![WP_20190915_20_26_30_Pro (2)[9776].jpg](https://cdn.head-fi.org/a/10353732.jpg)
![WP_20190915_20_26_30_Pro (2)[9776].jpg](https://cdn.head-fi.org/a/10353731.jpg)
Aplle
Head-Fier
- Joined
- May 8, 2015
- Posts
- 56
- Likes
- 12
Has anyone compared the dac that's buit into the 3U with other dacs, and if so how did it fare?
Introducing the Matrix M-Stage HPA-3U and HPA-3B, review to follow!
- Thread starter keanex
- Start date
raoultrifan
Headphoneus Supremus
http://www.head-fi.org/products/matrix-m-stage/reviews/13475
http://www.headfonia.com/review-matrix-m-stage-hpa-3u/
http://headphone.guru/the-new-matrix-audio-m-stage-hpa-3u/
So, DAC inside 3U seems decent, but the amp itself can do better with external sources.
http://www.headfonia.com/review-matrix-m-stage-hpa-3u/
http://headphone.guru/the-new-matrix-audio-m-stage-hpa-3u/
So, DAC inside 3U seems decent, but the amp itself can do better with external sources.
raoultrifan
Headphoneus Supremus
I think I'll buy some bigger heat-sinks for drivers transistors, because they heat up very high (I estimate at least 70C with case closed). I'll also buy a couple of heat-sinks for the OPAMPs as well (they heat up very well too, at least 50C).
By looking at these drivers closer I can see written on them .T-S and :H-S., but I can't really say what exact transistors are they, maybe a head-fier might help here, please. Anyway, after reading datasheets for some of these SOT-23 SMD transistors (hope I'm not wrong about the case type) are usually able to work up to until 150C internally, here's some examples of ST-marking transistors I found:
- 2SC3929A: Max. power dissipation: 150mW, max.junction temp: 150C, 55VCBO, 55VCEO, cont./peak current: 50mA/100mA
- 2SC2406: Max. power dissipation: 200mW, max.junction temp: 150C, 55VCBO, 55VCEO, cont./peak current: 50mA/100mA
Also, we can notice the 6 x OMRON relays, each costing 5 USD + VAT.
I had installed, after all, 4 x WIMA MKP 0.1uF/100V from each V+/GND & V-/GND for both OPAMPs, 4 x WIMA MKP 0.1uF/100V for each of the 4 x 330uF Nichicon capacitor from the middle of the board and 2 x AVX 10uF/35V tantalums for each of the 2 x 330uF capacitors from output of power regulators.
I measured again AC ripple and noise with my 10MHz PicoScope and max. peak-peak value was less than 1mV (first time I did that I used a 100KHz true-RMS VOLTCRAFT multimeter and the value was lower, because of lower bandwidth and accuracy). I find this being quite decent from 0Hz to 10Mhz bandwidth, especially that LME49860 has a PSRR and CMRR that exceed 120dB, so there should be no noise getting from power rails to headphones output.
I measured on the transformer the following voltages:
- 8.35V between the 2 remaining windings or 1x3.85V between GND and first winding and 1x12.3V between GND and second winding (perhaps used for the blue LED, for output protection circuit and also for +5V OMRON's)
- 2x21.70V used for power regulators
I found a 0.65V difference between the 2 regulators:
- JRC7818: 18.45V
- JRC7918: -17.80V
but this is not going to influence negatively headamp's functionality, because this circuit only knows about the total voltage of 36V between +Vcc and -Vcc. Also, the very low DC output speaks for itself.
I unscrewed for about 1 turn the 4 screws from bottom of PCB that keeps aligned the big heat-sink from output transistor (2 screws for each heat-sink), because of a wrong angle of the heat-sinks were not perfectly aligned with the aluminum case:
You can see that the thermal paste MX-4 is missing from the bottom-mid of the heat-sink because of the wrong angle. Now, after unscrewing the 2 screws for both heat-sinks I'm hoping the heat-sinks will perfectly get aligned with the case, for a perfect heat transfer.
