TECHNICAL
ULTRAFIDE EXPLAINED
As an introduction to any discussion on audio amplifier measurements (as a means of comparison between different models/output topologies), it’s very important that they are not taken as “complete truth”, for several reasons.
The test signals used to do most of the measurements are nothing like the signals that the amplifier was intended to be used for. In the case of music/speech reproduction, these are dynamic signals with typical Crest Factors of 4 or above (CF 4 equates to an average of 1/8th power). Sometimes heavily compressed or synthesized music may even be as low as CF 2.8 (1/4er power) but most material is actually much higher than CF 4.
The measurements are usually made with resistive loads which give very different results to the reactive, and sometimes negative impedance loads, that a loudspeaker presents (especially those with complex passive crossover networks). Different makes/models of loudspeakers have a huge variation in the loads they present, and an amplifier that sounds very good on a particular loudspeaker may not be so good on another.
There are also many figures stated on data sheets that either have no relevance in assessing audio quality or are very ambiguous. The philosophy at MC² is to keep specifications brief/relevant and try not to baffle users with technical jargon. We take pride in designing some of the best sounding amplifiers available in the world, which is verified time after time in blind listening tests. This is the best way of qualifying sound quality.
Every person’s hearing ‘profile’ is unique, which is why many professionals have different opinions about the merits of one ‘set up’ over another. So, the conclusion is, as always, to engage in listening tests in the environment that the equipment is likely to be used in.
Slew rate is a limitation that applies to opp amps and linear Class A or Class AB amplifiers where the rate of change at the outputs is finite and potentially causes distortion overtones or other nonlinearities with very high amplitude, high frequency signals.
The most important factor affecting this is that the amplifier must be able to handle all the signals that are thrown at it. In other words, there must be enough filtering to ensure that the amplifier’s ‘Slew Limit’ is not exceeded.
In common with many manufacturers, we do not quote a slew rate since it is not helpful in comparing different designs. There are so many ways of determining slew rate that a direct comparison of quoted figures is impossible. Mostly any figures quoted relate to the output amplifier and ignore the effects of preceding stages, which are often much more relevant.
Our unique linear topology (Current Driven Sliding Opp-Amp) ensures that the output stage slew rate is largely dependent upon the op-amp and RFI filtering which we use to drive it. Unfiltered, an MC² S1400 current driven output stage has a slew rate of well over 100V/microsecond but the amplifier bandwidth is restricted to –3dB at 100kHz.
Our latest Ultra Sigma amplifier topology is effectively devoid of slew rate limitations so it’s not relevant or possible to quote a figure for it. That’s why we quote a rise time. It takes 10uS for signal to change from one voltage to another be it 1V to 2V, 1V to 20V or -40 to +40V. Assuming the preceding equipment has no slew rate limitations (i.e. produces a perfect voltage step change) then it’s always 10uS. That’s because the Ultra Sigma design is the filter (set at 40kHz) and cannot ‘Slew Limit’ like linear Class A, Class AB designs.
Many manufacturers calculate the damping factor from the components in the output chain, and do not consider the various conditions which can influence the figure. We quote a damping factor of >400 (8R, 1KHz) for our live sound products, and this is measured at the speaker terminals, so external to the amplifier which we see as giving the correct figure. We regard anything above 400 to be inconsequential because other external factors will have more influence on this figure, such as the speaker cable and the connectors between the cable and the speaker cabinets.
Over the years we have noticed some of our competitors have various ways of rating the damping factor of an amplifier, using irrelevant testing means to sell a product by putting this onto a datasheet. It’s almost like some manufacturers feel a datasheet is a sales tool above an actual physical demonstration, a notion we completely disagree with. We are more focused on real world ratings, and ultimately how these translate to the end user in the sound presented to them. We have seen spec sheets with >5000 quoted and it would appear that this is because the measurement has been taken internally in the amplifier, so by the time it gets to the loudspeaker it has passed through various components that reduce the rating dramatically.
To clarify the difference between class A and class AB amplifiers, they both operate in the same manner, the difference being that in the class A design the devices that are not driving current into the load do not completely switch off, whereas in an A-B design they
(almost) do. Some A-B designs have a very abrupt switching point, and some are gradual. MC² use a gradual approach which tends to have much lower distortion but does generate more heat.
Tube (valve) amplifiers are a different matter, these are high impedance devices which require transformers in order to drive low impedance loudspeakers. They tend to have quite high distortion figures, but it is usually 2nd harmonic which is quite melodic. They also handle transients in a different manner to transistors, and also introduce dynamic high frequency filtering. This sounds particularly good when used for musical instrument amplification such as guitar amps.
