

Status
Not open for further replies.

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top