[moved] mini DRSSTC/Singing arc / Tesla Coils / Forums

Forums

4hv.org :: Forums :: Tesla Coils

« Previous topic | Next topic »

[moved] mini DRSSTC/Singing arc

1 2 next

Move Thread

LAN_403

EEYORE

Sat Feb 11 2006, 06:30AM

Registered Member #99 Joined: Thu Feb 09 2006, 06:10PM
Location: florida, usa
Posts: 637

Hello all,
continuing from the old forum...
My PLL 4046 driver is near ready to build(stupid UPS didnt make it on time today, so i have to wait for my expresspcb boards...)

I designed the board primarily for a singing arc. I plan to use a cheapo MP3 player to feed in the audio into the PLL...Ive never done this before so i have a few questions:
1-i assume fullwave smoothed DC input is in order?I am uisng IRFP460s for the higher freq secondary i have planned.The board is fully adjustable as far as phase adjust and freq goes....
2-Do i need an amplifier stage for the audio in?
3-any other tips?

The board will fully allow a scaled up coil for future systems...Such as a DRSSTC...Im looking for a mini coil, with "mini sparks", as the last was threatening at over 3 footers, and i have little space here:-P

My final question:
I have numerous 40n60IGBTs that have been "compramised" through priar failures. They test fine with a diode checker, as well as resistance check.I have heard a theory that while they may test fine, in operation they are doomed to fail. Is there a good way aside from tesla operation to test them?I tried applying 12vDC to the C-E and they wouldnt turn on until i applied 12vdc to the gate,turned off when gate was grounded...seems normal...Comments?
pics of the board assembled to come this week...

Matt

ragnar

Sat Feb 11 2006, 11:08PM

Registered Member #63 Joined: Thu Feb 09 2006, 06:18AM
Location:
Posts: 1425

If you've got some IGBTs that you *know* work, and some IGBTs that you *know* have failed, you can compare them and make up a table or remember the results.

The way I sort my MOSFETs is:
- trays for NEW mosfets,
- trays for second-hand mosfets that have survived in my projects,
- trays for random SMPS mosfets,
- trays for suspected dead mosfets, (second-hand, but not sure if I killed them or not/can't remember)
- trays for known dead mosfets.

If I'm about to build or prototype something, I reach for the random SMPS fets or second-hand fets.

If I'm going for the "suspected dead MOSFETs" I grab my multimeter. Set the range to 10Meg.

With the positive lead on the SOURCE and the negative lead on the DRAIN, it measures 9 megohms,
with the positive lead on the SOURCE and the negative lead on the GATE, I get out-of-range.

This test tells me the FET is working just fine, and has never let me down, ever. If I measure anything else (e.g. 2meg, hundreds of K, etc, or 0-ohms, 0-ohms, hehehe) then the MOSFET is just about guaranteed dead.

With a bit of experience metering dead and live IGBTs, you'll get a feeling for what's dead just by the voltage drop, or how many ohms you measure etc.

EEYORE

Sat Feb 11 2006, 11:53PM

Registered Member #99 Joined: Thu Feb 09 2006, 06:10PM
Location: florida, usa
Posts: 637

Yea, i always keep unused on hand and compare...The used ones compare identically to the unused, but a few here have reported failures despite passing the tests because they were in the bridge while another failed...I guess i just have to throw them into operation

Hope for the best...

Matt

EEYORE

Mon Feb 20 2006, 07:11PM

Registered Member #99 Joined: Thu Feb 09 2006, 06:10PM
Location: florida, usa
Posts: 637

The boards are made and all looks nice!Im having a weird issue. I have an LED on the enable input to the ucc chips as i wanted it to indicate when i was using the interruptor. It lights even though it is connected to nothing?Im guessing its picking up interference somehow?

The ucc chips will not pass the signal from the 4046 unless i tie the enable to the positive rail via resistor, or tie to to ground via resistor and then pulse it with interruptor. Im guessing this is normal for a PLL and it just spoofed my head when designing.

Other than that, nice crisp waves, and adjusts from around 160khz to 1 mhz...Very clean waves off the GDT too, nicest ive ever had of all the coils!

Soon i will power up the bridge and all and see how the coil likes it:-)Pics coming soon...

