I'm not surprised that replacing ceramic caps with film caps made such a difference. Ceramic caps are nearly all piezoelectric to some extent and distortion comes with the territory. Interestingly there's more of a connection between your circuit and the HP analyser than you may think. The Wien bridge oscillator with a light bulb in the feedback circuit to stabilise the gain was invented by Bill Hewlett as part of his degree thesis in 1939, he and Dave Packard then founded HP and Bill's oscillator was the basis of their first product, the HP 200A. I had the privilege of meeting Bill Hewlett in a restaurant in Palo Alto once. A really nice down to earth guy. Good luck with the oscillator,and greetings from Chipping Sodbury.
What a fascinating story, I remember reading about them making that oscillator in one of their garages when they started the company and of course nothing else came close to it in performance . I've long been a lover of hp having started of using it at work, then salvaging a 141t spectrum analyzer from the scrap heap, which I managed to fix but not the display, so I connected the output to a 'scope which worked well. What a shame that the name is now no longer used on the test equipment division. Yes I was surprised how much difference the capacitors made. I will make a pcb for it and maybe that will further improve it. You live near me then, I'm near Stroud. thanks for watching
Nice little circuit, and I'm sure things will be better once on a PCB and shielded. I suppose that wobble during start up is the lamp coming to temperature and stabilising then stabilising the sinewave; 73 Bob
Hi Bob, that will be the next part of the project making a pcb for it and maybe putting it in a metal box and yes the bulb is there for stabilization and hence why it takes time to settle, but putting it in thermal conductivity with the copper board should make it more stable. Thanks for watching
I'd think with an unshielded oscillator setup like that, with wires flying about, unless your lab is electrically very quiet (mine sure as hell isn't), you're going to see noise drive up the THD measurement.
Totally agree Matambale, or sorts of undesirables floating around in the workshop and the breadboard is not helping. When I get parts I ordered will make a pcb that should improve it and maybe stick it in a metal box. Thanks for watching
Put the final PCB version in a shielded metal enclosure. Looks like a lot of noise floor there. Filter the power leads. Coaxial out. Temp stability, well good luck with that. Great work on lowering THD, but I bet a lot is coming from other external sources. Anyone know if polystyrene capacitors would be better than Mylar films for Delta T?
Hi Danish, all good points. The final version will be in a metal box with a bnc on the front. Power supply not an issue as I've decided to run it on two 9v batteries. It takes minimal current and works fine at 9 volts, no difference in the output level or distortion. I'm now down to .0015% and a TLO74 (quad opamp) came today and produces exactly the same results. I've now got to make a pcb, which I'm not looking forward to! cheers Nick, ps, cold and snow today, what happened to spring?
Interesting how much difference swapping those caps made! I wonder if it's worth considering Mr Carlson's insistence on getting the outer foil on the earthy side? This should be a great circuit if built on a PCB with shielding. I've also found it's important to get small decoupling caps right next to the supply pins on any amplifier chip.
Hi Timothy, yes I'm surprised how well it works considering its built on a breadboard, I'm going to make a pcb for it and maybe put it in a metal box and also proper decoupling capacitors . thanks for watching
I'm not surprised that replacing ceramic caps with film caps made such a difference. Ceramic caps are nearly all piezoelectric to some extent and distortion comes with the territory. Interestingly there's more of a connection between your circuit and the HP analyser than you may think. The Wien bridge oscillator with a light bulb in the feedback circuit to stabilise the gain was invented by Bill Hewlett as part of his degree thesis in 1939, he and Dave Packard then founded HP and Bill's oscillator was the basis of their first product, the HP 200A. I had the privilege of meeting Bill Hewlett in a restaurant in Palo Alto once. A really nice down to earth guy. Good luck with the oscillator,and greetings from Chipping Sodbury.
What a fascinating story, I remember reading about them making that oscillator in one of their garages when they started the company and of course nothing else came close to it in performance . I've long been a lover of hp having started of using it at work, then salvaging a 141t spectrum analyzer from the scrap heap, which I managed to fix but not the display, so I connected the output to a 'scope which worked well. What a shame that the name is now no longer used on the test equipment division.
Yes I was surprised how much difference the capacitors made. I will make a pcb for it and maybe that will further improve it.
You live near me then, I'm near Stroud.
thanks for watching
Nice little circuit, and I'm sure things will be better once on a PCB and shielded. I suppose that wobble during start up is the lamp coming to temperature and stabilising then stabilising the sinewave; 73 Bob
Hi Bob, that will be the next part of the project making a pcb for it and maybe putting it in a metal box and yes the bulb is there for stabilization and hence why it takes time to settle, but putting it in thermal conductivity with the copper board should make it more stable.
Thanks for watching
Thumbs up to the old, reliable 741!
I'd think with an unshielded oscillator setup like that, with wires flying about, unless your lab is electrically very quiet (mine sure as hell isn't), you're going to see noise drive up the THD measurement.
Totally agree Matambale, or sorts of undesirables floating around in the workshop and the breadboard is not helping. When I get parts I ordered will make a pcb that should improve it and maybe stick it in a metal box. Thanks for watching
At a 2k load, an RC4558 can output a +13V off a 15V rail, or +14V at a 10k load. So 13V output sounds like its just below clipping.
Put the final PCB version in a shielded metal enclosure. Looks like a lot of noise floor there. Filter the power leads. Coaxial out. Temp stability, well good luck with that. Great work on lowering THD, but I bet a lot is coming from other external sources. Anyone know if polystyrene capacitors would be better than Mylar films for Delta T?
Hi Danish, all good points. The final version will be in a metal box with a bnc on the front. Power supply not an issue as I've decided to run it on two 9v batteries. It takes minimal current and works fine at 9 volts, no difference in the output level or distortion. I'm now down to .0015% and a TLO74 (quad opamp) came today and produces exactly the same results. I've now got to make a pcb, which I'm not looking forward to!
cheers Nick, ps, cold and snow today, what happened to spring?
Interesting how much difference swapping those caps made! I wonder if it's worth considering Mr Carlson's insistence on getting the outer foil on the earthy side? This should be a great circuit if built on a PCB with shielding. I've also found it's important to get small decoupling caps right next to the supply pins on any amplifier chip.
Hi Timothy, yes I'm surprised how well it works considering its built on a breadboard, I'm going to make a pcb for it and maybe put it in a metal box and also proper decoupling capacitors .
thanks for watching
Can we go down 50 Hz .
Why?