boost converter unstable / Electromagnetic Projectile Accelerators / Forums

Forums

4hv.org :: Forums :: Electromagnetic Projectile Accelerators

« Previous topic | Next topic »

boost converter unstable

1 2 3 next

Move Thread

LAN_403

loveHV

Wed Mar 04 2009, 04:23PM

Registered Member #1854 Joined: Wed Dec 10 2008, 06:09PM
Location: West-Flanders Belgium
Posts: 88

i recently start building a coilgun, kinda new at this.

but i started with my capacitor bank 2200uF 400V or about 160 joules
after i finished building that i builded my boostconverter

the first time i tried it was on a regular 400VDC foil cap

it worked fine, i could regulate the voltage with the comparator, i did experience some problems with the power supply though, my power supply could only supply 3amps so when i hooked it up to the boost converter the voltage dropped from 12V to 7V, so that was a little irritating.
but with some regulation i could get it at 400V

then i hooked up my capacitor bank to it, my voltage went up to 500V so i overcharged my caps wich could only handle 400V. so i regulated it but i can't go below 500V.

here's the schematic site:

but i didn't use the parts they use.

instead of an IRFP240 i used a IRFP450
and a MUR880 instead of an MUR810

also i didn't use a 125uH inducter, i experimented with some inductors and i used other one
with that i can regulate the voltage between 300 to 500VDC on a normal foil cap. but i can't get below 500V on my 400V capacitor bank.
is there anyway to regulate it over a wider voltage range.

any answers or advice is certainly welcome

KLH

Wed Mar 04 2009, 07:17PM

Registered Member #1819 Joined: Thu Nov 20 2008, 04:05PM
Location:
Posts: 137

loveHV wrote ...
also i didn't use a 125uH inducter, i experimented with some inductors and i used other one
with that i can regulate the voltage between 300 to 500VDC on a normal foil cap. but i can't get below 500V on my 400V capacitor bank.
is there anyway to regulate it over a wider voltage range.

The two semiconductor substitutions that you made probably are not affecting the operation of the circuit.

More specifics on the inductor (value, etc.) and how the capacitor bank is connected together would help. Also, what is your operating frequency? The inductor value, operating frequency, and capacitor bank size can make a big difference on how it charges.

loveHV

Wed Mar 04 2009, 07:59PM

Registered Member #1854 Joined: Wed Dec 10 2008, 06:09PM
Location: West-Flanders Belgium
Posts: 88

my operating frequency is 4 to 5 kHz (calculator)

my inductor value, i don't know. the only things that are written on it are

61111
05131
LSE
9810

thats all i know about my inductor

my capacitor bank is 2200uF by 400V

you said it could make a difference in "how" its charged

do you mean with how, how fast it charges, or at what voltage it will charge ??

rp181

Wed Mar 04 2009, 10:10PM

Registered Member #1062 Joined: Tue Oct 16 2007, 02:01AM
Location:
Posts: 1529

Both. If the inductor is mismatched, it could not "charge" completely, and give a low output. Too low inductance, and there will be dead time, so there is less output. Play with the frequency to find the best combo.

aonomus

Thu Mar 05 2009, 12:21AM

Registered Member #1497 Joined: Thu May 22 2008, 05:24AM
Location: Toronto, Ontario, Canada
Posts: 801

Well a few things...

1. I didn't even know that the schematic was accessible (it should be hidden in a directory I think.. or at least buried somewhere and non indexed?
2. As the input voltage sags, the comparator and 555 tend to act odd, I had to just max out the power supply to let it run properly.
3. 2200uF is under the size of the capacitor 'bank' I had (2 lytics) so it should be able to be charged.
4. The link to the actual inductor used:

and its datasheet:

5. The most striking modification that I'm wondering about now is your use of the MUR810, the way a boost converter works is the inductor will cause a voltage spike when the mosfet turns off. Once the voltage of the capacitors rise over 100V (the rating of the MUR810 is 8A 100V, the MUR860 is rated for 600V), some of that energy will avalanche through your mosfet depending on its rating. In my schematic I had IRFP240/250 written down, however using a larger voltage rating would be advisable too.

Thinking about the diode now too, if you exceed its rating, you may end up pushing 400V onto your 12V bus...