By looking at these drivers closer I can see written on them .T-S and :H-S., but I can't really say what exact transistors are they, maybe a head-fier might help here, please. Anyway, after reading datasheets for some of these SOT-23 SMD transistors (hope I'm not wrong about the case type) are usually able to work up to until 150C internally, here's some examples of ST-marking transistors I found:
- 2SC3929A: Max. power dissipation: 150mW, max.junction temp: 150C, 55VCBO, 55VCEO, cont./peak current: 50mA/100mA
- 2SC2406: Max. power dissipation: 200mW, max.junction temp: 150C, 55VCBO, 55VCEO, cont./peak current: 50mA/100mA
Also, we can notice the 6 x OMRON relays, each costing 5 USD + VAT.
I had installed, after all, 4 x WIMA MKP 0.1uF/100V from each V+/GND & V-/GND for both OPAMPs, 4 x WIMA MKP 0.1uF/100V for each of the 4 x 330uF Nichicon capacitor from the middle of the board and 2 x AVX 10uF/35V tantalums for each of the 2 x 330uF capacitors from output of power regulators.
I measured again AC ripple and noise with my 10MHz PicoScope and max. peak-peak value was less than 1mV (first time I did that I used a 100KHz true-RMS VOLTCRAFT multimeter and the value was lower, because of lower bandwidth and accuracy). I find this being quite decent from 0Hz to 10Mhz bandwidth, especially that LME49860 has a PSRR and CMRR that exceed 120dB, so there should be no noise getting from power rails to headphones output.
I measured on the transformer the following voltages:
- 8.35V between the 2 remaining windings or 1x3.85V between GND and first winding and 1x12.3V between GND and second winding (perhaps used for the blue LED, for output protection circuit and also for +5V OMRON's)
- 2x21.70V used for power regulators
I found a 0.65V difference between the 2 regulators:
- JRC7818: 18.45V
- JRC7918: -17.80V
but this is not going to influence negatively headamp's functionality, because this circuit only knows about the total voltage of 36V between +Vcc and -Vcc. Also, the very low DC output speaks for itself.
I unscrewed for about 1 turn the 4 screws from bottom of PCB that keeps aligned the big heat-sink from output transistor (2 screws for each heat-sink), because of a wrong angle of the heat-sinks were not perfectly aligned with the aluminum case:
You can see that the thermal paste MX-4 is missing from the bottom-mid of the heat-sink because of the wrong angle. Now, after unscrewing the 2 screws for both heat-sinks I'm hoping the heat-sinks will perfectly get aligned with the case, for a perfect heat transfer.
raoultrifan
Headphoneus Supremus
Like I said before drivers heat-sinks heat up pretty high: I measured with my IR thermometer between 62-64C without any signal input and case open. Output transistors heat up a bit lower...around 52-55C and OPAMPs a couple degrees less. So I ended up replacing drivers heat-sink with bigger ones and I also installed 2 x heat-sinks for OPAMPs as well.
I filed drivers heat-sinks by using a pile and a vise and I still got trouble with one heat-sink almost touching an output resistor of 10 ohms, so I removed the resistor and solder it 1-2 mm forward.
Initially I used thermal Scotch tape to stick heat-sinks to drivers transistors, but I wasn't pleased with thermal transfer because these heat-sink are 3 times bigger and 4 times heavier than the originals and I can't really stuck them in place properly, so I used Arctic Alumina thermal glue. Even if thermal transfer is now even better than original thermal tape and heat-sinks are bigger and thicker drivers temperatures are not much lower than it used to be before, just 2-3C better.
I still have 60-61C with case open and I'm not very comfortable with this, because this probably means that with case closed drivers heat-sinks will reach up to 75C and their internal junctions will most likely heat to 85C or even more.
So, I had installed a small fan from an old video card:
Power supply I got it from an existing 8.35V AC (white and black wires from transformer's plug) and I used a small rectifier bridge and a 100uF/25V capacitor. Final DC voltage was 10.6V and the little fan was powered through a 50 ohms/0.25W resistor and was sticked with 1mm thick 3M double tape directly on OPAMPs heat-sinks, perfectly centered. Now drivers heatsink are having between 42-45C with case open and no input signal applied, so I suppose with case closed and music singing drivers transistors will heat up no more than 60C or maybe less.
