Charge regulator / Electromagnetic Projectile Accelerators / Forums

Forums

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

« Previous topic | Next topic »

Charge regulator

Move Thread

LAN_403

Killa-X

Sun Oct 10 2010, 06:35AM

Registered Member #1643 Joined: Mon Aug 18 2008, 06:10PM
Location:
Posts: 1039

I wish to learn more on this circuit as my experience has gone off hand.

Yes, this is ripped from Uzzors 50W capacitor charger. I made this, using the same parts listed. I used a 10K pot instead of a 1K pot. I noticed at the higher ohms, i could have it stop at 25V.. 40V... but after 40V it went to some high voltage that didnt stop at even 400V. and the pot wasnt even half turned to 5K.

But, besides my issues. I wanted to know if it's possible to get this circuit to be a 10V - 450V automatic stop. So I can make a variac-like dial scale, and pre-set it before charge. so I can charge it to a given voltage while being far away from it.

I plan to enjoy these capacitors in my 1KJ bank that are surplussed. Old. I want to hook them to random items, and discharge 1KJ into the item, causing explosions. Understanding how dangerous that can be, i have a SCR that will have a very long gate wire. I need to have a variable settable voltage stop, so I can know it will stop at that voltage while im a couple hundred feet down range, then when it hits the voltage, it will stop, cut charge, and emit a tone. Then I know it's ready, hit the SCR trigger, Boom.

Safety is priority here, so I plan to stand back! Please help and suggest, Thanks!

Why in projectile accelerators? I'm also working on a 700J coilgun, and looking for the same exact type of system to be used. Dial-scale voltage stop.

GhostNull

Sun Oct 10 2010, 08:49AM

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

This part of the circuit, as you already know, turns off the ZVS circuit once the target voltage has been reached.

It consists a of voltage comparator comparing a reference voltage to the voltage of the capacitor bank. The inverting input pin is feed a reference voltage given by the TL431. The 1M, 5.6K resistors and 1K pot act as an adjustable voltage divider to the non-inverting input. It variably divides the voltage from the capacitor bank.

When the voltage given from the capacitor bank, divided by the resistors is greater than the reference voltage, the comparator outputs high, turning off the ZVS, stopping charging.

I'm not definitely sure about the 1nF and 2.2K and 1k Resistors

Killa-X

Sun Oct 10 2010, 09:00AM

Registered Member #1643 Joined: Mon Aug 18 2008, 06:10PM
Location:
Posts: 1039

Well, from what I got out of this circuit, a big enough pot SHOULD allow me to do 10V - 450V. And mark lines to make a dial scale. If its on 20V, it will charge to 20V and stop. However, if its on 400V, and it charges - stops at 400V, and i turn the pot down to 20V, it will fry/explode the LM393 because the voltage for 400V was safe for the LM393, but when you drop to 20V, its a couple hundred volts or so..and it dies.

Technically speaking, pin 1 and 3 of the TL431 can be placed any direction correct? the anode is always centered, and the reference/cathode combine as one...so pins 1 and 3 can be 3 and 1.

I'll try more tomorrow...Just oddly today it refused to work right.

GhostNull

Sun Oct 10 2010, 09:13AM

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

Do the math with the voltage divider, that will show you everything. You do know how voltage divider circuits work right?
Voltage Divider Circuits:

I'm not sure what you mean when you are talking about when you say "pin 1 and 3 of the TL431 can be placed any direction" but they are common, if that's what you are trying to say.

I think the 1nF capacitor is to prevent premature triggering/switching of the comparator and the 2.2K and 1K on the comparator are feed back so it doesn't keep switching back and forth once the target voltage is reached

Killa-X

Sun Oct 10 2010, 05:12PM

Registered Member #1643 Joined: Mon Aug 18 2008, 06:10PM
Location:
Posts: 1039

Well heres the thing. Correct me if im wrong.

