No such thing as good hum

[LarryLarry]

Just finished building a Rocket stock from the rocket_layout_03_09_09_860.pdf layout and I've got a bit of hum that I'd rather not have! It is playable/useable as is (and sounds awesome), but the Express I built has zero hum, as does my Laney AOR30. I used a set of Baldwin organ amp transformers that are a bit out of spec; the PT is a bit high at 290-0-290 but the OT is in there at 4k2 primary with only a 4-ohm secondary. All pots work great, with the intended function of each performing as expected.

The hum is the same with or w/o the PI (V3) tube but of course gets amplified with V2 and V1 installed. I think my voltages are all as would be expected given the PT. There are 2 center taps on the PT but neither seems to be for the 6.3 filiments so I am using a virtual CT via 2 100 ohm resistors.

I used the headphone trick for both the OT and choke, and actually unbolted each and moved them around to see if it made a difference, but it didn't.

Attached are a few shots as I know they are needed. It really sounds to me like it's a power section filter issue, but I tested all caps/resisters before installing them.

[LarryLarry]

adding a few more photos...

I've tried several different 12ax7s and EL84s, and rectifiers too.

I might add I've chopsticked the heck out of this...

[Cliff Schecht]

Check that ALL of your hardware is securely mounted. I've had two builds recently (both my Rocket and Express actually) that become very hummy recently. The Express had a loose output jack and the Rocket had most of the hardware not securely fastened down.. Both amps are nice and quiet again.

[catears]

Sounds like it could be a power section or maybe a grounding issue. Check all your grounds. The build looks solid. Did you try moving the heater winding going to the 100R resistors? Try crossing it over your OT CT and not over the anode lines going to your power tubes. Also check the B+4 connection at the cap/resistors, it looks a little shakey in picture 36. I hope it works out for you!

[rooster]

I think for one thing your filament wires are too close to the pins of the EL84s. You need to arch them across the four tubes or loop them in a half circle pattern flat against the chassis like Ken. I think anywhere but right through the middle of them. The next thing that seems odd is the ground wires from the PT. I get the red yellow - what is the green yellow?

Other than this things look OK. In regard to the 1 or 2 watt carbon pieces, remember that the bigger the carbon, the more the shot noise. So a 1/2 watt version is going to be quieter than a 2 watt version. Also, carbon film is quieter than carbon. I love carbon when they are good and in the right amp, don't get me wrong, but your use of metal film and then the carbon together makes me think that you should consider that what you gained with one you lost with the other. Know what I mean? I would definitely not use that 33K one there at V1 for example.

OK, sorry to rag. I hate hum, too, loved the title of this post!

[Cliff Schecht]

Shot noise in any resistor type is directly proportional to the amount of DC current flowing through it. The amount of current flowing through just about any stage of a preamp will be very low. Even the screen resistors will not contribute a significant amount of noise and even then, any noise contributed at this stage of any amplifier should be insignificant (the noise floor is dominated by the first stage). I've heard about the wattage of a resistor having an effect on this as Rooster mentions but I'm yet to see any real data to back this up..

Some of the noise we deal with is thermal noise or Johnson/Nyquist noise. This is proportional to the value of the resistor where a larger value has more noise. This is where carbon comps hurt an amps SNR the most, especially in the first stage, because they do inherently have a worse noise floor than more modern technologies.

There is also 1/f noise and popcorn noise. 1/f is inversely proportional to the frequency (meaning it gets worse as frequency response goes lower. We don't really see this because we roll off enough low frequencies to squash this noise typically. I don't know how much popcorn noise exists in tubes (I only "understand" it in terms of silicon junctions) but I don't think this is an issue with tubes.

Noise is one of those topics that is easily misunderstood and misinterpreted by not understanding the origins. I've been through the painful derivations and have beaten the topic to death with regards to silicon. While tubes are a different story in terms of how they amplify, they are still essentially charge controlled devices which are governed by the same laws of physics. The light speed electrons flying around in a sealed glass vacuum only being steered by electric fields are bound to get free every once in a while. I think this alone is a big noise contributor in tubes (shot noise) along with the obvious 60/120 Hz issues.

