another neophyte question about AC voltage

[Enzo]

Dai, I think we are talking at cross purposes. I think I am talking abuot how something COULD occur, and I think you are discussing how people make it not. I didn't realize Leo was a ham, that is where I learned my electronics 50 some years ago.

The Champ hum is nothing esoteric, it is simply that a single ended amp has no hum cancellation like a push pull amp does. To kill the hum you'd need to add more stages of filtration BEFORE the power tube, and Leo was NOT going to toss that extra money into them.

The alum foil "lid" on a Fender is far enough from the circuitry as to offer littel capacitive effects, IMO.

[dai h.]

Dai, I think we are talking at cross purposes. I think I am talking abuot how something COULD occur, and I think you are discussing how people make it not.

Yeah I suppose. I'm think more of why something does work when done in a way knowledgable people say don't (and WHY does that "wrong way" work). But another example: in the Aiken article, his ground DOES go through the chassis. One end is lower power star, other end higher power star, and connecting them is the chassis, so not using the chassis doesn't seem to be an absolute wrong. A lot of times you see DIYers looking at a Fender or something, thinking the scheme is wrong because there are multiple ground connections to the chassis but I don't think they understand and I'm saying it can be okay depending on HOW the points are connected.

The Champ hum is nothing esoteric, it is simply that a single ended amp has no hum cancellation like a push pull amp does. To kill the hum you'd need to add more stages of filtration BEFORE the power tube, and Leo was NOT going to toss that extra money into them.

that is what I figured after trying to think why the hum can be low in a typical 5F6-A/Marshall circuit when the filtering doesn't have a lot of stages (answer: the push-pull must help a lot).

re: the Champ: I've never built one or tweaked a Champ, so I don't have a real good physical/aural sense of how far you can (or can't improve) the noise (even with the single-ended output) compared to say a typical Marshall (which I think can be made reasonably quiet). Kind of suspect it could be improved but really, I don't know this.

The alum foil "lid" on a Fender is far enough from the circuitry as to offer littel capacitive effects, IMO.

yes, it could be more that it's helping to keep out the spk. signal coming back in. But kind of close to some of the pots and jacks though (Hi-Z = higher coupling effects I think just like possible sensitivity of the gtr. to input cable capacitance).

[Steve Conner]

Ptron, I've done a good deal of PCB layout too, and broadly speaking R.G. is right. The digitalness is what makes it not matter: as long as the ground noise is below the noise margin of the circuitry all will be fine. It takes a pretty good jolt of noise to corrupt an 0 into a 1 and vice versa.

Analog audio circuitry is far more sensitive, and sometimes "dropping a via to the ground plane" - which in Leo's time would have meant soldering your component leg to the chassis - just won't do it.

Because currents flow in the ground plane, and it's not a perfect conductor with zero resistance, the "ocean of ground" assumption (by which I mean the assumption, beloved of analog newbies, that the groundplane is exactly 0v everywhere) breaks down.

The classic example in a tube amp is if you screw up your grounding network so badly that the current spikes in the path between first filter cap negative and rectifier (which can be an amp or two) pass through the ground conductor of the preamp tubes. The result is a very loud and harsh buzz contaminating your audio.

If you soldered every "ground" to the nearest point on the chassis, this might well happen, since the chassis is steel with a relatively high resistance. So, you usually find that the PT center tap (or negative terminal of a bridge rectifier) is wired straight to the - terminal of the first filter cap with a separate ground wire. This is then connected to the rest of the ground network using another wire taken from the first filter cap - to the chassis, etc.

Another popular trick is to connect all the preamp circuitry to a separate ground bus, connected to the chassis at one end only, normally next to the input jack. The idea is that stray currents can't get induced in this bus if it's only connected at one end.

[Ptron]

Ptron, I've done a good deal of PCB layout too, and broadly speaking R.G. is right. The digitalness is what makes it not matter: as long as the ground noise is below the noise margin of the circuitry all will be fine. It takes a pretty good jolt of noise to corrupt an 0 into a 1 and vice versa.

Well, that's logical enough (no pun intended) and probably should have been obvious enough

Seems like whenever I assume something to be that simple someone comes along and throws the blauchengrubershmitzen effect in my face.

Just to clarify by the way, I was just responding to the PCB/ground plane thing and not trying to imply anything about guitar amps or audio in general.

What was this thread originally about anyway?

[mooreamps]

Tut 3

You want to hear from someone knowledgeable

Go to LondonPower.com and order TUT 3 from Kevin O'Conner. This book will detail every answer you are looking for on this subject.

-g

[dai h.]

sorry for digging up this old thread but I typed out a section for somebody to try to explain grounding a bit and I think this might show where Leo Fender (as a radio guy) was coming from when doing grounding schemes. This is a passage from a book that came out way back in 1936 (gosh, maybe even older than Enzo... ).

