• Disclaimer: This Vendor's Forum is a paid-for commercial area. Unlike the rest of diyAudio, the Vendor has complete control of what may or may not be posted in this forum. If you wish to discuss technical matters outside the bounds of what is permitted by the Vendor, please use the non-commercial areas of diyAudio to do so.

TSE 45, how do I remove background noise

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
There is a background hiss at some fixed frequency coming out of my tweeter. It is very faint, and only audible with my ear right next to the tweeter, but I prefer a black background. Any idea where the hiss is coming from?

My power supply has a 10H choke with all film capacitors and I'm using 2.5V DC regulated filaments (via Sharp PQ5EV5). 175V for 5842's and 295V/26mA for 45's. Power should be clean. I don't know what is producing the noise.

Is the noise inherent in the tubes? Or should I switch to a regulated B+ supply?
 
I just tried it without the volume pot, but unfortunately it made no difference and the hiss is still there. I want to drive sensitive headphones with the amp, so I want to make the background noise completely inaudible. I wonder if 5842's are inherently noisy?
 
Some 5842's do exhibit hiss. Remove the 5842's from their socket and power up the amp. Does the hiss go away? Remember to mark the tubes so they go back in the same socket, or you will need to adjust the plate voltage again.

If the hiss goes away, it is originating from the 5842 circuit, If you have a volume pot, does the hiss remain when the pot is all the way down? The volume pot shorts the grid resistor when all the way down, so if the hiss remains, the resistor is not the source of the noise (some resistors are noisy).

You may get less hiss at a different plate voltage. You can try anywhere from about 125 to 225 volts and see what happens. Most tubes have the lowest distortion around 175 volts, and much above 225 runs the tubes too hot.

You probably don't want to connect the headphones directly to the TSE's output. The usual method involves a resistive attenuator that presents a good load to the TSE and attempts to match the headphones. I have used a 10 to 15 ohm resistor across the TSE's output and a resistor in series with the headphone to set the level such that the headphones (or your ears) will not be damaged if the volume is accidentally cranked all the way up. The resistor value depends on the headphones impedance.
 
Wow, it's dead silent with the 5842 tubes removed. Absolute silence. These were brand new tubes too, by Amperex. So now we find the culprit. I tried 160V to 180V on the 5842 but it hisses no matter what. Maybe I will have to use a different tube. How about a 12AT7 variant externally mounted? If less gain is needed, a 12AU7 variant?

Thank you for taking the time for the response, George. This amp is the best I've ever heard, and I've heard very many high end amps including in my own setup.

My headphones are 20 ohms flat so I want to drive them directly out of the amp. I've tried the resistors but they change the sound. It sounds better without the resistors. Yes at full power, it will definitely melt my headphones. But my other headphone amps output over 50W into speakers, and I've never accidentally melted any headphones.

This amp design is perfect for headphones, so I really want to make it work and believe that I can.
 
I notice that my 5842 filaments are running at 5.5V instead of 6.3V. Could that be a problem?

Yes, I just ran into that issue in my preamp. Low heater voltage will definitely cause hiss. 5.5V is pretty close though, so hard to say how much difference it will make, but you need to get closer to 6.3. I ended up with ~5.9 and no hiss, need a pot to determine exact resistor value for the voltage reg. if I want it perfect...
 
So you added another C to the filter on the input tube's filament power supply.... I guess the question is why this is needed? The circuit is either picking up some extras on the way or your AC power may not be so great. Either way, happy to hear the solution is working for you!
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.