One thing's for sure: initially I had around 48-50C on top of 3B's case and now, with bigger heat-sinks for driver transistors and OPAMPs + a small fan inside, I have around 41-42C.
I filed drivers heat-sinks by using a pile and a vise and I still got trouble with one heat-sink almost touching an output resistor of 10 ohms, so I removed the resistor and solder it 1-2 mm forward.
Initially I used thermal Scotch tape to stick heat-sinks to drivers transistors, but I wasn't pleased with thermal transfer because these heat-sink are 3 times bigger and 4 times heavier than the originals and I can't really stuck them in place properly, so I used Arctic Alumina thermal glue. Even if thermal transfer is now even better than original thermal tape and heat-sinks are bigger and thicker drivers temperatures are not much lower than it used to be before, just 2-3C better.
I still have 60-61C with case open and I'm not very comfortable with this, because this probably means that with case closed drivers heat-sinks will reach up to 75C and their internal junctions will most likely heat to 85C or even more.So, I had installed a small fan from an old video card:
Power supply I got it from an existing 8.35V AC (white and black wires from transformer's plug) and I used a small rectifier bridge and a 100uF/25V capacitor. Final DC voltage was 10.6V and the little fan was powered through a 50 ohms/0.25W resistor and was sticked with 1mm thick 3M double tape directly on OPAMPs heat-sinks, perfectly centered. Now drivers heatsink are having between 42-45C with case open and no input signal applied, so I suppose with case closed and music singing drivers transistors will heat up no more than 60C or maybe less.
One thing's for sure: initially I had around 48-50C on top of 3B's case and now, with bigger heat-sinks for driver transistors and OPAMPs + a small fan inside, I have around 41-42C.
shatrukov
New Head-Fier
- Joined
- Jul 31, 2015
- Posts
- 20
- Likes
- 11
Code:
[color=rgb(34, 34, 34)] Вы обеспокоены температуре 75 ° С? Вообще допустимой температурой на МОП транзистора 150 °. 70 - 85C является нормальная температура для класса А.[/color]raoultrifan
Headphoneus Supremus
Thanks to Google translate: "Are you concerned about a temperature of 75 ° C? Generally allowable temperature of the MOSFET 150 °. 70 - 85C is the normal temperature for Class A."
Indeed, absolute max. temp. of these transistor drivers is probably 150C (perhaps 125C to be on the safe side), but this is calculated on internal junction temperature and not on top of transistor's heat-sink. Usually I don't care if a TO-126 or a TO-220 case transistor heats up to 100C, but a small plastic SOT-23 transistor having it's top heat-sink with a temperature of 75C worries me a little bit, yes.
Anyway, 75C measured on top of this heat-sink probably means 90-95C inside transistor's junctions, because original heat-sinks were stick with thermal tape and amplifier's case is closed anyway, so there's no ventilation inside. If the aluminum case would be made with proper vent haules on top and also on bottom, like Schiit Asgard is, then I wouldn't be worried about internal temperature anymore. I'm not saying M-Stage 3B has temperature issues, not at all actually, but all I'm saying is that I'm not feeling comfortable with a 50C top-case temperature.
If it was my speakers amplifier then I wouldn't probably bothered at all, but I don't really like to place my hand on top of a headphone amplifier and to almost get burned, if you know what I'm saying.
And same feeling have it OPPO HA-1 owners on DIY-AUDIO forum and we all know how good HA1 is and how well it measures.By lowering internal case temperature with almost 10C I had also decreased all internal passive components temperature as well, because PCB was initially hot and all capacitors, resistors and even the ALPS potentiometer were quite warm and we all know that warm resistors means more Johson/thermal noise, right?
