Previous in Forum: Control of home appliances through mobile phone.   Next in Forum: Potential Divider
Close
Close
Close
7 comments
Rate Comments: Nested
Associate

Join Date: Feb 2007
Location: India
Posts: 27

High Power High Speed switches

02/15/2007 6:56 AM

Hi

I am looking for an on/off switch that can switch 30Vdc_15Amps at 30KHz rate.

I need to control this using 5V TTL control signal. What sort of component should I use?

I have gone through the solid state relays, but they are bulky for my requirement.

So I am thinking of to use discrete components and build my own switch.

Any suggestions... Please?

__________________
I know not with what weapons World War III will be fought, but World War IV will be fought with sticks and stones. :Albert Einstein
Register to Reply
Interested in this topic? By joining CR4 you can "subscribe" to
this discussion and receive notification when new comments are added.
Guru
Engineering Fields - Electrical Engineering - New Member

Join Date: Sep 2006
Location: El Lago, Texas, USA
Posts: 2639
Good Answers: 65
#1

Re: High Power High Speed switches

02/15/2007 10:03 AM

What you want is a MOSFET transistor switch. Go to www.irf.com. They have devices that can be switched directly from your 5V level, although I would suggest using a CMOS gate to drive rather than a TTL gate, to get more drive current. MOSFETs don't require any drive current except what's required to charge and discharge the gate capacitance, and CMOS outputs are ideal for that. You can get MOSFETs in a variety of packages, TO-220 being the most popular, but smaller surface mount packages are available.

Register to Reply
Power-User
United States - Member - New Member Fans of Old Computers - PDP 11 - New Member Hobbies - HAM Radio - New Member

Join Date: Jan 2005
Posts: 381
Good Answers: 8
#2
In reply to #1

Re: High Power High Speed switches

02/15/2007 11:36 AM

Beware the statement: "MOSFETs don't require any drive current except what's required to charge and discharge the gate capacitance..."

While this is true, when you're trying to charge and discharge the gate at very high frequencies, the peak currents can be much higher than you would think. I've seen high speed MOS gated power modules that pulled 20 Amps of charge/discharge current.

Register to Reply
Guru
Engineering Fields - Electrical Engineering - New Member

Join Date: Sep 2006
Location: El Lago, Texas, USA
Posts: 2639
Good Answers: 65
#3
In reply to #2

Re: High Power High Speed switches

02/15/2007 11:58 AM

This is true. I prefer to use a gate driver chip that's made specifically for driving MOSFET gates, but if all you've got is a CMOS logic driver chip, you can drive the 3000pf or so that a part like the IRF1010E has on it's gate, and you'll be OK. The disadvantage is that the lower the drive current the slower the turn on time, and the greater the power dissipation (MOSFETs are more efficient the faster you switch them on and off, and in general, faster switching speeds means more drive current).

Register to Reply
Associate

Join Date: Feb 2007
Location: India
Posts: 27
#4
In reply to #3

Re: High Power High Speed switches

02/16/2007 6:18 AM

Suppose If i use IRF1010E to switch 30V_10A at 30kHz rate, what driver shall I use?

Hey ! dont scare me that Its going to draw around 20Amps for just driving the gate at this speed...Is it so ???

I guess it will be around 1A right ?

__________________
I know not with what weapons World War III will be fought, but World War IV will be fought with sticks and stones. :Albert Einstein
Register to Reply
Power-User
United States - Member - New Member Fans of Old Computers - PDP 11 - New Member Hobbies - HAM Radio - New Member

Join Date: Jan 2005
Posts: 381
Good Answers: 8
#5
In reply to #4

Re: High Power High Speed switches

02/16/2007 7:26 AM

20 Amps is too high for a 30V - 10A MOSFET Gate. I was referring to much larger IGBT modules (~ 1000V, 1000A). My point was just that the statement "MOS gates don't draw any current" can be misleading. The peak gate current will depend on how fast you try to switch the device. It's not that hard to calculate.

At such high switching speeds, if your load is at all inductive, you may need some snubbering as well to keep voltage spikes from killing the MOSFET.

Register to Reply
Guru
Engineering Fields - Electrical Engineering - New Member

Join Date: Sep 2006
Location: El Lago, Texas, USA
Posts: 2639
Good Answers: 65
#6
In reply to #4

Re: High Power High Speed switches

02/16/2007 10:28 AM

The IR2121 is a driver made especially to drive MOSFET transistors, and that's what I would recommend (or something similar) if you can fit it into your circuit. However, an ACT type CMOS logic driver that can source and sink 24 mA would probably work as well. Another easy option would be an OPAMP with a high current output, or discrete transitors (MOSFET or bipolar) in a high current (~100-500ma) push-pull configuration.

Again, the higher the drive current, the lower the power dissipation will be in your MOSFET - you want the MOSFET to be either all the way on (i.e., minimum drain-to-source resistance) or all the way off (i.e., no drain-to-source current). Where you burn up power is in the transitions from on to off, and off to on, so you want the transition time to be a short as possible, and that means you want as much drive current as possible. (An exception to this would be if you were driving a highly inductive load - in this case you might want a slower transition.) Remember, the more power you dissipate in the switch, relative to it's capacity, the less reliable it will be.


International Rectifier has some application notes on driving MOSFETs on their website.

Register to Reply
Anonymous Poster
#7

Re: High Power High Speed switches

02/16/2007 11:03 AM

A MOSFET is the way to go. What kind of inductance will the FET see? Is the Source leg tied to a ground plane or other low inductive path? This will become critical when trying to turn the transistor on or off quickly.

The FET driver must be tied as close as possible to the gate and source legs of the transistor. Any inductance in either side of the loop will cause ringing of the Vgs and possibly turn the transistor on or off unintentionally. The resulting power dissipation will destroy the part.

The drain may need to be snubbed in order to absorb the inductive "kick" when interrupting the current. There area variety of lossy passive snubber designs and also lossless active snubber designs that can accomplish this.

FETs can be paralleled to increase the amount of load to be switched. Gate resistance will need to be added in series with each gate lead if they are sharing the same driver source. Otherwise, it's possible for the individual cells within the MOSFETs to start oscillating and the devices will fail.

If the MOSFET is floating instead of being tied to a fixed ground then a gate drive transformer will be required. (more headaches)

Register to Reply
Register to Reply 7 comments
Copy to Clipboard

Users who posted comments:

Anonymous Poster (1); bhankiii (3); gayatrikumar_1 (1); Steve (2)

Previous in Forum: Control of home appliances through mobile phone.   Next in Forum: Potential Divider
You might be interested in: Switch Boots, Audio Switch Matrices

Advertisement