Go Back   Home > Forums > Amplifiers > Power Supplies

Please consider donating to help us continue to serve you.

Ads on/off / Custom Title / More PMs / More album space / Advanced printing & mass image saving
Reply
 
Thread Tools Search this Thread
Old 11th July 2009, 01:19 PM   #1
kvasefi is offline kvasefi  United States
diyAudio Member
 
Join Date: May 2009
Default 1kw+ SMPS...blowing FETS at full voltage

Hi there...
I've been working on a project for an SMPS in the KW range for some months now and I keep running into different problems. Much research in books and online solved some of them, but I'm stuck now and I thought the expertise here might give some insight. I studied EE in college but most of it was in digital design, so power circuits is more self-taught. I've done some linear power supply design and also a lighter SMPS with simple topology, but this high power and different design has been a beast. I'm also working on a PFC stage which I will include (although I haven't gotten that to work yet...any input would be greatly appreciated if you get a chance).

Any insight on my design is good. If you have a design example for this power range or a link to one, I'd be happy to try that, although I haven't found any through much searching.

My application:
Charging very large capacitors (.8 F x 3 in parallel) in a short time (<1s) up to 24 or 48V. To do that, I'd like to source up to 30-40A current.

Here are my specs:
Input voltage: 110Vac through a voltage doubler and rectifier (not shown), so 340 Vdc onto DC+ and DC- (as labeled on schematic)
Output voltage: 24Vdc
Max current: 40A

My problem:
I manage input Vac using a variac, and then either load it with an uncharged cap (~2000 uF) or a resistor 10R. Either way, when I increase input voltage to near the max (ie. normal US 110Vac), I start to blow the FETS. They are heatsinked and don't heat up too much, just suddenly I blow the fuse and testing shows that 1-4 of the FETs are now shorted on all 3 pins.
I have a cheap o-scope can it's ground line is grounded to earth, so I can't watch what's happening between the gate transformers and my output (the voltage doubler sends DC- below 0 and sparks flew when I tried before I read the o-scope manual).

I'm using a full-bridge topology with some modifications to improve the MOSFET switching and control noise on the gate. I don't know if those are causing problems. This includes the RC snubbers across the FETS and the transistor between the FET gate and source to drop Vgs quicker on the switch off (ie. bypass the transformer inductance)

Thanks in advance!
-Kevin


PS. I know my schematic is all backwards...sry this is what happens when you start a design with no knowledge and then try to add to it for months.
Attached Images
File Type: png smps2.png (65.5 KB, 1081 views)
  Reply With Quote
Old 11th July 2009, 02:19 PM   #2
infinia is offline infinia  United States
diyAudio Member
 
infinia's Avatar
 
Join Date: May 2005
Location: Californication
Default Where to begin

Hi
You need an 1:1 isolation transformer ie floating for the DUT.
The control loop/s... look at using current mode control (startup), and loop stability is key for large capacitive loads. You need to pick the right topology.
Magnetic design and switching freq? Try lower power or lower freq first.
Paralleling output rectifiers is problematic. Snubbers use 2 on the primary at the transformer not the fets. Snubbers across the output rectifiers.
This stuff is very complex for a beginner. You need a lot of guidance or time to complete this project with any hope of success.
__________________
like four million tons of hydrogen exploding on the sun
like the whisper of the termites building castles in the dust
  Reply With Quote
Old 11th July 2009, 09:44 PM   #3
infinia is offline infinia  United States
diyAudio Member
 
infinia's Avatar
 
Join Date: May 2005
Location: Californication
Default Re: 1kw+ SMPS...blowing FETS at full voltage

Quote:
My application:
Charging very large capacitors (.8 F x 3 in parallel) in a short time (<1s) up to 24 or 48V. To do that, I'd like to source up to 30-40A current.
Why does your application need such a large capacitance? It would be better to let the supply itself provide the longer term currents?
__________________
like four million tons of hydrogen exploding on the sun
like the whisper of the termites building castles in the dust
  Reply With Quote
Old 14th July 2009, 07:31 AM   #4
Eva is offline Eva  Spain
diyAudio Member
 
Eva's Avatar
 
Join Date: Oct 2003
Location: Near the sea
Send a message via MSN to Eva
Are you going to fed laser diodes with those capacitors?
__________________
I use to feel like the small child in The Emperor's New Clothes tale
  Reply With Quote
Old 14th July 2009, 01:06 PM   #5
star882 is offline star882  United States
diyAudio Member
 
star882's Avatar
 
Join Date: Mar 2007
It sounds like a spot welder application to me.
__________________
"Fully on MOSFET = closed switch, Fully off MOSFET = open switch, Half on MOSFET = poor imitation of Tiffany Yep." - also applies to IGBTs!
  Reply With Quote
Old 14th July 2009, 11:53 PM   #6
kvasefi is offline kvasefi  United States
diyAudio Member
 
Join Date: May 2009
infinia - thanks for the input. I'll keep working on it and see what I can do with modifying the design.