I don't know about other headamps, but Beyerdynamic A1 and Burson HA-160 have about 35C on top, Schiit Asgard 2 has around 45C on top (pure Class-A again), so original Matrix M-Stage 3B with its 50C is quite hot I'd say, perhaps because it's balanced and it has twice as many transistors.
shatrukov
New Head-Fier
- Joined
- Jul 31, 2015
- Posts
- 20
- Likes
- 11
Code:
[color=rgb(34, 34, 34)] Thanks for the translation!))) Google translated well. Interesting to know the quiescent current MOSFET . There is such a possibility?[/color]raoultrifan
Headphoneus Supremus
Well, I'm not sure the driver transistors are MOSFETs after all; could you please check this post from top of this page, please? Perhaps you know better, but I've seen on Internet that H-S and T-S notations correspond to bipolar transistors, but I can't be 100% sure of that.
Anyway, now that I glued the heat-sinks on top of driver transistors I can't really measure any voltage drop across their legs. In case you have some suggestions how to easily check quiescent current, please tell me.
Now I'm listening at moderate volume on my 32-ohms AKG K550 headphones and while case is completely closed, internal air temperature of M-Stage 3B is around 43C (sometimes more, depending on probe position inside M-Stage case). Measurement was done with the external sensor of UNI-T UT58C multimeter stuck inside M-Stage 3U through upside vent-holes for about 4-5cm.
While my IR thermometer shows an acceptable temperature of 43C on top of case (although it feels hotter when placing my palm on top), UT58C shows between 57-60C when temp probe touches drivers heat-sinks (with aluminum case closed, spot led lantern used to guide the small probe inside M-Stage vent holes). 60C seems quite OK to me, too bad I haven't done this measurement before installing the bigger heat-sinks and the small fan inside the case (I was only approximating the "case close" temp based on "open case" temp); perhaps drivers heatsinks had about 75C indeed when case closed.
Thank you!
Anyway, now that I glued the heat-sinks on top of driver transistors I can't really measure any voltage drop across their legs. In case you have some suggestions how to easily check quiescent current, please tell me.
Now I'm listening at moderate volume on my 32-ohms AKG K550 headphones and while case is completely closed, internal air temperature of M-Stage 3B is around 43C (sometimes more, depending on probe position inside M-Stage case). Measurement was done with the external sensor of UNI-T UT58C multimeter stuck inside M-Stage 3U through upside vent-holes for about 4-5cm.
While my IR thermometer shows an acceptable temperature of 43C on top of case (although it feels hotter when placing my palm on top), UT58C shows between 57-60C when temp probe touches drivers heat-sinks (with aluminum case closed, spot led lantern used to guide the small probe inside M-Stage vent holes). 60C seems quite OK to me, too bad I haven't done this measurement before installing the bigger heat-sinks and the small fan inside the case (I was only approximating the "case close" temp based on "open case" temp); perhaps drivers heatsinks had about 75C indeed when case closed.
Thank you!
raoultrifan
Headphoneus Supremus
In case you guys missed that: AUDEZE DECKARD headamp is quite similar with M-STAGE 3U (perhaps the same manufacturer?!?): http://www.head-fi.org/t/749486/audeze-deckard-class-a-amp-usb-dac/150#post_12160801.
raoultrifan
Headphoneus Supremus
Just realized that output power specs for THD+N=0.01% published by Matrix manufacturer (http://www.matrix-digi.com/en/specifications/76/index.html) are quite simmilar with the ones published by Texas Instruments on LME49860's datasheet graphs (http://www.ti.com/lit/ds/symlink/lme49860.pdf) and could be achieved with a gain of "only" +15dB gain (at least for 300ohms and 600ohms, because for 33ohms output buffers can't output so much power). So, I wonder if anyone has decent equipment to do some measurements for this amp and to publish some numbers.
I estimate that for an input source having 2V RMS/4V RMS (single-ended/balanced) full power specified by manufacturer will be achieved by using the max. gain setting of +20dB and volume knob around 3'o clock, at least for high impedance headphones.
Based on Matrix specs, max. output voltage for HPA-3B for a THD+N of only 0.01% would be:
- around 11.5V RMS for single-ended operation for high impedance phones (220mW in 600ohms)
- around 22V RMS for balanced operation for high impedance phones (800mW in 600ohms)
I used http://www.sengpielaudio.com/calculator-amplification.htm to find out that these figures are equivalent to a gain of +15dB if input source is 2V RMS/4V RMS (single-ended/balanced).
