• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required as is 20 years of participation in forums (not all true). Come here to have fun, be ready to be teased and not take online life too seriously. We now measure and review equipment for free! Click here for details.

Burson FUN (naked pics, some technical thoughts a.s.o.)

trl

Senior Member
King of Mods
Joined
Feb 28, 2018
Messages
371
Likes
222
Location
Iasi, RO
#1
FUN under test in the last month. Very good an powerful sound, no oscillation issues found on scope, so it's able to deliver up to 2W/30Ohms of RMS power without struggling. Also, @600Ohms it can deliver a bit over 170mW RMS power.

IMG_5177_.jpg


Sound is neutral, RMAA shows perfectly flat freq response, as it should be. However, I was able to see a tiny-little phase-shift on trebles, just enough to mellow and warm the sound a little bit, making the songs a bot more laid back.

FUN_RMAA_FR.png

Frequency response

FUN_RMAA_Dynamic.png

Dynamic

FUN_RMAA_SNR.png

Signal/Noise Ratio

Lot of power in a small package, I kinda like it. FUN is powered from 2 x dual +/-17V rails, so a total of 34V for the output stage to be able to kick the bass into planars perfectly. There are 4 x SMPS power regulators each one delivering + or - 17V, so each channel is independently powered.

Output DC-voltage was somewhere between 1.5mV and 3.5mV, depending a lot of the op-amps used and the burn-in time (usually after 20 minutes the DC-voltage could get lower). However, few mV are normal and pretty common with most amplifier, hence I consider this amp can be used with IEMs as well, especially that the background noise is even lower than PLAY's internal amplifier.

AdobePhotoshopExpress_0e038520b7f64c58ae2da7c4b30cb2ba.jpg AdobePhotoshopExpress_aad337ad202d46dd974acca187a09ffd.jpg
 
Last edited:

restorer-john

Major Contributor
Joined
Mar 1, 2018
Messages
1,333
Likes
1,649
Location
Gold Coast, Queensland, Australia
#2
There are 4 x SMPS power regulators each one delivering + or - 17V, so each channel is independently powered.
'Your' naked picture posted above shows nothing of the sort.

Look at your picture. There are two Linear Tech 5 volt, three terminal, low dropout linear regulators, clearly to power a volume/display etc, and two XLSemi XL6019 buck/boost switchers to power the amplifier stage from the one, common to both channels, +/- supply.

The picture you have posted above appears to be of the Burson Play with built in D/A and a digital display.

If 'your measurements' for the rails are correct at +/-17v, you will have an swing of around 11 Volts RMS max after output Tr CE drop and R losses. Into 600 ohms, the unit would struggle to hit 200mW, so your 170mW is probably the limit. It is difficult to determine the topology visually, due to the double layer board, it may be running BTL, but I doubt it.

The Burson FUN on the other hand, does have 4 x XLSemi XL6019 buck/boost PWM power rails and uses a completely different PCB to 'your' photograph above. It is apparently a Class A unit.

According to our friends at Burson : "Its symmetrical circuitry is powered by four sets of Max Current Power Supply (MCPS) developed by Burson. The revolutionary MCPS is far superior to traditional transformers delivering instant, clean, and maximum electric current to the Fun. " What a load of BS.

It seems Burson are at their usual tricks again, claiming they 'developed' a power supply by simply buying an off the shelf buck converter IC and using it in a decidedly non 'dual mono' design. But then again, clearly their market for these products are merely geeky, pimple-faced gamers who think lighting up the interior of their liquid cooled PCs with lurid colors of LEDs to match their keyboards and 'gaming' mice is the coolest thing in the world.

Hint: Dual mono means two completely separate power transformers, not a single power supply split up into rails inside the unit.
 
Last edited:

trl

Senior Member
King of Mods
Joined
Feb 28, 2018
Messages
371
Likes
222
Location
Iasi, RO
#3
This is happening when posting before the morning coffee, thanks for pointing this out, John. I will change that picture and post some more as well.

