12v to 450v+ Boost Converter / Projects / Forums

Forums

4hv.org :: Forums :: Projects

« Previous topic | Next topic »

12v to 450v+ Boost Converter

1 2 3 4 next last

Move Thread

LAN_403

aahz

Fri Apr 07 2006, 10:13AM

Registered Member #186 Joined: Thu Feb 16 2006, 07:22AM
Location:
Posts: 42

Newest Edit: This is a pretty terrible converter, I'd highly recommend using a pre-packaged switching converter regulator (Linear Technology has a pretty amazing line of regulators

,C1,C1003,C1042,C1031) instead of this piece of garbage. =)

Edit: More random junk added. Now has a switch and led for power, and a separate switch and led for charging. Schematic has been updated again.

Yet another edit: Added in a few misc caps to keep it from constantly humming to keep the voltage at the right spot.... instead, it chirps every few seconds. The schematic pic has been modified.

Edit: Soldered together my newer boost converter. After a few screwups(Putting IGBT in backwards, a short, and a few other little mishaps), I managed to get it working. Charges 4 450v 470uf caps in about nine seconds, stops at around 440v, and holds it there. Oh, and the LED comes on when it is done charging.

My hdieous soldering skills:

Size comparison: Old/new

Here's what I'm mounting it in:

Schematic:

24V input
50% duty cycle
~6khz operating F (I think)

OLD:
Some of you may remember a few questions on the boards a while ago, concerning boost converters and half-bridges. Well, I'm well on my way to completing my coilgun design. However, everything but the boost converter is still in test phase. Here is the schematic, part listings, where I ordered my parts, and total price of my boost converter.

66.6% duty cycle
1.5khz operating frequency
~17W output

Price / Qty / Component
AllElectronics.com
0.05 1 1M 1/4W resistor
0.05 1 100 1/4W resistor
0.10 1 10uf cap
0.10 2 4.7k 1/4W Resistors
0.12 1 .068uf cap
0.15 1 pc-mount switch
0.15 1 47uf cap
0.40 2 8pin IC sockets
1.00 1 Pc board
1.40 2 2pin connectors, with headers

Mouser.com
0.33 1 TS555CN
0.96 1 FFPF30U60S
2.78 1 680uH toroid

Digikey.com
3.83 1 HGTP20N60A4

Total: $11.42

This will charge a single 470uf 450v capacitor to 450v within 3.5 seconds (The charger takes about 1.5 to 2 seconds to transition down to its optimal frequency due to the large capacitance value of C1, once there, it zooms) and five 470uf 450v capacitors within 14 seconds. I recommend a T0-220 heat sink for the IGBT if you plan on using it to charge for more than 25 seconds, and perhaps a fan if using it continuously.

Charging components:
Allelectronics.com
2.25 470uf 450v snap-in capacitor
12.00 12v 2.3ah SLA battery

WaveRider

Fri Apr 07 2006, 10:17AM

Registered Member #29 Joined: Fri Feb 03 2006, 09:00AM
Location: Hasselt, Belgium
Posts: 500

I love these simple useful circuits.. Great work!

EDIT: What kind of conversion efficiency are you getting? Are you thinking of adding a regulating loop?

Steve Ward

Fri Apr 07 2006, 07:38PM

Registered Member #146 Joined: Sun Feb 12 2006, 04:21AM
Location: Austin Tx
Posts: 1055

I would get rid of the 100 ohm gate resistor (and maybe use a gate driver) for added efficiency. In a well designed converter, the losses should be about an even split between switching loss and conduction loss... im betting almost all your heating is due to switching losses because of the *slow* transition times with that 100 ohms. Be warned... faster switching times are likely to increase voltage spikes across the IGBT, so better keep your wiring between IGBT, diode, and cap short (add a small film cap right after the diode, connected to IGBT emitter to catch these fast transients... you should really have this no matter what).

aahz

Sat Apr 08 2006, 05:23PM

Registered Member #186 Joined: Thu Feb 16 2006, 07:22AM
Location:
Posts: 42

Steve Ward wrote ...

