VVR noise

[Bluesamp]

Hi there at ampgarage,

I'm new here, but already found lots of good information and thoughts to build me a better amp.
Already one year ago I built a TW express clone with two EL84. It sounds awesome but way too loud for use at home. So I dicided, to give VVR a try. I don't like the sound of MV Amps...
Everything seemed to work well at lower volumes and max. volume, but if the pot reached about middle, I got bad tone with ghost notes and a lot of hum. A friend gave me the advice, to clip out the 12V Zener and it worked! No more hum and bad notes.
Now I really would like to know the purpose of this zener. Is the zener really necessary? Or will I risk some damage?

Thanks a lot...

[ChopSauce]

I hope that the OP found the answer to his question - pertaining to the zener protects the MOSFET, I think - but does anyone know why the zener would create such a noise...

[strelok]

Can you post a schematic? If the diode you're referring to is the one I'm thinking of then its there for over current protection and is not strictly speaking necessary. Its connected from one side of the current limiting/sense resistor to the gate of the mosfet. The way it works is, under zero load the voltage on the gate and the other side of the resistor is roughly the same. As you pull more current there's a drop across the resistor, when it gets to 12v the diode starts conducting and pulling the gate voltage down until the current drops enough to cause the voltage to climb back up and the diode to stop conducting. In the one I designed a while back it was a 10 ohm resistor which gives a current of 1.2A when it starts conducting. Its mostly just there to protect the mosfet in case of a short to chassis or something like that.

As far as the noise is concerned sometimes zeners will generate noise if they're biased such that they're sitting right on the knee in the curve. As the supply fluctuates the zener goes in and out of conduction and can generate some high frequencies that can interact with other parts of the circuit. Its rare but I've seen it happen. However if this is the circuit I'm thinking of then you must be pulling a lot of current to get it do that. It shouldn't be conducting at all under normal conditions. So do post a schematic in case we're talking about different circuits. It might just be that the series resistor is a different value.

[ChopSauce]

This one

Variable_Voltage_Reg_with_current_limiting.JPG

with no D2, for I scale the whole amp. Any clue?

Actually I'm not 100% sure the VVR is the one and only responsible for the noise but the point is my previous builds used to be more silent...

[martin manning]

This one

Variable_Voltage_Reg_with_current_limiting.JPG

with no D2, for I scale the whole amp. Any clue?

Actually I'm not 100% sure the VVR is the one and only responsible for the noise but the point is my previous builds used to be more silent... :?

I would try replacing the Zener and see if that has any effect. The FET current is limited when V R3 + Vgs equals the zener voltage. Vgs will vary with the FET, but it is usually around 4.5V, so a 10R resistor will limit the current to about 0.75A. That limit will be reached on start-up if the output of the VVR is charging a large capacitor.

[Colossal]

When installing a VVR, a reservoir capacitance of some value must be placed after the rectifier and before the VVR. Some people just place the main reservoir capacitance in front of the VVR if scaling the whole amp. Without it, the VVR can oscillate. You can also use a small capacitor before and then the main reservoir capacitance after, as usual. The diode before and after the VVR are typically used when scaling the power amp or a section of the amp only and the supply rail is split and needs some isolation. If you are scaling the whole amp, you can leave out the diodes. The resistor below the pot sets the lower limit of the scaling voltage range. R3 limits current and is typically 5 to 10R.

[ChopSauce]

Well, I found this topic while searching the forum and didn't noticed that it was in the TW section...

So, I am scaling the whole amp and the VVR is connected between the first cap and the first dropping resistor - on my Tweedle Dee Deluxe...

I also connected the ground of the VVR to the power transformer center tap together with the power amp filter caps - as I could read here was recommended.

So, because the VVR is bolted at the bottom of a loaded 5E3 chassis, well changing the zener is something of my worst option - hence the reason of my post here.

Thanks you very much anyway...

[martin manning]

So, I am scaling the whole amp and the VVR is connected between the first cap and the first dropping resistor - on my Tweedle Dee Deluxe... :?

With the VVR output feeding the output transformer primary CT? You could try clipping the zener out, if that is easier.

[ChopSauce]

Ok, I'll try that possibly, thanks!

So, I am scaling the whole amp and the VVR is connected between the first cap and the first dropping resistor - on my Tweedle Dee Deluxe...

With the VVR output feeding the output transformer primary CT?

EDIT: Touché

[martin manning]

Like This:

[ChopSauce]

Ahh... Martin, you nailed it - one more time!

I (brainlessly) copied the 5E3 layout - with the OT center tap connected to the rectifier - without even noticing...

Am I wrong assuming that you already guessed it?

[martin manning]

I wasn't sure, but your wording led me to ask exactly how it was connected. All quiet now?

[ChopSauce]

Err... not really.

"Luckily" I am sort of unable to follow a plan/layout/ etc. I thought I made that mistake but actually the OT/CT is connected to the VVR output. I did not even thought about it while wiring. I just made it.

The reason why I thought I made a mistake is I didn't found the power attenuation very impressive.
I should be able to change the zener quite painlessly but I might as well end up throwing the VVR out.

Not quite sure about anything, yet...

[ChopSauce]

I think I'm going to try this grounding scheme for the power section - as an alternative
(that would be 3 ground points, the fine line being a representation for the chassis - I guess)

Plan18W_0V_bon.png

His author wrote(*) that it is an improvement over all the filter caps being connected to the same location on the chassis, the VVR being only connected to the power transfomer center tap (CT).
(he made no comments about the 6.3V artificial CT, though)
___

(*) https://www.projetg5.com/phpbb3/viewtop ... 888#p99888

[bepone]

conecting to the chassis is better to do in the first stage, not in the dirtiest point (first filter supply capacitor)

grounding is easy if you follow the current path.
current passing to one dedicated stage is starting from and ending to the same capacitor . so if anode current of output section started from capacitor xx, also curent from the cathode needs to return to this capacitor xx.

if the current through the first stage anode coming from the capacitor YY, also the return current from the cathode needs to go to this capacitor. so you need to think where you connect MINUS pole of dedicated capacitor, and all dedicated cathodes bring there. you will not have excessive hiss and humm from the grounding..

also there is humm from the magnetic field coupling , humm from the heaters etc.. all variables you need to fix