|
Post by tupisac on Mar 18, 2017 10:50:04 GMT
Hello everybody, New guy here. I'm looking for an advice. I've been using emu-404 for like 10 years but it started slowly dying (power up issues, another story), just as I finally bought me a pair of HD-600. So i acquired the Objective2 kit. Specs looked decent, blog and story behind it were really engaging, I thoroughly enjoyed my first soldering, amp worked on first try. But... I'm not satisfied with the end result. The difference is subtle, but clearly audible. Emu seems slightly more detailed and more precise in instrument placement. Bass seems more pleasant and tighter, more controlled. Contrary to what I've expected from the lower output impedance of O2. I'm using EMU RCA output, so there is the same DAC playing. Now, I'm not an audiophile. I cannot hear above 14 kHz. I cannot hear the difference between 24bit flac and nicely encoded 192kbps mp3. I do enjoy music and I sometimes fire up ableton as a hobby, but that was it, I was perfectly happy with my gear. Now I feel like I've entered this dark abyss, where there is no happiness and my wallet is always empty . Now I just cannot buy me a shiit stack and enjoy my music anymore. Now I lurk on the forums and read and I just get more and more confused. So, long story short, I'm looking for a decent amp project that can compare to EMU and is diy noob friendly. My budget is around 200-300 euro. Extensive googling led me to this forum, to the Kameleon project. I have tested Sonarworks plugin and I was really surprised with the results on the default curve. If the results are the same with hardware filter, than I'm in. But is the amp on board decent enough? Have anybody here had heard the EMU and could point me to the right direction? Here are the specs for emu amp: Headphone Amplifier Type: Class-A power amplifier D/A converter: AK4396 (shared with Line Out) Gain Range: 60dB Maximum Output Power: 20mW Output impedance: 22 Ohms Frequency Response: (20Hz–20kHz) +0.06/-0.035dB Dynamic Range: (A-weighted) 114dB Signal-to-Noise Ratio: (A-weighted) 114dB THD+N: (1kHz, max gain): 600ohm load -95.5dB (.0018%) Stereo Crosstalk: (1kHz at -1dBFS, 600 ohm load) < -85dB
|
|
|
Post by tupisac on Mar 21, 2017 0:43:21 GMT
All right. It was my fault. I was using it on gain 2,5. After removing some resistors the amp plays like it should.
My hope in DIY is restored.
I'm strongly interested in assembling Kameleon 2 amp. One question though - will I be able fit the desktop version into 40mm height Galaxy enclosure? Can I just put boards next to each other using cable connection? Will it be enough?
|
|
solderdude
Administrator
measureutternutter
Posts: 4,882
|
Post by solderdude on Mar 21, 2017 6:02:48 GMT
I use the EMU 0204 (which basically is the same device but less inputs. It is always connected to the Kameleon for headphone measurements.
There is no comparison between the amp section of the 0404 and the Kameleon 2 (Desktop) The Kameleon portable = on par with O2 power wise and can be used like an O2. Portable version is right between EMU and Desktop version in driving capabilities. The portable O2 is the same size as O2
The output devices used in both Kameleons are higher spec'ed than the output devices in O2
Basically the DT version is an O2 + Sonarworks on steroids.
The desktop version needs 50mm height when the boards are on top of each other. You can use a 40mm height enclosure but needs to be a longer (boards behind each other) or wider (board next to each other)
|
|
|
Post by tupisac on Mar 21, 2017 10:41:34 GMT
Now I'm perfectly ok with o2 sound signature on gain=1. I must admit it has a better bass control than EMU. Slightly harsher highs tho, but very subtle. Feels now like o2 has a slight "V" shape. But the scene and air is present. I might even say that o2 feels more musical, especially on rock/pop genres. I've googled a lot on the topic, and generally EMU is regarded as top notch in given budget with it's crazy specs, but many people stated that amp itself is a weak link. I just knew there must be something wrong, but I wasn't expecting this much of an audible difference from a simple gain setting. Kudos to HD-600 for being this revealing, each day I fall in love some more Anyway, back to Kameleon 2 (Desktop). How would you compare it's sound signature to o2, assuming no filters used? Can you hear the difference with better output devices? What are the exact dimensions of both of the boards?
|
|
solderdude
Administrator
measureutternutter
Posts: 4,882
|
Post by solderdude on Mar 21, 2017 13:26:59 GMT
The DT Kameleon can play louder without distortion. The output devices used in the Kameleon are designed to drive headphone alike loads.
The Kameleon is intended to be used with filters. It can also be used as a normal amplifier though but there are already enough 'normal' headphone amps around already.
amp board: 100mm wide, 81 mm deep Powersupply: 100mm wide, 118 deep thickness = 1.6mm (standard)
With the HD600 the Kameleon may be of interest if you find it to lack subbass (rumble) and 'body' and if you find the mids just a tad too clear.
|
|
|
Post by tupisac on Mar 21, 2017 21:41:40 GMT
What gain setting would I need? I understand that it's set by the filter?
Now O2 is set to gain=1 and it's perfect, with all my music sitting between 11 and 3 o'clock on volume potentiometer.
|
|
solderdude
Administrator
measureutternutter
Posts: 4,882
|
Post by solderdude on Mar 21, 2017 22:04:12 GMT
Gain is determined by the filter and cannot be set. Not to worry though... it's not like the O2 which can be overdriven when the gain is set too high. If need be gain could be lowered by changing 2 resistors.
|
|
|
Post by tupisac on Mar 26, 2017 7:27:12 GMT
I got used to the sound of o2. Very nice little puppy, going to stay with me for sure as an workplace semi-portable amp. It is definitely better than what is on the EMU, which now for me feels like no sub-bass at all.
