Hum with a twist?

[ashok]

Hi everyone,

I built an ST-35 Rev C in 2004 and have been enjoying it very much. It is now fitted with a GE 5751, RCA 12BH7 and Reflektor 6P14Ps, and sounds great.

The amp however, has had a distinct hum, which gets masked by the music, but is certainly noticeable during quiet passages. Shannon took a look at the amp in early 2005, and found that routing the speaker grounds directly to J1-5,6 instead of the ground terminals on J2 and J5 helped reduce the noise, but he did not find that there was any humming or buzzing. So, we concluded that there was something specific to my location that was causing a problem. After I got the amp back from Shannon, I basically just ran it as it was, and just ignored the hum.

Anyway, I would like to resurrect the issue now, and am hoping that I can draw on the experience of the people here to isolate the problem. The subject line says “humming with a twist” because the hum is audible only if interconnects are connected to the inputs of the amp. If the inputs are shorted at the RCA input jacks, the amp is almost entirely silent (speaker sensitivity is about 94dB/W/m). There is however a very, very slight raspy buzz that is audible with my ear right up by the speaker.

Would I be correct in saying that buzzing indicates a 60Hz (or 50Hz) problem, while humming (a sonorous sound) indicates a 120Hz (or 100Hz) problem?

Can the mere act of plugging interconnects to the inputs “overload” in some sense, the amp’s power supply?

Upstream of the interconnects is a 10K pot, no source connected, but with the pot turned all the way down. So, the amp’s inputs will be shorted to ground, albeit through the interconnects, as shown below (picture shows the wiper arm turned up a little bit from ground):



The funny thing is that if I unplug the interconnect from one input jack on the amp, and just short that input RCA to ground, both channels are now silent. The hum disappears.

My ST35 uses the 270HX power transformer, solid-state rectification. Instead of the 100 ohm parallel resistor combo, I am using the 156R choke. The amp is assembled on an aluminum plate, mounted on a wood chassis. RCA and speaker input jacks are isolated from the chassis. PCB is mounted on isolated stand-offs. All signal grounding is done on the PCB. Safety earth is connected to chassis at one of the mounting screws for the power transformer. J1-5 from the PCB is connected to this same point by a jumper.

Filament supply is connected to PCB ground on one side. (R40 and R41 have been lifted from the board.)

I thank you for reading so far and for any pointers that you can give me.

With best regards,

Ashok

[EWBrown]

First thing is to disconnect the inputs from any external connectins (shorting the inputs helps but isn't absolutely necessary for this, and turning the input pots all the way down accomplishes this for you). There should be no hum at all, at least audible hum, at all, if everything is good.

Since you have apparently already done this preliminary checking, it looks like your hum is externally induced. (Skip the next paragraph, go on to the following one).

If a hum is present, look for things like ground loops (using non insulated input jacks is a prime suspect in most cases), and loose connections on the terminal strip, if those were used on the board.

If the "stand-alone" amp is hum-free, then the input source or cables are the major suspect, look for AC ground loops, especially if the amp and the input source are plugged into separate AC outlets. Interconnect cables can have poor or incomplete grounding, even if they measure as being low resistance from end to end.

Hum (or buzz) comes in three distinct "flavors:

Hum is distinctive in that it is low frequency only, there are no "sharp" or higher frequency components, it will be either 60 or 120 Hz (in the USA and Canada and Mexico). Think of it as sounding like the lowest notes from a bass guitar (with no fuzz or other "effects").



60 Hz AC hum is from AC sources only, be they power line, or transformer windings. (which can also create a mechanical vibration, especially on a thin steel chassis). It can be electrically or magnetically induced. An example of the latter is having the two OPTs and the power trannie close together and with the cores oriented in the same direction. This is a particularly bad case, as no amount of circuit work will eliminate it, as it is magnetically coupled directly from power trannie to the OPT windings. This is not a common situation, but it CAN happen... The obvious give away here, is that hum can be present with no tubes in the amp.

