• WARNING: Tube/Valve amplifiers use potentially LETHAL HIGH VOLTAGES.
    Building, troubleshooting and testing of these amplifiers should only be
    performed by someone who is thoroughly familiar with
    the safety precautions around high voltages.

Ripple and chokes

Status
Not open for further replies.
Jump to Latest
Gluca

Gluca

Member
Joined 2004
#1
I am on my way to finish my new amp and ... yes ... I threw the parts with care and caution on the breadboard and everything seems to work fine.

PSU is: hybrid GZ34/SS bridge + CLCLC and 750V at the output. I am using hammond chokes for the time being (I have an option to upgrade them in the near future) and to be on the safe side I wired them on the ground leg (those chokes are rated at 500 or 600V max). I measured the ripple and found it to be 500mv (RMS). I was not that happy so I rewired the stuff and put a choke on the upper leg (exposed to the B+ high voltage) ... well the ripple fell to 20mV or so. Quite an improvement!

Here and there I read it is quite a common practice to use chokes on the ground side when they are rated for low voltage use but ... hemmm ... results seem to suggest this practice is not going to give you the best results.

Any comments? I'd like to undertand the facts in a better way.

Here is a schem.

Ciao
Gianluca
 
#2
Any comments? I'd like to undertand the facts in a better way.

Here is a schem.

Ciao
Gianluca
Click to expand...

Just a comment and I'm looking forward to seeing the other replies ... my understanding about chokes in B- lines is that this was commonly used when various B- values were required for multiple bias requirements (my experience is with tube radios).
It seems to me that if you are putting them in the B- line, then the total currrent requirement for the amp will flow thru' the chokes. It also seems to me that it shouldn't matter whether the chokes are in the + or -, and smoothing at the end of the filtration line should be the same.

ps : I didn't see any schematic
 
#3
It makes a difference whether the choke is in the + or - rail once you consider stray capacitance from the choke's coil to chassis. If it's in the + rail, that stray capacitance helps filter, if it's in the - rail it short-circuits the choke.
 
#5
Schematics

oooops ... missing pic. uploaded onto imageshack (I hope it works now).

I moved the 10H choke on the B+ leg.

Yes, I had your same ideas ... but my tests showed an unexpected result. I am pretty sure I took the measurements properly and I repeated it a numebr of times. I'll try again.

How much the stray C can help? The difference seems to be huge?

Any idea?


Ciao
Gianluca 😕
 
#6
Having seen the schematic, I will confess to being somewhat confused regarding the the +- AC voltages? and by the seemingly misnumbered pinout on the rectifier tubes? The rest looks pretty straightforward.
So, being very ignorant on things Audio (and being new here), I look forward to more discussion about this PSU schematic.
The inherent capacitance in the inductor is something I will play with and learn about as I head towards building an amp.

cheers
 
#7
Another possibility...

It might depend on how the measurement was made. If an oscilloscope was used and that diagram accurately reflects reality then the chassis connection at the reservoir capacitor means that the oscilloscope wouldn't see the chokes at all.
 
#8
Pins etc ... yeah ... but that's the sw I use, +- are just a reference (for the sake of clarity) to my power tranny.

@EC: I used the very method you suggested time ago (was it dec last year?) ... I mean 0.1uF wired to B+ and to a large resistor (240k, you said 1M) tied to ground (B- actually or signal ground if you like, does it make the difference? shall I ground the probe to the earth?) and my oscilloscope...

I couldn't listen to the amp (power stage to be completed) so I can't report on any audible hum ... just seen the traces on the o/s.

Ciao
Gianluca
 
#9
Your problem is that oscilloscopes measure between input and earth. If you move the point at which your 0V connects to chassis (earth) from the reservoir capacitor to after the last choke you should get a more sensible measurement.
 
#10
Mumble-mumble got it. Thanks. I was just measuring the charge and discharge of the last reservoir cap. Wasn't I?

But ... shall I return all the cathods (and reference the signals) to what I called signal gnd on the scheme or to earth? Star grounding here.

Gianluca
 
#11
With a ground at the (-) input of the chokes and at the (-) output they are shorted... unless one of the two ground symbols isn't connected anywhere...

Normally in this circuit, the ground is at the output of the filter, and the (-) input floats at some volts negative. A small additional R-C filter in the (+) lead (just a resistor - capacitor is already there) can take care of any noise that is coupled through common mode or via the choke's capacitance to ground.
 
#12
If you use a "common" choke in the neg to ground side of the power supply...you then disturb the ground for all the rest of the circuits.. You now introduce a ripple in ground ...so the ground is bouncing.... This may be fine in RF equipment...but in Hi-End audio the sensitive pre-amp stages will have a noisey ground as reference..

Chris
 
#13
Thanks.

Tom signal ground (see the scheme) is not connect to anything so choke is not shorted. I elected to connect to earth the (-) lug of the first cap (that's is the input to the first choke) for safety reasons.

Cerrem, not sure I got your idea properly ...


Gianluca
 
#14
What you have labeled signal ground is the 'real' ground. If you want to connect your chassis anywhere it should be there, otherwise your signal ground will be bouncing around relative to the chassis (earth.) You can make the connection to the chassis through a small value resistor; that's up to you. I've never found the resistor necessary unless there were other grounding problems like ground loops.

Where to connect the cathodes? Obviously to signal ground in any case.
 
#15
OH! Thanks all! So I was given a wrong suggestion ... I was told to tie to ground the first cap and let the cathode (signal etc...) return to the "floating" end of the last choke/cap.

I'll do that... I just need to swap a wire.

Gianluca (much happier now)
 
Status
Not open for further replies.
Top Bottom