IMG_5157_.jpg


IMG_5160_.jpg


IMG_5164_.jpg


IMG_5170_.jpg
 

trl

Senior Member
King of Mods
Joined
Feb 28, 2018
Messages
371
Likes
222
Location
Iasi, RO
#4
John, I totally fell you and understand your point, but giving enough power into such a small body and for such a price I would say that it's quite a good thing after all.

The primary SMPS is simmilar with the one used in PLAY, 12V/6A, just a better 12V plug to get a better grip inside the unit. It's very low noise actually, up to 10MHz I could only see on my scope just few "glitches" of few mV each, so way much better than most laptop SMPS:

Here are the 4 boost converters (2 x XL4015E1, 2 x XL6019E1), each being able to deliver up to 5 Amps:
IMG_5171_.jpg

IMG_5172_.jpg

IMG_5176_.jpg

I find these boost converters a very good approach for a device that should be used outside and inside the computer's case, depending on everyone's mood of the day. Basically, connecting it inside the computer and powering ON via the MOLEX plug will make the computer to sound a very good addition for a gamer and, why not, for a guy who want to listen to music under decent conditions. On the V1.6 of PLAY I was able to measure some strange noise (above 150kHz, but there were harmonics that got into the audible range, especially when IEMs were used), since V2.0 of PLAY the PCB was much improved by increasing the ground plane and the no. of caps. However, the FUN benefits of the same increased power plane and also lot of polymer caps to combat the ripple and noise, so this is why it's so quiet perhaps.
 

trl

Senior Member
King of Mods
Joined
Feb 28, 2018
Messages
371
Likes
222
Location
Iasi, RO
#5
'Your' naked picture posted above shows nothing of the sort.
[...]
If 'your measurements' for the rails are correct at +/-17v, you will have an swing of around 11 Volts RMS max after output Tr CE drop and R losses. Into 600 ohms, the unit would struggle to hit 200mW, so your 170mW is probably the limit. It is difficult to determine the topology visually, due to the double layer board, it may be running BTL, but I doubt it.
[...]
Screen Shot 2018-07-27 at 21.45.19 copy.png

10.15V RMS perfect sine-waves, without any signs of distortion, I would call this 171.7mW @600Ohms.

Screen Shot 2018-07-27 at 21.45.13 copy.png

At 11.12V RMS there are visible signs of distortions, but that means 206mW @600Ohms, like John said it's beyond amplifier's limit.
Good that my 600 Ohms DT880 are labeled to 100mW max. power handling. :) However, not sure what 600Ohms cans need more power anyway...

Screen Shot 2018-07-29 at 20.46.03 copy.png

7.638V RMS @ 29.5Ohms, so about 2W/channel.

Screen Shot 2018-07-27 at 19.59.18 copy.png

Red sine is the input signal, blue one is the output from FUN's output plug. I would call this the perfect bass with no roll-off to 10Hz.

Screen Shot 2018-07-27 at 20.00.48 copy.png

@20KHz there's a tiny phase-shift, most likely caused by an aggressive low-pass filter. Purists might say something here, but in the A/B tests I've done I got just perfect trebles and music sounded neutral, melodious and a bit laid-back. However, for frequencies below 20KHz there's negligible to no phase-shift; also, RMAA tests show no roll-off at all.

Screen Shot 2018-07-27 at 20.06.17 copy.png

20Hz square signal, output identical with the input (sorry about the non-perfect squares, but it's AC-coupled; with DC-coupled it looks perfect)

Screen Shot 2018-07-27 at 20.02.52 copy.png

20KHz square signal (a bit latency in raise and fall, perhaps this is why the sound is a bit laid-back, which actually I like it more)

'Your' naked picture posted above shows nothing of the sort.
[...]
It seems Burson are at their usual tricks again, claiming they 'developed' a power supply by simply buying an off the shelf buck converter IC and using it in a decidedly non 'dual mono' design. But then again, clearly their market for these products are merely geeky, pimple-faced gamers who think lighting up the interior of their liquid cooled PCs with lurid colors of LEDs to match their keyboards and 'gaming' mice is the coolest thing in the world.