A major warning: you *will* need a good amount of capacitors on the input of your boost converter. Have several large (tens of thousands of uF) electrolytic capacitors, and several film capacitors to prevent any ripple voltage from working its way back into your PSU and frying it.

loveHV

Thu Mar 05 2009, 04:53PM

Registered Member #1854 Joined: Wed Dec 10 2008, 06:09PM
Location: West-Flanders Belgium
Posts: 88

ok but i think you get something wrong (or i misunderstood it):

im not using a MUR810 but a MUR880 rated 800V

im not using a IRFP240 or 250 because those are rated 200V, im using an IRFP450 those are rated 400V 14 amps

Also why must that schematic be hidden ??? it works

and for my supply filtering i use an electrolyc capacitor of 50V 24000uF, but no film caps, is that a neccecarity ? because i don't get much trouble with my supply its a power supply of
HQ-Power.

i measured the resistance of the voltage regulation Pot in function of the charge voltage and when the resistance of the Pot goes up the voltage to where he charges goes down so i think i need a Pot with a resistance of 50K or so,

i hope it works to get the voltage to 400V

and i will also try to expiriment a little with the frequency to find a good combo like rp181 said

KLH

Thu Mar 05 2009, 06:38PM

Registered Member #1819 Joined: Thu Nov 20 2008, 04:05PM
Location:
Posts: 137

loveHV wrote ...

im not using a IRFP240 or 250 because those are rated 200V, im using an IRFP450 those are rated 400V 14 amps

Also why must that schematic be hidden ??? it works

and for my supply filtering i use an electrolyc capacitor of 50V 24000uF, but no film caps, is that a neccecarity ? because i don't get much trouble with my supply its a power supply of
HQ-Power.

i measured the resistance of the voltage regulation Pot in function of the charge voltage and when the resistance of the Pot goes up the voltage to where he charges goes down so i think i need a Pot with a resistance of 50K or so,

i hope it works to get the voltage to 400V

and i will also try to expiriment a little with the frequency to find a good combo like rp181 said

My experiments with the boost converter circuit (along with simple analysis) say that you don't need a high voltage MOSFET; the diode is what stops the capacitor from discharging, since there is no way for it to discharge thorugh the MOSFET until the diode breaks down.

About the frequency / inductor value issue: Set the operating frequency and duty cycle so that after the switch turns off, the inductor stores a peak current that you specify. I recommend a 50% duty cycle, so that the inductor does not charge too little or, worse, oversaturate at startup. Use this RL time calculator here (http://ourworld.compuserve.com/homepages/Bill_Bowden/rl.htm) to find your time period that will charge your inductor up to your chosen peak current. The reciprocal of the time period divided by 2 will become your 50% duty cycle operating frequency.

I also recommend that, in the future, use inductors from J.W. Miller's 2300 toroidal inductor series. There is an inductor with a value of 10uH and a current rating of 20 A.

The small film capacitors on the converter input are necessary to keep HF from feeding back into your supply and possibly destroy it, as stated earlier.

loveHV

Fri Mar 13 2009, 08:45AM

Registered Member #1854 Joined: Wed Dec 10 2008, 06:09PM
Location: West-Flanders Belgium
Posts: 88

i expirimented with different kind of inductors and different frequency's and my comparator

i now get 405V over my 400V capacitor bank, that extra 5V doesn't do any harm to the capacitors so its not a problem

i still have a small question:

my capacitor bank mainly consists of filter capacitors of PSU's its 2200uF 400V, but the problem is: yesterday i shorted it out with just 2 small wires (gave a bang). but i noticed after i shorted it there was still a charge of about 100V on the capacitor bank, wont that be a problem with suckback???

Sulaiman

Fri Mar 13 2009, 09:23AM

Registered Member #162 Joined: Mon Feb 13 2006, 10:25AM
Location: United Kingdom
Posts: 3140

Actually you only partially discharged the capacitor,
the copper wire probably vapourised rapidly,
some current may continue briefly in the ionised remains of the wire,
then current flow ceases, leaving some charge remaining in the capacitor(s)

IF you'd shorted the cap with e.g. a screwdriver (who? me? never ! ;)
all available charge in the capacitor would be used
and the capacitor voltage would be zero,
but even then,
when you remove the short-circuit
the capacitor voltage will 'magically' start to rise,
because not all stored charge in an electrolytic is instantly available,
some takes quite a while to move.
CAN SOMEONE CLEAR MY MENTAL-BLOCK AND NAME THIS PHENOMENON PLEASE ?

None of this is significant for a coilgun though!
Whether or not you get 'suckback' is independent of the above.

Bjørn

Fri Mar 13 2009, 09:43AM

Registered Member #27 Joined: Fri Feb 03 2006, 02:20AM
Location: Hyperborea
Posts: 2058

Dielectric absorption is the name.

What is important is to never assume that a capacitor is discharged, there are many possible reasons why it might not be.

1 2 3 next

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