• 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!

BURSON PLAY mods

trl

Major Contributor
King of Mods
Joined
Feb 28, 2018
Messages
1,967
Likes
2,523
Location
Iasi, RO
The device works very well by default, lot of power, class-A output stage, very good looking, but some mods might improve it a bit, at least for my taste.

With sensitive cans there might be some background noise (I was using board revision v1.6, but with board v2.1 and higher the noise settled down a lot!), especially when using 16 Ohms IEMs, so a linear power supply to replace the built-in SMPS/DC-DC converters could be an answer to resolve this. Basically, this combo it's getting power from a 70W 12V SMPS external PSU, then 2 x DC-DC boost converters are creating +/-15V; also the initial 12V is lowered by linear regulators to 5V (several RC filter are in there as well).

1. LPSU on the bench (PCBs are from ANT-X from diyforum, just a bit modified to adjust the Japanese NJR 7812/7815/7915 regulators):

WP_20180127_20_09_49_Pro.jpg


WP_20180127_20_10_45_Pro.jpg


WP_20180127_20_11_14_Pro.jpg


WP_20180201_08_11_42_Pro.jpg


AdobePhotoshopExpress_82c34cf6022143a3aa1c6d3fc08d5686.jpg


AdobePhotoshopExpress_1d3d8fd0b960492e89d782bec5d70fbb.jpg


Now there's no more noise on the 6.3mm headphones jack, even if 16-Ohms IEM's are used, so getting rid of DC-DC boost converters and upgrading to linear PSU did helped here.

For board revision v2.1 or higher you might not be needing this LPSU-upgrade mod, as the new board is having 20 more ELNA caps and ground plane has been increased to about 100% of the board, so noise is minimised.

2. RCA output seems to be connected directly to output stage transistors. To further minimise the noise on the RCA outputs I was able to cut the PCB traces that go to the RCA plugs and add a couple of shielded wires directly from the Low-Pass Filter op-amp (the one from the middle), via additional resistors (anything between 10 -50 kOhm should do).

WP_20180212_22_44_41_Pro.jpg


My cleanest and darkest headamp I have is the Objective 2; basically, with volume to the max, gain of 3.7X and 16 Ohms IEM's used I can't really hear any noise. Connecting the PLAY to the O2 didn't changed anything...same darkness. Of course, no music was playing, but PLAY was connected to USB of the computer.

3. Op-amp testing, besides the SS V5 and V6, I was able to test many more from the below pics. However, I choose AD8599 (or MUSES8920) for I/V and LPF and AD797 (or MUSES8920 with DIY dual-to-single adapter) for pre-amplification stage

WP_20180210_20_20_27_Pro.jpg


WP_20180210_20_21_00_Pro.jpg


WP_20180210_20_48_26_Pro.jpg


The dual-to-single adapter is DIY, because I was unable to find such thing on eBay...not sure it even exists somewhere honestly.

4. Gain adjust pins/jumpers could be easily soldered on the green ovals. It used to be 220pF caps in there on a former board revision, but manufacturer decided to remove them, so paralleling the existing 2.1 kOhms resistors with something around 5 kOhms will lower the output volume enough to make the device much friendly with high sensitive headphones.

BURSON_PLAY_v2.2_.jpg
 
OP
trl

trl

Major Contributor
King of Mods
Joined
Feb 28, 2018
Messages
1,967
Likes
2,523
Location
Iasi, RO
LEDs and metallic case are pretty much the same, thank you! :)
 
OP
trl

trl

Major Contributor
King of Mods
Joined
Feb 28, 2018
Messages
1,967
Likes
2,523
Location
Iasi, RO
In regard with my penultimate post, I got some "jumpers" soldered with ease on the PCB, right in place of the 220pF caps. I inserted a couple of 1KOhms value-matched resistors and gain decreased substantially...not sure how much, perhaps output volume is reduced now with 30-40% or so. However, with my very sensitive SUPERLUX HD381F (16 Ohms, 103 dB SPL) there's absolutely no background noise...it's like the Objective2 when low-gain is used.

WP_20180726_21_00_18_Pro.jpg

2 jumpers installed per each channel


WP_20180726_21_17_29_Pro.jpg

Added gain-resistor for one of the channels

In the past volume was around 7-9 for my IEM's, now it's somewhere around 15-20. Fostex T50-RP mk3 and Hifiman HE-560 max. volume is around 55%, perhaps 60% for only couple of seconds. So this "low-gain" mod works for me and it's quite easy to do.

