Veronica Pro-III/IV Frequency Response

[reverend]

First time poster, long time reader...

Have had a Veronica Pro IV sitting in the cupboard for a few years and have had plenty of time on my hands so decided to fire it up. I noticed that I couldn't get a flat frequency response, especially for the stereo pilot and RDS. A quick look over the circuit diagram and it seems that there is a 2n2 capacitor over the base of the transistor that forms part of the loop/modulator circuit. Together with the resistors around it, this gives a 3dB HF roll-off around 15 kHz. This can be corrected by replacing one of the 50/75 us pre-emphasis capacitors with a 470pF one. Using this as a 'flat' setting flattens the roll-off until nearer 100 kHz and lo and behold phase/frequency response for the pilot and RDS are restored. (Note the Pro III circuit is identical in this respect).

Now the LF - though the 3dB cut-off at the lower end is well below 10 Hz, no matter how much correction I apply I can't get a 20 Hz square wave not to over modulate. I initially thought this might be a consequence of the HF roll off but I'm now wondering whether the loop filter is acting at low frequencies.

On the BW 1 Watt exciter, the modulation is DC coupled to the varicap, so no LF issues at all. Can't quite work out if it's possible to do the same with the Veronica board.

The Rev.

[sinus trouble]

Hello Reverend

I am a bit lost with the issue you seem to be having, Are you having trouble with MPX?

The audio frequency response of the PLL PRO III is 20Hz - 75Khz

Just to check? You are disabling the Pre Emphasis before adding any MPX device?

Also its not good to use square wave for audio testing, its full of nasty harmonics!

[sinus trouble]

Just to add, I am not familiar with BW gear but when you say it is DC coupled? How do you mean?

[reverend]

Hi!

If you take a look at the input circuit, there is a 2n2 capacitor across the base of the emitter follower that feeds the Varicap. There are 3 resistors feeding this 2n2 capacitor, a 33K (from the audio input), an 8K2 (to ground and via the 10 uF capacitor) and a 100 K from the PLL. Put these in parallel and you get an equivalent feed resistance of 6.1K. The contribution from the base-collector, or base-emitter junction of the transistor would be negligible.

Calculate the 3dB roll-off point of a 6.1 K resistor in parallel with a 2.2nF capacitor (1/2*pi*R*C) and you get 11.8 kHz. So the frequency response of the input circuitry is not flat to 75 kHz, it is already 3dB down at 12 kHz. This could be changed either by:

* Reducing the 2.2nF capacitor to 220pf (which would change the roll-off frequency to nearer 120 kHz), or
* Putting a 470 pF capacitor in parallel with the 33K resistor feeding the audio.

At the LF side, the cut-off is primarily determined by the 10 uF capacitor, and the 100K resistor from the PLL, making the LF 3dB frequency 0.15 Hz which is plenty low enough.

Using a low frequency square wave to check modulation is a a good way of checking phase response and LF response as these all need to be correct for modulation to remain at 75 kHz. Any LF 'droop' or any HF phase response errors, will result in over-deviation.

My point about the BW driver is that from the audio input to the varicap, there are no DC blocking capacitors, meaning that the LF response is flat to DC, other than any LF pulling caused by the PLL.

Hope that helps explain!

Rev

[Krakatoa]

About the HF rolloff, if I remember well C1 and C3 should be 100pF and you are right to go. But I am sure I have never seen 2n2 at those positions.

About the LF response flatness, there was a discussion some months ago, and it was solved using a much longer lock time (slower loop filter). In the BW you can get the 20Hz square wave right because it uses dual time constant for the loop filter: one very fast to tune the oscillator in the vicinity of the target frequency and the slower one to keep it locked in the long term (and this is what benefits the LF flatness).

[sinus trouble]

Hello Again Reverend

Very interesting calculations you have performed.

Please forgive me for asking once more, Why is it necessary to have a flat response across 20Hz - 120Khz?