I would get rid of the 100 ohm gate resistor (and maybe use a gate driver) for added efficiency. In a well designed converter, the losses should be about an even split between switching loss and conduction loss... im betting almost all your heating is due to switching losses because of the *slow* transition times with that 100 ohms. Be warned... faster switching times are likely to increase voltage spikes across the IGBT, so better keep your wiring between IGBT, diode, and cap short (add a small film cap right after the diode, connected to IGBT emitter to catch these fast transients... you should really have this no matter what).

I've tried both of these. No gate resistor does nothing (it seems, at least). Charge time difference is unnoticable, with apparently the same amount of heat generated. A cap from the negative side of the diode to ground seems to slow the charge time... across the inductor as suggested in a different thread significantly slows charging. (This is a .01uf cap by the way)

WaveRider wrote ...

I love these simple useful circuits.. Great work!

EDIT: What kind of conversion efficiency are you getting? Are you thinking of adding a regulating loop?

Right now, simplicity. Once I get my half-bridge working, I'll start adding in conveniences. For now I'll just monitor it with mutimeter, and turn it off when the voltage gets to the right spot. As far as efficiency, it was reading about 1.7 amps input at 12v. When I tried to measure output, I got some crazy reading of 15-18 amps.... no way that can be right.

Anyway, it charges ~235j worth of capacitors in 13.5-14 seconds... so that would be about 17 watts output. With 1.7 amps at 12v, that's 20.4 watts input. So, *roughly* 83% efficiency.

thom

Sun Apr 16 2006, 06:53AM

Registered Member #239 Joined: Thu Feb 23 2006, 03:15AM
Location: canada
Posts: 23

Copied your design and built the circuit today, it works nicely.

I was hitting 700V+ at ~8V in, and killed a couple IBGT's (20N60B3, lower current rating)... Didnt realize they were so expensive.

I agree with your heatsink recommendation as well.

Mike

Mon Apr 17 2006, 05:57AM

Registered Member #58 Joined: Thu Feb 09 2006, 05:40AM
Location: Tri-Cities, Washington, US
Posts: 317

if it read 15-18 amps, i assume you just put your amp meter leads on plus and minus?
You need to have a load on it to put the multimeter leads in series with.
Mike

aahz

Mon Apr 17 2006, 08:50AM

Registered Member #186 Joined: Thu Feb 16 2006, 07:22AM
Location:
Posts: 42

The meter was put in series with the caps I was charging.

Quantum Singularity

Fri Apr 21 2006, 07:49PM

Registered Member #158 Joined: Sun Feb 12 2006, 09:53PM
Location: Central Ohio
Posts: 282

Hmmm would most amp meters be sensative to noise/spikes? That could probably throw off the reading?

aahz

Fri Apr 21 2006, 10:23PM

Registered Member #186 Joined: Thu Feb 16 2006, 07:22AM
Location:
Posts: 42

Haha. More than likely it was something along those lines. I have a $4 dmm.

Anyway, I'm made a couple of changes to the design... 24v instead of 12, smaller inductor... and planning on putting a comparator in to make it self-stopping.

Edit: All is working well. It's tuned it to reduce IGBT heating, while still maintaining a decent charge time. (Still charging ~235j in about 13 seconds). Added that comparator I was talking about... so it stops at around 440V. I'll probably solder it up sometime tomorrow, and post the results.

Edit again: Here's the schematic I'm using currently. Charges 235j in about 12 seconds, no heat felt from the IGBT.

aahz

Mon May 01 2006, 03:11PM

Registered Member #186 Joined: Thu Feb 16 2006, 07:22AM
Location:
Posts: 42

I've soldered it together, everything's working nicely. I've updated the top post with pics and the schematic. Now, time start working on my coil-gun logic.

Thanks a ton guys, for helping me with this.

1 2 3 4 next last

Moderator(s): Chris Russell, Noelle, Alex, Tesladownunder, Dave Marshall, Dave Billington, Bjørn, Steve Conner, Wolfram, Kizmo, Mads Barnkob