Simple Symetrical Amplifier

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Hi naf. The zeners seem to me to be 1/2 watt rated. You should use a 1watt one.

With 470ohm feed resistor, the dissipation is over 400mW; although less than 500mW, having a larger dissipation margin is good for long term reliability as 1W zeners have thicker wires that can help transfer much of the heat to the copper quickly.
 
Shaan, here my progress on building..
sorry don't have chance to source 1W zener :-(
 

Attachments

  • 20120923_160003_crp_2.jpg
    20120923_160003_crp_2.jpg
    189.3 KB · Views: 530
dear Esperado, thank you very much for your explanation
ie: for 15v, i could use 7v and 8v in serial, right?

regards
naf

Hi naf

You don't need 1 W zener since there's cca. 10 mA zener current, multiply 15 V, so you end at 150 mW power dissipation on one zener diode. 500 mW zeners are OK.

If you dont have 1,2 k resistor than parallel (one resistor on top other on bottom side) two 2,4 k or 5,6k parallel to 1,5 k etc. It will be easier later to adjust correct value in this way, idea is to parellel them rather than serial connection. ;)
 
Hi Jay,

I promised to report of the effects of using a 100R resistor between case earth and amplifier star earth. I recently changed my amps over to this arrangement and I am very happy with it. When I measure at the amps o/p on the scope now noise appears to be a little lower. I am not totally sure if that is a real improvement or whether this arrangement just makes it easier to measure noise properly.

Subjectively perhaps it brought a little more refinement but to be absolutely sure I would have to do more tests which I don't feel inclined to do.

Logically this is a very neat completion of my choke regulated scheme and clears up a lingering concern I've had for a while about creating beautifully clean voltage & earth rails and then connecting the amp star ground to what could be a horribly polluted domestic earth and also prevents the amps metal casing acting as some kind of RF aerial that is being injected in the amp.

Now there is strong filtration between the domestic earth, amp casing etc and the amps power supply system. Not sure if further cleaning up of the domestic earth would bring further improvements but for now I happy with things how they are.

Hope this useful

mike
 
Hi naf

You don't need 1 W zener since there's cca. 10 mA zener current, multiply 15 V, so you end at 150 mW power dissipation on one zener diode. 500 mW zeners are OK.

If you dont have 1,2 k resistor than parallel (one resistor on top other on bottom side) two 2,4 k or 5,6k parallel to 1,5 k etc. It will be easier later to adjust correct value in this way, idea is to parellel them rather than serial connection. ;)
Woaa!!! thank you very very much LC for explanation and hint on building this baby:up:
not many people/designer willing to explain much detail like this :worship: You're the man:cool:
 
Last edited:
I recently changed my amps over to this arrangement and I am very happy with it. When I measure at the amps o/p on the scope now noise appears to be a little lower.

Hi Mike

If scope measurement showed improvement at noise floor, than this is real improvement not just subjective. The whole idea to control ground currents or at least to know where their paths are in GND, GND star connection, that sounds like an improvement to me. Earth potential connection to the metal case is a must but to sensitive audio signals this brings also the whole spectra of mains noise currents to GND which is bad but almost unavoidable. In this way you gain control over the noise interference currents, directing them to flow away from sensitive GND areas. :up:
 
Hi naf

You don't need 1 W zener since there's cca. 10 mA zener current, multiply 15 V, so you end at 150 mW power dissipation on one zener diode. 500 mW zeners are OK.

With 1K2 resistor the current at 15v will be over 12mA into the feedback bridge. In my experience the zener current should be higher than this for effctive regulation, requiring 470ohm or so for the zener feed resistor and essentially a 1w zener(from my last year's messed up SSA attempt).

IMHO, the 1/2 watt zener can not regulate the voltage for the total required current, which can be as high as 15mA if the VAS devices have lower gain than usual.
 
With 1K2 resistor the current at 15v will be over 12mA into the feedback bridge. In my experience the zener current should be higher than this for effctive regulation, requiring 470ohm or so for the zener feed resistor and essentially a 1w zener(from my last year's messed up SSA attempt).

IMHO, the 1/2 watt zener can not regulate the voltage for the total required current, which can be as high as 15mA if the VAS devices have lower gain than usual.

There's 10 V voltage drop on 470 ohm resistor, meaning 21,3 mA current flow, minus 12 mA of current through 1,2 k resistor, what remains is 9,3 mA zener bias current. Zener current regulation required is 9,3 mA +/- 5 mA at max.

To my opinion 500 mW zener can do that without problem, at least what datasheet says. :)
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.