But why the gain of 2,5 made such difference? NwAvGuy says it shouldn't be the problem, especially with external power. It wasn't overdriven, I've set the output volume on EMU to match the "12 o'clock" position on o2 volume. The distortion was there on any combination of input/output level. This got me curious - what was the nature and (psycho)physics of the distortion I was hearing?
It was kind of like loosing some of the echo sounds, carrying some of the spatial information. Cymbals didn't ring properly. Bass was different, probably loosing bits from the highs component. It was like the compression, with the overall impression of sound being louder, but with all the nuances and presence flattened. I actually had trouble ear-matching volume on both amps, since EMU seemed more dynamic. Overall impression was that song was quieter, yet some instruments or sound hit a bit harder.
EDIT 1: Another theoretical question - what makes the o2 susceptible to to such distortion? Simply power levels or more complicated design principles, or just op amps used? How it differs in that regard from another op-amp amps, like kameleon?
EDIT 2: EMU outputs 3 V RMS on line-out according to specs. Theoretically o2 is designed for less. Yet with gain=1, the EMU is on full volume, o2 is on "12" and everything seems like perfect sounding.
|
|
solderdude
Administrator
measureutternutter
Posts: 4,882
|
Post by solderdude on Mar 26, 2017 11:49:47 GMT
I think the amp was overdriven at 2.5x gain and was what caused the sound to degrade.
The O2 is susceptible because of its design. The input stage does amplification regardless of the volume control setting. All signals thus are amplified even when the volume is turned down to almost nothing.
That input opamp (NJM2068) however does not have a rail-to-rail output swing. The max output swing is +Ub - 1.5V and -Ub - 2V.
The power supply voltages when using battery power is around +/-8.2V (with full batteries, lower when used for a short while) When using the adapter the power supply voltage is +/- 11.8V.
So clipping at the input stage appears at +11.8-1.5V (+ 10.3V) and -11.8V +2V = 9.8V And that is when mains fed ! This is theoretical. One should realize that at clipping point the distortion is already audibly high. For proper SQ the output voltage is just a tad lower 9.5V thus is about the limit that can be reached.
Well 9.5V doesn't seem that bad BUT this is peak level and not RMS level. To calculate the RMS level we need top divide by SQR 2. So RMS level just before distortion is 6.7V on the output of the gain stage.
When the 0dB level of the EMU is 3V RMS and you amplify this 2.5x you get... 7.5V which is higher than 6.7V so clipping occurs now and then or at least distortion will set in on this setting.
|
|
|
Post by tupisac on Mar 26, 2017 21:39:35 GMT
I think I got it. So now with EMU on full volume and with gain 1 on o2 I'm hitting peaks around 4,2 V right?
During the initial testing I've dialed down the output level on EMU to avoid the overdrive. But it was about 1 o'clock position, so maybe I was still hitting over 6,7 V you've mentioned.
|
|
|
Post by tupisac on Mar 27, 2017 20:27:38 GMT
Small update. I'm eagerly waiting for Polaris to arrive. Also, ordered Kameleon in-line boards.
I think amp (and bass-wise) side is covered for now, hopefully for another 5+ years. I don't like to change my gear too often.
|
|
|
Post by tupisac on Mar 28, 2017 7:41:38 GMT
Just today we had a short power shortage. Many new houses getting built in the area, so this is quite frequent. Now my EMU won't power up for like few hours. This reminds me that I've forgotten about replacement DAC.
What I'm really looking for in a DAC are stable win64 drivers, including ASIO. I have 20% of my music in 192 khz formats and I exclusively use foobar on PC. That's it. No other uses, not really need DSD but like the idea of having the ability to play this format if someone brings such file.
I've narrowed down my search to XMOS from DIYINHK (cheapest one) paired up with AK4990. Any other suggestions? Maybe same 2nd hand gear to look for. Just from the company that is know for not abandoning their support for discontinued products.
|
|
Javier
Administrator
Digital bytes
Posts: 987
|
Post by Javier on Mar 28, 2017 12:15:53 GMT
Once you account for all things needed for the build the cost difference with the iFi Nano LE (http://ifi-audio.com/portfolio-view/nano-idsd-le/) won't be that big and the latter performs very well.
|
|
jello
extremely active
Posts: 1,569
|
Post by jello on Mar 28, 2017 13:02:37 GMT
That looks interesting Javier and a degree of portability is never a bad thing.
The biggest thing that put me off the iDSD (non nano version) was that it couldn't be used whilst charging. Does the same limitation apply to this or can you charge while you listen?
|
|
solderdude
Administrator
measureutternutter
Posts: 4,882
|
Post by solderdude on Mar 28, 2017 13:09:13 GMT
I was planning to buy the LE but after reading the manual and talking to a sales person I decided not to.
The DAC's I currently have all work well with or without a resume in windows. the DAC is always 'on' and powered by USB. Starting the PC means the soundcard works.
With the LE this is not as simple. You cannot leave the DAC 'on' because it will deplete the battery when USB power is turned off by the PC. To use the LE from USB power the PC has to be started and THEN the DAC must be switched on after that. This means starting up the PC, wait and then turn on the DAC after that.
Far from user friendly and not suited for the way I use DAC's.
Good DAC, cheap but this 'feature' made me decide to not buy one.
Sometimes my EMU 0204 doesn't work properly when the PC is switched on with the power already on though. In that case I have to switch the DAC off and on again. All my other DAC's do not exhibit this behavior.
|
|