Need Advice: Vinyl Hum Issue

AI AI its off to work I go 

lanx, Did you tell AI that the hum was not present when the subwoof was connected via bare wire but only appeared when spade connectors were added (presumably to the ends of the bare wires)? Because AI response does not seem to take that into account.

@lewm  Sorry, I didn’t. That’s my fault. Last time I only shared the “detailed symptoms.” Here’s what I told the AI this time. It’s impressive how AI analyzes the symptoms and narrows them down to the most likely root cause...

I have been using the KEF KC62 via the high-level (L connector) for some time without any hum issues. The problem only appeared after I replaced the bare speaker wire connections with spade connectors at the Willsenton R8 binding posts...

Detailed Symptoms  

• Hum is present only on vinyl

• Hum increases with volume

• When the KC62 is disconnected, the hum disappears

• However, with the sub disconnected and volume above ~10 o’clock:

  • Audible vibration / rattling

  • Woofer cones fluctuate excessively, especially on low-frequency passages 

• Touching the RCA plugs does not eliminate the hum

• I have not yet tested touching the tonearm

• Digital playback remains silent under the same conditions

This is a very good, well-observed symptom set, and the fact that it started immediately after switching from bare wire to spades on the Willsenton R8 is the key clue.

Below is a cause-and-effect diagnosis, then a step-by-step isolation plan.

---

What the symptoms tell us (put together)

1. Vinyl only

• Phono stage has 40–60 dB more gain than digital

• Any grounding, reference, or mechanical issue will be massively amplified

• This rules out the KC62, amp, or speakers as primary faults

2. Hum increases with volume

• This is upstream of the volume control

• Classic grounding / reference issue, not EMI pickup

3. Hum disappears when KC62 (high-level) is disconnected

• KC62 is participating in a ground/reference loop

• The sub itself is likely fine — it’s completing a loop

4. Switching to spades triggered the issue

• This strongly points to:

  • ground reference change

  • accidental chassis contact

  • loss of “floating” behavior that bare wire allowed

5. Woofer cones pumping + rattling with sub disconnected

• This is infrasonic energy, not hum alone

• Typically caused by:

  • turntable rumble

  • acoustic feedback

  • or phono grounding instability

• Tube amps + no sub = no LF damping → visible cone motion

6. Touching RCA plugs does NOT kill the hum

• This suggests:

  • signal ground is not the main issue

  • chassis / earth reference mismatch is more likely

7. Digital is dead silent

• Confirms:

  • amp is fine

  • sub is fine

  • speaker wiring is fine electrically

  • problem lives in phono + grounding + KC62 interaction

---

The most likely root cause (very high confidence)

🔴 The spade connectors are breaking the Willsenton R8’s floating ground behavior

Many tube amps (including the R8):

• do NOT have a true common speaker ground

• rely on transformer isolation

• behave best with floating speaker connections

When you used bare wire, the connection:

• had higher contact resistance

• did not mechanically reference chassis ground

• allowed some “give” electrically

With spades, you likely introduced one of these:

• spade barrel touching the binding post nut or chassis

• tighter coupling between speaker negative and amp ground

• slight short between L– and R– via sub’s high-level network

👉 When the KC62 is connected via high-level, it bridges the speaker negatives internally, which tube amps often hate.

Bare wire masked it.
Spades exposed it.

---

Why the KC62 triggers it

KEF KC62 high-level input:

• is NOT purely passive

• references signal ground internally

• expects solid-state, common-ground amps

Tube amps with:

• output transformers

• floating secondaries

• asymmetric grounding

→ can create a ground loop through the sub’s input network

---

Why the cone pumping happens (even with sub disconnected)

This is classic vinyl infrasonic feedback:

• Phono stage amplifies sub-20 Hz junk

• Tube amp has no subsonic filter

• Speakers unload at very low frequencies

• Result:

  • cone pumping

  • cabinet rattling

  • exaggerated when grounding is unstable

The KC62 normally helps by damping this — but the grounding issue forces you to disconnect it.

---

What I would do — in this exact order

Step 1 — Immediately revert to bare wire (test)

Just for diagnosis.

• If hum disappears → 100% confirmed spade / grounding interaction

• If hum remains → proceed, but this is very likely the fix

---

Step 2 — Inspect spades VERY carefully

Look for:

• any part of the spade barrel touching:

  • binding post nut

  • amp chassis

• oversized spades that bottom out

👉 If you want spades:

• use narrow, insulated spades

• or heat-shrink the barrels

---

Step 3 — Test tonearm grounding

This will tell us if the phono ground is also contributing.

While hum is present:

• gently touch the tonearm tube

• then touch the phono ground wire

• then touch the amp chassis

Results:

• hum reduces → grounding reference issue confirmed

• no change → loop is speaker-side, not signal-side

---

Step 4 — Temporarily lift the KC62 high-level negative

(Only for testing)

• Disconnect only the negative lead from one channel going to the KC62

• Leave main speakers connected normally

If hum drops dramatically:

• confirms KC62 bridging floating grounds

---

Step 5 — Long-term stable solutions (pick one)

✅ Best solution (tube-amp friendly)

• Use bare wire or fully insulated spades

• Keep KC62 on high-level, but:

  • route wires cleanly

  • avoid chassis proximity

✅ Alternative

• Use line-level from phono preamp (if available)

• Avoid high-level entirely with tube amps

✅ Highly recommended for vinyl

• Add a subsonic / rumble filter

  • even a gentle 18–20 Hz filter will stop cone pumping

  • this also reduces feedback and hum exaggeration

---

 

@veridian

AI can’t even use the common term "acoustic feedback"?

But it will take your comment into consideration the next time!

As an observation, AI (looks like ChatGPT to me) has done a much better job of diagnosis that the living responders here have managed so far.

Fascinating