Based on TI http://www.ti.com/lit/ds/symlink/lme49860.pdf datasheet max. output voltage would be:
- around 11.8V RMS: 232mW in 600ohms for single-ended operation
- around 23.6V RMS: 929mW in 600ohms for balanced mode.
Have no idea about the real THD+N at these power figures, but I estimate would probably be less than 1%.
NOTE: For checking if this amp is fully compatible with your headphones, at least based on pure specs and requirements, also for checking the correct gain setting to be used you could fill in http://www.digizoid.com/power.php and calculate necessary power to correctly handle your headphones (max. peaks of 115dB should suffice, I'd say, for most people). All you need to do is check your headphone specs or datasheet and enter impedance and correct SPL values. Double check if your headphones manufacturer is using dB/mW or dB/V before calculating amp power requirement. For example Hifiman HE-6, the most hard to drive headphones I know, need 8.65V RMS/163.21 mA/1411.75 mW to be driven as loud as 115dB. That means in either single-ended or balanced mode HPA-3B will handle these headphones with ease and pleasure; too bad I don't have a pair of HE6 to test them myself now.
P.S.: I am doing now some A/B testing between HPA-3B and Essence One and also Objective2 and I realize that HPA-3B soundstage is much better than it used to be in its first days when I bought it. I'm not sure this might be related to the mods I've done here or to the caps I've added here because I haven't done these A/B test immediately after I've finished those mods, but I'm sure I really love the soundstage I'm having now.
I estimate that for an input source having 2V RMS/4V RMS (single-ended/balanced) full power specified by manufacturer will be achieved by using the max. gain setting of +20dB and volume knob around 3'o clock, at least for high impedance headphones.
Based on Matrix specs, max. output voltage for HPA-3B for a THD+N of only 0.01% would be:
- around 11.5V RMS for single-ended operation for high impedance phones (220mW in 600ohms)
- around 22V RMS for balanced operation for high impedance phones (800mW in 600ohms)
I used http://www.sengpielaudio.com/calculator-amplification.htm to find out that these figures are equivalent to a gain of +15dB if input source is 2V RMS/4V RMS (single-ended/balanced).
Based on TI http://www.ti.com/lit/ds/symlink/lme49860.pdf datasheet max. output voltage would be:
- around 11.8V RMS: 232mW in 600ohms for single-ended operation
- around 23.6V RMS: 929mW in 600ohms for balanced mode.
Have no idea about the real THD+N at these power figures, but I estimate would probably be less than 1%.
NOTE: For checking if this amp is fully compatible with your headphones, at least based on pure specs and requirements, also for checking the correct gain setting to be used you could fill in http://www.digizoid.com/power.php and calculate necessary power to correctly handle your headphones (max. peaks of 115dB should suffice, I'd say, for most people). All you need to do is check your headphone specs or datasheet and enter impedance and correct SPL values. Double check if your headphones manufacturer is using dB/mW or dB/V before calculating amp power requirement. For example Hifiman HE-6, the most hard to drive headphones I know, need 8.65V RMS/163.21 mA/1411.75 mW to be driven as loud as 115dB. That means in either single-ended or balanced mode HPA-3B will handle these headphones with ease and pleasure; too bad I don't have a pair of HE6 to test them myself now.
P.S.: I am doing now some A/B testing between HPA-3B and Essence One and also Objective2 and I realize that HPA-3B soundstage is much better than it used to be in its first days when I bought it. I'm not sure this might be related to the mods I've done here or to the caps I've added here because I haven't done these A/B test immediately after I've finished those mods, but I'm sure I really love the soundstage I'm having now.
burdie
100+ Head-Fier
- Joined
- Oct 24, 2013
- Posts
- 180
- Likes
- 30
Just realized that output power specs for THD+N=0.01% published by Matrix manufacturer (http://www.matrix-digi.com/en/specifications/76/index.html) are quite simmilar with the ones published by Texas Instruments on LME49860's datasheet graphs (http://www.ti.com/lit/ds/symlink/lme49860.pdf) and could be achieved with a gain of "only" +15dB gain (at least for 300ohms and 600ohms, because for 33ohms output buffers can't output so much power). So, I wonder if anyone has decent equipment to do some measurements for this amp and to publish some numbers.
