Hi ! I bought the black-blue tpa3116 amplifier board. Connected to the RPI B + PiCobberDAC PCM5102 + rotary encoder software volume... and have some problems with sound 🙁(((
If you set a small volume, very like to hear some buzz. But there is nothing to pause it. Perhaps it is a matter of audio drivers?
It looks like a high-frequency noise is heard on quiet passages only when playing music. Also audible clicks when switching music tracks.
Can someone faced with this problem and can help me ? Thank you. [emoji3]
If you set a small volume, very like to hear some buzz. But there is nothing to pause it. Perhaps it is a matter of audio drivers?
An externally hosted image should be here but it was not working when we last tested it.
It looks like a high-frequency noise is heard on quiet passages only when playing music. Also audible clicks when switching music tracks.
Can someone faced with this problem and can help me ? Thank you. [emoji3]
Last edited:
I have a very similar configuration : RPI B + hifiberry+ tube buffer + black-blue tpa3116
For me, no bad noises. But I already have this issue with bad connections or ground loop or bad volume controller.
I don't think it is a driver issue.
Do you have same problems with another source like a PC ?
Do you have same problems without the rotary encoder?
Regards,
Christian
Sounds like a ground loop problem. Try using a longer audio connector between the RP and the amp to physically increase the distance and see if the noise pickup via RF is reduced. The ground loop will require that you use separate power supplies or have isolated DC-DC converters to power the RP from the same PS as the amp. I get similar noises when using a Bluetooth module to driver the amp. Solved with either separate PS or using the DC-DC converter (Murata MEE series).
Just cut one GND and make the new GND star point at your inputs. Or use a modem transformer. 🙂
I
Btw, has anyone seen this amp on Amazon, what variant is this? I like ordering from Amazon if possible.
http://www.amazon.com/SMAKN%C2%AE-TPA3116-Amplifier-Board-Class/dp/B00KCEHS54/ref=sr_1_1?ie=UTF8&qid=1419956186&sr=8-1&keywords=tpa3116d2
Seems like a nice compact board with candydrop tantalums on the LC filter, and film caps on inputs, SMD ceramics on bootstraps.
I finally got this board a few days ago. I bought his board from a shop at TB that claimed to be the designer. However, the board has different components as shown in the picture above. Instead of those orange caps, regular box caps (probably film caps) were used for the output filter. And the input caps are actually those "orange beads" instead of box film caps. I guess they used whatever parts that were available at the time manufacture.
My initial impression is favourable. The sound reminded me of the Audiobah green board. Human voice and string instruments sound natural, and has a lot of "air". The high is not bright subjectively and sounds clean. The base is good but, based on my memory (I blew my 2 Audiobah boards and no longer have them for comparison), not as strong as the Audio board. The soundstage was quite wide and deep, no complain. Overall, I would rate it above the original YJ red board and on par with the Audiobah board. For someone looking for something simple and sounds good, I recommend this board.
I do have some criticism of the board 1) the 3-pin input plastic connector was a bit smaller than the standard one used on the Audiobah and YJ black boards. Therefore, one need a new input lead. I decided to desolder it and replaced it with the regular sized one that I removed from the Audiobah board. Unfortunately, the solder blinded the 3 holes and that was not possible. I ended up solder the regular-sized connector side way to the input pad at the bottom of the board. 2) there is a light pop when the amp is switched on 3) the power on blue LED is very bright and distracting.
The volume pot on the board has an on-off switch which is convenient. The output inductors are 22 uH and suits 8 ohms speaker better (not that the 10 uH ones has any problem). The is an additional input in the form a phone jack. The soldering job was quite clean and well done. There are a lot of instructions (all in Chinese) on the bottom side of the board explaining how to turn the board into PBTL configuration (there are 5 links to join and one to break) and information on other features of the board (where and how to tap the 5 V supply). The board is set for a gain of 32 dB instead of the usual 26. Now even I use my TVC (with the volume pot on the board turn to full) as the "preamp", I have to turn it down. With the Audiobah, the TVC has to be turned up at least half way to get to similar sound level. That 6 dB makes quite a difference.
As a side note, I also ordered a version of the YJ Black from another vendor, and plans to mod it. I hooked up this board and listened to it briefly and I was disappointed. I rate it below the Red YJ. The base was vague, the human voice was flat. After 15-20 minutes, I stop listening. Looking at this board, I realised that any attempts to alter the board will very likely damage the board seriously (the soldering pads are very small). At this point I do not think I will use it for anything. I think I am going to get DUG's and/or wait for GMASH's board.
Hope these information are helpful to someone.
Regards,
Have a small problems with the YJ 2.1 amp. The sub output gives steady 300mVAC output, without any input signal. My Fluke tells it is 117kHz, so there is obviously something wrong? Of course I don't hear anything, but could it be danger to the loudspeaker? Or early sign of coming electric smoke?