Hint: Dual mono means two completely separate power transformers, not a single power supply split up into rails inside the unit.
I actually started to like their trick, though I'm into the Linear PSU if you'd ask me. However, FUN measured pretty well and sounds very good with all my cans, from 16Ohms IEMs to 600Ohms Beyers, so their Basic version I recommend it for it's price/performance ratio and the 5-years warranty. I would probably replace the NE5534 with AD797, but nothing else.

P.S.: John, I don't want to even think what you'll say when I'll post thoughts about the BANG. :)
 

amirm

Founder/Admin
Staff Member
CFO (Chief Fun Officer)
Joined
Feb 13, 2016
Messages
16,218
Likes
11,466
Location
Seattle Area
#6
Lot of power in a small package, I kinda like it. FUN is powered from 2 x dual +/-17V rails, so a total of 34V for the output stage to be able to kick the bass into planars perfectly.
You read my mind as I was thinking of designing an amp with high voltage rails (relative to typical headphone amps) to get much more power. Well done!
 

trl

Senior Member
King of Mods
Joined
Feb 28, 2018
Messages
371
Likes
222
Location
Iasi, RO
#7
The easiest one I can think now it would be LME49860+LME49610 (+/-22V) and on outputs get a small resistor and an inductor, not the 10Ohms resistor (Ventus headamp perhaps). This should output >15V RMS on headphones plug and Benchmark guys love LME chips enough to trust them.
 

restorer-john

Major Contributor
Joined
Mar 1, 2018
Messages
1,333
Likes
1,649
Location
Gold Coast, Queensland, Australia
#8
as I was thinking of designing an amp with high voltage rails (relative to typical headphone amps) to get much more power.
Pretty much all the top model integrated amplifiers from the late 80s onwards used high gain power stages with no active (read: noise generating) preamplifier stages up front.

Used in 'direct' or 'bypass' mode, and removing the series headphone resistors (put there to prevent burning out low R headphones), you are getting exactly the same thing. Not only that, you can run speakers as well.

Realistically though, this pursuit of 'more power' for headphones is not valid- it's really about more voltage swing into high impedance cans- the actual 'power' (e^2/R) is still only in milliwatts.
 

March Audio

Major Contributor
Manufacturer
Patreon Donor
Joined
Mar 1, 2016
Messages
2,805
Likes
1,997
Location
Perth Western Australia
#10
FUN under test in the last month. Very good an powerful sound, no oscillation issues found on scope, so it's able to deliver up to 2W/30Ohms of RMS power without struggling. Also, @600Ohms it can deliver a bit over 170mW RMS power.

View attachment 14568

Sound is neutral, RMAA shows perfectly flat freq response, as it should be. However, I was able to see a tiny-little phase-shift on trebles, just enough to mellow and warm the sound a little bit, making the songs a bot more laid back.

View attachment 14560
Frequency response

View attachment 14559
Dynamic

View attachment 14561
Signal/Noise Ratio

Lot of power in a small package, I kinda like it. FUN is powered from 2 x dual +/-17V rails, so a total of 34V for the output stage to be able to kick the bass into planars perfectly. There are 4 x SMPS power regulators each one delivering + or - 17V, so each channel is independently powered.

Output DC-voltage was somewhere between 1.5mV and 3.5mV, depending a lot of the op-amps used and the burn-in time (usually after 20 minutes the DC-voltage could get lower). However, few mV are normal and pretty common with most amplifier, hence I consider this amp can be used with IEMs as well, especially that the background noise is even lower than PLAY's internal amplifier.

View attachment 14557 View attachment 14558
Can I ask what's the issue with all the mains and harmonics pickup? Measurement problem or the amp?
 

trl

Senior Member
King of Mods
Joined
Feb 28, 2018
Messages
371
Likes
222
Location
Iasi, RO
#11
It gets injected from mains into the RCA cable. Funny thing, if I use an isolation transformer the mains hum disappears from the measurements.
However, that hum is non audible, even if it's picked up by my RCA cable.

