"N Cook" wrote in message
...
Arfa Daily wrote in message
...

"N Cook" wrote in message
...
Broken input socket but I noticed a burnt 2W resistor.
Uses 100W TDA7293 , no mention on ST PDF manual of this R-C snubber?
between speaker output and ground. 100V,1uF cap in series with 2W
resistor,
measuring existing sections maybe was 3.9 ohm , anyone happen to know
correct value .? Or hazard a suggestion , obviously not too critical as
probabably months of use with it being at near infinite resistance,
metal
oxide , does it increase or decrease in value on excessive heating ?


The R-C combination is a Zobel network, and can be quite important to the
correct operation of the amp. It's purpose is to neutralise the
inductance
of the speaker voice coil, with the intention of flattening its impedance
curve at the HF end. However, without this network in place and doing its
job, I have known output stages to burst into spontaneous ultrasonic or
even
RF oscillation. The values of R and C are calculated according to some
esoteric formula that I'm sure real audio buffs could quote at you
chapter
and verse, but the general rule of thumb is that you start with a
resistor
of equivalent value to the voice coil's nominal impedance, and then
calculate C from there. Assuming that Carlsbro ( who are actually quite
helpful if you phone them ) know this, then there's a good chance that if
the speaker is 8 ohms, the resistor will also be in this region.

Arfa

2 different Carlsbro this one Beaver and the other one, Hornet, with 2 x
3055
I'd not heard of Zobel.
Does it relate to a specific type of speaker as there is no mention on the
ST sheet relating to the TDA7293 and this R-C



Ah, OK.

I didn't look at the postings quite closely enough to spot that thay
referred to different amps. I just saw the Carlsbro bit and assumed both the
same job. Zobel networks are absolutely standard on the ends of amps. As far
as I am aware, they are nothing to do with any particular type of speaker,
so long as we are talking one with a normal voice coil, which obviously
possesses inherent inductance. I guess that " Zobel " is actually some guy's
name. It must be germanic, as it is normally spelt with an umlaut-ed "o".
Anyway, one job or two, what I said about such networks is valid anyway. Try
a Google on " Zobel Network ". I just did, and there is cartloads of info. I
had a quick scan of one article picked at random, and it said basically the
same as I did.

Arfa