Tubelab SE Noise Problem

[cddeluca]

I have been running a new 300B TSE for about 2 weeks now and there has been some low level buzzing on my speakers that I've diminished to tolerable with more capacitance in the pi filter. Some is still there, but it's now only evident from the treble units (BMS 4592 on 400HZ Speakerlab horns) which doesn't surprise me because their sensitivity >105dB. It also seems >120Hz which evidently was the problem before the added capacitance.

Here's the weird thing - last night I discovered that scrolling the screen of my Mac Air from across the room was modulating the buzzing I'm guessing that the laptop is radiating more than it ought to be but I'm also guessing that there's a problem in the 5842 circuit.

I know almost nothing but the way I read the schematic there is no grid stopper, just a grid leak resistor - is this true? Could it be that it's a problem because I'm running without a volume control? Do I need a resistor to buffer the input in place of the volume control?

I'd appreciate any ideas on where to start.

 

[Edwind]

Hi cddeluca.
I had the same experience with you before.
May be your chasis not grounded so any electromagnetic interfere your internal components and amplified.
As I know grounded chasis will become as Faraday shield and attenuate any electromagnetic or radio interferences (EMI & RFI).
Please correct if I'm wrong.
The noise can interfere your equipments through power lines also.
Switching devices in computer (PSU unit) or switching AC to DC adapter that always used in notebook computers.
Microprocessor inside the computer/notebook draw more currents while working than in idle condition.
To attenuate noise from power lines, you can use power line noise filter or use power line conditioner (which more expensive but better results).

Hope can help.

 

[cddeluca]

Thanks, Edwind

Hi cddeluca.
May be your chasis not grounded so any electromagnetic interfere your internal components and amplified.
To attenuate noise from power lines, you can use power line noise filter or use power line conditioner (which more expensive but better results).
Hope can help.

The chassis is actually bamboo, so grounding isn't the issue and yes, you're probably right, backfeed of PS noise into the power lines is a distinct possibility - makes more sense than the laptop radiating so badly the amp picks it up 10 or 12 feet away. The amp is fed from a Furman PL-8 that's supposed to incorporate filtering, but it may have no effect at SMPS frequencies.

Thanks for your reply.

 

[gootee]

Photos of your implementation might help.

If you add a grid stopper, maybe you should try making an RC lowpass filter, there, with a cutoff frequency of a few hundred kHz.

 

[cddeluca]

Photos of your implementation might help.

If you add a grid stopper, maybe you should try making an RC lowpass filter, there, with a cutoff frequency of a few hundred kHz.

Thank you, that suggestion probably has merit (though it's beyond my ability to evaluate) but my real question relates to the TSE circuit as implemented on George's circuit board - my implementation is as-designed except for the parts substitutions necessary (2SK3565 mosfets in the 300B grid circuit, higher voltage caps as recommended by George) and the elimination of the volume control, also as approved by George in other posts - so it seems if there's indeed an issue it's within the circuit as-designed (which I doubt) or my best guess an unforeseen consequence of eliminating the volume control.

 

[Edwind]

Ah, one more.
What player do you use?
If use computer or notebook, the noise may come together with input lines and amplified.
High frequency ground loop by computer's SMPS happened while using unbalance interconnect.
Causing by player like this, adding a grid resistor will not much different.

 

[cddeluca]

Ah, one more.
What player do you use?

The result is the same no matter the input - I have a Vincent SA-31 preamp driving it - and it doesn't matter which (or if any) source is active.

But since you brought it up you made me think - I just fired it up with the inputs shorted and it doesn't make a sound; well a sound barely at the threshold of my hearing with my ear stuck inside the horns - so you're right it's something to do with the input - the noise isn't in the circuit itself - though on other amps (notably my Simple SE) - the Vincent's no problem.

 

[cddeluca]

Progress of a sort

So I pulled the Vincent and replaced it with an ADCOM SLC-505 (a volume pot with source selectors, all passive) and the noise level is 'zero' with no sources active and when the CD player is on there's some low level buzz but at the level I would expect (very nearly inaudible at the listening seat ~8 ft). So the TSE doesn't like the Vincent, or my best guess now, the TSE has way more gain than I can use with 105 dB+ loudspeakers.

Thanks for your help,
Charlie

 

[Edwind]

You're welcome.
Could you post again if your problem solved?
It'll be useful for us as a lesson.
Thanks.

Regards,
Edwind

 

[Alan Hope]

Some weird - but good things happening to my TSE noise-wise ...

Still bread-boarded. Initially on switch-on there was a faint thump, then a quiet hum from both speakers. Then at 10s with B+, a louder hum audible from listening position between tracks.

After a week or so regular and long playing sessions, the initial thump+hum disappeared completely, probably with me repositioning the power components a bit, but the B+ hum remained the same.

Tonight, I switched off the music and went to turn off the TSE and noticed that it was silent. Completely and utterly silent! Pushing my ear to the driver (Fostex FE206E) I had to hold my breath to hear the faintest of hum. Fearing the worst, I switched on the Squeezebox - and ... music.

I was resigned to building the amp case, then chasing ground-loops and eventually living with some hum. This level of silence from a valve amp has taken me completely by surprise. Wow!