PLL circuit wont lock / Tesla Coils / Forums

Forums

4hv.org :: Forums :: Tesla Coils

« Previous topic | Next topic »

PLL circuit wont lock

Move Thread

LAN_403

Dago

Sun Dec 23 2007, 08:28PM

Registered Member #538 Joined: Sun Feb 18 2007, 08:33PM
Location: Finland
Posts: 181

I've built Conners mk2 PLL circuit (

) and theres something wrong with because it wont lock to any frequency. I've checked the circuit and board layout a few times and I cant find anything wrong with it (I guess I've missed something). The circuit is a bit stripped down version from Steve's original, I left out the synchronizer part (because I dont think I'm even going to use an interrupter and if I will I doubt the synchronizer will make much difference in an SSTC) but I DO have 4013 and 4011 chips there doing the "divide by 2 and shift 90-degrees" thing.

Here are some pictures

The eagle images are not the current ones (added a NAND chip for interrupting the gatedrivesignals and theres a bug on the board on the target phase pot which has been fixed on the board (the streamer picture is from a friends SSTC, not mine (my hand though)).

My CT is made on a Ferroxcube 3E25 core with an Al value of ~5000 and a diameter of ~25mm (about an inch), wound with around 100 turns. The secondary ground wire is wrapped around the toroid a couple of times and the CT output is clamped with two fast 3A diodes in antiparallel.

The circuit around the 4046 is pretty much identical although I left the pin 12 unconnected to anything (this should give the widest locking range). I also left out all the analog switches so there is no connection over IC10D and IC10C IS connected.

The circuit oscillates and I can adjust the frequency with the frequency pot on pin 11 and adjusting the "target phase" pot changes the frequency a bit aswell. The loop gain pot makes no difference to the waveform.

I'll try to be as thorough as I can describing what I know about the circuit. What I have scoped (all pulse lengths mentioned are with the same frequency pot setting):

There is feedback on pin 14 (signal in pin) when running.
There is a 50% square wave with a pulse length of around 2Âµs on pin 2 (phase comp 1 out).
There is a 50% square wave with a pulse length of around 2Âµs on pin 3 (comparator in),
There is a 50% square wave withe a pulse length of around 1Âµs on pin 4 (VCO out).
There is a waveform that looks almost like a square wave (theres small spikes on the middle of the pulse and on the middle of the low-time of the waveform) at the smallest phase pot setting with an amplitude of around 0,8V on pin 9 (VCO in). When I turn the phase pot to the other way the signal "squarewavish" waveform stays pretty much the same but there forms small pulses ("ditches") between the pulses that grow its amplitude to around 8V or so when I turn the pot to the other way.
The signal from the first flip flop is phase shifted by 90degrees like it should.

When running the circuit the frequency pot just changes the frequency just like in a VCO with no "oddities". Turning the phase pot changes the frequency abit. Loop gain pot makes no difference when the circuit is in operation. I have fiddled alot with the pots and changed the primary phasings etc. But anything doesnt seem to make much difference, the circuit just wont lock to the resonance frequency (streamers get smaller when I bring my hand closer, FETs heat up alot, and the CT signal and the gatedrive signal arent in phase). I also tried running without the CT and it made no difference.

I'm running out of ideas, anyone have any suggestions what to measure or can spot whats wrong my circuit? If you need more pictures or more measurements or something, just ask and I'll get em.

thedatastream

Sun Dec 23 2007, 10:03PM

Registered Member #505 Joined: Sun Nov 19 2006, 06:42PM
Location: Yorkshire!
Posts: 329

I found that I had wired my op-amp in the VCO feedback was wired incorrectly so it wouldn't lock.

Once I built the circuit fine it was OK

Dago

Mon Dec 24 2007, 12:53PM

Registered Member #538 Joined: Sun Feb 18 2007, 08:33PM
Location: Finland
Posts: 181

I made a small mistake when describing the circuit behaviour, the loop gain pot and target phase pots got switched in my mind so when I'm talking about loop gain pot I mean the target phase pot and vice versa (=target phase pot does nothing).

Edit: Yay! Found a cut trace that made it so that the cap C20 wasnt properly connected to the opamp. That might have fixed it though at first try I was still able to use it more like a VCO than a self-resonant circuit but the waveform was "doubled" and some of the "components" of the waveform seemed to lock because they moved when I brought my hand closer etc. but the "main" waveform stayed still (Edit: I realized that this "doubling" is propably due to the fact I'm running half-wave rectified and it cant lock when theres no feedback). Didnt have time to experiment with primary phasings etc. yet.

The target phase pot seemed to work more like a switch, it doesnt do anything when you turn, then suddenly at some point the other led shuts and the other lights up and the streamers stop and I have to retune it again.

Steve Conner

Mon Dec 24 2007, 06:20PM

Registered Member #30 Joined: Fri Feb 03 2006, 10:52AM
Location: Glasgow, Scotland
Posts: 6706

Dago wrote ...

The target phase pot seemed to work more like a switch, it doesnt do anything when you turn, then suddenly at some point the other led shuts and the other lights up and the streamers stop and I have to retune it again.

Try reversing the phasing somewhere in your feedback loop, see if that makes a difference.

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