PLL sstc problem (note i have no scope). / Tesla Coils / Forums

Forums

4hv.org :: Forums :: Tesla Coils

« Previous topic | Next topic »

PLL sstc problem (note i have no scope).

Move Thread

LAN_403

Arcstarter

Wed Apr 22 2009, 06:23AM

Registered Member #1225 Joined: Sat Jan 12 2008, 01:24AM
Location: Beaumont, Texas, USA
Posts: 2253

I had my PLL sstc running, even with audiomod. Here is a video of it running audiomod from a Zen, which is now dead, but completely unrelated to the sstc! It was a dead battery.

The audio was rather good, i thought. I decided it was time for a more permanent lash up, so i put the ucc's on the board like they where meant to be in the first place. The schematic is from Uzzor's website, which some credit also to Steve Conner for the schematic. It is the pll halfbridge

There are many differences. I used a 1nf timing capacitor, resistor 1 was i believe 10k and the other was i think 12k. This got me to a perfect 320khz, which is *just* above my secondary with the topload. I believe it locked, as the frequency would fall to about 200khz, and as i brought my hand close it would drop to about 280Khz, and it did this very reliably. I also made a small modification for audiomod, which was just a 180 ohm resistor in series with the 10nf capacitor on pin 9, and then to ground. The positive for the audiomod was 'in between' the 10nf cap and the 180 ohm ressitor. The negative was pin 5, which i ungrounded. This worked reliably, and i never killed a mosfet on low voltage. I believe the feedback however sucked, and the GDT is still likely the wrong kind.

I have checked the whole circuit, and it looks fine! I might just rebuild it, though.

Now for the problem. It will not oscillate at all, not even the VCO. I was probing it one day with my frequency meter, and i accidentally shorted pin 2 and 3. Pin two is the phase comparator 1 output, and three (tied with four which is the comparator input) is the output. When this happened, the oscillations started, at about 600khz. This was weird, i never changed the timing capacitor! So, i got a 2.7nf capacitor, put it in, and the frequency was PERFECT. On the dot, perfect :D. So, i decided to connect it to the halfbridge and see if this was really the case. I shorted the two pins for a split second, and it would give some nice streamers, much longer than before, for awhile. It was usually about 2 seconds or so at a time. Notice i work near a treadmill, and i had my ear arc to it.

It was CT while it ran, and i changed to antenna, which of course changes nothing, because the VCO is still misbehaving like a child with ADHD that just had extraomgchocolate cake

This truly puzzles me, and i am wondering what might be the problem. It is late now, so this might be a bit vague, i will edit tomorrow if needed. Thanks ahead of time.

Also, a thanks to Avi for help understanding the PLL and other misc help with the circuit, including a thorough description of audio modulation, and Uzzors and Steve Conner for giving suggestions and lending a hand!

Myke

Wed Apr 22 2009, 07:16AM

Registered Member #540 Joined: Mon Feb 19 2007, 07:49PM
Location: MIT
Posts: 969

When exactly did it fail? (from what I can tell is that you shorted pins 2&3 together accidentally and it doubled the freq) The only thing I can see that happened is that you killed the 4046. Do you have one that can replace it and see if that works? I noticed on Steve's page, there were some dead gate drivers. Maybe one of those died?

Arcstarter

Wed Apr 22 2009, 01:45PM

Registered Member #1225 Joined: Sat Jan 12 2008, 01:24AM
Location: Beaumont, Texas, USA
Posts: 2253

Well, it failed right after the ucc's where put on there, but i have taken the ucc's out, and it acts exactly the same. Also, when it did oscillate the gate drive was fine, the mosfets where getting enough voltage to drive them, and i am sure they where full on because there was not much heat.

I have changed the 4046 3 times, and the exact same happened.

The shorting of the pins was purposeful, after finding that it started the oscillations. The confusing part is that the VCO only oscillates for a few seconds after i short the two pins. And i have checked all of the joints, and nothing is shorted on the board that i could see. I even washed the board with alcohol and compressed air to clean the flux and all soldering debris, and i still could not see a a short.

EDIT: I was using CT at first as stated here, and someone said i should use an antenna instead. Well, i am remaking the circuit and i do like CT more, but would an antenna be better? The antenna is already on there, but if you guys think CT would work just as well, i will change.

Also, what i did with a CT is ground one side, the other side has a 10k resistor and then a .1uf poly cap to the 74ch14 or CD4046. Should i use a burden right across the CT? If so, what resistance? 100 or so ohms? I ask because my 10k resistor burned out after a few short runs. This happened with the mini as well. I do not want to lower the resistance to keep from the 1n4148's having to clamp too much current.

I would prefer to just use less burden and keep the 10k resistor, if that would be required to get a strong enough signal. By the way, the 10k has only been a fourth a watt, but why use more when you don't need more than 1/4 watt for the drive? That would just require the diodes to clamp more often! The CT was thanks to Tribologist, it was for a mini brute, and im not sure how many turns the secondary is. I am guessing 50-100 and the primary is usually 2-3 turns.

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