The kind of noise that comes in through the AC mains supply is usually like buzz riding on the 60Hz (or 50Hz outside of N America). Ground loop hum inside an amp is just a low-pitched hmmmmmmmm sound. First make sure you know which you're dealing with here.
yes my amp hums with no tubes in v1 or v2.
That means that him is getting in at the PI stage.
In fact if I have the TMB volume and master dimed with no v1 or v2 tubes it has a little hum, and it is much worse with v2 in, a little worse still with v1 in plus more hiss as well.
Sounds like you've contradicted what you just said....
I am dropping the b+ because of that with a series of diodes on the CT that are grounded with the power section there.
Not good. You should have a good solid connection between the PT secondary CT and ground. No zeners or switches or other such junk in the path. A much better place to use seners is in series with the B+ line after the first filter cap. This will also help the zeners to last longer.
Look at the schematic and the layout, and it should be perfectly obvious. It's the first filter cap that comes after the rectifiers. Also bear in mind that Randall Aiken is devotee of star-grounding, while we usually use buss grounding with 18W amps, although we apply the same basic prnciples of star grounding.
At Paul Ruby's site I picked up this tidbit on killing filament hum - the trinity layout shows and my amp has the filament center tap grounded but he suggests grounding it to the Cathode resistor positive as even quieter on a high gain amp.
That may make a small improvement. Nothing big. I do it with some of my amp builds.
The question I have is this.. he says "if you have a nice big bypass cap..." my amp has a smallish 50uF bypass cap, If I were to try this should I add a bigger cap first? I don't understand the nuances enough to go ahead without a bit of input or explanation here.
You won't know without trying.
FYI, here's one of my posts from 18watt.com on the subject of grounding, which you may find helpfull.
A few points and observations.
- As I've said before, successful grounding is 90% science and 10% voodoo.
- On the 90% science side, good grounding practice is based on the principles of star grounding, even if your ground system doesn't physically look like it's a star. This is the way that grounding is done in professional electronics design. A ground buss is just a stretched out star point.
- So stages should be thought of as individual entities with their own ground or buss section, with their relevant filter cap also connected to the same point (see Phil_S' comments).
- It's vital to understand that the the preamp and PI stages typically have very small ground currents and high sensitivity to any noise being injected into their grounds and power rails.
- Power amp and power supply stages typically have large ground currents, low gain levels, and fairly high amounts of ripple on the rail. If these large ground currents are allowed to circulate through the preamp/PI ground system, they can induce noise which the preamp/PI stages will amplify and send to the rest of the amp.
- In general, ground return currents belonging to one stage should never cross ground currents belonging to another stage, and especially between currents of the "light" and "heavy" ends of the amps.
- Avoid running a single buss wire for all the pots which is separate from the preamp grounds, as typically these pots are associated with different preamp stages. So break the pot grounds between groups of pots which belong to the same stages. Then connect a single wire between each group of pots and the preamp ground buss section (or star point) that belongs to that stage.
- As long as you keep these rules in mind, a number of different grounding systems will work well. You could even ground the whole amp to a single buss wire if you wanted.
Phil_S wrote:
Separate the stages for grouping purposes. Each end of stage is the cathode of a tube. The first stage is the input jack, filter cap, and cathode. If the B+ node is shared, then all of the cathodes are part of the stage (like V1A and V1B, typical). Pot-related grounds are included in the stage that ends with the next cathode.
Some excellent advice here - and it applies equally for star or buss grounding, although it's based on the star grounding concept. With a buss it's a good idea to split the buss into virtual sections, so all the grounds from each stage go to the same part of the buss. Ground currents must not cross between stages.
Phil_S wrote:
...don't use the chassis as a buss. It is a shield. For that reason, don't establish a number of ground points on the chassis.
This is a very important point.
Phil_S wrote:
On shielded wire...avoid it's use as there is thinking that the added capacitance is a tone thief. I can't say one way or the other to be honest.
Personally, I wouldn't tell people to avoid shielded wire. There are some places it's pretty much essential, such as from the input jacks to the first preamp tube grid(s). It's often needed for connections to gain and volume pots to help prevent squeal.
Phil_S wrote:
It is not placed on the buss. See above regarding chassis as shield, it is part of the shield, so any convenient chassis bolt can be used.
I always ground shielded wire to the ground buss, at the part of the buss which belongs to the relevant stage. The one exception is the input jack(s), where I usually ground the shield to the jacks' ground lug(s), which will be connected to a chassis ground lug very close to the jacks, where the buss also connects.
GibsonGM wrote:
PI is grounded to the power section, all preamp gnd connections on the same buss...
Personally I prefer to see the PI (at least of the LTP kind) grounded to the preamp side. The PI is a gain stage and has very small ground currents, which we don't want heavier ground currents mixed with.
GibsonGM wrote:
I question if I should eliminate the preamp-to-chassis connection and run a wire over to the power grounds, or is it ok to just ground preamp right to chassis??
It's very OK to ground the preamp buss right near the input jack, as that will help reduce interference. It's also *not* a good idea to run a long ground wire back across the amp to the heavy current side. It increases the possibility of noise getting being picked up.
GibsonGM wrote:
Gonna replace all input jacks with Cliffs soon to avoid 'the loop'.
That should help a lot.
Phil_S wrote:
I've seen people swear by the system of making a chassis ground near the input jack, the way you've done it. I don't do it that way. That doesn't mean what you did is wrong.
I've done it both ways, and it's worked fine. These days I go for having the preamp ground near the input jack, as I think it's a tad quieter.
Phil_S wrote:
I don't have any builds with a MV. My feeling is, if the MV is tied in after the cathodes of the PI (post PI MV), then ground the MV with the power section. If you have it at the tail end of the preamp and feeding the PI, then it belongs with the preamp grounds on the buss.
I usually keep the MV pot grounded with the PI/preamp, even when it's a post PI MV. I tend to visualise the MV pot's wiper as the natural break point.
rjgtr wrote:
I use a simple 3 point grounding scheme.
I prefer to think of that as a 2-point grounding scheme, plus one safety ground connection, since that latter one isn't part of the actual amplification circuit, although it's critically important for safety.
Although in theory we shouldn't use the chassis for a ground return, which is what happens in a 2-point system with separate preamp and power amp lugs, it appears to work well when done in this limited way, at least with an aluminium chassis. I have built a number of amps using this scheme and they've been real quiet.
It doesn't always seem to work so well with a steel chassis. However, I did build one amp in a steel chassis with separate busses for the PSU/PA "heavy" end, and for the preamp/PI, and then connected both busses to a single lug on the chassis. It worked real well, although with hindsight I now wish I had positioned the lug closer to the input jacks.