> You can find the Ignore function under the User Control Panel at top of the screen.
Thank You!!
______________
> what a bad design for a headphone amp.
Not marketed as a headphone amp. Assuming "line output" (generally Power amps), it is a fine design.
> It's not a grounded-grid amplifier.
That bothered me at first. GG is "all wrong" for most hi-fi work.
It is of course a grounded plate feeding a grounded grid feeding something on top (SRPP?). The top-bit is grounded plate and will act as "Cathode follower" for some range of load impedance. GPGGGP? GPGGSRPP? Looks like a cat on the keyboard. Not a marketable name.
> short tailed pair
Tail voltage is ~~twice tube voltage. Pretty "long" by many standards.
_____________
me> Say upper cathode internal impedance 1/Gm is 1K
12AU7 at 100V 2mA has Gm of 1,100uMho. 909 Ohms. Nominal voltage gain is 6. 12AU7 as a straight voltage-amp gives gain above 15. 6/15 is 0.4. 0.4 times 909r is 360-something. The SSRP(?) may add something to this. I could stimulate but it really is not important. It sure will drive 10K line loads with authority, but as hpeter says it "has not big enough balls to drive phones." 2mA max peak in the proposed 150 load is 0.3V peak, 0.21Vrms, 0.3 milliWatts totally clipped. Audible, but not satisfying.
Thank You!!
______________
> what a bad design for a headphone amp.
Not marketed as a headphone amp. Assuming "line output" (generally Power amps), it is a fine design.
> It's not a grounded-grid amplifier.
That bothered me at first. GG is "all wrong" for most hi-fi work.
It is of course a grounded plate feeding a grounded grid feeding something on top (SRPP?). The top-bit is grounded plate and will act as "Cathode follower" for some range of load impedance. GPGGGP? GPGGSRPP? Looks like a cat on the keyboard. Not a marketable name.
> short tailed pair
Tail voltage is ~~twice tube voltage. Pretty "long" by many standards.
_____________
me> Say upper cathode internal impedance 1/Gm is 1K
12AU7 at 100V 2mA has Gm of 1,100uMho. 909 Ohms. Nominal voltage gain is 6. 12AU7 as a straight voltage-amp gives gain above 15. 6/15 is 0.4. 0.4 times 909r is 360-something. The SSRP(?) may add something to this. I could stimulate but it really is not important. It sure will drive 10K line loads with authority, but as hpeter says it "has not big enough balls to drive phones." 2mA max peak in the proposed 150 load is 0.3V peak, 0.21Vrms, 0.3 milliWatts totally clipped. Audible, but not satisfying.
200 may be stretching it, but it won't be as bad as 2k. The modified CF action of the upper valve will ensure a resonably low output impedance. Having a highish cathode resistor up there (3k) will reduce output impedance, provided that the valves still carry enough current - although I'm not sure they do in this design.DavesNotHere said:
If I was asked to make a wild guess at the output impedance of this design I would go for 300-400 ohms. I can't be bothered to actually calculate it. It would be interesting to know on what basis they claim 200: measurement or calculation? If the latter, did they do it correctly? For example, did they take account of the fact that this is not a standard SRPP but modified by the cathode-coupled drive?
Ah, I hadn't spotted the negative rail for the tail. So it is a long tail pair.PRR said:
Last edited:
I was curious and simulated the circuit. Current through each valve is about 2ma, anode voltage of the bottom valve is about 80V, gain is 3.8, output impedance @ 1KHz is 1.7 KOhm. The 2ma current moves the bias into a region with rather high rp which explains the high(ish) output impedance.
I wonder why they use such a low current? Seems a poor choice for a line stage, and no hope of driving headphones. Their claimed 200R seems somewhat optimistic.
it won't be able to drive enough current.
eg. a LM4562 op amp has an output impedance of .01 ohm until you hit 40 ma of current
then it shuts down the output. Same thing happens here when the bias current is reached.
eg. a LM4562 op amp has an output impedance of .01 ohm until you hit 40 ma of current
then it shuts down the output. Same thing happens here when the bias current is reached.
In the literature I have on this preamp, the designer says 300 ohms output impedance. This comes from the writeup he did in his book Audio Reality.
He also doesn't claim it as a SRPP, but rather the use of the extra triode in the envelope as a pseudo current source, noting it's not even very good at that purpose because it doesn't offer high enough impedance.
Specs as the designer has stated them:
By the way, there is no reason why the three terminal regulator used to power the filaments cannot be totally reliable and last for decades as long as the heat issues are addressed. It's not shown heat sinked on the schematic but he makes specific refererence to it in the text that it must be heat sinked.
He also doesn't claim it as a SRPP, but rather the use of the extra triode in the envelope as a pseudo current source, noting it's not even very good at that purpose because it doesn't offer high enough impedance.
Specs as the designer has stated them:
- bandwidth: flat from a few herts to 300 KHz.
- Voltage gain: 12 dB
- 20V RMS output (I assume that is into a typical tube amp load of say ~100K)
- Feedback: 7 dB
By the way, there is no reason why the three terminal regulator used to power the filaments cannot be totally reliable and last for decades as long as the heat issues are addressed. It's not shown heat sinked on the schematic but he makes specific refererence to it in the text that it must be heat sinked.
well its a really not a nice thing for a regulator to begin working into a near short when the tubes are cold.
current regulators (1 per tube) would be a better choice since they would soft start the heaters.
current regulators (1 per tube) would be a better choice since they would soft start the heaters.
btw here is another example of low capacitance power supply. instead of high inductance, it is resistance. resistance and capacitance only is calculated in this filter.
Last edited:
It is a PSU for a low current line stage. Given that, 22uF is not a particularly low capacitance. It has parallel zeners, so the supply impedance will probably be relatively low. Not relevant to the discussion in the other thread about SET supplies.
Yes of course, because 22uF is a largish capacitance for a circuit drawing just a few mA. However, the series resistance might be a problem - but if the zeners were removed then the resistors could be reduced in value.
Whether a capacitance is considered large or small depends entirely on context.
Whether a capacitance is considered large or small depends entirely on context.
- Status
- Not open for further replies.