PLL problem - No Bass and full loading

I built a PLL FM Transmitter using discrete VCO and RF stages and a ML145170 digital PLL IC.. The transmitter it getting locked to the desired frequency but there are two major problems in the overall circuit.

1) When I first connect the PLL IC to the transmitter's driver output I do not observe any loading effect across the node. I use LED powermeter circuit to estimate the power level by observing intensity of light it emits. Suddenly after some time, I find heavy loading effect across the driver's output. I verifed all connections, solder joints, power supply voltage but could never figure out the problem. The loading goes off after I disconnect the PLL from the driver.

In short, PLL detects the transmitter signal, divides it down to the phase reference frequency and functions properly only after sucking everything out of the driver stage and leaving the output stage defunct. Wonder what is wrong!

2) Low frequency components of the input audio get attenuated severely. I get no bass effect. Phase Comparator frequency is set to 100Khz.

PLL Loop bandwidth issues...

Hi all,

After taking a closer look at the design, I just wonder if the loop bandwidth of the PLL circuit is causing the problem of low-frequency attenuation..

The loop bandwidth in my case is 2KHz. Beats frequency spectrum is primarily made of components lying in a 60-600Hz range. These components in the audio source can be easily attenuated by a PLL with such loop-bandwidth specs as mine.. So I think of reducing it down to around 100Hz although it will result into greater lock times. Worth a try??

Besides this, I also want to know if I can use the PLL circuit with the famous Veronica 1 Watt FM. I have a PCB sample of it. It is a transmitter which runs a push-pull oscillator pair to have better stability and phase margin. Here is the link:

Hope to find some help from you guys..

The bass attenuation problem is solved. It was due to the high loop-bandwidth of PLL LPF that caused negative feedback for frequencies modulation produced by input low-frequency audio.

However, ML145170 still loads the exciter far too much. I think I should go ahead with introducing a Class-A buffer stage between the exciter and the synthesizer. Will give you an update on it soon..

Meanwhile, I have discovered a new issue with the synthesizer.. When the exciter output voltage amplitude exceeds a certain value, ML145170's reference frequency counter starts to malfunction and no longer provides a stable reference out which ultimately destabilizes the exciter output.. Any ideas about this strange problem??



2006-12-28 12:19 am
you shpuld be using a buffer/limiter for the phase detector input, this will eliminate loading of the exciter, and keep the amplitude to the phase detector constant. you should have a limiter on the exciter output anyway, since FM signals should NOT be amplitude modulated at all.