New to Coilguns - Coilgun Buildlog (Charger apparently fixed) / Electromagnetic Projectile Accelerators / Forums

Forums

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

« Previous topic | Next topic »

New to Coilguns - Coilgun Buildlog (Charger apparently fixed)

first previous 12 13 14 15 next last

Move Thread

LAN_403

Artikbot

Sat Oct 30 2010, 09:55PM

Registered Member #3247 Joined: Mon Sept 27 2010, 09:42AM
Location: Spain
Posts: 137

Yeah, I realised about that after burning two pots... One just shorted itself, the another one (this last one) caught spontaneous fire as if it was a magnesium stripe lol!

1M makes the small pot to be about 15.8k, leading to a wattage across it of not even .1W. Last burnt pot had a wattage of 1.2W across it, I believe it's rated at half watt or even less. It's a regular carbon linear pot, those that can be bought for a few cents.

I'll see if I have any 1meg or something similar fixed resistor lying around, measure it for more accurate adjustement and put a 100k pot to serve as the small resistor (I don't have any values between 10k and 100k readily available, and I don't think there are any 10k left.. I burnt both I had lol)

I feel so fulfilled ^^

Next stage, the triac triggering circuit! Although that doesn't seem to hide much trouble. Feed thru input 1, connect input 2 and gate through a button and put a HUGE fixed resistor before gate.

And then... Optical triggering and multistage, there we go!

Now a small reflexion... If I had done all this simple math previously, the potentiometer values, the inductor guesses, the voltage regulator circuit and this stuff... I could've possibly avoided burning two potentiometers and spending two weeks heating my bwainz to make this thing work.

Laziness is a bad friend, guys xD

Thanks!

Edit: I found an 1/8 1M resistor, real value of 972kOhm. Now time to fine tune the pot to 15.5k

And this is the burnt pot. I guess the copper brush makes a too light contact, and the current has heated up the carbon until the point it has burnt.

Dsc00069o

[Edit: Image size]

ScotchTapeLord

Sun Oct 31 2010, 02:40AM

Registered Member #1875 Joined: Sun Dec 21 2008, 06:36PM
Location:
Posts: 635

Potentiometers are generally pretty wimpy... but then again, anything over a half-watt is a bit much for the logic part of any circuit.

With regards to duty cycle, a larger one will build up more current. This does put more stress on the MOSFET and decrease efficiency, as will create shorter charge times.
Anything that increases the power in the system will do both of these: Lower efficiency and lower charging time. One of those is bad and one of those is good. You just have to remember why you're doing this whole thing: lower charging time. If you want maximum efficiency and minimum charging time and as little stress as possible on the MOSFET, you'd have to take all the work you've done so far, put it in a closet where it'll be safe, and series up enough batteries to get the desired output voltage, and put that directly across your capacitor bank to charge it. Your MOSFET will never get even above room temperature, because you aren't using it, and your capacitor bank will charge as quickly as the battery lets it. ;)

By making a circuit like this, you've chosen to trade in some electrical efficiency for convenience and usefulness. You just have to find a balance between power and efficiency, as they work against each other. Three seconds isn't bad at all. Now you can experiment with taking a few windings off the inductor at a time and seeing how that changes things.

quote:

"6: One second later, that same potentiometer spontaneously burnt with a flame of about 3cm high.
7: Then I cut the power"

You are now officially a 4hver.

Artikbot

Sun Oct 31 2010, 09:42AM

Registered Member #3247 Joined: Mon Sept 27 2010, 09:42AM
Location: Spain
Posts: 137

Well... This time it's the big resistor the one that has exploded. It's one of that very very very tiny ones, must be rated at 1/16and be f*cked up, because I drove 0.0016W (math doesn't lie) trough it and has gone BAM (with a pretty loud explosion I can say).

According to the circuit, well, its main goal was avoiding an uber amount of batteries in series (obvious xD) and providing a relatively small and lightweight, yet powerful, solution of charging my banks.

It's not very small (it's mounted on a 60x70 perfboard) but it's small enough, compared to the space photoflash boards took in order to charge at a comparable speed (12 boards to be precise).

I guess my inductor is just excessive, I can take windings off when everything works in a stable way, that's for sure. It has three times the original windings (on a thicker wire aswell) and a toroid with double the radius and double the thickness.

ScotchTapeLord wrote ...

You are now officially a 4hver

Sir, that's an honour for me

GhostNull

Sun Oct 31 2010, 09:52AM

Registered Member #2648 Joined: Sun Jan 24 2010, 12:45PM
Location: Australia
Posts: 291

Maybe you it drove it beyond the voltage rating??

Artikbot

Sun Oct 31 2010, 01:57PM

Registered Member #3247 Joined: Mon Sept 27 2010, 09:42AM
Location: Spain
Posts: 137

Don't know what's it rated for (salvaged from an old power supply), but yes, it's a possiblity. It was driven about 200V through.

