PLL un-equal square wave?

Hello Group,

I have been playing around with a 4046 Phase lock loop! I am using it to drive two mosfet drivers one inverted and the other non inverted which then drive a gate transformer connected to a Half Bridge.

Now when I view the square wave output with a scope from the PLL vcoout it appears slightly un-equal. The negative portion of the wave appears slightly shorter than the positive portion. This is exaggerated even more when viewing the output(s) of the mosfet gate driver (tc4421 / tc4422). I believe this is causing one side of the half bridge to work harder which creates much more heat in one mosfet than in the other (Well that's what I think is happening as one mosfet becomes much hotter than the other) .

Additinal info :- I am using a current sense transformer located around a parallel tank circuit to send the correct phase signal to the Phase Lock Loop. This appears to lock the PLL / VCO at the self resonant frequency of the tank circuit, which should then drive the Half Bridge mosfets at ZCS / ZVS causing the desired affect of soft switching.

The output on the scope suggests that soft switching is almost taking place all though there is slight ringing which at this point I am unable to minimalise any further and which may be connected to the above waveform issue?

Does anyone have any idea why the initial waveform could be un-even even when just running from the fixed VCO (ie with no PLL feedback)?

Hope this makes some degree of sense! ;-)

Regards Steve.

Reply to
SRG
Loading thread data ...

Can you run the PLL 2 times as fast and use a divide by 2 to get the original frequency with perfect duty cycle? GG

Reply to
Glenn Gundlach

Thanks for the feedback! Not sure if that would work I could be wrong ofcourse as this is the first time I have played with a PLL.

The VCO on the 4046 is currently set to around 150khz with a shift allowance from aprox 100 - 200 khz. The output drives a parallel tank circuit consisting of 1.5uh inductor and 564nf worth of capacitors giving a frequency of around 171khz. Additionally the output of the inverter is connected to DC blocking cap then a 45uh inductor in series(for impedance matching though having trouble finding the correct value for this!) this then connects to the aforementioned Parallel tank circuit. I have connected a ferrite (Clip on type) with aprox three turns which acts as a current sense transformer (CT) which passes its output to the PLL input. I then tune the VCO until I get a perfect phase alignment (well almost) for both current and voltage so as to achieve the golden Zero Current Switching for the mosfets in the inverter (half bridge). That's the plan anyway ;-)

So in effect the VCO (on the 4046) acts as a starting frequency to drive the inverter which in turn feeds the tank circuit AC that acts like a filter ie only producing significant power when the inverter is tuned to the tanks resonant frequency. The current transformer feeds back voltage and phase to the PLL which follows any resonant frequency change that's produced from the tank circuit thus theoretically keeping everything running sweetly in tune and consequently soft switching mosfets at ZCS / ZVS which should make them run cooler and more efficiently with the exception of inherent mosfet losses.

So I guess by upping the frequency would not help in this case? I still am at a loss why the square wave output from the VCO and gatedriver is not evenly spaced for the positive side compared to the negative side of the waveform?

My God did I just type that? LOL scratching head......

Reply to
SRG

MosFet's have Capacitance on the gate verses what you normally see on bipolar, you maybe current sinking ? you may want to test the output unloaded to see if you are still getting the problem? if not, then the problem is most likely current sinking due to external cap. you could try a current driver Tansistor (emitter flower)? or some high speed switching device to output more current with much less input current and capacitance.

the ringing you see may just be the effects of your scope that is common, also you may want to experiment a bit with your scope probe in 1:1 mode and 1:10 mode, you may not hage the cap set correctly in the probe.

Reply to
Jamie

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.