The problem with really big power amps...

AVForums

Help Support AVForums:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

JimGore

AVForums Grandmaster
*
Joined
Jul 8, 2005
Messages
4,488
Reaction score
32
Location
Jhb
I have been busy working on something to supersede the SM-150 power amplifier for a number of months now, and things get interesting quickly.

The basic topology will remain symmetric complementary differential (same as the SM-150), but implemented a bit differently and optimized in other ways.  Furthermore it will be fully balanced mono-block design  and will most likely use a buffered input stage so I can get away with a couple of other wild ideas I have.

Anyway, one of the other interesting things is to push for more power.  More power is somehow always better (at least in the mind), so why not?

With increasing power comes very interesting design and engineering challenges.  Here are just a few of them:

1.  The transformer becomes huge very quickly (4kVA +).  Space is a problem and powering it up is a problem (huge    HUGE      H.U.G.E    inrush current).
2.  With such a large transformer or pair of transformers the bridge rectifier becomes a problem - your average 35 and 50 Amp rectifiers are no longer quite strong enough to cope with the massive current, especially when the transformer becomes overloaded such as during power-up and in the event of a dead short or internal fault.
3.  The standard IEC power inlet you find on nearly all audio equipment becomes a problem as it can only really do 16 Amp, and that not continuously.  With this kind of power we will need something much more serious.  There are very serious power inlets to be had in the industrial side, but they leave me somehow uninspired.  Perhaps direct connection would be best.
4.  The power dissipation in the output transistors becomes a problem, so you need many, many pairs.  Perhaps 40 large transistors or more will be needed.
5.  Getting rid of the heat generated by the output stage is a problem.  Your average heatsinks simply can not deal with the huge amount of heat generated.  Fans could work but are not kosher, so other ways are needed.
6.  With so much power on tap, the protection circuitry becomes paramount.  It must work fast and with an iron fist in the event of disaster or else your speaker could melt and your nice expensive speaker cable simply vaporized, leaving a black trail on the carpet where it used to lie  :EGrin:

I guess the only question is how much power do we really need, and how far down should it double down?  My gut feel is that it should double down into 2 ohms at least, but who is to say what the real answer is.

So this makes things kind of difficult on many levels, but a challenge is always fun (to me at least).

Feel free to share your thoughts and / or experience.  Let's keep the discussion light-hearted as far as possible - no need for lines drawn in the sand or boxing gloves here  :2thumbs:

Kind regards,
Ian.
 
Top