What do you think? I am still working on the 6GM8 low power amp... just got bored. Instead of a BUF634 (original design) which is limited to 250mA current (peak), why not use a single MOSFET that *can* do 10A peak output? So, we end up with something like the attached file. If nothing else, it could make one evil heaphone amp. 
But, if it can do decent output (on 12V supply, an opamp could swing about 9V, can a tube as well?) it might make an intreresting car amp. Maybe parallel two halves of 6GM8 instead of the one half, then run heaters in series to avoid creation of a heater supply.
Any comments, concerns, etc.?
Thanks
EDIT: The 1N5310 is a 3.3mA CRD. Also, the LM317 and 5-ohm res (actually a 25-ohm trimpot to adjust between about 50mA and 500mA) act as a 250mA CCS.
EDIT again: Just had a thought... 470uF may not be best for coupling for 8-ohm speaker....
But, if it can do decent output (on 12V supply, an opamp could swing about 9V, can a tube as well?) it might make an intreresting car amp. Maybe parallel two halves of 6GM8 instead of the one half, then run heaters in series to avoid creation of a heater supply.Any comments, concerns, etc.?
Thanks
EDIT: The 1N5310 is a 3.3mA CRD. Also, the LM317 and 5-ohm res (actually a 25-ohm trimpot to adjust between about 50mA and 500mA) act as a 250mA CCS.
EDIT again: Just had a thought... 470uF may not be best for coupling for 8-ohm speaker....
Attachments
You won't get much HF since the MOSFET has a lot of gate capacitance, likely more than a tube can push against at 12V (made for it or not, it's going to be a relatively feeble output). Besides that, it won't be capable of rail-to-rail operation since tubes never turn on or off completely.
Oh, and you'll dissipate a lot of power in the output. For 250mA you won't need much of a heatsink but you'll be running pretty deep class C if you actually try to drive the MOSFET into its 10A capability. Of course, that's no worry since most MOSFETs I know take at least a few volts G-S to turn on very far, a near impossibility given your circuit conditions (unless the source were grounded, for instance)
Tim
Oh, and you'll dissipate a lot of power in the output. For 250mA you won't need much of a heatsink but you'll be running pretty deep class C if you actually try to drive the MOSFET into its 10A capability. Of course, that's no worry since most MOSFETs I know take at least a few volts G-S to turn on very far, a near impossibility given your circuit conditions (unless the source were grounded, for instance)
Tim
Hi Needtubes;
That idea of yours is perilously close to one I had a while back, though I was going to drive the valve with a stack of 9V batteries to permit some real voltage swings. As far as the solid state follower is concerned, I was thinking about using a 12V lantern battery (for some current capability). I have frequent enough power outages here that a smallish battery powered hifi would be a very nice thing indeed.
However, I've been a bit preoccupied over here trying to resolve some pesty computer hassles to permit construction and testing of new audio circuits.
If you try it out, do give us all a shout as to how it worked out! I, for one, am curious as to the merits and flaws of this approach.
All the best,
Morse
That idea of yours is perilously close to one I had a while back, though I was going to drive the valve with a stack of 9V batteries to permit some real voltage swings. As far as the solid state follower is concerned, I was thinking about using a 12V lantern battery (for some current capability). I have frequent enough power outages here that a smallish battery powered hifi would be a very nice thing indeed.
However, I've been a bit preoccupied over here trying to resolve some pesty computer hassles to permit construction and testing of new audio circuits.
If you try it out, do give us all a shout as to how it worked out! I, for one, am curious as to the merits and flaws of this approach.
All the best,
Morse
> something like the attached file.
Output swing is limited not by your monster 10A MOSFET, but by tube plate swing and LM317 drop.
> on 12V supply, an opamp could swing about 9V, can a tube as well?
You can find good op-amps to swing 11V p-p on a 12V supply.
Tubes at high voltage tend to give peak swing about 20% of supply voltage (p-p swing about 40% of supply). At low voltage it is tough to even do that,
Without doing a full work-up, I suspect that even after extensive tweaking you won't get even 2.8V swings at the Plate and Gate. Assume the MOSFET has unity gain (close enough). That's 2V RMS into the load.
2V RMS in 8 ohms is 2^2/8= one-half Watt.
BTW: car speakers are more often 4Ω. That brings us to One Watt.
The cheapest car radios give 4 Watts per speaker. This class of radio is almost obsolete; most radios today give 16 Watts RMS in 4 ohms (often listed as 20-25 watts, at 10%THD 14.4V over-charging battery).
Therefore your one-watter will be the softest car radio on the block. Maybe not much softer (when fully optimized) than a $50 Radio Shack universal replacement radio. Significantly softer than the average new-car factory radio (these are mostly 16-25 watts per channel, and often 4 channel). And of course the kilowatt booster-amp boys will laugh when they hear you coming.
My opinion: use the tube if you must, but follow it with any standard car-audio booster amp to get the sound of the tube up above car noise and car-speaker inefficiency.
> one evil heaphone amp.
Not amazingly evil. In 32Ω cans the 2V RMS is over 100 milliWatts, really quite powerful. But in the hi-impedance phones, 2V RMS is not even peak normal listening level.
Output swing is limited not by your monster 10A MOSFET, but by tube plate swing and LM317 drop.
> on 12V supply, an opamp could swing about 9V, can a tube as well?
You can find good op-amps to swing 11V p-p on a 12V supply.
Tubes at high voltage tend to give peak swing about 20% of supply voltage (p-p swing about 40% of supply). At low voltage it is tough to even do that,
Without doing a full work-up, I suspect that even after extensive tweaking you won't get even 2.8V swings at the Plate and Gate. Assume the MOSFET has unity gain (close enough). That's 2V RMS into the load.
2V RMS in 8 ohms is 2^2/8= one-half Watt.
BTW: car speakers are more often 4Ω. That brings us to One Watt.
The cheapest car radios give 4 Watts per speaker. This class of radio is almost obsolete; most radios today give 16 Watts RMS in 4 ohms (often listed as 20-25 watts, at 10%THD 14.4V over-charging battery).
Therefore your one-watter will be the softest car radio on the block. Maybe not much softer (when fully optimized) than a $50 Radio Shack universal replacement radio. Significantly softer than the average new-car factory radio (these are mostly 16-25 watts per channel, and often 4 channel). And of course the kilowatt booster-amp boys will laugh when they hear you coming.
My opinion: use the tube if you must, but follow it with any standard car-audio booster amp to get the sound of the tube up above car noise and car-speaker inefficiency.
> one evil heaphone amp.
Not amazingly evil. In 32Ω cans the 2V RMS is over 100 milliWatts, really quite powerful. But in the hi-impedance phones, 2V RMS is not even peak normal listening level.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.