Op-amps used in the pre-amp/voltage-amplification stage are AD797. Haven't tested with other op-amps, but if an op-amp is designed for unity-gain I see no reasons not to work fine.

P.S.: I believe same type of jumpers are installed inside the Objective2 headamp, so it's quite useful to easily adjust the gain this way.
 
OP
trl

trl

Major Contributor
King of Mods
Joined
Feb 28, 2018
Messages
1,967
Likes
2,523
Location
Iasi, RO
After seeing the results from https://www.headfonia.com/review-burson-play/3/ (PLAY v1.6 vs. PLA v2.1) I did myself an A/B test between the original v1.6 and the modded one from above and seems that my modded v1.6 is still having a bit lower background noise than the new v2.1. However, I was able to notice the difference between the 2 mainboard revisions only on a completely silent environment (1 AM @night) and using sensitive cans like AKG K550 (32-Ohms, 114dB SPL/V), Beats Solo2 (32-Ohms, SPL better than K550) and Superlux HD381F (16-Ohms IEM, SPL 103dB SPL, 1mW). With GRADO or other headphones I couldn't get any background noise at all.

I'm also uploading some RMAA results taken with my good old ASUS U7 ADC/DAC: PLAY v1.6/V6 Classic with SMPS vs. PLAY v1.6/MUSES8920 with SMPS vs. PLAY v1.6/V6 or MUSES8920 with LPSU.

PLAY-SMPSvsLPSU-RMAA(V6-MUSES8920-LPSU).png

3.6dB lower noise when using the LPSU mod with MUSES8920 opamps in I/V and LPF and 5.7dB lower noise when using LPSU with V6 Classic opamps. It seems that the V6 opamps are affected by the noise induced by the SMPS, probably they're more sensitive because their bigger size/shell.

PLAY v1.6/MUSES8920 with SMPS vs. PLAY v1.6/V6 Classic or MUSES8920 with LPSU
PLAY-SMPSvsLPSU-Freq_resp.pngPLAY-SMPSvsLPSU-SNR.pngPLAY-SMPSvsLPSU-Dynamics.pngPLAY-SMPSvsLPSU-THD.pngPLAY-SMPSvsLPSU-IMD.png

I see that the background noise is indeed lower when using Linear PSU mod and RCA-outputs connected to the LPF opamp, instead to the 6.3mm jack, but the 50Hz and 150Hz mains induced by the Linear PSU transformers is affecting the performance. However, increasing the distance between the LPSU and the PLAY is lowering these harmonics, but as long as I can't hear this "mains hum" at all I'm somehow comfortable with it.
 
OP
trl

trl

Major Contributor
King of Mods
Joined
Feb 28, 2018
Messages
1,967
Likes
2,523
Location
Iasi, RO
I don't have access to an original PLAY V1.6 board to test it, but I do have access to the newer V2.2 board, so I can compare it with my modded board V1.6. Subjectively, when A/B-ing between the two boards, the V1.6 modded is a little bit lower noise then the V2.2 (tested with 16 Ohms IEMs, 100dB/mW sensitivity).

ASUS_U7_threshold_ASIO4ALL.png

ASUS U7 noise threshold

PLAY_orig_noise_threshold_RCA_ASIO4ALL.png

PLAY board V2.2 - RCA output

PLAY_noise_threshold_RCA_ASIO4ALL.png

Modded PLAY board V1.6 - RCA output (less noise and spikes, but mains hum coming from linear PSU's toroid appears in the graph)

PLAY_orig_noise_threshold_Headphones_ASIO4ALL.png

PLAY board 2.2 - headphones output (identical graph with the one from RCA output, as there are the same PCB traces)

PLAY_noise_threshold_Headphones_ASIO4ALL.png

Modded PLAY board V1.6 - headphones output (definitely less noise and spikes, also the 6.3 mm jack is 25 cm further from the toroid than RCA plugs are, hence I see no mains spikes here).

PLAY_orig_1KHz-RCA_Stereophile02-25.png

PLAY board V2.2 - 1KHz sinewave RCA output

PLAY_1KHz-RCA_Stereophile02-25.png

Modded PLAY board V1.6 - 1KHz sinewave RCA output (ignore the skirt...it's from ARTA, the noisefloor is indeed lower, but with the expense of harmonics)

PLAY_orig_1KHz-Headphones_Stereophile02-25.png

PLAY board V2.2 - headphones 1KHz sinewave (identical with the RCA-output graph)

PLAY_1KHz-Headphones_Stereophile02-25.png

Modded PLAY board V1.6- 1KHz sinewave (about the same harmonics, but lower background noise)
 
Last edited:
Top Bottom