I estimate that for an input source having 2V RMS/4V RMS (single-ended/balanced) full power specified by manufacturer will be achieved by using the max. gain setting of +20dB and volume knob around 3'o clock, at least for high impedance headphones.
Based on Matrix specs, max. output voltage for HPA-3B for a THD+N of only 0.01% would be:
- around 11.5V RMS for single-ended operation for high impedance phones (220mW in 600ohms)
- around 22V RMS for balanced operation for high impedance phones (800mW in 600ohms)
I used http://www.sengpielaudio.com/calculator-amplification.htm to find out that these figures are equivalent to a gain of +15dB if input source is 2V RMS/4V RMS (single-ended/balanced).
Based on TI http://www.ti.com/lit/ds/symlink/lme49860.pdf datasheet max. output voltage would be:
- around 11.8V RMS: 232mW in 600ohms for single-ended operation
- around 23.6V RMS: 929mW in 600ohms for balanced mode.
Have no idea about the real THD+N at these power figures, but I estimate would probably be less than 1%.
NOTE: For checking if this amp is fully compatible with your headphones, at least based on pure specs and requirements, also for checking the correct gain setting to be used you could fill in http://www.digizoid.com/power.php and calculate necessary power to correctly handle your headphones (max. peaks of 115dB should suffice, I'd say, for most people). All you need to do is check your headphone specs or datasheet and enter impedance and correct SPL values. Double check if your headphones manufacturer is using dB/mW or dB/V before calculating amp power requirement. For example Hifiman HE-6, the most hard to drive headphones I know, need 8.65V RMS/163.21 mA/1411.75 mW to be driven as loud as 115dB. That means in either single-ended or balanced mode HPA-3B will handle these headphones with ease and pleasure; too bad I don't have a pair of HE6 to test them myself now.
P.S.: I am doing now some A/B testing between HPA-3B and Essence One and also Objective2 and I realize that HPA-3B soundstage is much better than it used to be in its first days when I bought it. I'm not sure this might be related to the mods I've done here or to the caps I've added here because I haven't done these A/B test immediately after I've finished those mods, but I'm sure I really love the soundstage I'm having now.
Sorry, I thought LME49860 was used as I/V stage and the high current amplication was done by the Toshiba A1358/C3421?
raoultrifan
Headphoneus Supremus
Yes, LME49860 OPAMP is used for Voltage-Gain stage, so the final voltage of amplifier is given by the OPAMP itself. Output stage made by transistors only provides enough current to drive low impedance headphones.
I believe 3B's THD is mostly given by the OPAMP used, especially for high impedance headphones, but it would be great if somebody could take some measurements on this amp with dedicated equipment.
I believe 3B's THD is mostly given by the OPAMP used, especially for high impedance headphones, but it would be great if somebody could take some measurements on this amp with dedicated equipment.
raoultrifan
Headphoneus Supremus
I did a couple of tests last evening to roughly check for max. output power and also for THD or any possible oscillations when driving the Matrix M-Stage 3B amplifier to max. power.
Source: Balanced outputs from ASUS Essence MUSES MKii + audiocheck.net FLAC file -3dB 1KHz sinusoidal signal connected to desktop computer (1.2mV noise+ripple on USB port)
Scope: PicoScope 2204A, latest app. version, connected to Lenovo T430 laptop on battery (for galvanic decoupling probes from AC mains)
Dummy load: 4 x 240ohms/1W paralleled resistors (about 60ohms/4W) connected to one end of original Beyerdynamic DT880 cable (I had this one left after I balanced my headphones). So, as you can see, there's also a real cable capacitive load as well on output circuit, so it's almost like driving regular headphones. The cable has about 2 ohms: 1 ohm for signal wire and 1 ohm for GND. I've only published results from one channel.
Volume knob when starting to clip was around 3 o'clock and moving the knob even 1mm more or less made the output voltage to increase/decrease with at least 0.5V RMS. So, I estimate that with 60ohms headphones this baby can get up to 1.5W RMS power on single-ended and up to 4.5W RMS power in balanced operation, which theoretically should be good enough to drive any headphone from this planet, HE6 included. I'm very curious if someone can pair this amp with HE6 balanced headphones, to see about quality of the sound.
