Meanwell EPP-200-27 whining/hissing

Yes, cld hear the PSU, so add a dummy load resister to drain 10% full-load?

Is there any real concern if operate the whinning PSU very low load for long term, likely to cause failure?

No, I don't think it will be a problem for the PSU. However, there might problems associated with the response of the PSU to rapid changes in current demand. Voltage dips and overshoots that are not within specification. I can't quantify them; best measured if possible.
 

tomchr

Member
Paid Member
2009-02-11 12:58 am
Calgary
www.neurochrome.com
Just a note: I'm listening to a pair of RPS-400-27 powering a pair of Modulus-686 power amps. I hear no hissing from the supplies. Not even when they provide no load (i.e. I unplug the amp modules). Each MOD686 channel draws about 300-350 mA in idle current.

The EPP and RPS series are identical from what I can tell. The EPP is $5 less and the commercial grade. The RPS is the medical grade.

Tom
 
SMPSs with regulation typically operate on the principle of energy transfer in small (variable) lumps but at high frequency. With the usual PWM, a minimum lump exist. If the load cannot consume the minimum lump times the cycle frequency, a problem arise and low energy operation has to be foreseen.
In order to regulate the SMPS in low energy mode, the designer has to choose for another modulation principle, often pulse frequency or pulse skipping schemes. Varying the number of pulses per second, less energy can then be transferred.
But, entering such low energy modulation schemes may generate an audible frequency or excite an audible resonance. You may hear it, but it is not certain you will hear it. This low energy regulation is an exceptional operation mode of the SMPS with reduced performance.
At low loading, the SMPSs may run into discontinuous operation or light-mode which again reduce performance.

As an audio amplifier (not class A) has a current consumption ranging from the quiescent current until full power current, the dynamic consumption varies enormously. Not what the SMPSs are initially intended for.
As a minimum, you should ensure that the quiescent current is above the low energy range. Then, the SMPS performance is better.
Tom has 300-350mA on his new power monster - fine.

The linear power regulator is conceived to supply any current, from close to none and until its design maximum. This is an important advantage of the linear regulator for use with audio power amplifiers. It has other important drawbacks.

SMPSs are very useful circuits and their modulation principles become steadily more advanced. In concept, they are just not the ideal solution for audio power supplies.
 
There are 2 possible conditions for a SMPS to become audibly noisy at low load:

- Working in pulse skip mode: Minimum pulse width is too long for current load (as explained in previous post). Some pulses are skipped. This usually manifests as a soft "hissing" in the transformer. Moderately high noise floor in the operational amplifiers and comparators in use makes that noise "spread-spectrum" (no strong content a particular frequency). As load is increased the hissing vanishes, or is replaced or modulated by 50/60Hz humm/buzz.

- Low-load instability: The control loop runs out of phase margin and/or gain at low load. This manifests itself as a "whinning" whose pitch varies with load. It disappears past a certain load level. This condition is not dangerous on itself, just unpleasant sounding, but it may indicate the presence of another problem, like a cold solder joint or a dry capacitor. There is also the case where the "whinning" increases as load is increased, this is a serious warning sign, the SMPS needs to be serviced before further use.