bias ground

[joesatch]

2204 circuit. where should the bias circuit get grounded to?  at the reservoir cap ground   the first filter stage?

[Latole]

Many bias circuit built ; ground to reservoir or first stage. I've never seen a difference.

[joesatch]

thanks, i have it grounded it to the reservoir cap ground. a well respected "genius" on YT mentioned only the PT CT, rectifier and reservoir cap should be grounded there. anything else at the first filter cap. glad i dont have to re-run a lead to a different spot

[Willabe]

i have it grounded it to the reservoir cap ground.

No.

a well respected "genius" on YT mentioned only the PT CT, rectifier and reservoir cap should be grounded there. anything else at the first filter cap.

He's correct.

That 1st filter cap has the highest ripple current in the amp, because of it's charging cycle. So it's the noisiest ground in the amp.

Only the PT CT or if no CT using a solid state FWB, it's ground and reservoir cap should be grounded there, nothing else at the first filter cap. Then run a wire from that connection to the the local ground star. This way the reservoir cap can charge up/down, charging cycle/ripple, all by itself without disturbing anything other ground.

And this reservoir cap ground should be on the eyelet/turret board, not made directly on the chassis. Then the wire run from that ground connection goes to the local power amp ground star  on the ground buss or local power amp ground star that then goes to the chassis if using 2 chassis grounds, power amp and preamp chassis ground.

Look at the attached drawing below and read the link to Merlin's grounding page. 

https://www.valvewizard.co.uk/Grounding.html

[joesatch]

ok , grounding the bias at the first filter stage (chassis next to the second filter cap).  Reservoir cap will have it's own chassis ground with the rectifier (fwb) grounded there. no other grounds at the reservoir cap ground. interesting the recommended ground scheme (metro forum) is having the bias ground isolated

[Willabe]

ok , grounding the bias at the first filter stage (chassis next to the second filter cap).

No.

Look at the drawing from Merlin I posted.

NOTHING else gets grounded at that 1st filter cap. You have to have a single wire separating/isolating that 1st filter cap ground from everything else.

How many chassis grounds do you have? The less chassis grounds the better, with only 1 chassis ground for the entire amp being best, next to that only 2 chassis grounds, power amp/preamp, that works well for most amps. You don't want 'random' chassis grounds. 

But, you have cap cans, so you really can't ground just the 1st cap, you have to ground the screen grid cap with it. It will work like that, but not recommended. Merlin, Kevin O'Connor, and others who use a bussed. local ground star grounding system, say not to use multi cap can's because of the cans shared ground.

 

Reservoir cap will have it's own chassis ground with the rectifier (fwb) grounded there. no other grounds at the reservoir cap ground. interesting the recommended ground scheme (metro forum) is having the bias ground isolated.

1st you write "grounding the bias at the first filter stage", then you write "Reservoir cap will have it's own chassis ground with the rectifier (fwb) grounded there. no other grounds at the reservoir cap ground."   

Which is it? 

If your building the same amp as the metro layout amp you posted, why are you asking how/where to ground the -bias?

That "Larry" ground schem is supposed to work, why not use that? 

[joesatch]

reservior cap and first filter cap are different caps.  in the 2204 the first 50/50 is parellel'd for 100uf and takes the rectified DC it's considered the reservoir cap.  The first filter cap would be the second physical cap but first cap in the filter stage.

[Willabe]

reservior cap and first filter cap are different caps.  in the 2204 the first 50/50 is parellel'd for 100uf and takes the rectified DC it's considered the reservoir cap.  The first filter cap would be the second physical cap but first cap in the filter stage.

No, not exactly. Just using different names, semantics. Might be an English term, just like the English call tubes 'valves'.

The reservoir cap is the 1st filter cap and is part of the 'filter stage'. It's the 1st part. It does the most work filtering the dc ripple out, because it's the 1st filter cap. Their calling it the 'reservoir cap' because it has the deepest charging cycles, gives up/supplies the most dc voltage of the filter cap string, functions as a dcv 'reservoir'. And the 2nd filter cap is the 2nd filter cap.

Think of when the filter caps are labeled as node A, node B, node C, node D, etc. There's no node labeled 'reservoir' node.

Anyway, since your reservoir cap/1st filter cap is a || cap can, you can use a bussed local ground star system just like in the layout drawing I posted from Merlin.

Just use that exactly like he shows it, then you should be safe.

[Willabe]

Reservoir cap will have it's own chassis ground with the rectifier (fwb) grounded there. no other grounds at the reservoir cap ground. interesting the recommended ground scheme (metro forum) is having the bias ground isolated.

1st you write "grounding the bias at the first filter stage", then you write "Reservoir cap will have it's own chassis ground with the rectifier (fwb) grounded there. no other grounds at the reservoir cap ground."   

reservior cap and first filter cap are different caps.  in the 2204 the first 50/50 is parellel'd for 100uf and takes the rectified DC it's considered the reservoir cap.  The first filter cap would be the second physical cap but first cap in the filter stage.

Ok, I get it now what you were saying.

[Willabe]

How many chassis grounds do you have? The less chassis grounds the better, with only 1 chassis ground for the entire amp being best, next to that only 2 chassis grounds, power amp/preamp, that works well for most amps. You don't want 'random' chassis grounds.

The reason I wrote this is because you can't control where the ground return signal goes in the chassis. And different grounds can cross each other in the chassis and modulate the other tubes circuit ground and thus modulate that tubes circuit. That creates humm.

But you can control where the ground return signal goes with a wire.

So we're trying to get away from 'random' chassis grounding as much as reasonably/feasibly possibly, in a given build, by using a planed out wired ground scheme.   

The less chassis grounds the better. And use a local ground star for each filter cap's circuit that it feeds dcv to. Then use a ground buss to connect them in order as laid out in the circuit and chassis. Then ground the buss at 1 point in the chassis.

[Willabe]

The -bias, whether it's taken from a tap on the high voltage B+ wind or from the B+ wind end itself, like on the Marshall amp your building, it's still part of that wind. So technically it could be grounded with that secondary wind's ground, CT or FWB ground.

But, I would still not ground the -bias circuit ground directly to reservoir cap/1st filter cap and CT/FWB ground.

Let that reservoir cap/1st filter cap and CT/FWB ground, which is part of the PT B+ wind, charging pulses go round and round in a loop by themselves. And tie the -bias circuit ground to the power tube local ground star on the buss. 

[pdf64]

Foe a capacitor input rectifier, the reservoir cap is the first filter cap.

The reservoir cap is subject to the most (ripple current) stress because it turns the V DC from the average value to (near) the peak value.

But whatever, it's still the first HT node.

In a 2204, as with most amps, the filter HT node supplies the output valve anodes (via the OT primary).
The 2nd HT node feeds the output valve screen grids.