I hope someone will be able to do some decent RMAA or AP-precision tests, because I can't trust the THD+N of my scope (see note below).
Note: On the FFT graphs the resolution and granularity of 2204A scope sucks, because I've used a gain of 10x for my probes and dB and THD are definitely not accurate. I need to use the +/-50mV peak-peak scale to get actual dB THD+noise (I can only do this with volume pot. to the min.), but that means I'm unable to check for max. power or for harmonics because I'll blow away my scope. So, this is why I used the +/-200V peak-peak scale (actually +/-20V with a gain of 10X setup to my probe).
L.E.: I must confess that the dummy resistors were extremely hot during these tests, untouchable by fingers, but none of them changed their color to brown or black.
Source: Balanced outputs from ASUS Essence MUSES MKii + audiocheck.net FLAC file -3dB 1KHz sinusoidal signal connected to desktop computer (1.2mV noise+ripple on USB port)
Scope: PicoScope 2204A, latest app. version, connected to Lenovo T430 laptop on battery (for galvanic decoupling probes from AC mains)
Dummy load: 4 x 240ohms/1W paralleled resistors (about 60ohms/4W) connected to one end of original Beyerdynamic DT880 cable (I had this one left after I balanced my headphones). So, as you can see, there's also a real cable capacitive load as well on output circuit, so it's almost like driving regular headphones.
The cable has about 2 ohms: 1 ohm for signal wire and 1 ohm for GND. I've only published results from one channel.Not-clipping SE: a bout 1.425 W RMS power(26.65V p-p, 9.402V RMS) | Clipping SE: a bout 1.687W RMS power(28.14V p-p, 10.23V RMS) |
Not clipping SE: about 1.5 W RMS power - no strange harmonics | Clipping SE: about 1.7 W RMS power - lot of harmonics |
Not clipping BAL: about 4.327 W RMS power (46.7V p-p, 16.38V RMS) | Clipping BAL: about 4.617 W RMS power (47.58V p-p, 16.92V RMS) |
Not clipping BAL: about 4.4 W RMS power - no strange harmonics | Clipping BAL: about 4.6 W RMS power - lot of undesired harmonics |
Volume knob when starting to clip was around 3 o'clock and moving the knob even 1mm more or less made the output voltage to increase/decrease with at least 0.5V RMS. So, I estimate that with 60ohms headphones this baby can get up to 1.5W RMS power on single-ended and up to 4.5W RMS power in balanced operation, which theoretically should be good enough to drive any headphone from this planet, HE6 included. I'm very curious if someone can pair this amp with HE6 balanced headphones, to see about quality of the sound.
I hope someone will be able to do some decent RMAA or AP-precision tests, because I can't trust the THD+N of my scope (see note below).
Note: On the FFT graphs the resolution and granularity of 2204A scope sucks, because I've used a gain of 10x for my probes and dB and THD are definitely not accurate. I need to use the +/-50mV peak-peak scale to get actual dB THD+noise (I can only do this with volume pot. to the min.), but that means I'm unable to check for max. power or for harmonics because I'll blow away my scope. So, this is why I used the +/-200V peak-peak scale (actually +/-20V with a gain of 10X setup to my probe).
L.E.: I must confess that the dummy resistors were extremely hot during these tests, untouchable by fingers, but none of them changed their color to brown or black.
BlutoSlice
100+ Head-Fier
- Joined
- Mar 24, 2010
- Posts
- 257
- Likes
- 11
Im thinking of upgrading my hpa1 to b3 can someone suggest which adapters I would need for a Sony mdr-z7 it has a balanced cable but with mini jacks. Would I need xlr > twin female phono?
raoultrifan
Headphoneus Supremus
I believe you can try doing this by yourself, because it may be hard to find such an adapter. 2 x 3.5mm female jacks to XLR male should be fine or you can get rid of existing small jacks and solder directly a high-quality 4-pin XLR-male.
Users who are viewing this thread
