What about savings on a smaller heatsink and smaller psu ?
To me class D sounds just as good as class AB.
With a little care class d is fairly straight forward.
Vendors not only provide working schematiocs but pcb layouts too, I dont see what more they could do to be helpful.
I have made 5 class d designs which all work very well.
One was a class d with a valve front end.
Well, Gentlemen, let's go back to the item !
That's the point. You hit the nail on it's head!
If you'd think about using two tubes per channel, one at each side of each PP arrangement, there might be a chance to choose a pair out of a greater lot with the same sums of transconductance and idle current per bulb. And this would also let you think of ultra linear operation mode. But, due to the faster degeneration of the more powerful section within one bulb you'd have to expect re-biasing the tubes regularly, not to say very often. Just my thoughts...
Best regards!
That's the point. You hit the nail on it's head!
If you'd think about using two tubes per channel, one at each side of each PP arrangement, there might be a chance to choose a pair out of a greater lot with the same sums of transconductance and idle current per bulb. And this would also let you think of ultra linear operation mode. But, due to the faster degeneration of the more powerful section within one bulb you'd have to expect re-biasing the tubes regularly, not to say very often. Just my thoughts...
Best regards!
After studying both blocking distortion and crossover distortion in a P-P tube output stage (EL84's in my case) and trying several things to reduce both, I am starting to believe that blocking distortion is pretty much always bad sounding, and crossover distortion without the blocking distortion component, isn't necessarily that bad sounding, and may even be good in a guitar amp if done right.
What's right? Biasing the tubes to run hotter reduces crossover distortion, but often pushes the tubes power dissipation beyond the max rating (unless you reduce B+ which quickly drops max output watts, since voltage is a squared term in the power formula), and makes the blocking mechanism more active (the output tubes saturate sooner).
It seems like the best thing is to use the Paul Ruby method to drastically reduce the blocking mechanism (it did work well for me - certain others apparently did it wrong), and put a "conjunctive filter" across the speaker terminals (20 ohm 5W in series with a 0.22uF cap in my case - if you use negative feedback this may open another can of worms by becoming the dominant pole - I prefer no feedback in a guitar amp).
Since the impedance of probably any guitar speaker (and most Hi-Fi spkrs) rises from nominal as you go up in frequency, especially above about 8kHZ (for guitar speakers), the output transformer will ring when it receives high frequency energy (maybe 10kHZ and up). The OT ringing is triggered by the high frequency spectra of bad crossover distortion (and sharp cornered clipping). Crossover distortion caused by blocking distortion seems to have by far the most high frequency spectra, which can cause quite a bit of OT ringing, which is highly unlikely to be pleasantly musically related to the notes being played on the guitar. After using the Paul Ruby zener method of dumping the blocking distortion, and using dummy load R's (without conjunctive filter - doesn't need them), I find that the crossover distortion that is left is very smooth on the scope; darn little if any high frequency spectra being generated. It may actually sound good in a guitar amp, and provides a more gradual transition from totally clean to clipping. Has anyone else had any experience with this?
What's right? Biasing the tubes to run hotter reduces crossover distortion, but often pushes the tubes power dissipation beyond the max rating (unless you reduce B+ which quickly drops max output watts, since voltage is a squared term in the power formula), and makes the blocking mechanism more active (the output tubes saturate sooner).
It seems like the best thing is to use the Paul Ruby method to drastically reduce the blocking mechanism (it did work well for me - certain others apparently did it wrong), and put a "conjunctive filter" across the speaker terminals (20 ohm 5W in series with a 0.22uF cap in my case - if you use negative feedback this may open another can of worms by becoming the dominant pole - I prefer no feedback in a guitar amp).
Since the impedance of probably any guitar speaker (and most Hi-Fi spkrs) rises from nominal as you go up in frequency, especially above about 8kHZ (for guitar speakers), the output transformer will ring when it receives high frequency energy (maybe 10kHZ and up). The OT ringing is triggered by the high frequency spectra of bad crossover distortion (and sharp cornered clipping). Crossover distortion caused by blocking distortion seems to have by far the most high frequency spectra, which can cause quite a bit of OT ringing, which is highly unlikely to be pleasantly musically related to the notes being played on the guitar. After using the Paul Ruby zener method of dumping the blocking distortion, and using dummy load R's (without conjunctive filter - doesn't need them), I find that the crossover distortion that is left is very smooth on the scope; darn little if any high frequency spectra being generated. It may actually sound good in a guitar amp, and provides a more gradual transition from totally clean to clipping. Has anyone else had any experience with this?
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.