mochi
valued member
Posts: 24 
|
Post by mochi on May 17, 2018 20:32:16 GMT
Hi,
I was just wondering if there would be a simple way of improving the sound of these pretty cheap amps, I'm using an external DAC and there seem to be a high pass filter for the line in, so ¿would it be possible to remove/bridge these caps without risk (the two in the upper left corner)?.
(Link with all pictures) imgur.com/a/GxDWR#5WA7Q8C
This seems to follow one of the sample TI designs for the TPA6120A2 with some modifications, but there also seems to be more filtering in the output stages.
Also, is there a way to purchase Garage1217 products (Ember for example), directly from Europe?, when buying from USA shipping costs and customs increase prices a lot.
Thanks and best regards. |
|
solderdude
Administrator
measureutternutter
Posts: 4,886
|
Post by solderdude on May 17, 2018 21:19:54 GMT
Alas no G1217 products from European soild.
There were plans but they did not materialize.
It's difficult in the pics to see but the small caps may well be on the input.
Can't see the values.
Maybe they are in a feedback loop maybe in the signal path.
In both cases they should not be polar.
Looks like the circuit has an +/- internal voltage rail those input caps may be bypassable.
The thicker ones are power supply caps.
Can't see if there are more caps in the signal path.
Most likely the BW limiting caps are somewhere in the signal switching path.
At one point the DAC output and line-in will be either switched or mixed.
The TPA also could have input caps.
Perhaps follow the traces from pin4 and pin17.
Maybe hard to follow as the PCB is multilayer... 2 layer ? 4 layer ?
|
|
mochi
valued member
Posts: 24
|
Post by mochi on May 17, 2018 23:13:59 GMT
|
|
solderdude
Administrator
measureutternutter
Posts: 4,886
|
Post by solderdude on May 18, 2018 4:41:28 GMT
Yes seems to be very close to the FiiO E09K.
Changing the input caps for a jumper may not do much for LF extension.
Input R needs to be < 10k for this capacitance to become a problem.
I think it may be safe to bypass it.
Would put a 100 R resistor across it first or measure to see if there is any DC across it.
When there is no DC over it under operation it is safe to replace it with a wire.
It is impossible to follow the signal as it jumps layers a lot.
The design seems to have a lot of PS decoupling with inductors and local decoupling caps.
Can you do an RMAA measurement or is there one out there that you know ?
|
|
mochi
valued member
Posts: 24
|
Post by mochi on May 18, 2018 6:34:59 GMT
Hi, I don't have a decent ADC so I won't be able to do any reliable measurements, there's a reddit thread where they say that distortion is high when using the líne outs:
https://www.reddit.com/r/headphones/comments/76f77u/attention_all_monoprice_dacamp_owners/
What I don't like very much is that they seem to use polar electrolytics all through the audio path which I suppose degrades sound quality.
Thanks.
|
|
solderdude
Administrator
measureutternutter
Posts: 4,886
|
Post by solderdude on May 18, 2018 6:58:36 GMT
The usage of electrolytics is not a problem as long as there is a DC bias across it.
In this case I don't think there is.
For small signals it isn't a problem but at line level signals (6Vpp easily) distortion could well increase.
What can be done is to mount a second (similar type/brand) in anti-parallel to the existing ones or simply bypass it or replace it with a higher value ceramic (at least X7R and > 50V).
Uttery pointless to bypass it with say 100nF for instance as 99% of all current will still pass through the electrolytic.
Just check if there is no DC across it and it may be safe to bypass it when there is no DC or just a few mV.
|
|
mochi
valued member
Posts: 24
|
Post by mochi on May 19, 2018 18:02:26 GMT
Thanks for the advice, I may leave it as its, the problem with this unit is that the internal DAC is just OK, and the sound can be improved with an external unit, I've tried with two units, one of them is a cheap chinese PCM5102A DAC :
I've removed and bridged the two output electrolytic caps with good result, sound is certainly cleaner after that, and this DAC sound pretty amazing for the price, with a warm and relaxed sound and pretty good soundstage. The problem is that the USB implementation is extremely picky, so the SA9227 is just detected in firmware update mode when the PC powers on and you need to unplug and replug it everytime. Moreover it picks a lot of usb noise unless it's plugged into a USB hub.
So I've tried another alternative, a cheap Topping D30, which seems to measure really well, but the sound character is completely different to the PCM5102A DAC, surely more neutral (I can't measure the cheap DAC), this combination is certainly more detailed but a little painful to listen long term to with bright headphones (AKG K7XX ), etc.
But as you say this amp doesn't seem to be really suited for modding, so I may well leave it as it is and look elsewhere, sound with the D30 is too lean for my taste. |
|
