Curiosity got the better of me. I did check the Supertex. In the to252 package it can dissapate a maximum of 2.5 watts. A 50v drop at 20mA would be 1 watt, so should be ok. But careful heat sinking would be in order and I'm not sure how it's done with that type of package.
Sheldon
Sheldon
Re: Voltage Regulator Example
Coming in late from the side here (in rugby one gets a penalty for that), I am just a little hesitant about the amp being for dc, not ac.
A regulator is by definition there to keep the supplied output voltage constant, which translated means that it has a low internal impedance. For amplifiers this has to be so over the whole audio band, though, which implies that the regulator's internal amplifier has to be good for all audio frequencies, or the internal impedance would rise at h.f. This could lead to instability depending on the load. Now there is usually a capacitor over any regulator's output which would probably make the output impedance low enough. But just a thought: It is not necessarily good to attenuate internal amplifier response with too large a Miller effect.
Regards
Johan
Miles Prower said:
Coming in late from the side here (in rugby one gets a penalty for that), I am just a little hesitant about the amp being for dc, not ac.
A regulator is by definition there to keep the supplied output voltage constant, which translated means that it has a low internal impedance. For amplifiers this has to be so over the whole audio band, though, which implies that the regulator's internal amplifier has to be good for all audio frequencies, or the internal impedance would rise at h.f. This could lead to instability depending on the load. Now there is usually a capacitor over any regulator's output which would probably make the output impedance low enough. But just a thought: It is not necessarily good to attenuate internal amplifier response with too large a Miller effect.
Regards
Johan
Hmmm... my call, is it? MY call?!? I think we all know what happens when I make the call. Remember the oil-for-food thing?
Anyway, I likes me the learnin', so I think I'll play with PSUD, but I might also just run with the suggested choke / cap.
I think I'll stay away from this. Thanks for the input.
Well, I'm obviously too dumb to respond to this, but I know some of you ain't and I just bet there's a seventeen-posting-long debate awaiting this one.
Let's watch and see, shall we?
I'll get together a final parts list by this weekend and post it. Any other suggestions would be most welcome.
Thanks again!
Kofi
Re: Re: Voltage Regulator Example
The voltage regulator circuit is basically a hollow state version of the same thing implemented with BJTs. In the solid state case, the gain of the error amp is much greater than what you get even with a high g(m) pentode like the 6CB6. That means an even bigger Cmiller.
Solid state or hollow state: I have never had any sort of instability issues arising out of the use of such regulators. In the application where it's used, it works great.
Johan Potgieter said:
The voltage regulator circuit is basically a hollow state version of the same thing implemented with BJTs. In the solid state case, the gain of the error amp is much greater than what you get even with a high g(m) pentode like the 6CB6. That means an even bigger Cmiller.
Solid state or hollow state: I have never had any sort of instability issues arising out of the use of such regulators. In the application where it's used, it works great.
An externally hosted image should be here but it was not working when we last tested it.
Check out this link for a possible Supertex application:
http://www.antiquewireless.org/otb/resto0504.htm
🙂
http://www.antiquewireless.org/otb/resto0504.htm
🙂
No sure Miles,
I was talking very generally; more an additional point than criticism. Just the old stuff of a super regulation quoted but which is only valid at dc. I have also never had instability problems but then as said one usually also uses a capacitor across the output so that there is low (enough) impedance also at h.f. But I have read about cases (more generally with BJTs) where parasitic oscillation was traced to the finite output impedance of a heavily compensated regulator chip. So there is not normally a problem; just something to be kept in mind depending on the application.
I am in agreement with your previously set out filosophy.
Regards
I was talking very generally; more an additional point than criticism. Just the old stuff of a super regulation quoted but which is only valid at dc. I have also never had instability problems but then as said one usually also uses a capacitor across the output so that there is low (enough) impedance also at h.f. But I have read about cases (more generally with BJTs) where parasitic oscillation was traced to the finite output impedance of a heavily compensated regulator chip. So there is not normally a problem; just something to be kept in mind depending on the application.
