Best grounding scheme

[MickeyB]

Doing a little research on the best grounding in my 6G16 vibroverb clone. Have the brass plate that came with the chassis from Ted Weber. Some say this was a good idea from Leo Fender, others say no, go with the star ground technique. So my head is spinning and thought I would check and see if any of you guys have any experience with both and what worked better. I just want the least amount of hum possible.

[tubeswell]

Well I just built a 5F2A, and I don't know how similar it is to your build, but MJWB suggested that I ground the preamp (i.e. everything in the preamp including vol and tone knobs, inputs etc - including the preamp-filter cap) to the input ground. The power amp (i.e. the power tube cathode the two power amp filter caps) went to the only other ground along with the HT CT ground, PT shield, heater CT, and Mains ground (on one of the PT bolts). And it works just dandy.

[TheTinMan]

There are almost as many opinions about grounding schemes as there are amp builders. That said, here are some references that I found useful:

http://www.aikenamps.com/StarGround.html
http://www.paulrubyamps.com/info.html#Ground
http://www.el34world.com/charts/grounds.htm

The approach tubeswell describes worked well in a 5F2-A for me too, but I used the Hoffman approach in a Princeton Reverb. Kevin O'Connor describes a "galactic" grounding approach which makes a lot more sense <to me> than a true star ground - grounds for each stage bundled together at the cathode ground, then all cathode grounds collected and grounded to a single point. Hoffman's and other buss grounding methods are related to the galactic ground, but not exactly the same.

Leo Fender built a ton of amps using those brass plates and relatively random grounding. However, the risk of corrosion and/or bad connections between the brass plate & the chassis makes it seem like a bad idea. (FWIW someone suggested drilling a small hole in the brass plate for each ground to get some mechanical connection before soldering.)

Hope this helps,

Chip

[MickeyB]

Thats why my head was spinning. Every well known amp builders site showed a different style. These are good ideas. Thanks for the input guys.

[dai h.]

Leo Fender built a ton of amps using those brass plates and relatively random grounding.

I don't see a "random" grounding scheme. The general groupings there make sense in terms of theory.

However, the risk of corrosion and/or bad connections between the brass plate & the chassis makes it seem like a bad idea.

this is according to Doug Hoffman, however according to Bruce (Collins) here, he (Bruce) has never seen one in a corroded condition. (Perhaps the different locations have something to do with this however.)

If you search the archives, you should find a lot of info on grounding as well as a ton of links.

edit:

also, another point. "Starring" points is for the purpose of eliminating or reducing common resistance. What happens when you have two points on a plane? Look at a ultra thin piece of foil (plane) as a resistor with a (very low) resistance. What happens if you keep adding foils of the same exact size and keeping doing that? The resistance goes down therefore the resistance between the two points goes down. "Reducing common resistance" between the two points has been achieved using a plane. So a star between two points and a plane could appear different yet be achieving the same goal of low common resistance between points. You can also see that the lower the resistance of the plane, the better.

[cminor9]

I am still new, but I think I can contribute something interesting to this discussion:

I built a 5e3 in a 12x8x2 chassis and used star grounding. I then re-built the same 5e3 (same everything) into a 16x8x3 chassis and this time used the grounding method described at http://www.el34world.com/charts/grounds.htm.

Here are the changes I made from one build to the next, as well as all mods I have made (in the name of full disclosure): in the first build I had a switchable feedback loop. It seemed next to worthless so I omitted it from the second build. On the first build, I left out the bypass cap on V2. On the latter build, I made this bypass cap switchable. On both builds I lowered the first two filter caps to .022 from .1. On both builds I added 470 ohm screen resistors to the 6V6 tubes. Lastly, I upped the 5W resistor on the cathode from 270 ohm to 330. Both builds are in aluminum chassis.

A 14ga bare copper wire runs across all pots, to the input jacks, and then to the chassis. The preamp and PI grounds run to this ground. The power tubes and filter caps ground to a lug near the PT. The PT center taps ground to another lug near the PT. Earth ground goes to yet another lug near the PT.

This is a much cleaner build for it, and it dead quiet. It's nice to hear an amp that has no hum, it only hisses a bit (the carbon comp resistors). The star grounded amp was quiet, but there was just a little hum and it was a mess with ground wires running everywhere. Some of this could be better lead dress the second time around. But what is for certain, all other things considered equal, buss grounding made for a MUCH neater amp.

