• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

Review and Measurements of Emotiva XPS-1 Phono Amp

Xulonn

Major Contributor
Forum Donor
Joined
Jun 27, 2018
Messages
1,828
Likes
6,311
Location
Boquete, Chiriqui, Panama
Does the guy doing the review know anything about good sound? I know that YouTube is not hi-res audio, but that is about the worst YourTube audio quality I have ever heard.
 

invaderzim

Senior Member
Joined
Aug 30, 2018
Messages
320
Likes
563
Location
NorCal
Amir please do a review of the Emotiva BasX a-100 I would love to see how it is as a headphone amplifier and the old school engineering of it seems sexy as hell.

Although I am quite curious about how it would test, now that I've ordered one I'd rather be ignorant and happy :)
 

phoenixdogfan

Major Contributor
Forum Donor
Joined
Nov 6, 2018
Messages
3,297
Likes
5,075
Location
Nashville
Looks like they're using the same case they used for their XDA2 DAC's remote control.
 

TKO

Member
Joined
May 5, 2019
Messages
21
Likes
11
Although I am quite curious about how it would test, now that I've ordered one I'd rather be ignorant and happy :)

I would just really love to have an all in one unit for my living room for 2.1 speakers and a headphone out, with a well measuring dac/amp combo.
 

SIY

Grand Contributor
Technical Expert
Joined
Apr 6, 2018
Messages
10,386
Likes
24,749
Location
Alfred, NY
Power supply rejection of the LT1115 is well over 100 dB at the ripple frequencies. What you have there is a layout and grounding issue. Changing power supplies is a band-aid, the real fix is rerouting the grounds and power supply buses.

edit: If the BOM cost allowed it, the S/N on MM with an actual cartridge attached (!!!) would be improved with some FET buffers. The 1115 has a pretty high input current noise.
 
Last edited:

garbulky

Major Contributor
Joined
Feb 14, 2018
Messages
1,510
Likes
827
FWIW this is the phono unit found on the XSP-1 balanced stereo preamp albeit with a different power supply.
The phono specs from Emotiva's website:
  • Input Impedance:
    MM: 47k Ohms.
    MC: user selectable between 47, 100, 470, 1k Ohms.
  • Gain:
    MM: 40 dB.
    MC: 60 dB.
  • Frequency Response:
    MM: RIAA + / – 0.12 dB, 20 Hz – 20 kHz.
    MC: RIAA + / – 0.25 dB, 20 Hz – 20 kHz.
  • Signal to Noise Ratio (S/N; A weighted):
    MM: > 96 dB.
    MC: > 79 dB.
  • THD:
    MM: <0.0075 % @ 1 kHz; < 0.02%, 20 Hz – 20 kHz.
    MC: <0.006 % @ 1 kHz; < 0.05%, 20 Hz – 20 kHz.
https://emotiva.com/products/xps-1
 

WLVCA

Member
Joined
Jan 16, 2019
Messages
44
Likes
43
Location
Tucson, AZ
Amir please do a review of the Emotiva BasX a-100 I would love to see how it is as a headphone amplifier and the old school engineering of it seems sexy as hell.

I have two of the Emotiva A-100 amps.

The most current one has the built-in headphone out and that is the one I am using now exclusively for headphones with the jumpers installed. Works well for me - subjectively. :)

The older one is idle for the moment. I used speaker taps to drive balanced headphones.
 

JaySlice

Member
Joined
Mar 6, 2019
Messages
10
Likes
2
i absolutely love this website. I wonder how much weight is given to the subjective experience? I am one that leans to the measurements, but I do wonder when I should just enjoy the sound for what it emotionally offers
 
OP
amirm

amirm

Founder/Admin
Staff Member
CFO (Chief Fun Officer)
Joined
Feb 13, 2016
Messages
44,368
Likes
234,394
Location
Seattle Area
i absolutely love this website. I wonder how much weight is given to the subjective experience? I am one that leans to the measurements, but I do wonder when I should just enjoy the sound for what it emotionally offers
The emotion should come from quality and nature of the recording, not the electronics. :)
 

JaySlice

Member
Joined
Mar 6, 2019
Messages
10
Likes
2
unfortunately, my early life gave me a tunnel vision of music, but when I have good equipment I think I get touched more by music that on lesser equipment I would not. This situation is me just needing good equipment for me to break my emotional reserve, so I do agree with you of course, although for some thick skinned folks like me I need a little extra that the equipment can help convey the music
 

LTig

Master Contributor
Forum Donor
Joined
Feb 27, 2019
Messages
5,760
Likes
9,442
Location
Europe
Oh, I guess I forgot the dashboard for MC. :) Here it is:

View attachment 26317
I think it was not wise to use a single opamp for this much gain, even if it is an LT1115 where the data sheet shows a 40/60 dB switchable phono preamp. I would expect lower THD with two opamps sharing the gain load.
 

wynpalmer

Active Member
Technical Expert
Joined
Dec 14, 2018
Messages
175
Likes
214
Here's a shot of the interior:

View attachment 26320

It appears to implement the Linear Technology LT-1115 ultra low noise opamps. Clearly it also is running a couple of buck style converters to take the DC input and generate the +/- rails.