Some photos about the monster (the resistor that burnt is that microscopic one right in front of the mosfet heatsink):

[Edit: Image size]

ScotchTapeLord

Sun Oct 31 2010, 03:50PM

Registered Member #1875 Joined: Sun Dec 21 2008, 06:36PM
Location:
Posts: 635

Um... what is this resistor connected to? Is this still the one going to the op amp? Can you direct me to the schematic you are using and point out which resistor this is?

Artikbot

Sun Oct 31 2010, 04:13PM

Registered Member #3247 Joined: Mon Sept 27 2010, 09:42AM
Location: Spain
Posts: 137

Oh yes, sure.

That resistor has been burnt and replaced by two carbon composite rated at 1W in series, giving a total of 1.6MOhm output, therefore the pot has been risen to 25.3kOhm to get an output lying at 315V (that allows system a 15V variation for the ripple not to damage the capacitors).

Those resistors (pot+fixed) are the ones going from caps to inverting, to stop the 555.

This is my current schematic (yenka image, sorry >.<):

Edit!

No Zener is present, instead of it, there's a 7805 IC providing precise 5V sample.

Edit2: Awkward!! My resistors have different values when measured from tolerance ring to color ring that when measured from color strips to tolerance ring!!
To be precise, 1.6MOhm and 800kOhm. I never encountered this O.O

ScotchTapeLord

Sun Oct 31 2010, 05:25PM

Registered Member #1875 Joined: Sun Dec 21 2008, 06:36PM
Location:
Posts: 635

How can you measure a resistor from one side to the other? Are you implying that you have a polarity sensitive resistor?

I would use a different resistor!

Try to get a voltage measurement across the resistors that get hot. That should tell us how much power they are really dissipating, then maybe we can diagnose the problem.

Artikbot

Sun Oct 31 2010, 05:40PM

Registered Member #3247 Joined: Mon Sept 27 2010, 09:42AM
Location: Spain
Posts: 137

I don't know what does it imply, but I know that they're changing values if I measure it using reversed polarity.

I've found that one resistor changes its value from 880kOhm to 1.2MOhm when polarity is reversed. Don't know what happens if I apply power, but I can say it comes from a CRT TV from the '96 and it worked until today. Well, hasn't been plugged for almost two years now, but it worked flawlessly. And it's taken from the tube interconnection board, so I doubt it's broken, otherwise the TV wouldn't have worked properly.

I'm powering it again and see if it drops, bangs again or what in God's sake happens :P

Btw, I've found a "puffed" capacitor in the bank. It's not broken, or leaking, or anything, it just has the top cover slightly puffed. I'm taking it out just to ensure it's not the cause of any problems.

Edit again:

It doesn't explode anymore! Removing the capacitor and inverting the resistors worked :)

I now have to apply continued power and see if it cuts, bangs, or what does it do lol.

Edit 3: Charged full 3.6mF bank to 260V in about 20 seconds, MOSFET got hot, but not enough to hurt my finger on touch. Diode got hot as well, the rest of the circuit was cool.

Will start to remove windings as it seems it's limiting current too much. When I cut power voltage was rising in an 8V per second scale. The good of it, is that the bank reaches 150V in less than a second.

Also note that a bleeding resistor was connected across the leads, 150K 1W fixed resistor. Maybe that was slowing charge times aswell. Will try without it, but I don't think it's safe to have the capacitors charged without any bleeder for safety.

Will take out like 5 turns, power circuit on and try to reach voltage threshold to see if it cuts power itself or what does it do.

Edit 4: Something has gone bonkers. During capacitors self-discharge (battery disconnected), I heard a beep from the charger as if it was charging, but during a few milliseconds, and now the circuit won't even start to charge, no beeping from the inductor, no voltage rising... I think something has burnt...

I guess I'll have to re-check all components just in case what has burnt isn't a resistor from the op-amp and isn't letting the 555 oscillate.

ScotchTapeLord

Mon Nov 01 2010, 10:43PM

Registered Member #1875 Joined: Sun Dec 21 2008, 06:36PM
Location:
Posts: 635

I think you should use a higher resistance for bleeding, 150k sounds a little low. Motor run capacitors for 220V motors use 1 Meg from what I've seen. At 260V it will be dissipating about half a watt.

It turned on while it was discharging? I have a theory.

Your op amp tells the 555 to turn off when it's full right? That would mean it tells it to turn on when the capacitor's voltage is too low. I'm guessing that's what happened. Though, this would only happen if the op-amp and 555 were connected to a power source. My guess is that your decoupling was keeping your logic circuit alive even after the battery was disconnected. Perhaps instead of cutting the power off to turn it off, you could put a resistor between the comparator's output and the reset pin and put a switch directly from the reset pin to ground. The advantage of this is that it guarantees that no output from the 555 will take place when the switch is off, but the circuit will always be drawing a very tiny amount of current, even when switched off. There's probably a better way. I'll think about it.

first previous 12 13 14 15 next last

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