Latest Thread Images

Competition Closed...but what is in the Retro – Stereo 50?
Apr 14, 2015 at 9:39 PM Post #257 of 517
Just reviewed the digital board picture - if I am not wrong, the DAC portion of the Retro only has a single PCM1793 chip? Just wondering what effect does this have compared to say the micro which has 2 PCM1793? 
 
Apr 14, 2015 at 11:49 PM Post #258 of 517
I bought my Retro a week ago. It has two levels of 3D and bass that the micro does not have. The micro is designed as a dual mono DAC with two PCM1793 chips and the Retro is designed as a stereo DAC with one chip. Even the Oppo HA-1 that I have uses just one DAC chip. With two  PCM1793 chips you may get better S/N ratio and channel separation providing that the implementation is the same. Tube amps do not always have the very best S/N ratio as compared to SS counterparts. However the Retro DAC section has a Dynamic range of 113 vs 117 of the micro. Not that far apart in that regard.
 
Apr 15, 2015 at 9:13 AM Post #259 of 517
John57: very interested to hear your impressions, setup, possible comparisons! Let us know what you think in the coming days :)
 
Apr 15, 2015 at 4:09 PM Post #260 of 517
Competition Closed...but I was still wondering how many components were on the boards.

My guess was 631, which was wrong.
So I disassembled my unit and counted again...



...still think it is 631 :smile:
 
Apr 22, 2015 at 10:07 PM Post #261 of 517
heywaj10 said:
John57: very interested to hear your impressions, setup, possible comparisons! Let us know what you think in the coming days :)
Click to expand...

Yes, no problem. I had a problem with the right channel crackling and IFI send me a new tube set which fix the problem. It now sounds very very good! I have been experimenting with high efficiency near field speakers since I wanted a choice. I will say more this weekend.
 
Apr 23, 2015 at 2:29 AM Post #262 of 517
john57 said:
Yes, no problem. I had a problem with the right channel crackling and IFI send me a new tube set which fix the problem. It now sounds very very good! I have been experimenting with high efficiency near field speakers since I wanted a choice. I will say more this weekend.
Click to expand...


Mine is in running in 24/7 and the sound is opening up. My wife, daughter and I like it very much :) its sound so sweet (thx the tubes) and powerful given its tiny size! My daughter loves it's incredible performance when she listens to Korean pop and Canton pop! I will provide my feedback as soon as I take the retro back from these ladies :)
 
Apr 25, 2015 at 8:42 AM Post #263 of 517

Retro Stereo 50

There are Watts and there are Watts
 
 

 
 
 

Question: How does the Stereo 50 achieve 25W+25W?

The short answer is that the Stereo 50 is measured using a music signal into real speakers (the LS3.5 to be precise) as opposed to a sine wave test into a resistor.
 
This has been outlined in the technical specifications area of the Retro Stereo 50 webpage:
http://ifi-audio.com/portfolio-view/retro-stereo-50/
As this uses a music signal, this we feel is relevant in the real world and differences tend to be large. For the technical reason behind this, please read on.
 
 
 

The technical backdrop

Modern audio testing is carried out generally using steady sine waves and test sets such as the Audio Precision System Two.
 
 
 

 
 
These test systems and the methodology came long after the advent of transistor amplifiers. So they are geared to testing such amplifiers and they are fair and equitable to such amplifiers.
Some fifteen years ago, Peter Van Willenswaard1 investigated connected issues and found that while testing transistor amplifiers with resistive load and steady state sine waves and driving speakers with music led to results for maximum power and output from the speakers that are closely matched between the two states. Yet when testing tube amplifiers the results from the two different tests were manifestly different2.
 
 
"I hooked up my oscilloscope to one of the speaker outputs of my 300B amp and observed the screen while Touch (a specific CD – Ed) was playing.

Holy Moses. I saw something like 30V peaks from an amp that, when driven with sine waves and loaded with an 8 ohm resistor, never showed more than a 14V peak—more than twice the voltage technically supposed possible. You'd need a 50W transistor amp to realize the same peaks my 9W 300B launched without wincing at my speakers."
 
 
In the letters resulting from the publication a reader recounted a case where he replaced a 15 Watt rated Leak ‘Point One’ EL84 Tube amplifier (likely the Leak Stereo 20) with a Solid-State amplifier rated 75 Watt per channel and found that at levels that did not bother the Tube amplifier the least, the Solid-State Amplifier would run out of steam and shut down3!
Most extreme perhaps was the finding that a tube amplifier capable of no more than four watts continuous output power in sine wave testing produced a subjectively undistorted output from the speakers used in the test equal to that of a 25 watt transistor amplifier. Yes, on the same speakers a 4W tube Amplifier matched a 25W transistor amplifier for actual playback levels4.  
This seems to break all laws of physics, though in reality this is not so. There are two key underlying reasons for this behavior of Tube Amplifiers.
 
 

1) Speakers are not Resistors
Modern Loudspeakers are designed to be driven by low output impedances, in effect by voltage (with current being drawn by the speaker as required). A typical 2-Way speaker such as the Audio Note Model-E as used by Van Willenswaard has an impedance magnitude that is very variable with frequency and often has very high impedance in the bass region, where normally much of the power demand of music happens.
 
Here the impedance curve for the AN-E (from Stereophile5) as used by Van Willenswaard:
 

 
 
C:\Users\Owen\AppData\Local\Temp\msohtmlclip1\01\clip_image003.png
We observe that while the impedance minimum is 4 Ohm at around 220Hz, the mean impedance of the speaker is much higher. A crude analysis in excel suggests a geometric mean of 8.5 Ohm and an average of 9.5 Ohm Impedance.
A transistor amplifier usually clips (read: distorts heavily) when its output approaches the power supply voltage. It can simply no longer increase the output voltage.
 