I am in agreement with your previously set out filosophy.
Regards
^^^^
I really had no idea where you were going with that. Naturally, you do bypass for high frequencies. Anyone wants to criticize, they can come over for a listen. That'll put a stop to that. 😀
I really had no idea where you were going with that. Naturally, you do bypass for high frequencies. Anyone wants to criticize, they can come over for a listen. That'll put a stop to that. 😀
Hey-- can I use a volume pot / attenuator for volume control with the Bevois Valley or do I need to drive it from a preamp? If I can use the pot / attenuator, what value would be best?
Putting together a parts list now and I'd like some input on this.
Kofi
Putting together a parts list now and I'd like some input on this.
Kofi
Kofi Annan said:
The Bevois Valley has a sensitivity of 2VRMS, which is what a typical CD player outputs. So, no need for a preamp unless you have a low output source, such as a turntable or some FM tuners. The pot value would be determined by your source output impedence. It should be at least 10 times that value, 20x is better. Many sources will have an output impedence of about 100 ohms, so a 20k pot is a good starting point.
Sheldon
Kofi,
On my "Bevois Valley" I run separate 50K Log pots, one for each channel. That way there is no need for a balance control. You get used to two handed adjustment of volume very quickly.
Cheers,
Ian
On my "Bevois Valley" I run separate 50K Log pots, one for each channel. That way there is no need for a balance control. You get used to two handed adjustment of volume very quickly.
Cheers,
Ian
So... had some mishaps and a LOT of work and I'm about ready to order some parts for this thang. So about those potentiometers...
So, there's a shunted 1M resistor at the input of the Bevois Valley amp. If I am instituting a potentiometer, do I do away with this or do I just add the pot so that the input sees the pot first, then the 1M resistor?
Things are getting odder at the Annan household lately and I don't have much of a window to order these things, so the sooner you reply...
Kofi
So, there's a shunted 1M resistor at the input of the Bevois Valley amp. If I am instituting a potentiometer, do I do away with this or do I just add the pot so that the input sees the pot first, then the 1M resistor?
Things are getting odder at the Annan household lately and I don't have much of a window to order these things, so the sooner you reply...
Kofi
Kofi,
Just leave that 1M resistor there. It won't do anything much as far as changing the response curve of the pot BUT it does provide some protection against "colateral damage" if the wiper connection to the pot fails for any reason (old age, abuse, shoddy part, terrorist or Mosad actions etc.).
Cheers,
Ian
Just leave that 1M resistor there. It won't do anything much as far as changing the response curve of the pot BUT it does provide some protection against "colateral damage" if the wiper connection to the pot fails for any reason (old age, abuse, shoddy part, terrorist or Mosad actions etc.).
Cheers,
Ian
Kofi Annan said:
Keep the resistor and add the pot. Take your source output to one side of the pot and connect the other side to signal ground. Wiper goes to the input as drawn (junction between 1M resistor and the 330R grid resistor). I think this is what you meant above. Assuming you have the book, look at the figure on pg. 484, "daughter of beast", for an example. Picture, thousand words, etc..
Sheldon
Kofi,
I would suggest 2W resistors for the 47R Screen Resistors.
Did'nt see ECC88 or EL84 Tubes listed - JJ Electronics perfectly OK for these too.
Cheers,
Ian
I would suggest 2W resistors for the 47R Screen Resistors.
Did'nt see ECC88 or EL84 Tubes listed - JJ Electronics perfectly OK for these too.
Cheers,
Ian
The 47Rs are actually 2W jobs-- should have detailed that. Also, I already have the EL84s and the ECC88s-- they're all JJs.
Any other issues? I know there's a catastrophic one in there somewhere. Its me, after all.
Kofi
I'm ordering this stuff this morning. Any other words of wisdom? Any other issues with parts selection? Anyone have any reason why this man and this amplifier should not be joined together?
Kofi
Kofi
- Status
- Not open for further replies.