Hi Gents,
As many of you will know I have been learning electronics by means of building amplifiers.
I have been working on this new design which I have up and running now (after a number of fuses blowing up of course).
The new design seems to work pretty well and faithfully reproduces just about anything I can throw at it between 1.0 Hz and 80kHz. Ringing, oscillations, etc all sorted out so the thing seems stable from that point of view.
The one thing that is giving me a bit of concern (and I don't know whether this is really an issue or not) is this:
- When the amp starts up from cold, the output transistor bias across one of the 0R22 emitter resistors sits at around 0.7mV
- As it warms up (takes a while because the heatsinks are large - thanks fdlsys) the bias level will slowly make its way up to around the 35mV mark.
- All good so far, except for this: After reaching its stable operating temp, the bias levels will drift by about 10mV in total. What this means is that it will slowly edge up to perhaps 40mV, then come back down to 35mV, then go back up to 37mV, then go down to 30mV, then ...
I tracked this with my multimeter (which has a data capture and graphing function) over a period of around 45 minutes after ensuring everything is warmed up, and the average seems pretty much around 35mV, but this drifting has me wondering about whether this is truly problematic or not.
BTW: I do know what's causing it. I have some current sources fighting each other and this drift is just those devices having a fist-fight to determine some dominance. I can make it behave better, but at the expense of the performance so I also need to find out what a suitable middle ground would be. In order to completely get rid of the problem, I will need to completely redesign the input stage and VAS. No problem in doing that, but first I want to check out how this thing works, sounds, etc.
Thanks,
Ian.
As many of you will know I have been learning electronics by means of building amplifiers.
I have been working on this new design which I have up and running now (after a number of fuses blowing up of course).
The new design seems to work pretty well and faithfully reproduces just about anything I can throw at it between 1.0 Hz and 80kHz. Ringing, oscillations, etc all sorted out so the thing seems stable from that point of view.
The one thing that is giving me a bit of concern (and I don't know whether this is really an issue or not) is this:
- When the amp starts up from cold, the output transistor bias across one of the 0R22 emitter resistors sits at around 0.7mV
- As it warms up (takes a while because the heatsinks are large - thanks fdlsys) the bias level will slowly make its way up to around the 35mV mark.
- All good so far, except for this: After reaching its stable operating temp, the bias levels will drift by about 10mV in total. What this means is that it will slowly edge up to perhaps 40mV, then come back down to 35mV, then go back up to 37mV, then go down to 30mV, then ...
I tracked this with my multimeter (which has a data capture and graphing function) over a period of around 45 minutes after ensuring everything is warmed up, and the average seems pretty much around 35mV, but this drifting has me wondering about whether this is truly problematic or not.
BTW: I do know what's causing it. I have some current sources fighting each other and this drift is just those devices having a fist-fight to determine some dominance. I can make it behave better, but at the expense of the performance so I also need to find out what a suitable middle ground would be. In order to completely get rid of the problem, I will need to completely redesign the input stage and VAS. No problem in doing that, but first I want to check out how this thing works, sounds, etc.
Thanks,
Ian.