[martin manning]

The next thing that seems odd is the ground wires from the PT. I get the red yellow - what is the green yellow?

I think that might mean that there is both an artificial filamant CT and a real one. If so, one or the other should be removed. Which other winding does the green-yellow wire have continuity with?

[surfsup]

I think for one thing your filament wires are too close to the pins of the EL84s. You need to arch them across the four tubes or loop them in a half circle pattern flat against the chassis like Ken.

Probably a non issue but looks like the green is wrapped around the black going to the first el84 and second el84 too...which makes em uneven/uncancelling.

Your grid run on the PI is kinda running parallel to the cathode, not sure if that would be better running straight out a bit along the chassis then turn to go back to the board.

remember that the bigger the carbon, the more the shot noise. So a 1/2 watt version is going to be quieter than a 2 watt version.

I heard exactly the opposite. The larger, higher watt resistor is going to cause less noise.

[hans-jörg]

Hi,
1+ for filament Ct. If green/yellow is the CT you have to remove it.
And filament wireing at powertubes!

Hans-Jörg

[RevD]

I don't know if this is relevant but it was on the Aiken amp site:

"Use the largest practical wattage resistors (unless you are using wirewound resistors) because contact noise is decreased in a larger geometry material."

And:

"The material and geometry of the resistor can greatly affect the contact noise. Therefore, if you double the power rating of the resistor, which increases the size and area, you will reduce the contact noise generated by the resistor."

Page: http://www.aikenamps.com/ResistorNoise.htm

Just something I remembered reading and had in my little personal journal of amp stuff...

Regards,

Don

[hans-jörg]

Yepp,
I´m reading the same aiken pages and follow this advises. Until now sucsessful regarding noise.
Hans-Jörg

[LarryLarry]

Well, thanks for all the suggestions and information!!

I thought the green/yellow s/b the ct for the filiments myself, and when I first built it I did not do the 2x100r virtual ct - it hummed really loud!! Then I looked back at my previous build with this transformer set (36 watt lite) and I did the virtual ct on it as well. Also, it did not have any hum so I know I can get rid of what I have.

I'll re-read all the replies and start trying stuff tonight!

[leadfootdriver]

Could be a ground loop. Get some wire, or test leads with gator clips, and attach all of your grounds to where your main boards ground buss attaches to the chassis.

It could be noisy preamp tubes as well. My Express really revealed the tubes in my stash that had a lot of hum, and some of them hummed a lot!

[selloutrr]

did you clean the chassis where you are making grounding contacts? It make have a film that needs to be removed for a better contact.

Have you metered your ohm (impedance) to ground as well as your other solder joints?

The Carbon Comp resistors are also naturally more noisy. I noticed you used them in the preamp section.

Have you tried tube rolling? some tubes in this circuit are more prone to noise then others.

[rooster]

OK, I realized I did present just the OPPOSITE of what Aikens states, and that is that the larger the resistor, the SMALLER the shot noise. Eh, I thought about that as I walked away from the computer to watch the Grammys... I was wrong stating that according to Aiken's data, and yes, I had read that there myself.

On the subject of carbon comp resistors I will also add that very recently I revisited an amp I built 9 years ago. It's a Fender SR clone and I used cc throughout the circuit, notably on the plates of the pre stages. Measuring the ccs, I see that they have all gone up in value 7-10% from when I put them in. (And yes, I measure everything before I use it.) OK, so, as things would have it, I am also repairing a 62 brown tolex Deluxe. As I measured the ccs in it, they are all pretty much spot on for value, maybe 1% either way. Which leads me to conclude that not all ccs are created equal and that they definitely were made better back in the day. The ones in my SR were all 1/2 watters purchased from CEDist, BTW. I still have a supply, and as I measure them - unused - they are spot on. Which leads me to conclude that it was the applied voltages over time that changed their value. ...OK, something else to fret about.

Since I'm here, is there a cc brand available today, being made currently, that some of you might suggest? Something that might hold it's resistive value like the old ones? Thanks.