Simplified MrEvil / PMI Capacitance Multiplier

Have you measured any of your components to verify the two rails have the same (withing reason) values? Does it exhibit this when unloaded or with equal resistive loads?

If there are significant differences in component values, especially relating to the rail filtering parts then one rail might rise at a different rate than the opposing rail.

If the load draws more heavily from one rail during startup then the PSU output may rise at a lower rate on the more heavily loaded rail.

My experience has been very symmetrical behavior between the rails. I have not had any issue with the PSU causing a turn-on thump.
 
I'd be very surprised to find a production transformer produce a 0.5 volt ac difference on two secondary windings, particularly if it's a torroid - that voltage difference might be caused by a 'dud' diode, for example.

If you do indeed have a difference in the ac secondary voltages, and they have a common connection in the centre (centre tapped), you might consider adding an extra series diode, for example, to absorb the 0.5 volt ac, or insert a resistor, etc or just replace the transformer

One thing that may produce a 'turn on/off thump' is possibly where you connect the 0 volt point on the supply - see if you have connected the centre 0 volt on the output of the C-Multiplier to the amplifier, speaker (headphs), etc including the one to the chassis earth - connecting it otherwise can produce all sorts of problems, particularly with the F5 headamp

If all this is okay, just do as Andrew suggested and attach a speaker protection cct on the output of the headamp- it'll delay the headph connection on startup and quickly disconnect it when turnoff - most of them operate with a dc voltage over 1 volt that isn't too kind to the headphones, but it will stop that 'thump' okay.

if you want to use a 'proper' headphone protection cct, there's a cct and pcb available on the AMB website (Epsilon 12 project) and a Cavalli Audio variation on their SOHA II project pages
 
Founder of XSA-Labs
Joined 2012
Paid Member
Hi JasonKeutemann,

I am using this circuit now and it seems that it takes a 3.7v drop in order to sustain 3.7amp draw with low ripple and I have 22mF bulk capacitance.

The setting was confirmed with oscilloscope - before this I set it at 2.5v drop thinking that was sufficient but about 240mV of sawtooth was leaking through.

The more I look at the schematic, the more it looks like a pass through voltage regulator and not a capacitance multiplier. The usual dropout of a 78xx regulator is about 3.5v if IRC.

Which cap is being “multiplied” here? The 2.2uF C1? Usually the cap being multiplied is circa 220uF.

Thanks,
X

446038d1414720404-simplified-mrevil-pmi-capacitance-multiplier-capacitance-multiplier-schematic-png
 
Member
Joined 2011
Paid Member
The general rule is: Diffamp_IN_+ == Diffamp_IN_-

[R1/(R1+R2+R5)] * Vin

is (by negative feedback) made equal to

[R3/(R3+VR1+R4)] * Vout

Notice that a 10% increase of Vin, produces a 10% increase of Vout.

It's not a "Voltage Regulator" because Vout is not independent of Vin.

Notice that it is a 2 stage amplifier and both stages have considerable amounts of voltage gain, but there is no explicit frequency compensation circuitry to guarantee stability. Minus fifteen points to Slytherin.