He might not know. You clearly don't know. We do.DavesNotHere said:
No. NFB can, and usually does, vary both input and output impedance. In this circuit there are two types of NFB used; both will reduce the output impedance. That is how they can quote 200 ohms from a valve circuit.DavesNotHere said:
Look at the circuit - Google will find it for you. That is what I did.
If you understand circuits and can do algebra then you can work this out for yourself. If not, you will just have to believe what we and the textbooks are trying to tell you. Or you can remain ignorant of basic electronics. Your choice. Sorry to be blunt, but you are spreading a lot of misinformation in this thread.DavesNotHere said:
No, just truth and error. Unfortunately you are not yet in a position to judge which is which. It happens to be the case here that the majority are right (4 against 1?), but that is not a reliable method for determining technical questions.sharkey22 said:
whatever. 1/(A+B) is not significant enough and is not guaranteed. Because it fluctuates through the life of the tube. That is why it is ignored.
I don't recognise 1/(A+B) as a feedback formula. Perhaps you omitted something?
Reduction of output impedance by feedback is a major technique used in lots of circuits in audio and other areas of electronics, using both valves and semiconductors. It is certainly not ignored - except perhaps by those who don't understand it.
Reduction of output impedance by feedback is a major technique used in lots of circuits in audio and other areas of electronics, using both valves and semiconductors. It is certainly not ignored - except perhaps by those who don't understand it.
Too bad there isn't an ignore button (or at least I haven't found it yet)
jumping in the middle of someone else's rebuttal when you don't know what is being said is really rude DF96
jumping in the middle of someone else's rebuttal when you don't know what is being said is really rude DF96
wow, what a bad design for a headphone amp.Is this the heart of the beast in question?
Some simple assertions suggest it may be 200 Ohms output impedance for SMALL signals. Say upper cathode internal impedance 1/Gm is 1K, lower right triode is gain of 50, NFB sets final gain to 10. 1K*(10/50) is 200 Ohms. Some other best-guesses lead to the same area.
But the LARGE signal impedance is still the plate resistances. For hotter tubes run at higher current, 10mA-20mA, we could get down to 3K. 3K driving 150r is a really heavy load. You will not get 20Vrms. But if supply voltage is say 300V, 150V per stacked tube, you could get much of 150V*(150r/3,000r) or 7.5V peak, 5Vrms. >150mW in 150r. However the high-impedance loading the circuit needs for good forward gain and thus low output Z is shot by the heavy loading. I'll wild-guess >5%THD before you hit 100mW. That may be listenable. But I think there are much better headphone drivers.
The output cap has to be over 10uFd or bass will be extra strained.
It would be more reasonable if the output coupling circuit was moved to the plate of the bottom triode.
Some corrections:
It's Transcendent. It's not a grounded-grid amplifier. And it was never intend to be used as a headphone amplifier.
You can find the Ignore function under the User Control Panel at top of the screen.
It's Transcendent. It's not a grounded-grid amplifier. And it was never intend to be used as a headphone amplifier.
You can find the Ignore function under the User Control Panel at top of the screen.
No. More misinformation from DavesNotHere. Making the change you suggest will raise output impedance; have you heard of the SRPP circuit? It is not a headphone amp but a line stage which the OP is considering using as a headphone amp.DavesNotHere said:
Which rebuttal? All I see in this thread is misinformation from you and wise advice from others.
Based on experience with certain other forum members it could turn out that he is not only sincere in his misunderstanding but makes his living from audio! Unfortunately it can be difficult to distinguish between such a person and an experienced troll.
I know it's ridiculous and sad.
I personally appreciate all the information you share so thanks.
As been said here, preamp has not big enough balls to drive phones.
For example lundahls line out irons are (small) matchbox sized <0,2kg, but those capable of driving phones (or interstage) are for example LL1689 sized (0,75kg)
For example lundahls line out irons are (small) matchbox sized <0,2kg, but those capable of driving phones (or interstage) are for example LL1689 sized (0,75kg)
If it had an op amp as the output stage, it would drive it most definitely.
if it was balanced out, or even transformer out it would most likely work.
I would have to see resistance values and tubes used in the output to sign off on the 200 ohm output. Maybe its the lowest guaranteed impedance. but it would be a stretch.
if it was balanced out, or even transformer out it would most likely work.
I would have to see resistance values and tubes used in the output to sign off on the 200 ohm output. Maybe its the lowest guaranteed impedance. but it would be a stretch.
others may correct me. but way i understand it,
when you have a source that has 200 ohms output
impedance, then it is best to load it with at least 2000 ohms
or better so that most output voltage goes to the load...
i know this as a rule of 10.....
when you have a source that has 200 ohms output
impedance, then it is best to load it with at least 2000 ohms
or better so that most output voltage goes to the load...
i know this as a rule of 10.....
Valves are 12AU7. You can see the circuit here. I haven't done the sums for the SRPP output with further negative feedback, but 200 ohms seems plausible. Very roughly, SRPP achieves something like rp/4 to rp/3 (which would be in the region of 600-800 ohms) then the feedback reduces it.DavesNotHere said:
Certainly no good for driving 150 ohm headphones.
I also fail to understand why this is called "grounded grid". Other than that, it's a decent design but I wouldn't load it with anything less than 10kOhm considering the valves used. The feedback is very "mild", just a few dB, a 12AU7 doesn't have much µ to work with.
Last edited by a moderator:
Thanks for the replies. It sounds like investment in a headphone amp is the necessary solution..
I think marketing is the reason. Someone decided to call a cathode-coupled amplifier (short tailed pair) a 'grounded grid' for two reasons:Groucho2004 said:
1. in its simplest form the second grid is grounded - but it is almost never used in this form for audio as feedback is applied here instead;
2. a genuine grounded grid amplifier was used in VHF/UHF circuits for good engineering reasons including wideband response and it was thought this might implant the concept of 'wideband' for audio uses into the minds of purchasers who don't understand electronics.
If the 12v heater power supply can deliver an extra 150mA, it would be easy to add a pair of darlington at the outputs, each loaded on 6.3v heater section of one tube.
quite frankly, I think it is false advertising calling it a "grounded grid amplifier" and the circuit is like this.
and I think the 200 ohm is a typo too. Just by looking at this, even 2K might be stretching it. because of the cathode resistor being 3K.
The use of two power transformers must have been cheaper than using a standard center tapped transformer and a bridge to make a +/- supply.
another icky thing is the use of a voltage regulator on the heaters as that I can see it being a weak link of the unit's durability.
Last edited:
- Status
- Not open for further replies.