Content-Transfer-Encoding: 8bit
Message-ID: <5.1.0.14.0.20030429113850.00b0b6d0@gemini.herts.ac.uk>
Date: Tue, 29 Apr 2003 12:17:01 +0100
To: rsgb_lf_group@blacksheep.org
From: "James Moritz" <j.r.moritz@herts.ac.uk>
In-reply-to: <000d01c30dc2$7c758940$cecdfc3e@ian>
References: <5.1.0.14.0.20030416120614.024e6ea0@gemini.herts.ac.uk>
MIME-Version: 1.0
Subject: Re: LF: BK amps.
Content-Type: text/plain; charset=us-ascii; format=flowed
Sender: <owner-rsgb_lf_group@blacksheep.org>
Precedence: bulk
Reply-To: rsgb_lf_group@blacksheep.org

Dear Ian, LF Group,

At 21:11 28/04/2003 +0100, you wrote:
>Hi Jim,
>
>in your posting of 16 April on 'The Mystery....' you mentioned that the BK
>input circuitry could be modified to provide a bit more gain at 136.
>
>What needs changed to do this?
>
>73,
>
>Ian

In the Maplin audio amp, (and judging from the circuit diagram in the old 
LF handbook the BK amps have an almost identical circuit), the gain is 
actually adequate, but the slew rate limits the output at high frequencies. 
This means that the output voltage cannot change fast enough (which is what 
slew rate is, in V/us) to produce a sine wave at the full output amplitude, 
so what you get instead is a more or less triangular waveform with a 
somewhat smaller amplitude. The slew rate is limited by the current 
available from the input stage to charge and discharge the frequency 
compensation capacitors in the second stage, which are required for 
stability. The solution is to increase the bias current in the input 
transistors - however this in itself will make the amp unstable because of 
the increased loop gain it produces. Fortunately, the status quo is easily 
restored by adding emitter degeneration resistors to the input transistors 
to bring their gain back down to the original value. The end result is an 
amplifier with the same gain and small-signal bandwidth, but able to 
produce full power at higher frequencies.

I can't remember the exact details of the circuit and the changes I made - 
I will check at home later. I think the input stage was originally biased 
to 500uA and I increased this to 5mA in each transistor - the emitter 
resistors required would be about 45ohms in this case, but I'm not sure if 
this is correct at the moment.

Cheers, Jim Moritz
73 de M0BMU