Thank you for clearing up the BW version, I get what you mean now.
If the device you connect to the BW input has DC Blocking? Would that then in turn determine your actual frequency response?

[radium98]

sinus trouble freind this is the technical schematic of the famous 1w pll by broadcast warehouse .hope it serve you to be clarified .thanks.
i still waiting in pm

[mikroman]

Hello Again Reverend

Please forgive me for asking once more, Why is it necessary to have a flat response across 20Hz - 120Khz?

generally without a flat response, what comes out of the mixer and processor will not authentically come out of the radio. The two main problems are: -overshoots- The cause can be anywhere in the spectrum as a result of IMD or THD, but mostly the problem lies in the field up to 100Hz. For a thorough examination, it is necessary to introduce a square signal into the mpx channel. The presence of overshoots necessarily means quieter sound than the competition in order to stay in the + -75KHz range. Stereo separation- Another common problem related to flat response is the lack of stereo separation (anything more than + -0.5 dB is a problem).

[nrgkits.nz]

I’ve found the main cause of high frequency mpx roll off on exciters I’ve worked on is due to the Varicap diode(s) that do the modulation. You’ll find in a lot of designs there is a resistor in the path to the varicaps and this along with the capacitance of the varicaps acts as an RC filter, hence causing a gradual roll off of high mpx frequencies. Modulating a 400Hz sine may get you 75KHz deviation, then doing the same again with a 57KHz sine same amplitude may get you 35KHz deviation. Note this is not affected by the PLL loop constant/lock time. A slow lock time is only needed for modulating a square wave less than around 200Hz to prevent tilt, you get little to no tilt, if at all for anything above 400Hz.

[sinus trouble]

sinus trouble freind this is the technical schematic of the famous 1w pll by broadcast warehouse .hope it serve you to be clarified .thanks.
i still waiting in pm

Thank You Radium

Much appreciated!

[sinus trouble]

Thank you all for clarifying a few points

Flatness across the audio spectrum is desirable, I totally get that!

I was just puzzled why a few DB reduction in the MPX region would be such a massive issue?

I guess we all learn something new every day which can only be a good thing!

[radium98]

thanks for you Sinus

[reverend]

About the HF rolloff, if I remember well C1 and C3 should be 100pF and you are right to go. But I am sure I have never seen 2n2 at those positions.

You're correct, if C3 was 100pF then there wouldn't be a problem. The parts listing that comes with the PRO IV is attached, which clearly shows C3 as 2n2.

It seems, though, that the 2n2 C3 is the capacitor used for pre-emphasis AND the one that sits across the base of that transistor. A part numbering problem maybe?

Rev.

pll_pro_iv_parts.jpg

[Artifact]

Been lurking here for a while. I'm not big on posting but I think I see the issue here.
As reverend suggests, this seems to be a part numbering problem in the PLL PRO III schematic.
The parts list seems to match up against the PCB-markings but three components have the wrong number in the schematic.

C3 in the schematic is marked as C6 on PCB, value for this part should be 100 pF
Also note that C3 is on two places in the schematic

C6 in schematic is marked as C8 on PCB, value for this part should be 56 pF (same as C7)
Additionally, C7 and C8 should switch places, but it really doesn't matter as they have the same value.

C8 in schematic is marked as C10 on PCB, value for this part should be 22 pF.

[radium98]

good artifact thanks

[rigmo]

HI guys do you have schematic for final stage RD06?

[Novaro]

Mine is the easy tune pll3 version.

[radium98]

the easy tune is the last best one of the veronica pll ,and after went 0 down.

[Novaro]

In my pcb c1 = c3 101 capacitors.
But c15 near the pre emphasis is 2n2 .
I wonder is that right? Or should i change it to 470pf?
Also any easy modification to slow down the pll lock time to fix the bad low frequency response?

Thank you.

[Novaro]

the easy tune is the last best one of the veronica pll ,and after went 0 down.

Thanks for the response.
Glad i was lucky finding one complete with 12w amplifier and the coder for cheap.
I did not know it was good.