#124 started

[norburybrook]

yes that's the right one.


I'm fixing a computer for someone now for a few hours but will get back on this after.



Marcus

[norburybrook]

that was quick...job done..

OK I'm back.


M

[martin manning]

Marcus, is the cap you grounded to kill the hum the 0.1 at the lower-left? Which end?

[norburybrook]

yes at the node with the resistor. It's the same effect with the other 0.02 node.

Which I suppose is the same as grounding the grids directly.

[larsvictor]

that was quick...job done..

...computers don´t have tubes

[norburybrook]

indeed......

I'm quite an expert when it comes to computers........which makes being a rank amateur with tubes quite humbling!!!


I'll add that yesterday I chop sticked around every connection, moved wires around and it made no difference at all. There was no noise at all when prodding so all the physical connections are good.


M

[martin manning]

There are resistors at both ends! But if you grounded the end going to the grid then that's going to be about the same as pulling the PI tube. At least the hum seems to be in the PI.

[norburybrook]

indeed it was at the grid end , the 0.02 only has a resistor at one end


I've tried different tubes , same difference. gets worse when NFB is connected.

[larsvictor]

Please measure the voltages of V3 again.

Especially in the network (4 resistors) and between the 24k and 390R.

Ground from the junction 0,1uF/390R/24k/ (4k7 if the nFB is connected) - kind of "bypassing" the presence/NFB circuit. Is that altering the hum ?

Lars

[norburybrook]

Lars,

V3 voltages
pin1; 276
2; 34
3; 54
4,5
6; 265
7; 34
8; 54


grounding either of the two nodes you suggested kills the hum.

M

[norburybrook]

Looking at the schematic of the presence circuit , following it through to the speaker outputs. I noticed once again that the speaker jack is grounded through the chassis . I have no continuity to ground from the speaker jack. Just the connection to the OT

where should I put a ground to?

Fingers crossed this might be it..

M

[martin manning]

indeed it was at the grid end , the 0.02 only has a resistor at one end :D

Seems like semantics, but it's important: One end has a 1M resistor and the grid connected to it, and the other has a 4k7, a 24k, a 390, and a pot. A circuit node includes everything that is (theoretically) connected electrically and therefore has the same voltage. In reality a node may include several pieces of wire and several solder connections so there is always resistance, capacitance, and inductance that does not appear on the schematic.

[martin manning]

Ground the speaker jack with the output cathodes. NFB won't work without that ground, and it could put a noise voltage on the PI tail. The PI DC voltages look good, except recall you can't measure the grids directly.

[norburybrook]

gentlemen... That did it........I now have a whisper quiet amp........


Lordy lordy that was a journey.

Lars,Martin, Taylor,Erwin,aaron et al thanks so much for hanging in there with me.


so for anyone in the future who may read this.

1: Always check your components before putting them in place, you cannot rely on markings, manufacturers bagging etc CHECK FIRST


2; Anodised chassis will not ground easily through pot's , sockets, etc make sure you have a separate ground if it's showing a chassis ground on the schematic.




now onto the good stuff.

What should I bias to, I think I'm at 30mv at the moment, I haven't checked my plate voltage properly yet.

I cant remember if I do this with standby on or off


OD trim; seems quite OTT unless I have it turned way down on the trim I have a 500k with a 1.4M across it.


Marcus

[larsvictor]

Does grounding here kill the hum ?

edit: Ok, fine.