 

By comparison Tube Amplifiers invariably are current limited, meaning if the load impedance (speaker) is higher than the nominal match, they can drive much higher voltages into the speaker load because of the higher average impedance of the speaker.
As the speakers’ sound output is actually directly proportionate to the applied VOLTAGE (and not the power drawn by the speaker) we get more sound pressure as would normally be possible using a transistor amplifier of equal output voltage limit tested using sine waves into a resistor load.
This is however only half the story and still does not account for in effect turning a 4 Watt Tube Amplifier into the equal of an over 6x more powerful transistor Amplifier. Another ‘hidden’ mechanism must be accounted for.
 
C:\Users\Owen\AppData\Local\Temp\msohtmlclip1\01\clip_image004.png

2) Music is not a sine wave
If one looks at the waveform of music we find that that in fact the average levels are quite low, while peaks close to the maximum occupy only a small fraction of the time:
 
 
 

 
 
So, whereas a sine wave has a constant demand of power, music has a low average power requirement, but a high peak requirement.
Both classic Tube Amplifiers as well as modern ones (like the Retro Stereo 50) often employ a method to control output stage quiescent current called variously self-bias or cathode bias. This method has a great advantage in that it makes the resulting amplifier self-adjusting, with no need to perform maintenance nor regularly adjust the tube bias.
 
For a Push-Pull Amplifier (such as the Retro Stereo 50) this means that if the Amplifier output stage is driven out of the Class A operation the bias voltage will change. In effect, the bias will increase thus turning the output tube off progressively. Another mechanism is the onset of so-called ‘grid current’ which creates an additional negative grid bias; again, in effect turning the output tube more off, when it should be turned more on.
 
There is timing linked to this. In the Retro Stereo 50 for example, the cathode circuit will delay this self-adjustment by around 0.2 seconds (the grid circuit even longer), long enough to deal with largely all musical peaks, but so short that with steady state sine waves the adjustment seems instant to the experimenter or audio test set. The result, using sine waves, the amplifier seems to ‘saturate’ at fairly low power, turning up the input signal produces more distortion but no more output. In effect the amplifier power with steady sine waves is self-limiting, yet the limits do not apply to short high-power events.
 
 

This particular mechanism was well known in the ‘Golden years’ of tube audio and widely discussed6. It was also widely known that optimising an Amplifier output stage for sine wave testing would increase distortion with music or speech signals7. Equally known and discussed was the fact that sine wave measurements were not representative of the maximum power output available from a Tube Amplifier playing music into speakers8:
 
 
"To provide an agreed standard of measurement, all amplifiers are tested with a pure sine wave input signal. This never occurs in speech or music except, remotely, in the case of the flute. From the point of view of power handling capacity a pure sine wave is one of the most difficult waveforms for an amplifier to reproduce and invariably gives an inferior result. It can, therefore, be taken that an amplifier rated at 30 watts continuous sine wave output will give an instantaneous peak output of speech and music of at least double this figure."

 

Real world audibility over scientific measurements

So, there you have it. Testing Tube Amplifiers with sine waves and resistor loads will give a power reading that is artificially low and does not fairly reflect what the amplifier is capable of. This gave rise to such audiophile fairy tales that stated that Tube Watts are “Louder” or “Bigger”, they are not. They simply have been measured in an inappropriate way.
 
Thus we specify the ‘Output Power’ based upon testing with music and speakers attached to the amplifier. We connect an Oscilloscope in parallel with the speaker and observe the waveform. We keep turning up the level until peaks just barely flatten out (onset of true clipping). Tested like this most Tube Amplifiers show much larger signal swings than with sine waves into resistors. Transistor Amplifiers on the other hand show no such effects.
 
In other words, sine wave testing does allow us to make meaningful determinations about how much signal the Amplifier will drive into the speaker for transistor amplifiers, but it severely understates what a Tube Amplifier can do. Our rating is fair insofar in real life, a 25W solid-state amplifier will play equally loud on the same speakers as the Retro Stereo 50.
 
We hope we have shed more light on this and explained why we have done what we have done.

_
 
References:
1) Tubes Do Something Special, by Peter van Willenswaard • Stereophile Sep 2000
2) ibid
3) Letter to the Editor, by Bob McIntyre • Stereophile Nov 2000
4) Tubes Do Something Special Followup by Peter van Willenswaard • Stereophile June 2001
5) Audio Note AN-E Lexus Signature loudspeaker Review, by Art Dudley • Stereophile May 2006
6) Circuits for Audio Amplifier, Mullard Limited • Second edition August 1960
7) ibid
8) Avantic DL7-35 Amplifier Manual, Beam Echo Ltd.; Witham, Essex • ca. 1960 
 
For more information:       ifi-audio.com/products/retro/
C:\Users\Owen\AppData\Local\Temp\msohtmlclip1\01\clip_image005.jpg
C:\Users\Owen\AppData\Local\Temp\msohtmlclip1\01\clip_image006.jpg

For a PDF copy of this, keep a look out on our website. Thanks!
 
iFi audio Stay updated on iFi audio at their sponsor profile on Head-Fi.
 
https://www.facebook.com/people/IFi-audio/61558986775162/ https://twitter.com/ifiaudio https://www.instagram.com/ifiaudio/ https://ifi-audio.com/ https://www.youtube.com/@iFiaudiochannel comms@ifi-audio.com
Apr 28, 2015 at 4:49 PM Post #267 of 517
Anyone who has bought a Retro have a further review? I just ordered my Retro system from "Night and Day Audio" in North Carolina! The owner Shane is an amazing guy, I cant wait till next week to open the box!
 
You must log in or register to reply here.

Users who are viewing this thread

Total: 2 (members: 0, guests: 2)
Back
Top