It is for a spot welding device, which I know isn't audio, but the high power needed made this a good forum in my mind. The high capacitance is for high energy in a short discharge, which ends up being many times more amps than an SMPS could do.

As for the laser diodes - not sure what to do with that, I haven't run into it before.

I know it's a tough project, we'll see how it goes. Thanks for the help!
  Reply With Quote
Old 15th July 2009, 04:11 AM   #7
Eva is offline Eva  Spain
diyAudio Member
 
Eva's Avatar
 
Join Date: Oct 2003
Location: Near the sea
Send a message via MSN to Eva
I asked because professional epilation lasers based in 808nm LEDs use 48V too, and up to 150A pulses, and the capacitor bank method too. These are stacks of massive water-cooled LEDs, not a joke
__________________
I use to feel like the small child in The Emperor's New Clothes tale
  Reply With Quote
Old 15th July 2009, 09:39 PM   #8
TechGuy is offline TechGuy  United States
diyAudio Member
 
Join Date: Apr 2006
Location: Connecticut
Default Re: 1kw+ SMPS...blowing FETS at full voltage

Quote:
Originally posted by kvasefi
Charging very large capacitors (.8 F x 3 in parallel) in a short time (<1s) up to 24 or 48V. To do that, I'd like to source up to 30-40A current.
.
Using a small set of very large capacitors is not a good idea for a pulse discharge system. Becareful to note the max ripple current of the caps you are using. Excessive current discharge will destroy the caps as the current rush vaporizes the thin gauge metal film near the cap terminals. Imagine pulsing a large current through a long, thin peice of aluminim foil. The current will vaporize the metal and burn out like a blown fuse.

You need to either use a large set of caps or a set of "pulse" capacitors (Big $$$) that are designed for large current pulses. Super caps are not designed for Pulse discharging. BTW when the capacitors fail, the go out with a flash and a bang. When the foil vaporizes it creates rapid high pressures inside of the can causing the can to fire up at rapid velocity (enough to knock you out if it cap is large in size). The foil has a 50% of bursting into flames and the caustic electrolytic can cause chemical burns if it sprays on your skin. Have fire extinguisher handy, and wear eye protection when you test.

The best way to charge a cap bank is with a current limiting power supply that limits the amount of power. One simple option is to place an AC cap in series with the transformer primary. The Cap will limit the amount of current on each half cycle. The cap does not need to be large and you can increase the current load by increasing the switch frequency even if the cap has a fixed capacitance.

A more complex method is to use PWM current control starting off with a very low duty cycle and increasing the duty cycle as the caps near full charge. You could either do this on the primary side or as an additional circuit on the secondary side.

You can also use SMPS controllers with built-in current limiting. They usually use a current sensing resistor to monitor the current.
  Reply With Quote
Old 16th July 2009, 12:51 AM   #9
infinia is offline infinia  United States
diyAudio Member
 
infinia's Avatar
 
Join Date: May 2005
Location: Californication
Hi TechGuy
Good catch about using banks of pulse caps instead of 3 super caps.
A supply like this is going to have to use current mode control for sure. The best way to sense current esp big converters is using a current transformer. ( too much common mode and layout problems with using a resistor.)
AFAIK An AC cap in series with the primary does nothing to limit current in a pulse to pulse basis. It's used to balance voltseconds to prevent the saturation of a (half bridge)transformer core due to non symetrical pulses over many cycles.
__________________
like four million tons of hydrogen exploding on the sun
like the whisper of the termites building castles in the dust
  Reply With Quote
Old 16th July 2009, 03:32 AM   #10
star882 is offline star882  United States
diyAudio Member
 
star882's Avatar
 
Join Date: Mar 2007
What about use a bank of higher voltage capacitors and a high frequency transformer driven by an IGBT bridge to step down the voltage? Spot welding doesn't care if it's AC or DC, so the secondary winding can just directly connect to the electrodes.
__________________
"Fully on MOSFET = closed switch, Fully off MOSFET = open switch, Half on MOSFET = poor imitation of Tiffany Yep." - also applies to IGBTs!
  Reply With Quote

Reply


Hide this!Advertise here!
Thread Tools Search this Thread
Search this Thread:

Advanced Search

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off


Similar Threads
Thread Thread Starter Forum Replies Last Post
What about FETs for Cellular Base Stations in Audio Cascode Voltage Gain Stage (line tiefbassuebertr Solid State 8 23rd August 2009 02:53 AM
power supply fets keep blowing Clipped Car Audio 23 22nd January 2006 06:48 AM
SMPS - Why are my FETs cooking? Bosium Car Audio 42 17th August 2005 03:13 PM
high voltage fets kaukasion Solid State 11 25th December 2003 06:20 PM


New To Site? Need Help?

All times are GMT. The time now is 11:42 PM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2014 DragonByte Technologies Ltd.
Copyright 1999-2014 diyAudio

Content Relevant URLs by vBSEO 3.3.2