Output protection

[mikeks]

Folks...digikey has a UK subsidiary....or at least a UK website....

http://dkc1.digikey.com/uk/digihome.html

 

[mikeks]

.........low voltage transformer. 240:10:10V should work nicely, and under 5VA would be dandy.

I would suggest 20V-25V secondarys......with the excess DC output dropped across a rudimentary zener-based regulator....

........to ensure minimum Rds(on) MOSFET should ideally be driven to 15V....

.........12V should be seen as an absolute minimum....

 

[ShinOBIWAN]

I would suggest 20V-25V secondarys......

........to ensure minimum Rds(on) MOSFET should ideally be driven to 15V....

.........12V should be seen as an absolute minimum....

Mike, who's the bird in your sig?

I'm sure I've seen her before, for some reason I keep thinking she was an athelete

Sorry nothing to do with output protection but just trying to satisfy my curiosity.

 

[mikeks]

One delectable bird to be sure....

 

[ShinOBIWAN]

One delectable bird to be sure....

Ah, judging by that response I'd say its your GF or wife 😉

 

[Upupa Epops]

To Mike : 25 V AC give cca 35 V DC - do you know any mosfet, which have breakdown voltage of Vgs over 20 V ? You don't need for this circuit regulated voltage, only rectifier + filtration 😎 .

 

[mikeks]

Many modern mosfets have a continous breakdown voltage of 30V....

Of course you don't want to even approach that.

22V secondarys (quite common), should give roughly 28V across your reservoirs...(ripple troughs!)

Allow for 15V gate drive, and drop the rest across your zener regulator and gate drive circuit.....

 

[mikeks]

You don't need for this circuit regulated voltage, only rectifier + filtration 😎 .

I wouldn't mind a bit of precision in my gate drive supplies....the cost of a zener+resistor tends to zero compared to the other stuff...

 

[mikeks]

To Mike : 25 V AC give cca 35 V DC - do you know any mosfet, which have breakdown voltage of Vgs over 20 V ? You don't need for this circuit regulated voltage, only rectifier + filtration 😎 .

Remember transformer regulation can be as bad as 30%...!!
😱

This has to be accomodated...

I would say minimum 18V rms secondarys...

 

[anatech]

This type of protection is getting complicated compared to a relay. Sense circuits being equal and all.

-Chris

 

[mikeks]

Not really...intend to get a prototype up and running within the week.....hack it into an amp. for testing...

 

[anatech]

Hi mikeks,
I didn't say tough to build. Just more complexity since the detection circuits are the same. Anyhow, I'm positive it'll work as advertised.

Any though to leaving the front end energized to eliminate turn on and off thumps? Just an idea. Esp. if you have a regulated, separate supply for the front end. I would expect that from a design with this much forethought.

Build on McDuff!

-Chris

 

[mikeks]

Any though to leaving the front end energized to eliminate turn on and off thumps?

Front end should always have supplies independent of those of the output stage...they don't have to be regulated...

 

[anatech]

No, but I've found regulation really helps the noise floor. More important when the volume goes up with low impedance loads. But you are right, it's not neccessary.

-Chris

 

[mikeks]

No, but I've found regulation really helps the noise floor.....

It should...if the gain block has poor PSRR...

A recent design of my mine has >100dB PSRR across the audio band...for both +ve and -ve rails....Probably a record... 😎

No need for regulation there...no noise either....

 

[anatech]

That's really good .... but it could be better with regulation. 😉 With so much attention on every detail why leave something so basic out?

-Chris

 

[mikeks]

Unregulated supplies give vastly greater dynamic headroom.....

This is a vital ingredient for the 'realistic' reproduction of transients...

 

[mikeks]

That's really good .... but it could be better with regulation. 😉 With so much attention on every detail why leave something so basic out?

-Chris

....regulation does not improve the amp's PSRR....

-100dB is attenuation of 100 000 times....cannot be audibly improved upon......

 

[anatech]

I will disagree on two points.

Firstly, there are no great current transients in the front end. Therefore no line sag and no lost in headroom.

Secondly, dynamic supply shifts can cause small operating point shifts. High frequency noise is another nasty we like to block. Before you start with current sources, I use them too and have made some mighty quiet preamps and front ends. Each of them improved with supply regulation and lead dress.

Please understand I'm not trying to be a critic. It's just my practical experience differs from yours apparently. With most of us chasing the last bit of performance on barely audible issues, I believe this qualifies.

Another point. You will not find a single piece of good electronic test equipment made without attention paid to the supply and it's regulators. Why should an audio amplifier deserve any less? Or should we just copy the analog stages of test equipment and call those superior to what we are building?

-Chris

 

[mikeks]

I will disagree on two points.

Firstly, there are no great current transients in the front end. Therefore no line sag and no lost in headroom.

More to do with voltage transients...an unregulated supply can give vastly greater voltage swing,...on a transient basis, beyond its nominal voltage....than a regulated supply....

Hence the increase in dynamic headroom...

Secondly, dynamic supply shifts can cause small operating point shifts. High frequency noise is another nasty we like to block. Before you start with current sources, I use them too and have made some mighty quiet preamps and front ends. Each of them improved with supply regulation and lead dress.

On the contrary, current sources should never be entirely relied upon for good psrr...

This is because their effectiveness declines pro rata with frequency, due to their internal parasitic reactances...

The active loads must therefore be complemented with good old fashioned C-R filtering...