I'd bet you can make an equally quiet amp using either method. But if you want a clean build, ride the buss.

[MickeyB]

Grounding

Thanks for the reply. I actually did redo the grounding to a buss grounding and it eliminated the little hum that was there. Like you said, just the carbon comp hiss. I have never in my 30 years of playing have ever had such a clean smooth sounding amp like this. And I have owned Peavey, Marshall, even Mesa. I was hesitant at first to take on something like this. I am an electronic amateur hobby guy who didnt know one thing about tube amps. I did my homework, asked a lot of questions on this forum, read a lot of posts on this forum, and when at it. What a feeling that was when I plugged in my strat and heard the sound come out. Had to work out a few bugs, but in the end, its awesome. Parts to build it, $700, building a hand wired tube amp yourself, PRICELESS!
When the funds come available again, I am going for a high gain style head.
Can't wait :^)

[Bruce / Mission Amps]

For what it's worth... the brass plate is a star ground... it is just a large one.
The only thing odd that Fender did in some builds was to move the filter cap grounds to a spot close to the high voltage center tap ground point.
I was around to ask but I'm thinking this was done as a matter of "after a few prototypes, this works OK" and easy to assemble on the small assembly line.
Remember, as the story goes, those old Fender amps were almost all built by women and a few men who were paid by the piece, not hourly.
That means make it work OK make it simple and make sure it is fast.
The brass plate is a stellar way to always insure the audio path stuff is all connected to a large star ground and the power supply leads from under the dog house has to go to a place that is very close to the power transformer's HI-V center tap.
Plus all those star washers under the pots and jacks have something soft to dig into for a really good ground point.... the brass.

[PRNDL]

this is according to Doug Hoffman, however according to Bruce (Collins) here, he (Bruce) has never seen one in a corroded condition. (Perhaps the different locations have something to do with this however.)

Actually, a little while back I repaired a VibroVerb that had this exact problem - the copper plate was not grounding to the chassis. It was a silverface to blackface conversion, and I did see a black stain of what looked like oil that went under the plate.

The fix was simple. I made a large solder blob on the chassis that went onto the copper plate. It needed it on the preamp side, but I put an extra one on the other side just in case.

IMHO, the copper plate is fantastic for the multiple grounds in the preamp of most Fender builds (i.e. those with reverb and/or tremolo). It's not needed in simple amps, such as the Champ, since you can ground the preamp section with one wire to the pots.

[jrfrond]

I learned the hard way (well, I did a lot of research too) when building new designs to NOT bring everything back to a common ground point, a true "star ground".

The way to visualize it is this: all of your current flows from ground. This is called "electron current flow" (negative to positive), whereas many people think in "conventional current flow". This is positive to negative flow. In order to see how important grounding is, you need to deal with electron flow. Anyone who has dealt with poor PC board designs where ground traces burn up like a big fuse will know that the ground is the business end of the current flow system.

At the ground point, all of the current (electrons) needed for a particular circuit is pulled from ground. For example, let's say you have a preamp circuit, which we want as quiet as possible. If you were to ground the heater winding center tap to the same EXACT point as the preamp ground, you run the risk of superimposing the current flow demanded by the heaters (a fair amount of current at 60Hz) onto the current flow needed by the preamp (relatively small DC current flow). You might very well wind up with hum in the preamp stage. Moreover, you will NEVER see the hum on the B+ line, though that is where most people suspect hum to leak in as AC ripple. You WILL see it on your preamp stage cathodes though. That is your point of entry. However, if you ground the heater center tap to a point well AWAY from there, the possibility of shared current flow is reduced. Despite the fact that this is, as Bruce pointed out, an "extended Star" system, the small amount of resistance in the ground plane is enough to isolate these circuits from each other.

Grounding is perhaps THE most critical design feature in any electronic or electrical circuit, and it is part science, part black art. It can make or break a design, and many times does. Oftentimes, grounds must be played with and changed to get it all right.

Oh yeah, the brass plate used in Fender and other amps? There's no magic. Brass is just easier to solder to than steel. You could use copper as well if that's all you have.

Fender amps SEEM randomly grounded, but it was well thought out. That is why they are always dead quiet.

