You’re not interested in the peak voltage anyway - that could be 1000 volts with a very narrow spike, and still have less overall energy (spikes not recommended though!).
It’s the area under the curve that counts for increasing stored energy in the capacitors - the regulators take care of the final voltage anyway so no under-volting is going on 
Multiwave does NOT reduce the peak voltage.
Instead multiwave adds in a third order harmonic to widen the waveform at the apogee and perigee, see below:
Hi,
Not thinking of some very short peak, but the peak of a somewhat round waveform.
As volt RMS out of the powerplant remain the same with multiwave turned on and off, and we know that mutiwave provide a wider waveform, this show that the peak is lower.
Also, many power amplifiers do not have what we typically consider to be voltage regulators.
Hi Paul,
You are obviously the person who know all details about this.
But if charge time longer i.e. the wave get “fatter”, and peak voltage remain constant at 310V for a 220 V unit, shouldn’t then volt RMS increase?
Reduction of peak of the wave further confirmed with my measurements, both with a multimeter that can measure both peak and true RMS and oscilloscope. Also by the fact that power usage on my amplifier is reduced when turning on multiwave.
Would be interesting to hear what your measurements conclude.
That said, there is of course the option of turning on multiwave and increase output voltage of the Powerplant (which is measured in RMS or something RMS like) to compensate for the drop in peak voltage, to both get the fatter multiwave wave form and to maintain peak voltage.
No intending to criticize your product, I really like the powerplan and it have clear benefit in my system. The title of this tread chosen for the intention to get the discussions going. Maybe these discussions even can help to make your products even better?
Or maybe my thinking and measurements are wrong altogether…
Much appreciate your videos and and products! Hope i get an opportunity to visit PS audio at some time.
Hi,
It really looks to me as if your illustration show that peak has been reduced, doesn’t it?
Try plot the second graph, but exclude the third order harmonic and see where you get the peak.
Hi,
Not thinking of some very short peak, but the peak of a somewhat round waveform, like sine or multiwave.
As volt RMS out of the powerplant remain the same with multiwave turned on and off, and we know that mutiwave provide a wider waveform, this show that the peak is lower.
Also, many power amplifiers do not have what we typically consider to be voltage regulators.
As I said previously, multiwave does not reduce the peak voltage but instead adds in a third order harmonic to widen the apogee and perigee of the waveform, see below:
This is then smoothed, again still keeping the apogee and perigee at 120volts they are now simply wider.
To prove this - In your PSA regenerator change the ‘MW Strength’ to 6 in the setup. Go to the oscilloscope and view the output waveform. Watch the output waveform in the oscilloscope when you change from ‘sine’ to ‘multi’ waveform. You will notice the apogee and perigee is wider when in the ‘multi’ waveform than when in the ‘sine’ waveform. You will also notice that the aptitude and wavelength is unchanged when changing from ‘sine’ to ‘multi’.
I’m not an electrical engineer…but this is how I understand it - someone please correct me if I’m wrong.
(Focus on the top drawing below) After the current travels through the full bridge rectifier(D1,D2,D5,D6) the current will now be flowing in one direction but the voltage will pulse on/off and needs to be smoothed. Smoothing will occur with a capacitor(C1). The wider the apogee and perigee of the original AC waveform the longer the capacitor(C1) has to charge(remember it still has the same discharge rate as before). This is a beneficial design I assure you.
With the capacitor having a longer charge time to due to the wider apogee and perigee of the original AC waveform this results in lower DC ripple effect(Lower DC ripple means more linear power) after the voltage exits the capacitor but before it reaches the voltage regulator/device rail.
Now, many power supplies in expensive amps will have an inductor and another capacitor after C1 resulting in even more linear DC power and you can continue this design…adding another inductor and another capacitor. My guess is this is why you see amps with banks and banks of capacitors they probably have extremely linear DC power even without an AC regenerator.
Hope this makes more sense.
Ryan
Hi,
Thanks for comments.
Well, actually, the peak for 120V is around 170V. The 120 V we usually talk about is a RMS measurement.
