power amp proctection ckt!!!

[arasuk]

Hi Everybody!!

I was just going through Elektor published Millennium crescendo, not intreseted in the amp though, I was intrested in the Proctection ckt, published along with the millen cres amp.

I would like to know out there if somebody has implemented this proct Ckt to leach amp & super leach amp or would it be possible to implement this proctection ckt to leach amp.

As this ckt works on AC 2 x 35 volts , 50 volts DC applied , how about higher power ratings,

Thanks
Arasuk

 

[dherr555]

Protection

What about the circuit on prof . Leach's page for low tim . Link Is here . Leach

 

[Centauri]

For design info on GOOD dual-slope SOAR protection design, see the National Semiconductor LM391 datasheet http://www.national.com/ds/LM/LM391.pdf

Whilst the protection transistors are built into the chip, the design and calculations can be applied to any power amp. I have used this setup quite a lot and it works much better than simple systems.

Cheers
Graeme

 

[arasuk]

Proct ckt!!

Hi Graeme!!

Thanks for the reply, but what i am talking is the protection ckt separately built for the Millennium crescendo, which inturn if we can use it for our power amplifier.

Hi dherr555!!

The ckt of prof will work for any short ckt , but I am trying to put some prot ckt at the O/P of the amp in case of a Dc , overload etc...

This protection ckt has some good features like power on delay, overload prot, DC prot, voltage detector at the secondary of trans,

i am not able to post the pic here, if some body can please!! so that we can further discuss the issue here .

Thanks

Arasuk

 

[sam9]

It's a pretty common circuit. Wheter it's good or not depends on the implementation. The key values are the RE resistors and the base resistors that you would use be adapting it to a discrete power amp. It's a pain to try to hand calculate these and prone to simple error. I find a spice simulation is quiicker and you can test the action of the circuit onder various conditions including when the amp is clipping.

 

[mikeks]

I find a spice simulation is quiicker and you can test the action of the circuit onder various conditions including when the amp is clipping.

What method do you use to simulate your protection circuits?

 

[zcolic2]

Crescendo Millenium Edition protection circuit:

An externally hosted image should be here but it was not working when we last tested it.

Regards,
Zoran

 

[arasuk]

Prot Ckt!!

Hi Zoran!!!

Thanks a lot for the posting the schematic

tp1 , tp2 will measure the voltage across Re , tp3 will check for the O/p Dc,

this ckt is intended for AC 2 x 35 volts , 50 volts DC applied v+, v-

how about higher rail voltage applied region like leach amp.

Thanks
Arasuk

 

[sam9]

What method do you use to simulate your protection circuits?

Max non-clipping input singnal. Then run loads in decreasing steps (8, 4,3,2,1,.5, .1 -ohms) and check if the the presction transistors are turning on when they should but but when they shouldn't. It is usually in the low load but not true short range that there may be a problem.

I do the same thing with a hard clip signal, I don't really understand what's going on, but when clipping starts to resemble square waves, the VI relationships start looking wierd. By this I also mean that I can't generalize, -- different devices and different topologies seem to behave differently. What was safe just below clipping is suddenly unsafe with just a little more input -- and sometime the opposite. I would not be surprised if this has more to do with the Spice models than anything "real".

Anyway I plot the worst case VI relationships on to a derated SOA chart. If they stasy withing the envelope and the circuit doesn't become active at normal loads (>2.5 -ohm or pick your own), I'm happy.

A couple of observations - At normal listening levels the amp has a significantly greater margin for survival than if you are blasting the walls down. -Some Mosfet models bog down and even hang the simulation when very lowe loads are specified; I've no idea if this corresponds to any real world phenomonae. A dead short (Screwdriver across the binding posts) is actually easier to protect against than a seriously low load 1-3 ohms that a few speakers offer.