[sdiy] transistor op-amp problems

Elain Klopke functionofform at gmail.com
Mon May 28 04:29:39 CEST 2012

Ah, ok, thank you for that explanation, Phil. That makes a lot more sense now.

On Sun, May 27, 2012 at 3:58 PM, Phil Macphail
<phil.macphail at liivatera.com> wrote:
> Hi Elain,
> Sorry for the delay in replying, I have been away from home for a few days
> but here goes
> Breaking the problem into steps for clarity -
> 1. The resistor Rprg is 100k with a little over 10Volts across it. This
> gives a little over 100uA into Q5, and is the bias circuit I was referring
> to. Reduce this resistor value to run the design at a higher current.
> 2. This 100uA is mirrored by Q6, so each side of the differential pair
> will carry 50uA when balanced, so with no load the current mirrored by
> Q1,Q2 will be equal to the collector current of Q4 and the output voltage
> doesn't change.
> 3. Any load on the output (such as the resistors in red) will unbalance
> the differential pair - the limit is when Q4 isn't conducting and Q3
> carries the full 100uA. In this case the 100uA is mirrored by Q1,Q2, and
> there is no current flowing through Q4.
> 4. The maximum current into the feedback resistors and load is therefore
> 100uA.
> 5. With the resistors you chose (10k + 1k) with 100uA flowing through them
> gives a voltage of 1.1Volts - assuming all transistors are perfectly
> matched - which is pretty much the voltage you measure.
> So if you want a larger output voltage range you will need to follow the
> suggestions in my first mail. To make a useful amplifier from this circuit
> you need to add a buffer to the output, so this circuit is really of
> educational value only (with the limitations others mentioned too).
> On 26/05/2012 05:47, "Elain Klopke" <functionofform at gmail.com> wrote:
>>On Fri, May 25, 2012 at 2:46 PM, Phil Macphail
>><phil.macphail at liivatera.com> wrote:
>>> If you have used the same bias circuit, then the amplifier can only
>>>source/sink around 100uA. Into 10k that is about 1 volt,
>>> Either increase the bias current or increase the load impedance ( or
>>>both), add a buffer or other gain stage and it should start to work,
>>> Good luck,
>>> Phil
>>You're probably going to have to show me a diagram of what you're
>>talking about Phil.... Just so I can make sure I get it.
>>Synth-diy mailing list
>>Synth-diy at dropmix.xs4all.nl

More information about the Synth-diy mailing list