My school coilgun poject / Electromagnetic Projectile Accelerators / Forums

Forums

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

« Previous topic | Next topic »

My school coilgun poject

1 2 3 4 next last

Move Thread

LAN_403

Skynet

Tue Nov 12 2013, 02:47PM

Registered Member #11734 Joined: Thu Mar 21 2013, 08:44PM
Location: Brno Czech republic
Posts: 35

Hi all,
at first i apologize for my english (IÂ´m fom Czech republic)
So few weeks ago i started to work on my school project i gathered some big caps, IGBTs and SCRs. And last week i finished my triggering circuit. Also you can have questions on me

I would post to this thread progress of my wok

Skynet

Mon Jan 20 2014, 02:10PM

Registered Member #11734 Joined: Thu Mar 21 2013, 08:44PM
Location: Brno Czech republic
Posts: 35

A small update here.

1. I updated a half bridge switching of coil according to application notes

2. I created a logic that responds to output pulses from the optical gate. Pulses that are longer than 1ms are reduced to 1 ms and pulses that are shorter than 1ms leaves shorter. That prevents to damage of my IGBTS

3. I created a simulation with realistic parameters components, where I tried to simulate circuit when the current is switched.
C=4700uF
L=30uH
U=450V

Now there was only to conect everything together and make the test. It would by probably next week becauce I have exams this week.

The last photos are from the lab where I measured the total circuit delay and testing a new barrel.

Yandersen

Mon Jan 20 2014, 08:57PM

Registered Member #6944 Joined: Fri Sept 28 2012, 04:54PM
Location: Canada
Posts: 340

Finally some halfbridge is being made here. Good job, Skynet, I keep my fingers crossed for your success! :)

Skynet

Tue Jan 28 2014, 07:50PM

Registered Member #11734 Joined: Thu Mar 21 2013, 08:44PM
Location: Brno Czech republic
Posts: 35

Just a small photo update of progress

IÂ´ll probably test it tomorow

So today I made some tests...

At first at 150V with 25ms pulse it worked perfectly:

Then I increased the voltage to 250V. Everything looked good, but then IR2113 started smoking.

These parts were destroyed, and I do not know why. Could anyone advise me what could have happened.
Eegd

Yandersen

Sun Feb 02 2014, 03:45PM

Registered Member #6944 Joined: Fri Sept 28 2012, 04:54PM
Location: Canada
Posts: 340

1) The circuit like yours I see first time. As you did not specified the IC used, I thought that sch is some IC-specific one, because I do not clearly understand the purpose of R,Q1,Q2,D2. Please, post IC part number and explain the schematic. I guess, you are trying to hold Vs pin at the ground level until the shot. But because of R the closing time of Q1 will be delayed, so when High IGBT got opened, Q1 may still be somewhat conductive and may burn causing chain reaction for all other elements. Replace all this BS with a 100K resistor between ground and Vs.
2) The breadboard you use may let contacts shake, especially due to vibrations during the shot. That may have cause the problem. Use the soldering tool to connect all elements in the circuit.
3) 25ms is a way too long pulse. Drop it down to 10ms at most.
4) If you time the shot with a 555 or 556 timer IC, be aware that those devices are very sensitive - they may glitch and trigger due to any noise on the supply rails. I also suspect my 556 timers to react even on EMI caused by switching of the coils, but I may be wrong.

Skynet

Mon Feb 03 2014, 02:43PM

Registered Member #11734 Joined: Thu Mar 21 2013, 08:44PM
Location: Brno Czech republic
Posts: 35

I use IR2113 driver => http://www.irf.com/product-info/datasheets/data/ir2110.pdf

Scheme that I posted here is from Application notes that I found on the manufacturer's website on page 24 here => http://www.irf.com/technical-info/appnotes/an-978.pdf
Q1 and Q2 transisotrs are there for charging of bootstrap capacitor

I dont know if the 100K resistor from Vs to GND is good idea. It could cause difficulties for IC

I tinked about another reasons why it it blowed up, and I discovered that I used quite slow diodes P600M. So I think that could cause some voltage peaks when are IGBT switched. Datasheet => http://www.vishay.com/docs/88692/p600a.pdf or IC could be China shit ....

Finally, I drew the actual schematic with all now used components. Its working with bug above, and for more problems logic is unreliable sometimes it goes wrong and creates a very long pulse and capacitor is discharged completely with no react to optical gate.
L953

http://imageshack.com/a/img33/5761/l953.png

Thanks for reply

Yandersen

Mon Feb 03 2014, 09:50PM

Registered Member #6944 Joined: Fri Sept 28 2012, 04:54PM
Location: Canada
Posts: 340

The resistor between Vs and ground will work as good as your sophisticated charging circuit, but you also must add zener diode between Vs and Vb to prevent overcharging. The same zener between COM and Vcc is recommended also, as well as schottky diodes between LO->Vcc and HO->Vb. The capacitors are either ceramic or metal film - electrolythics do not work for short pulse applications.
Actually, use 10K resistor, not 100K. Just try - check what will be the voltage between Vs and Vb. If it is 10V or over - you are good. Normally, the separate power supply is used for the high switch, but in a single pulse setup even resistor may do the job. Zener is a must in any case.

The glitches of triggering are probably caused by noise on +5V rail. Do not connect photodiode here directly - bypass it with RC circuit with ceramic cap, otherwise photodiode acts as a cap translating all noise directly to the base of BC337. The noise is probably generated by EMI of the switching coil.

Skynet

Mon Feb 03 2014, 11:46PM

Registered Member #11734 Joined: Thu Mar 21 2013, 08:44PM
Location: Brno Czech republic
Posts: 35

Ok i get 10K resisor and zener diode... but I dont know the purpose of shottky diodes..

I remaked scheme.. did I get it right?

Pmr8

Yandersen

Tue Feb 04 2014, 03:24AM

Registered Member #6944 Joined: Fri Sept 28 2012, 04:54PM
Location: Canada
Posts: 340

Yep, right.
Schottky may not be needed - they just prevent overshoot that may or may not happen - depends on topology and switching performance. Better to be safe, right? I just mentioned zeners D2 and D3 on your schematic, so moving them to Vs-Vb & COM-Vcc and adding schottkies makes a close equivalent.
Just try.

Skynet

Tue Feb 04 2014, 08:04PM

Registered Member #11734 Joined: Thu Mar 21 2013, 08:44PM
Location: Brno Czech republic
Posts: 35

OK looks like the 10k resistor works with 13,5V supply IÂ´ve got 13V on 100uF low ERS elyt cap

Tomorow IÂ´ll make some shoot tests but I dont know how to bypass the photodiode with RC...

1 2 3 4 next last

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