Matt

EEYORE

Wed Feb 22 2006, 06:56PM

Registered Member #99 Joined: Thu Feb 09 2006, 06:10PM
Location: florida, usa
Posts: 637

First light!!!
3 in hissing, hot streamers...
Power in is a voltage doubler consisting of 2 10,000uF400volt capacitors fed via 10amp variac....
Half-bridge consisting of IRFP460's...

Boy does everything get hot quick...I need to switch to halfwave rectified while i get an interruptor built....

Question-My volt regulator and ucc chips all get HOT HOT while running, even when the bridge is powered off...I turned down the freq on the 4046 and they ran cool again. I know that higher freq stresses everything, but shouldnt everything be speced to run a little high?(668khz)

Arcs are neato!Today i will remove primary turns(17 right now) and run half wave to see how things perform...

Matt

Steve Ward

Wed Feb 22 2006, 09:15PM

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

Running 2 IRFP460s from a pair of 9A drivers at 688khz will indeed make things very hot! You need to heatsink those drivers (and definately the voltage regulator) if you are running CW.

I think the TI spec sheet gives an equation to calculate power dissipation in the gate drivers for a specific load and frequency. I would think that even dissipating .5W per driver would make them quite hot to touch. Glue some heatsinks onto the top of them

. A miniature fan is nice too...

EEYORE

Wed Feb 22 2006, 10:02PM

Registered Member #99 Joined: Thu Feb 09 2006, 06:10PM
Location: florida, usa
Posts: 637

Steve Ward wrote ...

Running 2 IRFP460s from a pair of 9A drivers at 688khz will indeed make things very hot! You need to heatsink those drivers (and definately the voltage regulator) if you are running CW.

I think the TI spec sheet gives an equation to calculate power dissipation in the gate drivers for a specific load and frequency. I would think that even dissipating .5W per driver would make them quite hot to touch. Glue some heatsinks onto the top of them

. A miniature fan is nice too...

Hehe, yea i thought so...I dont plan to run them CW except when audio mod them. I just wanted to see some sarks

I will see what ive got in my junk box as to cool down the drivers...Vlt regulator now has a nice finned heatink and just gets warm...

Pics coming soon, i promise!
Matt

EEYORE

Fri Feb 24 2006, 03:57AM

Registered Member #99 Joined: Thu Feb 09 2006, 06:10PM
Location: florida, usa
Posts: 637

Some pics....This is the driver

Here is a spark pic...Crummy pic, but you get the idea. About 8inches with half wave rectified variac full blast through a half bridge of IRFP460's...Not bad, no?

Next comes an interruptor, voltage doubler, and full bridge(in that order)
Secondary is 10in tall and 3.5in diameter with 27awg wire...Quite high freq but will be winding a new one soon with 30awg wire and something like 4in wide by 12 in tall...

[Edit: Fixed double post]

EEYORE

Wed Mar 01 2006, 07:27AM

Registered Member #99 Joined: Thu Feb 09 2006, 06:10PM
Location: florida, usa
Posts: 637

Well, a slip of the finger while adjusting the freq range sent a mosfet to mosfet hell...I decided to design an expresspcb board for 40n60's and put that to use with this. My current means of a bridge is soldered wires to the mosfet and just dont feel like soldering anymore

A half bridge of the IGBTs should give me more than plenty of output. And will fit perfectly in the space left on the perf board.This will be my most "attrative" coil yet!

I plan to fit the whole circuit minus HVPSU on a 8in by 6in board. Here it is so far...Tidy no?

Planned half bridge using 40n60s

Here is my setup up to date...It will probably stay like this as the half bridge of 40n60s is doing fine wired rather than on a pcb...I have a question. I would like to scope the output from the bridge for adjusting the phase adjust on the driver for best looking waves. Can i make a 1:1 xfmr with say a GDT ferrite torroid and put this in series with the primary?

[Edit: Fixed triple post and oversized picture]

HV Enthusiast

Fri Mar 03 2006, 02:21AM

Registered Member #15 Joined: Thu Feb 02 2006, 01:11PM
Location:
Posts: 3068

wrote ...

I have numerous 40n60IGBTs that have been "compramised" through priar failures. They test fine with a diode checker, as well as resistance check.I have heard a theory that while they may test fine, in operation they are doomed to fail.

Yes, this can be a problem. In my demonstration coils, if i blow even a single IGBT, i replace all four just to be safe.

1 2 next

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