Phonodude problem

[Miniwatt]

Hi all,

I finished my phono amp yesterday, well, not completely, because there is a problem

After power up all is fine, for a few minutes. Then the left channel fades out. The right channel keeps playing minutes longer, than it fades out too. It comes back though, after about ten seconds. It keeps playing, and then the left channel comes up again too. This only lasts a few seconds, then it's gone again. This keeps repeating.

Here's the amp:



Here's the powers supply:



First I thought one of the capacitors was faulty, so I redid the PS with new ones, but it's still the same.
I checked with the scheme, but everything seems to be like it's supposed to be. Due to my system set-up it's hard to measure anything whilst playing, but I'll have to do it in the end I'm afraid.
I think it's something in the PS, since one channel works pretty good, and the other one on and of, the PS is split in two separate channels at the end.

I don't think there's much you can do from out there, but you never know, so I post this anyway.
Help is much appreciated

[singa]

Why is the +12V heater ground return( blue line) going between R1 and R2?
Sould be at bottom of R2? To 0 of 230V- 0 -230V?

[AndreasS]

the first tube has grid leak bias, but where is the grid resistor and the blocking condenser?

Andreas

[Miniwatt]

There's no grid resistor and blocking condensor, because a 5755 is made for this, and in this application with very small signals you won't get a positive grid anyway.

(I'm sorry, I'm not really an expert on this, and the english jargon is difficult)

You're german, maybe can make something of this, it's what the author wrote:

de eerste trap, waar de kathode van de 5755 koud aan massa ligt. Dus zonder de gebruikelijke kathodeweerstand om een negatief rooster te krijgen. De 5755 is gemaakt om zo te gebruiken. Met 180 volt op de anode en de kathode aan massa loopt er iets van 2 mA. 100 volt op de anode stelt de buis in op 1 mA. Omdat er heel kleine signaaltjes aan het stuurrooster worden aangeboden, een paar millivolt uit het element, zal het rooster toch niet positief gestuurd worden. Dat scheelt niet alleen een weerstand die de signaal/ruisverhouding nadelig beïnvloed, maar ook een klankbepalende ontkoppelcondensator in deze gevoelige trap.

[Alain Dupont]

Miniwatt,

1)- Did you check the heater voltage right on the tube pins ?
{during the fade out}

2)- Otherwise your schematic is changed in the RIAA part from the original...
Why?

3)- And Changed capacitors C4/C5 also ?

Is this your ref schematic?

http://www.dddac.de/files/phonodude-hard-wired-41.pdf

Alain.

[Miniwatt]

Hi Alain,

Miniwatt,
1)- Did you check the heater voltage right on the tube pins ?
{during the fade out}

2)- Otherwise your schematic is changed in the RIAA part from the original...
Why?

3)- And Changed capacitors C4/C5 also ?

Is this your ref schamatic?

http://www.dddac.de/files/phonodude-hard-wired-41.pdf

Alain.

The ref schematic is the one I put in my first post, by Triode Dick, he designed this amp together with Doede Douma from the scheme you posted. I thought they would be the same, but there are differences. I always focussed on Triode Dick's one.

I didn't measure much so far, all the heaters were glowing all the time.
I'll have some dinner now, and get back to it tonight (it's 7 pm here), do some real measuring, and post about it when i know more.
Thanks so far everyone!

[Miniwatt]

It's done..

It was the 78S12 in the heater supply that got too hot. I just put a small metal clip on it for cooling. Now I screwed a larger metal plate on to it, and everything is fine! That plate gets quite hot too btw...

Now I can finally listen to my new toy


Thaks again for the input

[kevinkr]

I'd add whatever additional heating sinking you have space for on that filament regulator. Don't forget to use thermal grease/mica insulating washer or a silpad too.

You can also add a small value power resistor in parallel with the regulator - the regulator control loop will still dominate, but some of the power will be dissipated in the resistor instead of the regulator.
If you are dropping say 6V across that regulator and you have 600mA load you are dissipating 3.6W in the regulator, you can halve this by putting a 20 ohm 5W resistor in parallel which will reduce your heatsink requirement by half.

I usually put somewhere between 1/3 - 2/3 of the load current through the resistor. You just need to know the loaded raw supply voltage, the output voltage (to get the delta across the regulator) and the load current to calculate the resistor value where R=E(delta)/I(load current)

Note this is a great idea only if you know that the load current is not variable. Works well with filament supplies.

And finally how does it sound?

[Miniwatt]

It's still burning in, all the parts are new. Yesterday I was amazed with the sound, but it sounds a bit dull today. It only played for six hours so far, so there's more to come

[kevinkr]

I noticed long ago in a far away neighborhood that at certain times of the day nothing seemed to sound that good.

Power quality/line voltage might just be a factor here.

I have dedicated lines which feed my electronics which definitely helps with household noise sources, but doesn't do much for external noise sources sharing the same pole pig (transformer) or noise from elsewhere in the distribution system.

Some line filtering might be a good idea down the road.

See if your listening impressions at different times of the day correlate, that is what I did and determined that peak usage times resulted in worse sound. Looking at the line voltage and ac waveform distortion was quite interesting at these times..