[CitizenCain]

I've found the info in Paul Ruby's link noted above to be the breakthrough point for me in understanding grounding.

I used that method in my first big amp project and that one is very quiet until you crank it way up.

[BackwardsBoB]

V=-IR?

jrfrond wrote:

The way to visualize it is this: all of your current flows from ground. This is called "electron current flow" (negative to positive), whereas many people think in "conventional current flow". This is positive to negative flow.

So which shoe do you take off to compute the force on a current-carrying wire?

I find that lately I get nervous with less than four ground planes, but...

jrfrond is right. You really need to be the electrons, visualize the flow, and realize that any current will generate a voltage offset in just about anything.

You don't want to take a grid resistor back to a star ground, you want to take it to the bottom of the cathode resistor. And you want a single wire from you're first tube stage to your second tube stage. But you don't want your output tubes' cathode current stopping by the cathode resistor on the way to the bottom of you B+ caps, etc., so you should wind up with ground clusters, starred into your ground reference point.

Loops are bad things, not just for grounds, but power, high-current wires, small signal wires, etc. If they don't pick up radiated AC, they'll generate it. Loops don't have to be DC connected. From the top of a power cap, through a circuit, out the ground, and back to the cap is a loop. Gotta keep 'em small or isolated or something.

Keep AC current flow local with capacitors. If you put a cap across the ground and the supply for a circuit, the scary current will go through it, not a big loop back to your ground reference, though low-frequency current in that path is unavoidable.

Ground planes are wonderful things. If you run a wire close over them, they eat the radiated magnetic field effectively, leaving significant field only within a distance of a few of the spaces between the wire and the ground plane. They also help protect the wire from radiated fields from other wires. The field cancelling current in the ground plane does disturb the plane's ground, but unless your plane looks like swiss cheese, it should eat this current with minimal voltage disturbance, even if its only a few mils thick.

[MattT]

Where does the preamp 'handoff' to the power amp?

Take a cathode-biased no-NFB amp:
The input jack, preamp, and preamp filter grounds go to the preamp ground buss.
The OT center tap filters and screen filters go to the power amp ground.
The power tube cathodes and PT center taps go to the power amp ground.

But where does the PI filter ground?
The PI tail gets grounded to the same spot as the PI filter...correct?
Where do the power tube grid returns ground?

I've seen the PI grounds done several different ways...not really sure where they're supposed to go.

I've always used the fattest wire for the grounding just because I always figured that's where the actual 'flow' is originating and you don't want a bottleneck/logjam right at the entrance. Is this necessary?

[jrfrond]

You don't necessarily want to take your PI ground (or any other) back at the filter cap ground, because currents flow in that ground. You might need to take it from a "quieter" area in the chassis, right next to the PI circuit. Sometimes, you need to try several different points to find the quietest one.

I use a modular approach in my amps. The power supply, preamp and PI/power amp all have their own turret boards, and they are close to their associated tubes. I ground each board right at the mounting point in the chassis, soldered to brass busses I have soldered into the chassis. The heater winding is grounded right next to the power transformer (also to a buss bar) and the B+ winding a few inches downstream from that. This ground distribution scheme decouples the stages nicely and keeps everything quiet and stable, but I didn't happen upon it overnight. It took a month of solid work to debug it. This is just MY approach, but it might be helpful.

[mbratch]

A 14ga bare copper wire runs across all pots, to the input jacks, and then to the chassis. The preamp and PI grounds run to this ground. The power tubes and filter caps ground to a lug near the PT. The PT center taps ground to another lug near the PT. Earth ground goes to yet another lug near the PT.

This is a much cleaner build for it, and it dead quiet.

I just finished building my first 5E3 kit, and it's the first time I've played a 5E3. After reading lots about grounding and all the methods and varying (and sometimes conflicting) opinions, I did exactly as described above.

My amp, however, has a bit of 60 cycle hum that follows the volume control. At low volumes, it's practically inaudible. When the volume is up high, it's more noticeable. It's higher on the Normal channel than the Bright channel. I tried a couple of quick mods to see if I could alter it: (a) I tried moving the back-of-pot ground buss chassis tie from the jack side to the PT side, and (b) I tried a shielded audio cable (ground on one side) from the first volume pot to V2. In both cases, there was no distinguishable change to the hum.

So now I'm curious: how can I totally get rid of the 60 cycle hum?