The schematic above, lets say you have it set to stop at 130V. If its under 130V, the 2N3904 has power applied to the gate, thus making it charge. However, if it's at 130V, the gate turns off. stopping the charge.

With mine, I got 12V, and i have a pullup 1K resistor from +V to the output of the LM393. When I'm below 130V, its 0.68V. When I'm at 130V, its 12V. Shouldnt it be the opposite? I need it to keep a flow going, but once it hits 130V, cut the flow.

uzzors2k

Sun Oct 10 2010, 07:15PM

Registered Member #95 Joined: Thu Feb 09 2006, 04:57PM
Location: Norway
Posts: 1308

The 2.2k resistor limits the current so the TL431 can regulate. Without it the TL431 would try to load your stiff 12V supply down to 2.5V. I think you know who would win the struggle. The 1k resistor on the LM393 output is simply a pull-up resistor. If you check the LM393 datasheet you see that it has open-collector outputs, meaning it can only sink current, not source it. The 1nF cap filters the signal. I'm not sure how much good it does considering how slow the voltage on a capacitor bank tends to change, but it certainly doesn't hurt. If you want a huge range, placing a zener on the LM393 input is probably a good idea to keep the voltage there within spec.

Killa-X: You've measured the correct voltages but misunderstood how the circuit functions. As long as the switch is open, or the 2N3904 is off (open-circuit), the diodes are kept at 12V and allow the IRFP250 gates to function as normal. Once the switch is closed or the 2N3904 is on (closed circuit) the diodes are pinned to ground and hold the IRFP250 gates at 0V. If this circuit were used to control the current directly in some load, it would have to be the opposite like you said.

Btw, 2N3904's have bases, not gates.

Killa-X

Mon Oct 11 2010, 03:29AM

Registered Member #1643 Joined: Mon Aug 18 2008, 06:10PM
Location:
Posts: 1039

Thanks for the explanation Uzzors, helped. My last test, using the diagram but with a 50K pot (just what I had off hand) I was able to do 130v and up. Using a solenoid custom DYI PNP style switch, I was able to make it regulate and maintain 130..140..150..200..etc

So rigt now, I still have the 1K pullup. It gives the gate of my MOSFET below 1V. Once it hits the set voltage, it goes 12V. So really, given a PNP MOSFET will solve my problem easily. This way, when it gets 12V, the charge is cut. Under 12v it charges. Also is a nice system to keep the bank at the set voltage. Usually my tiny cap goes from 130 to 0 in 10 sec, this keeps it 139 solid

uzzors2k

Mon Oct 11 2010, 11:48AM

Registered Member #95 Joined: Thu Feb 09 2006, 04:57PM
Location: Norway
Posts: 1308

Simply switch the inverting and non-inverting input, then the output will be inverted too.

Killa-X

Tue Oct 12 2010, 02:25AM

Registered Member #1643 Joined: Mon Aug 18 2008, 06:10PM
Location:
Posts: 1039

Uzzors wrote ...

Simply switch the inverting and non-inverting input, then the output will be inverted too.

Ok, I switched pins 1 and 2. So heres the deal.

First, I set it to 40V stop. At 0V, it outputs 12V. Relay allows capacitors to charge. Once at 40V, charge is maintained to 40V, and the relay spazzes out. Fast rapid clicking. So, Simple....Put a pull down resistor on the mosfet because It's having sensitive gate issues. When I touch the pot with my hands, the relay acts differently, but still rapid clicking.

So I put an equal value (and tried 10K and 1K) resistor as a pulldown. Still, had sensitive gate issues. Not sure why...

Now. The circuit DOES work. I got 40V - 468V range. When set to 240V, it stops at 240V, making a ton of relay clicking, but it MAINTAINS 240V. So, It's not a NEED to fix this rapid clicking, but I would like to so it's not so ANNOYING! Why relay? AC switching. I'm considering a triac...

Happy to see it works though, Thanks!

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