i am in the process of designing my 6L6 and i cannot make up my mind if i should run the amp in AB1 or AB2..
i have this problem because i would love the cool 47watts of power a AB2 pair of 6L6s can push out.
what are the short commings of a class AB2 amp, sonically and electrically?
i was looking at the datasheet and it states seems the only difference between the 6L6 operating in AB1 or 2 is the impedance of the opt. it is between 9k and 3.8k. could i get poly to wind 2 sets of taps on the opt and let my ears decide later?
regarding the above, the datasheet states that the tube will draw 88ma on idle, it also states that a 250ohm cathode resistor should be used. an inteligent guess tells me that the cathode would have a potential of 22v. on the same datasheet, it states that if i were to operate Class AB2, the trans needs to be 3.8 insted of 9k and it requires a cathode potential of 22v. but, its maxium current draw is 200ma insted of 100ma for AB1. would i need another 22v tap on my mains trans or would a 250ohm resistor do in both cases?
Thanks!
i have this problem because i would love the cool 47watts of power a AB2 pair of 6L6s can push out.
what are the short commings of a class AB2 amp, sonically and electrically?
i was looking at the datasheet and it states seems the only difference between the 6L6 operating in AB1 or 2 is the impedance of the opt. it is between 9k and 3.8k. could i get poly to wind 2 sets of taps on the opt and let my ears decide later?
regarding the above, the datasheet states that the tube will draw 88ma on idle, it also states that a 250ohm cathode resistor should be used. an inteligent guess tells me that the cathode would have a potential of 22v. on the same datasheet, it states that if i were to operate Class AB2, the trans needs to be 3.8 insted of 9k and it requires a cathode potential of 22v. but, its maxium current draw is 200ma insted of 100ma for AB1. would i need another 22v tap on my mains trans or would a 250ohm resistor do in both cases?
Thanks!
Something like that. For normal drive, the output stage grids draw essentially no current. So the load is mostly just the output stage grid resistors. In AB2, when the drive voltage takes the output stage grids above zero volts, the grids start drawing current. So the load's resistance drops quite a bit and it starts to draw current. It's a very nonlinear resistance and it takes some power to drive. All bad juju for a normal driver stage.
Two common solutions are cathode follower drive and interstage driver transformers. The former is what I prefer (actually, I use a source follower), but you have to pay close attention to how all the coupling and power supply time constants are set in order to prevent blocking distortion and slow recovery from overload. You also need some pretty stiff supply rails, negative and positive, for the follower. The latter is a common approach (I first saw it in the classic Fisher 50A), but requires more drive voltage from the driver stage. A transformer that's capable of this service is neither small nor cheap. And like all interstage transformers, new rolloffs are introduced which can make stabilizing feedback schemes quite tricky.
For a beginner-to-intermediate designer/constructor, stick with AB1. There's just not that much audible loudness difference between 45 watts and 35 watts.
Two common solutions are cathode follower drive and interstage driver transformers. The former is what I prefer (actually, I use a source follower), but you have to pay close attention to how all the coupling and power supply time constants are set in order to prevent blocking distortion and slow recovery from overload. You also need some pretty stiff supply rails, negative and positive, for the follower. The latter is a common approach (I first saw it in the classic Fisher 50A), but requires more drive voltage from the driver stage. A transformer that's capable of this service is neither small nor cheap. And like all interstage transformers, new rolloffs are introduced which can make stabilizing feedback schemes quite tricky.
For a beginner-to-intermediate designer/constructor, stick with AB1. There's just not that much audible loudness difference between 45 watts and 35 watts.
You cannot use a cathode resistor for the output stage bias in class AB1 or 2 because the tube current varies widely depending upon the drive (output power). Although cathode bias is seen in some examples of AB1 designs (e.g. Leak and Mullard). it won't give you full ouput power.
You should used fixed bias, i.e. OP tube cathodes grounded and a fixed negative voltage applied to the grids. A convenient way to achieve this is by using direct-coupled cathode follower (CF) drivers (best for AB2 and not a bad idea for AB1). The CF cathode resistors are connected to a fixed and stable negative voltage and a fixed (but pre-adjusted) negative voltage is applied to the grids of the CFs. The CF cathodes, in turn, will provide the negative bias required at the grids of the output tubes. The negative voltage applied to each CF grid is adjusted until the quiescent current through each OP tube is correct.
You should used fixed bias, i.e. OP tube cathodes grounded and a fixed negative voltage applied to the grids. A convenient way to achieve this is by using direct-coupled cathode follower (CF) drivers (best for AB2 and not a bad idea for AB1). The CF cathode resistors are connected to a fixed and stable negative voltage and a fixed (but pre-adjusted) negative voltage is applied to the grids of the CFs. The CF cathodes, in turn, will provide the negative bias required at the grids of the output tubes. The negative voltage applied to each CF grid is adjusted until the quiescent current through each OP tube is correct.
Yes and no. Cathode bias will result in a less efficient amp with lower power. In that sense, "it won't give you full ouput power." But it will still be running AB. The key is that the cathode resistor is bypassed.
So-called fixed bias will give you more power capability, but it's often trickier to ensure good overload capability and there's a higher parts count. Additionally, fixed bias circuits are more susceptible to output tube runaway and will often require lower value grid resistors (which loads down the driver stage more). As with everything, it's an engineering choice as to which tradeoffs are more acceptable to the designer.
SY,
Aren't you describing the difference between class A1 and A2? I thought AB1 and AB2 differed according to how much of a conduction angle each tube was biased to supply? Or said another way, how close to class B. I think the push pull stuff on Steve Bench's site goes into this nicely.
I was wondering how you would notate AB operation drawing grid current, A2B1? Et tu Brute? R2D2?
Aren't you describing the difference between class A1 and A2? I thought AB1 and AB2 differed according to how much of a conduction angle each tube was biased to supply? Or said another way, how close to class B. I think the push pull stuff on Steve Bench's site goes into this nicely.
I was wondering how you would notate AB operation drawing grid current, A2B1? Et tu Brute? R2D2?
Well, there goes another misconception. Thanks.
It seems that I'm not alone this, however. I see quite a bit of information actually stating both sides of that at the same time, often stating at the beginning that grid current was the defining factor, then later using conduction angle. Odd, as the conduction angle has to do with the more negative portion of the grid swing and grid current the more positive.
It seems that I'm not alone this, however. I see quite a bit of information actually stating both sides of that at the same time, often stating at the beginning that grid current was the defining factor, then later using conduction angle. Odd, as the conduction angle has to do with the more negative portion of the grid swing and grid current the more positive.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.