120Hz "hum" (ripple) is from post-rectifier sources, the two diodes (or rectifier tube) effectively" "double" the frequency, the resulting "ripple" is then 120 Hz.
(Ripple isn;t just a cheap wine anymore Yellow_Light_Colorz_PDT_09 )

Buzz will have a sharp "edge" to it, the most likely source is through the inputs, it can come from light dimmers, CCFLs, nearby switching power supplies, TV sets, CRT monitors and RF sources. A flaky or bad rectifier can generate a similar buzz, as well, just look for any weird blue, greenish or purple glow inside the rectifier tube. If it is not a mercury vapor rectifier, then the glow is a very BAD thing... :o

Blue glow inside the glass (but not inside the plates) in a power tube is OK, some tubes are more colorful than others. Russian 300Bs can look like plasma globes, sometimes . Yellow_Light_Colorz_PDT_11

HTH

I once had a particularly persistent and nasty low level AC hum when I had TV, VCR, tape deck, CD player and a cheap SS tuner/amp all hooked up together.

The culprit turned out to be the incoming TV cable, as its "ground" was referenced to the main cable on the pole outside, and the indoors end of the cable was "hot" on its outer shield, with a few VAC, when referenced to house wiring neutral and ground.

I had to gin up an RF isolation transformer to eliminate the hum, or else just disconnect the cable altogether, toeliminate that hum.

/ed B in NH

[ashok]

Thanks for the tips Ed. I will be sure to try your suggestions. I now have an oscilloscope (old Heathkit, solid state), but have not used it yet on the ST35. I will see if the scope shows me anything useful.

Regards,

Ashok

[TomMcNally]

Hi Ashok -

I have a feeling your preamp is somehow connected to
ground via Cable TV (hooked to a TV set or FM tuner)
or to a computer sound card, which gets a ground via
the computer power cord. It's not impossible to get\
rid of, but can be tricky to do and maintain quality.
Those $ 16.95 isolation transformers from Radio Shack work,
but they are low-fi ... probably like 100-10K response.
There are baluns that go in the cable line that will do it
also, but they aren't cheap.

... tom

[kheper]

Following on Tom's suggestion: Can you find
a particular input source which is causing the
hum?

Take the amp and speakers into a different
room, then one by one, hook up each source
individually. For example, if you have a CD
player, hook it up directly rather than going
thru the pre-amp, etc.

[Ty_Bower]

Interconnect cables can have poor or incomplete grounding, even if they measure as being low resistance from end to end.

I'll second that. It's possible your interconnect cables might not be getting the job done. Have you tried a different set?

In your picture, at the source end of the interconnect cable, you show a 10k pot with one end tied to ground. Are you sure the ground reference at that end is the same as the ground reference at the amp?

It doesn't seem like the hum is in the amp. It seems like a ground loop, either due to an external piece of equipment or bad cables.

[Shannon Parks]

Hi Ashok,

I forget - was the power cord three-prong?

[ashok]

Thanks for the tips and suggestions everyone. I appear to have solved the problem, and here is what I did. First, a picture of the preamp and amp connection. In my earlier post, I did not show the source side in full detail. I apologize for that, I did not think that it was relevant.



Basically, there is a 1:1 transformer on the input. On the secondary of the transformer is the 10K pot, which is connected to insulated chassis-mount RCA jacks. Notice that the grounds of the chassis-mount jacks are connected to each other, shown by the dashed line. The interconnects connect to the ST35. Within the ST35, the grounds are tied together, and also to safety earth.

It turned out the the connection between the grounds of the RCA jacks in the preamp chassis was the problem. And you can see the loop it makes through the ground in the ST35 and the interconnects. Removing the jumper between signal ground and safety earth inside the ST35 chassis made no difference to the hum.

Removing the connection shown by the dashed line in the preamp chassis fixed the hum. With volume turned down completely, I now have to put my ears right next to the speakers to hear any residual hum/buzz.

But in all commercially available devices (at least the single ended ones), the grounds of all the chassis mounted jacks are probably tied together. Such is the case with a NAD preamp I was using some time ago and sure enough, the hum appears when it is connected to the ST35.

The interconnects are good quality Blue Jeans cables. Made with Belden cables and Canare RCA plugs. The transformers in the preamp are 1:1 audio coupling transformers made by Magnequest. Also, on the input side of the transformers, the signal grounds are connected to safety earth inside the preamp chassis (connection not shown).

I thank all who responded. For now, the problem appears to be solved.

Regards,

Ashok