Grounds

The way in which "ground" or return connections are made may have a considerable effect on the operation of the circuit. Ideally, all ground connections in the transmitter should have the same r.f. potential. Unfortunately however, all leads, no matter how short, possess some inductance and resistance, and it often happens that two points theoretically grounded may show a difference in potential in practice. The nearest approach to the ideal ground would be a large metal sheet of high conductivity on which all apparatus is mounted, the grounds to the metal sheet being made in all cases with leads of negligible length. In practical construction, a metal chassis makes a satisfactory ground; the large surface reduces both inductance and resistance to very low value in most cases. When such a chassis is used, ground returns should be made directly to the chassis with short direct connections; avoid, if possible, grounding any two circuit elements to the same point on the chassis, and never use a single ground wire for two or more bypass condensers, even though the common ground wire be quite short. With two condensers grounded through the same wire, the wire provides a coupling between the two circuits which the grounding is supposed to avoid.

This principle is illustrated in Fig. 907, adapted for construction on insulating material such as a breadboard. In this case, since no metal ground of large surface is available, it is best to bring all ground connections for each stage to a single point, connections between stages being made also by a single wire between the common ground points. In wooden frames or racks, where several stages may be mounted one above the other, loops and circuitous returns up and down the opposite sides of the frame should be avoided. In such cases, it is advantageous if a coupling system which does not require the ground connections to carry r.f. current is used; for example, link-coupling between stages on different shelves will avoid grounding troubles.

If the transmitter units are mounted on bakelite, pressed wood or plywood base, a satisfactory ground can be made by converting the bottom of the base with a thin sheet of aluminum or copper and making ground to this sheet just as though it were a metal chassis. This type of construction is used in some of the transmitter units illustrated in this chapter.

Poor grounding can account for many transmitter eccentricities, including inadequate excitation in a transmitter layout which on paper has ample driving power, and inability to neutralize a triode amplifier. It is often also reponsible for oscillation in screen-grid amplifiers.

from The Radio Amateur's Handbook, by the Headquarters Staff of the American Radio Relay League, 1936, pp.180-181. (you can download it from the PMillet site.)

Resistance in a Conductor

In general it is important to realize that:

* if you double the length of a wire, you will double the resistance of the wire.
* if you double the cross sectional area of a wire you will cut its resistance in half.

source:

http://www.regentsprep.org/Regents/p...st/default.htm

related thread:

http://music-electronics-forum.com/s...hp?t=16&page=2

[martin manning]

it's possible to get a quiet amp with AC heaters. One good example--the Soldano SLO100. There is some literature on having a quiet amp with AC heaters in the Radiotron book as well. If you do DC heaters and do it wrong, it can actually turn out worse with lots of hum. It can be more complicated too with more parts, huge caps, regulator, etc. The notion that running AC heaters is fundamentally wrong and therefore the answer to get a quiet amp is DC heaters is understandable but not really correct.

The Marshall 4001 Studio 15 I've been fooling around with has DC on the filaments of its two gain stages (a 12AX7). There are only two additional parts, as I see it: a FW bridge across the PT's filament winding, and a 1000 uf cap across the DC output. The cap is physically small since the voltage is low. There are two 100 Ohm resistors, one from each side of the cap to ground so the DC is + and -3.25v, but those would be there anyway. The PI and output tube filaments are connected directly to the transformer.

BTW, the socket for this tube is insulated from the chassis so that it and it's shield don't form a loop. Someone was really trying to keep this thing quiet, eh?

MPM

[dai h.]

yeah that might be a good option. At least one Boogie IIRC has DC to one input tube stage tube only.

BTW, the socket for this tube is insulated from the chassis so that it and it's shield don't form a loop. Someone was really trying to keep this thing quiet, eh?

it's not for shock mounting or something like that? Off the top of my head, doesn't seem like it'd be a problem to make multiple connections there. If there is just one connection I would think that was because only one was necessary so taking the effort and another lug, wire to make more would not have made sense to do. (Actually lots of connections might be better rather than worse since closer to zero ohms, zero impedance the harder it should be for a noise voltage to develop across it.)

[martin manning]

>>it's not for shock mounting or something like that?

Of course I don't know for sure (and I barely know a load-line from a load of bull), but this is the only tube that is mounted this way. The PI tube right next to it is bolted directly to the chassis.

The chassis has multiple ground points arranged just as you would expect, and so there is an E field of some sort throughout. The metal ring that holds the socket wouldn't contact the chassis except at the mounting points, so the ring and the shield would be in effect a loop of metal from one mounting point to the other, wrapped around the tube.

This is an unusual amp. The two gain stages are decoupled on the B+, too.

MPM

[Bruce / Mission Amps]

Grouding

Adding another response from one of those old time ham guys, ...me:
with respect what I think is a very small problem in most decently made guitar amps and their signal or filament grounds.... I always assumed the metal shield on the open end of a chassis was merely a type of Faraday shield.... you'll see these shields in old amps wrapped around the input jacks, soft metal plates vertically mounted in front of power tubes or sometimes in front preamp tubes or above high Z circuits and grid wires., etc etc...