As you spoke about the scope on the powerplant, i just checked, and as you wrote, the wave is clearly “fatter” but its also easy to see that the peak is lower.
So so far, everything I have seen, tested or measured point in the direction that the powerplant actually reduces the peak voltage which is feed into components.
This was discussed 6 years ago.
The peak voltage is reduced.
From that thread:
“I measured the B+ on my mono block tube power amps.
With Multiwave off, I measure 424 volts on each amp.
With Multiwave on, I measure 406 volts on each amp.
Is what I am seeing normal? And if so, please explain the electronics involved.
An explanation from Bob Stadtherr:
You are correct that the peak voltage is about 5% lower in order to maintain the same RMS value. However the softer wave shape extends the amount of time that the rectifiers conduct and charge the filter capacitors. This means the peak current from the transformer is lower, and thus the IR drops in the tranformer are lower. This offsets the lower peak voltage, but only under load.
The wave shape was chosen so that in typical equipment (with a constant DC draw) the lower peak voltage and the improved transformer efficiency just cancel out and the dc is about the same.
In a power amplifier, usually we see a slightly lower DC voltage at idle, and about the same with typical loads. When the demand on the amp is high, we usually see much less sagging of the DC voltages than with a sine wave input.
Another benefit of the extended charging time is lower ripple current in the filter capacitors. This reduces the internal heating and extends the life of the parts.”
Ah the reduction in R loses in Transformer and Rectifiers is a good point too, but I still think the main advantage is not about the voltage - speakers require energy to move them, voltage without sufficient energy store is not enough, and with the capacitors storing more energy between peaks of the AC waveform (and I refer you to my previous comment regarding the area under the curve) then more short term high current demand can be satisfied.
Or - use whichever sounds best - trying to estimate what is best from first principles is risky, as there are always more variables than we can account for in our theoretical ruminations
Very interesting! ,So this confirms that mutiwave indeed undervolt as I suspected and measured.
I now did some measurements on my amplifier and can conclude that it loose output power by turning on multiwave (clipping occur at lower power).
I would want to see or do some measurements on the argument with R-loss in transformers before I completely believe in that.
I think there may also be a marketing and communication challenge that the undervolting address. The benefit of the multiwave shape is easily to intuitively understand, but I think many clients would be nervous to see measured output voltage (which is measured in RMS) increase 10% when turning on multiwave, being affraid that they will fry their equipment.
In my case, I hear no difference with multiwave on or off (at normal listening levels), but can do (the mentioned) measurement that the amplifier loose some power. A “benefit” i see with multiwave is that power consumption of the power amp is reduced from 300 watt to 270 watt as living in the tropics, it already generate too much heat that need to be air-conditioned away…
So now when we have concluded that multiwave undervolt, it would be interesting to hear what listening experiences people have by running multiwave at standard undervolt VS compensating the undervolt by increasing output voltage with 10%.
I will do some tests on that.
I do not believe for one minute that multiwave reduces the voltage enough to affect the clipping point of an amplifier operating on it! I bet anyone measuring the DC rails off the amp will see there is no change at all!
Exactly. The rms voltage remains the same. It would be remiss of us to alter the power supply voltage of connecting equipment. We are only interested in power quality.
Interesting…
Paul,
So, I’m confused now. I provided what I thought was going on in my explanation above…but it sounds like I was incorrect.
Is aronake/st50maint partly correct that the peak of the voltage curve is reduced?..BUT, the RMS(root mean square - area under the curve) stays the same?
Am I understanding that correctly?
“Is aronake/st50maint partly correct that the peak of the voltage curve is reduced?”
What I posted was a quote from Bob Stadtherr.
He is the gentleman that designed the Power Plants.
It may be a fact but it is not relevant to the performance of the Power Plant or the equipment that is connected to it. The RMS power stays the same and ripple is reduced in the power supplies of the connected equipment the DC voltage out of those power supplies is not changed.
That is not necessarily true in all cases. As I stated earlier:
“With Multiwave off, I measure 424 volts on each amp. With Multiwave on, I measure 406 volts on each amp.”
Where are you measuring these voltages and what are you using to do so.
How much do you want to bet? 
