This amp sounds great, and it can perfectly drive the Sennheiser HD-650 headphones. But the bass is rolled off, probably because of the low value (40 uF) of the output cap. What could be a reasonable value? 220 uF? I will have to replace it with an electrolytic cap, because there is no space for an oil cap.
The values in red is what I have measured
Another issue is that it gets extremely hot. I guess that it is the cathode resistor (1k/25W). Next week I will borrow a thermal camera and I will have a hot look at it.
The values in red is what I have measured
An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.
Another issue is that it gets extremely hot. I guess that it is the cathode resistor (1k/25W). Next week I will borrow a thermal camera and I will have a hot look at it.
Last edited:
A closer view:
An externally hosted image should be here but it was not working when we last tested it.
Not an issue at audio frequencies. Nice layout, and very well executed. (Relatively speaking)
Hi Ian. Thank you for the information.
I am using 300 Ohm headphones (Sennheiser HD-650).
Is there some formula for caltulating the otuput cap? I would like to learn the maths
Hi briasnune. It is not soldering flux. The manufacturer has covered it with some kind of laque.
It is not my design; I did buy it from a Chinese manuafcturer. Now I am doing reverse-engineering and trying to learn something about cathode followers, in order to improve the design.
I will rewire the filament wires twisting the cables. I have also noticed that the cathode bypas cap of one of the the input tubes is not wired to the common ground, but to the RCA input. This is possibly creating a ground loop.
Thank you for your answers
It is not my design; I did buy it from a Chinese manuafcturer. Now I am doing reverse-engineering and trying to learn something about cathode followers, in order to improve the design.
I will rewire the filament wires twisting the cables. I have also noticed that the cathode bypas cap of one of the the input tubes is not wired to the common ground, but to the RCA input. This is possibly creating a ground loop.
Thank you for your answers
Thank you for the information Sofa.
I have found some pictures, and it seems that there is a woman manufacturing this amplifier. This could be the reason why it looks nice. There was a need of a femenine touch in the art of point-to-point wiring
url=http://www.diybuy.net/thread-245720-1-1.html
BTW: Thermal camera in hand, in only five minutes I have measured 180 degC at the 100Ohm/10W power supply resistors, and 165 degC at the 1k/25W cathode resistors
Time to start replacing components.
I have found some pictures, and it seems that there is a woman manufacturing this amplifier. This could be the reason why it looks nice. There was a need of a femenine touch in the art of point-to-point wiring
url=http://www.diybuy.net/thread-245720-1-1.html
BTW: Thermal camera in hand, in only five minutes I have measured 180 degC at the 100Ohm/10W power supply resistors, and 165 degC at the 1k/25W cathode resistors
Time to start replacing components.
Last edited:
pingfloid - sofaspud has already given you the formula for the capacitance for a given cut-off frequency but I'd thought I'd send a bit more detail.
Above is the equivalent circuit "looking into" the cathode. You don't need to worry about the left hand side resistor network since it doesn't impact frequency response. On the right, you have 300 ohms in parrallel with 10K so you can approximate that to 300 ohms. The two capacitors of 40uF and 2200pF are additive but that can be approximated to 40uF. Looking at the headphones, the frequency at which half the maximum power is being delivered to the headphones is given by f-3dB = 1/(2*pi*R*C)
= 1/(2*3.142*300*40x10^-6) = 13.26Hz. The CR combination is a high-pass filter so everything above 13.26Hz is deemed to pass through.
If you want to calculate the capacitance required for a different -3dB frequency (let's say 1Hz), the rearrange the above formula to give: C = 1/(2*pi*1*300)
= 530uF
An externally hosted image should be here but it was not working when we last tested it.
Above is the equivalent circuit "looking into" the cathode. You don't need to worry about the left hand side resistor network since it doesn't impact frequency response. On the right, you have 300 ohms in parrallel with 10K so you can approximate that to 300 ohms. The two capacitors of 40uF and 2200pF are additive but that can be approximated to 40uF. Looking at the headphones, the frequency at which half the maximum power is being delivered to the headphones is given by f-3dB = 1/(2*pi*R*C)
= 1/(2*3.142*300*40x10^-6) = 13.26Hz. The CR combination is a high-pass filter so everything above 13.26Hz is deemed to pass through.
If you want to calculate the capacitance required for a different -3dB frequency (let's say 1Hz), the rearrange the above formula to give: C = 1/(2*pi*1*300)
= 530uF
Grid Stopper resistor to cancel RF interference
I've tried everything to remove the RF interference: grounding, shielding, ferrites... Finaly, a 1k GS resistor is the only thing that do the job. No more news on the radio and crispy sparks from the elevator motor while I listen music! And I do not notice any change on the performance.
About the filaments wiring, the amplifier is dead quiet, so I will follow rule number one: "if it works, don't touch it".
I've tried everything to remove the RF interference: grounding, shielding, ferrites... Finaly, a 1k GS resistor is the only thing that do the job. No more news on the radio and crispy sparks from the elevator motor while I listen music!
And I do not notice any change on the performance.About the filaments wiring, the amplifier is dead quiet, so I will follow rule number one: "if it works, don't touch it".
Last edited:
No, Add some small value RF bypass caps from your grids and PS +V to ground. When I worked at a FM station we had a guy on a hill by us whose stereo played the station WHEN IT WAS UNPLUGGED. (RF received by speaker wires, rectified by output transistors and sent back to speakers as audio.)
Doc
Err.. I have to correct myself: there is a reduction in the volume.
I am going to try your suggestion, Doc. I have few old tubular capacitors of 1n, 1n5 and 2n7. I will start with 1n.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.