Hello folks. I have also posted this in the "Headphones" section. I hope this isn't against forum rules or etiquette. Apologies if is.....
I have a Trancendent grounded grid preamp and am unsure whether I can use it as a headphone source. The relevant specs are:
Max Signal Out: 20 volts RMS.
Output Impedance: 200 ohms.
Headphone impedance is 150 ohms.
Any advice much appreciated. Thanks, Ian M.
I have a Trancendent grounded grid preamp and am unsure whether I can use it as a headphone source. The relevant specs are:
Max Signal Out: 20 volts RMS.
Output Impedance: 200 ohms.
Headphone impedance is 150 ohms.
Any advice much appreciated. Thanks, Ian M.
Last edited:
The distortion will rise, and the maximum output voltage will decrease.
That low output impedance is due to feedback, which will be much less
when the output is heavily loaded, as with headphones.
Also, the output coupling capacitor should be much larger, at least 50uF-100uF.
Last edited:
In addition to the impedance mismatch, you probably don't have enough power.
In short, you need a headphone amp.
In short, you need a headphone amp.
50 ohms difference is not a big issue.
and the lower roll off is only going to be affected by a few hertz that you wouldn't notice anyways.
and the lower roll off is only going to be affected by a few hertz that you wouldn't notice anyways.
No. (to DavesNotHere)
Although the output coupling cap (1uF) is inside the loop, there is not much spare loop gain to cope with the drastic fall in gain due to seriously overloading the output - 150R load on a 200R output impedance is a serious overload. LF distortion will rise, and the LF rolloff will significantly rise too.
This question is a good example of the difference between output impedance and driving ability.
I note that in this circuit none of the grids is actually grounded. It is actually a hybrid of a cathode-coupled amp (i.e. short tailed pair) and an SRPP, with some feedback.
Although the output coupling cap (1uF) is inside the loop, there is not much spare loop gain to cope with the drastic fall in gain due to seriously overloading the output - 150R load on a 200R output impedance is a serious overload. LF distortion will rise, and the LF rolloff will significantly rise too.
This question is a good example of the difference between output impedance and driving ability.
I note that in this circuit none of the grids is actually grounded. It is actually a hybrid of a cathode-coupled amp (i.e. short tailed pair) and an SRPP, with some feedback.
Last edited:
50 ohms is not the end of the world since its near a short. Granted the circuit inside should be rated appropriately.
someone piped up and said it has a 1uf cap. are they sure this is on the output?
Like I said before, it shouldn't change much. It will be measurable but I seriously doubt it. Especially when its not going to driving the stage hard because maybe 2-5V in a 20V system.
but I doubt it is a 150 ohm output unless its some sort of transformer output. The only exception is if its some sort of low mu output (6as7).
someone piped up and said it has a 1uf cap. are they sure this is on the output?
Like I said before, it shouldn't change much. It will be measurable but I seriously doubt it. Especially when its not going to driving the stage hard because maybe 2-5V in a 20V system.
but I doubt it is a 150 ohm output unless its some sort of transformer output. The only exception is if its some sort of low mu output (6as7).
so the answer is he doesn't know until he tries it.
It is not going to blow up. it will either work or sound bad in worst case scenarios.
It is not going to blow up. it will either work or sound bad in worst case scenarios.
Yes, it's inside the NFB loop so it "appears" larger, but not under heavy loading conditions.
Overall negative feedback lowers the output impedance, provided the circuit has "excess gain" to use.
Last edited:
Thanks for all the responces. This seems to be a somewhat contentious subject with varying opinions. It sounds like I will just need to try it out. I appreciate that the GG pre may not have the power to drive headphones in a controlled fashion I could possibly split the signal before the 1uF caps and make a dedicated switchable headphone output with its own capacitors. Although I constructed the preamp from a kit I have no real understanding of electronic theory so am really looking for a yes/no/are you crazy/this is how to do it solution.
Sorry, no. This might help. http://www.allaboutcircuits.com/tec...rt-3-improving-noise-linearity-and-impedance/
Last edited:
It won't work. Well, it will play music but it will sound anemic. I know, I've tried a very similar setup. My solution was to use a dedicated headphone amp when I wanted to listen through headphones.
> Overall negative feedback lowers the output impedance, provided.....
Overall negative voltage feedback lowers the output impedance.
Alternative connections make current feedback which increases output impedance.
Current feedback is not real common in audio, because the *general* trend is to go for low-Z outputs (and high-Z inputs). One common exception is an unbypassed cathode resistor, roughly doubles Rp. Another special exception is guitar amplifiers using voltage-feedback chips, re-jiggered to add some current feedback and raise the output impedance (reduce speaker damping, more cone-tone, approximating a wood-body instrument).
Overall negative voltage feedback lowers the output impedance.
Alternative connections make current feedback which increases output impedance.
Current feedback is not real common in audio, because the *general* trend is to go for low-Z outputs (and high-Z inputs). One common exception is an unbypassed cathode resistor, roughly doubles Rp. Another special exception is guitar amplifiers using voltage-feedback chips, re-jiggered to add some current feedback and raise the output impedance (reduce speaker damping, more cone-tone, approximating a wood-body instrument).
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.
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.
Attachments
- Status
- Not open for further replies.