Datasheet link (with RIAA preamplifier application note design):

https://www.analog.com/media/en/technical-documentation/data-sheets/lt1115fa.pdf
It uses the LT1125 quad opamp in a 16 pin package, not the LT1115 which only comes in an eight pin package.
The LT1115 is quieter than a single 25 opamp, but I suspect that they are paralleling up the LT1125 devices.
 

wynpalmer

Active Member
Technical Expert
Joined
Dec 14, 2018
Messages
175
Likes
214
I think it was not wise to use a single opamp for this much gain, even if it is an LT1115 where the data sheet shows a 40/60 dB switchable phono preamp. I would expect lower THD with two opamps sharing the gain load.
It is NOT a single opamp.
There are four opamps available per channel. I wouldn't be at all surprised if they used two or three in parallel at the input with an additional one or perhaps two in the equalisation/buffering roles. The LT1125, which I believe the device is, has a fairly high level of distortion.
The LT1115 is really inadequate as a MM amp due to the input bias current noise- just like the AD797. A single LT1125 would be better.
I suspect that both the XPS-1 and the Cambridge Duo are actually pretty poor in a real world applications with MM cartridges and their c. 500mH inductances, despite their "impressive" specs that avoid much that is actually important.
I note that the XPS-1 seems to lack a reference level for the specified S/N ratio.
 
Last edited:

restorer-john

Grand Contributor
Joined
Mar 1, 2018
Messages
12,580
Likes
38,281
Location
Gold Coast, Queensland, Australia
It uses the LT1125 quad opamp in a 16 pin package, not the LT1115 which only comes in an eight pin package.
The LT1115 is quieter than a single 25 opamp, but I suspect that they are paralleling up the LT1125 devices.

It is NOT a single opamp.
There are four opamps available per channel.

Oh, OK. Right. You'd better tell Emotiva the LT-1115 they used and pictured doesn't exist.

1569362201709.png


Oh, and tell LT the package they specify in their datasheet (and the one pictured on the Emotiva board) they actually don't manufacture. You know, the one with one (1) opamp inside it and a whole lot of NC pins...

1569362324528.png


Maybe also tell Mouser the 136 they have in stock don't actually exist.

1569362484570.png


:facepalm:
 

wynpalmer

Active Member
Technical Expert
Joined
Dec 14, 2018
Messages
175
Likes
214
Ok.
Oh, OK. Right. You'd better tell Emotiva the LT-1115 they used and pictured doesn't exist.

View attachment 34395

Oh, and tell LT the package they specify in their datasheet (and the one pictured on the Emotiva board) they actually don't manufacture. You know, the one with one (1) opamp inside it and a whole lot of NC pins...

View attachment 34396

Maybe also tell Mouser the 136 they have in stock don't actually exist.

View attachment 34397

:facepalm:
Whoops :facepalm:
Ok. My apologies, I saw a picture that looked like it included the LT1125 as a 16 pin package, poor resolution to blame.
Mea culpa. In any case, unless they have some other device that is used instead of the LT1115 I believe that the opamp is unacceptable for the input stage of a MM stage due to the current noise of the input devices and will have problems maintaining RIAA accuracy at high MC gains.
Oh, and the distortion will be poor. (wait, that is all true...)
By the way thanks for the nice way you pointed out my idiocy. I should know better...
 
Last edited:

wynpalmer

Active Member
Technical Expert
Joined
Dec 14, 2018
Messages
175
Likes
214
fullpreamp20ohmssource250uvip.JPG


DIY MC phono stage rated 250uv rms input @1kHz, 69dB of gain, -0.2dBv full scale. Measured using an RME ADI-2 PRO FS.
Internal SMPS used for this with IEC connector in the same box as preamp. Spur at far right due to SMPS.
+/-15v internal rails. Generator coupled to amp through isolating balun. Balun output impedance in parallel with input load R on board provides 20 ohm equivalent source R, to mimic the APX source resistance.
Effective 40dB attenuation from generator- hence the low hum and small contribution of generator to the noise floor.

fullpreamp20kHz harmonicdistortion at+19dBvop.JPG


Same phono stage. 20kHz at +19dBv (just under the clip point) output to illustrate TIM performance. RIAA compliance simulated to be nominally <+/- 10mdB. Measured crudely by hand to be +/-50mdB with off the shelf components from Mouser. DC servo'd offsets.
FFT not the same settings as Amir's APX so not directly comparable. Overload margin c. 28dB 20Hz-20kHz.
Balun removed so generator at full power/noise output into preamp input. Loopback test not performed to establish the generator baseline as it seemed good enough as it was.
I'd love to have someone measure it properly...
A MM version exists with 2 paralleled OPA1642s at the input for operation with actual MM cartridges with 500mH typ. inductances and 20dB less gain.
No measurements as yet.
 
Last edited:

wynpalmer

Active Member
Technical Expert
Joined
Dec 14, 2018
Messages
175
Likes
214
Maybe send it to @SIY?

Very impressive low level performance at 0.25mV. I don't think I've seen anything quite like that.
As I implied, the FFT bin widths are probably not identical to his. I'm using 128k bins with a 192K sample rate and a rectangular window.
It also has a slightly unusual architecture (not from a data sheet) which can result in a near perfect theoretical RIAA compliance, including the DC servo.
The characteristic below is with simulated "real" components from a Mouser BOM.
It's not clear to me that this is something that Amir does, after all this is a DIY not a commercial product, and I tried e-mailing to check, with no response as yet.
It has been built, and is in use by several "audiophiles" in parallel with or replacing multi $K phono stages, and they apply the usual terms (imaging, depth, spaciousness, etc. etc.) to it so it seems it passes the "does it sound good" test in addition to the "does it seem to measure well" test.
Of course, I don't really know if it measures well. However, it has, for what it's worth, won a couple of phono stage shoot outs with the local audio society against a few pretty notable opponents.
I actually find the whole thing pretty amusing as the design was done essentially as a "dare".
The approach does not translate immediately to MM- a number of components have to change, however the relationships are mathematically well defined and the trick is in finding sufficiently accurate Mouser in stock parts.



RIAA.PNG
 
Last edited:
Top Bottom