See the 2 photos from https://www.audiosciencereview.com/forum/index.php?threads/isolation-transformers.3672/#post-137346, first is amp connected to mains, 2nd pic is same amp connected via isolation transformer. Under both circumstances there's no audible noise on my sensitive IEMs (vol. to the max., inputs shorted).
 
Joined
Nov 6, 2018
Messages
87
Likes
66
Location
.de
#12
You could probably replace the switch-mode wall wart with a linear supply and find the same. I would assume it's mains filter leakage current that is making its way downtown downstream to the PC's protective earth connection (presumably while humming along to Vanessa Carlton's hit "A Thousand Miles" ;) [1]). Depending on how the filtering is done, turning the wall wart around may make a difference.

I hope you have gotten the sample rate issue obvious in the first post sorted out by now. If not, I have some experience with cranky sound drivers...

[1] Those of you wondering what she's doing nowadays, she's still around and sounds way more "indie" these days. Her last one Liberman is a great album IMHO.
 

trl

Senior Member
King of Mods
Joined
Feb 28, 2018
Messages
371
Likes
222
Location
Iasi, RO
#13
Moving the wall wart is indeed changing the spikes on the ARTA/RMAA, but this noise it's inly induced in the RCA cables, not into the audio chain.
Not sure if a LPSU will help much, but I'll probably give it a try, thanks!
 
Joined
Nov 6, 2018
Messages
87
Likes
66
Location
.de
#14
This is the scenario in a nutshell:
Code:
                                             _
L o------+------/~/...     /|------o RCA sig  \
         |       OUT o---+| |                 |
        === C            | \|-+                > --+
         |   flt         +--|-+               |    |
         +-------[GND]------+------o RCA GND _/    |
         |                                         |
        === C                                      |
         |   flt                                   |
N o------+------/~/...                             |
                                                   |
                  +---------+                      |
                  |         |      _               |
                  |     OUT o-----o \              |
                  |         |       |              |
                  |  PC     |        > ------------+
                  |         |       |
PE o--------------+-----GND-o-----o_/
                  |         |
                  |         |
                  +---------+
Mains filter capacitors C_flt attempt to pull secondary-side GND to (V(L)-V(N))/2 (half mains voltage), while the source is convinced it belongs connected to protective earth. Some current flow is unavoidable. A similarly-sized transformer would still have some parasitic coupling between primary and secondary but probably just a few hundred pF at worst, while we are talking caps in the order of 1-2.2 nF here.
 

trl

Senior Member
King of Mods
Joined
Feb 28, 2018
Messages
371
Likes
222
Location
Iasi, RO
#15
Thanks for the comments, but there's no Earth on this scenario. There're only L and N, so GND is floating.

I wonder if I should connect a wire from Earth (PE) to amplifier's case and use Y caps between L/PE and N/PE; of course, I can use X2 caps between L & N as well, but I'm sure there're already installed these caps inside the SMPS power brick, so...it would be redundant.

AFAIK, regular speakers amplifiers are usually using L & N only and the GND written on the metallic case is only to be used for audio GND and not for PE. However, computer cases are using the Y2 caps between PE and L/N just to ensure safety for low current leakage, so...me again: should I try connecting FUN's metallic case to PE or not? :)
 

trl

Senior Member
King of Mods
Joined
Feb 28, 2018
Messages
371
Likes
222
Location
Iasi, RO
#16
[...]
Mains filter capacitors C_flt attempt to pull secondary-side GND to (V(L)-V(N))/2 (half mains voltage), while the source is convinced it belongs connected to protective earth. Some current flow is unavoidable. A similarly-sized transformer would still have some parasitic coupling between primary and secondary but probably just a few hundred pF at worst, while we are talking caps in the order of 1-2.2 nF here.
I was reading again your message. Tests were done with no inputs/outputs connected to the amplifier; also, tests were done with a floating laptop, on battery. Thank you!
 
Top Bottom