Class D amplifiers operate by directly switching power, on and off, at extremely high speed & frequencies using advanced MOSFET’s (or other high-speed devices). A modulator is used to vary the size or density of the switch pulses to achieve an average output power that represents the audio; and a low pass filter (typically a coil and capacitor) is used to remove the high frequencies generated by the modulation, leaving only the audio signal to feed the speaker.
Class D topologies inherently eliminates the midpoint cross over distortion associated with Class AB designs so that linear operation, similar to Class A, can be achieved; and yet the power switching devices are only ever operated in a linear mode for a fraction of the time making the amplifier much more efficient (reducing the need for massive heatsinks).
In early Class D designs the maximum switching speeds possible (then) could cause resonances or limit the audio performance at high frequencies, but modern designs from MC² (ULTRA SIGMA) using ultra-fast switching components (operating hundreds of times faster than audio frequencies) totally eliminate these drawbacks and often extend bandwidth beyond what is possible with linear designs.
When selecting amplification for a loudspeaker there are a few schools of thought when it comes to power requirements. Watts are often quoted in RMS or AES and relate to amplification requirements as Program Power. Headroom is the power available in the amplifier above the power rating of the loudspeaker. We have always advised that the amplifier should be approximately 1.5 to 2 times the power of the loudspeaker/driver rating to ensure good dynamics and a safe operating level for the amplifier.
Headroom is important and must be looked at carefully on different drivers/loudspeakers for both safety and sonic purposes. When listening to high resolution audio with high dynamics the amplifier should be able to successfully recreate the transient peaks with little or no effort, to ensure that there is no limiting/clipping of the signal. All the listening tests we have done with Studio monitors, PA, and Hi-Fi speakers has shown that there is a very discernible difference between the sound quality when an amplifier has lots of headroom, especially in passages with high dynamics. We would consider headroom as extremely important when specifying a system, particularly when using source material with a high crest factor.
A bullet proof, classic linear design was chosen with a low noise custom wound 500VA rms transformer, capable of delivering enough power for an amplifier 4 times the size! This ensures that the U500DC is never stressed and has next to no losses. In essence the stereo amplifier performs as if it were two separate mono blocks; in fact, from here on in it is…. Four 35A 600V rectifiers are usedbut each are only working at 80V, but able to deliver peak power with ease anywhere from 5 to 5,000 Watts! This is then delivered into a huge 27,000uF bulk storage capacitor bank… and that’s just for the left channel! The whole process is then repeated completely independently for the right channel ensuring no interaction between the two is possible and confirming ultimate L R separation and uncompressed powerful bass propagation.
Special attention to detail has been made with the component placement ensuring no 50Hz noise can be picked up from the power supply. This includes a physical separation barrier between amplifier/power supply but goes beyond that with use of separate ground return current paths and differential interstage coupling, ensuring total silence between music tracks.
A completely brand new push/pull output stage has been developed specifically for this product, and refined over the last two years through multiple listening tests. It uses supersonic switching techniques, 20 times higher than human hearing, to reduce resistive losses 100 fold compared to older class AB designs and totally eliminates crossover distortion (like a Class A amplifier). Overall control of the push pull power stage is achieved by a single feedback loop; this “ultimate in simplicity” is one of the mechanisms to the open, transparent, and expansive sound the U500DC delivers.
For the last 50 years or more it has been commonplace to have capacitors in between the separate stages of any audio product. This is because they eliminate small errors that are typically present which can cause clicks or pops or even damage speakers. But no matter how good the capacitor quality they can not improve the sound, they can only minimize how much they degrade it (skewing the phase of bass content in transients). To overcome this there are no interstage capacitors inside the U500DC; it only uses direct (copper tracks or wires) DC coupling, and it eliminates any errors that may be presented on its inputs using a single overall outer loop DC servo. This ensures total alignment of transients, amazing phase coherence across the entire audio bandwidth, and strong focused low-end (bass) response.
The power amplifier is fed directly from a Class A driver stage that collects differential signals from the input connector PCB and adds voltage gain and soft clipping protection. For those wanting to maximise power delivery (for bigger systems) the gain can be increased by 3dB and the soft clipping circuits removed using an internal link. Additionally, for those wanting a more vintage type of sound delivery a “TIMBRE” switch on the rear changes the input circuits from “Pure” to “VINTAGE” to operate more like traditional class AB amplifiers. This is a very subtle change, but it can be heard. Our in depth blind listening tests in development of the U500DC gave a 50/50 split as to what was preferred!
WHAT PEOPLE ARE SAYING
GET IN TOUCH
If you have any questions or require information about any of our products, please don’t hesitate to get in touch.
