Firstly I note you are using a double ended power supply, so R40 still seems pointless.
You seem to be doing a lot of design for someone with "basic" knowledge. Can you detail your knowledge a little more?
Does your oscillator stop? I'm not sure exactly where you see the oscillations stop.
R40 is for impose more or less 100μA since at emiiter I'll have about 150mV ac signal peak to peak.
Well I have started to learn electronics on my own during one broken leg recovery. Instead of being at bed without doing nothing I thought that it could be great if I could maximize that time and study a little bit. Electronics it's one big paisson I have since child, but I never had the possibility to study it. Instead, when I was young I have studied electricity, so I have knowledge in basic series, paralel circuits but for high currents, ohm law, and so on. I never was electrician and believe it or no, I'm just a simple tram driver. So I started to learn 2 books entitled "Electronic Principles Vol1 and Vol2" from Malvino, published by McGrawHill. About ESR, I saw many circuits in the Internet and have decided to create my own circuit, trying to improve some aspects I thought I could deall with it... And this was how my adventure in Electronics have started.
My oscillator in fact stops. I know that because this simulator (National Instruments Multisim 10) have osciloscopes, so I'm always monitoring the signal, both at the oscilator and at any point of the circuit I need. Today I have been thinking in this and I just have created one theory. I dont know if it is correct or no, but here it goes. I think that oscilations are stoping because the peak detector is demanding too much current. At ground we have 0V but we have currents running trought it. Current at peak detector is bidirectional and probably at some part of the cycle may be demanding too much current from oscillator via ground (the only common point the circuit have). Probably there is also some phase issue like Kris have said in his last post. I dont know if using one transformer with the primary connected to the oscilator via one transistor and the secondary to the rest of the circuit, will isolate the problem... or who knows using some photo coupler circuit... well I have to test it, but its only a theory I just dont know...