Grounding hum on lm3886 kit

[encode]

I'm fairly new to the world of amplifier building, so I started with a kit from chipamp.com - the lm3886 stereo kit with the snubberized power supply.

I constructed it as per the instructions, with a 300VA 25V twin secondary toroidal transformer. A picture of my construction can be found here, although the wiring is only temporary at this stage. Since the photo was taken I have also connected the second amplifier module. I chose to omit the feedback capacitor when building the amplifier modules.

In normal operation, there is a slight background hum. If I touch the input leads connected to IN and SG on the amplifier module, the slight hum increases dramatically.

Any suggestions for areas to investigate this grounding issue?

Thanks!

[gainphile]

I had the same frustration before and fixed it:

Gainclone HUM, THUMP problems - The solution

[encode]

Yes, I saw that post.

Unfortunately I think the issues are somewhat different, given that I'm using a PCB with groundplanes etc. Also, my input signal wires are some distance away from my transformer, and from any mains-voltage carrying wires.

[gainphile]

The concept of star-grounding still the same with PCB, only easier. Use (much) bigger cable for path to PSU. Isolate all inputs from chassis too (later)!

[AndrewT]

I absolutely disagree with the mains wiring layout shown in your pic.
The two single insulated mains wires pass over the low voltage circuitry.
A disaster waiting to happen.

Take all your transformer wires to terminals in a fixed terminal block and then run your mains input completely separate from the LV/isolated side so that you or your equipment cannot accidentally touch the mains side.
Double insulate your mains wires. Not just the Safety Earth.

[encode]

Originally posted by AndrewT
I absolutely disagree with the mains wiring layout shown in your pic.
The two single insulated mains wires pass over the low voltage circuitry.
A disaster waiting to happen.

Take all your transformer wires to terminals in a fixed terminal block and then run your mains input completely separate from the LV/isolated side so that you or your equipment cannot accidentally touch the mains side.
Double insulate your mains wires. Not just the Safety Earth.

The wiring is temporary, and I didn't have the insulated mains wiring passing over the rectifier PCB when I had the second channel connected (the mains wiring is currently at the left of the amplifier).

I will try changing the cable layout, and beefing up the grounding wires to the central ground point.

[AndrewT]

Temporary is not a valid excuse.
We want you to survive your temporary experiments.

[valleyman]

A good way to avoid dying in test setups which aren't properly safety grounded is to not touch them while they're plugged in, and I can't really see much danger if you do that.

I'm not saying it's ok not to properly safety ground your equipment, I do think people sometimes get a little too anxious about it. A finished article should always be properly safety earthed, but I'll bet many, many people here don't in the testing phase.
I have received 4 240V shocks in my life, all scary and very unpleasant, but I think mostly serious problems are only likely to arise if you have a heart problem or particularaly weak heart. Certainly if your mains is 115V personally I wouldnt really be worrying.

I suppose what I'm saying is really that very many people here will use improperly grounded test setups which is a reality that must be faced. Educating people not to do it is very important, but equally important is making sure people take the necessary precautions when the safety earth is not implemented as inevitably it will happen