Another issue, the sub filter FR is just crap. Peak at 100Hz, -4dB @60Hz.....
Has anybody modded that filter part of the amp?
-Erkki
Another issue, the sub filter FR is just crap. Peak at 100Hz, -4dB @60Hz.....
Has anybody modded that filter part of the amp?
-Erkki
Isn't switching very fast, if other outputs you do measure ~400khz you could replace resistor/capacitor for synch, Master clock doesn't get to Slave chip then. Amp is working in idle too, but 300mV seems much to me
I realize I'm a bit late in picking up on this, but its music to my ears nevertheless 🙂 I've been thinking about PLLXOs too, I even have a paper design for one worked out.
Just using R and C won't work if you want more than first order (which you do appear to want). You'll need inductors and they are hard to drive because in the main they are low impedance devices to get the size manageable. High impedance inductors are possible but are bulky and expensive and need to be custom designed.
Fortunately there is a solution - use transformers to translate from high impedance to low. It won't be purely passive beyond the XO because there'll be some gain that needs to be made up. So with a good sounding gain block after the XO that's one solution that I'm working on. I have in mind AD603 for the gain block - that also comes with a built-in DC controlled volume element. So no need for any fancy audiophile pot and also the DC control allows remoting and what's more paralleling additional channels. So those who need more than just stereo are catered for too.
I have to check if I measured the frequency correctly, by checking the half bridge output directly. Stereo channels show only 2 or 3mV RMS noise, and DVM cannot detect that frequency properly (64kHz..).
You can get 2nd order with a pair of caps and resistors as shown here:
TLS.org | Passive Line-Level Crossover
If you add a gain stage in between and cascade them you can get 4th order without any inductors. But I agree that using small signal level high inpedance inductors is another possibility. I couldn't find a calculator for that topology other than to use a XO sim and use high impedance loads.
What we need is for someone to make a new 2.1 amp board from the ground up that has an active XO circuit with a high pass on the stereo channels and a low pass on the sub. Xo point should go from 50 to 350Hz for FAST systems.
Last edited:
That's what I would call two cascaded first orders rather than a second order. I agree the slope ends up being 2nd order but the Q is lower than a normal 2nd order. It looks highly kludgey - for a start I'd not want any preamp to see a 5k load and I'd also not want an XO's corner frequency to depend on the poweramp's input Z. Seems too many compromises there to be sure of getting the best sound.
I'm ignorant of what 'FAST' systems are - the sub channel is summed L+R and low-passed with a setting between 50 and 350Hz? Does that frequency need to be infinitely variable (as might be achieved with an OTA) or are a few preset frequencies OK? And does the high-pass need to track the low-pass setting? If so perhaps high-pass by subtraction and compensating delay from low-pass might be the best solution.
I'm ignorant of what 'FAST' systems are - the sub channel is summed L+R and low-passed with a setting between 50 and 350Hz? Does that frequency need to be infinitely variable (as might be achieved with an OTA) or are a few preset frequencies OK? And does the high-pass need to track the low-pass setting? If so perhaps high-pass by subtraction and compensating delay from low-pass might be the best solution.
I'll pencil that one in after the Boominator card and 'proper TPA' designs get done 🙂
First project, CSR8670 + TPA3118 2.1 system
Hi there,
I have zero experience in DIY audio, though I did some work for a company that designs Bluetooth speakers (am a software guy with some physics background).
Now I'm trying to build my own 2.1 soundbar based on CSR8670. The amp could be either TPA3118 or CSR's own DDFA amp.
I honestly don't even know where to start, because I can't design PCBs myself, so I'd imagine I'd need to find ready-made boards for both the Bluetooth/DSP chip and the amps. Or, if I find someone to design the board I could order it on e.g. CircuitHub.com. Software is not a problem, I've done it before.
To make it even more complicated for myself, I want to add VU meters and nixies, vintage style.
The end goal is to go Kickstarter if the design proves to be good.
I understand this is a very broad request, but I'd appreciate any tips and suggestions. I'm really itching to do this project 🙂
Hi there,
I have zero experience in DIY audio, though I did some work for a company that designs Bluetooth speakers (am a software guy with some physics background).
Now I'm trying to build my own 2.1 soundbar based on CSR8670. The amp could be either TPA3118 or CSR's own DDFA amp.
I honestly don't even know where to start, because I can't design PCBs myself, so I'd imagine I'd need to find ready-made boards for both the Bluetooth/DSP chip and the amps. Or, if I find someone to design the board I could order it on e.g. CircuitHub.com. Software is not a problem, I've done it before.
To make it even more complicated for myself, I want to add VU meters and nixies, vintage style.
The end goal is to go Kickstarter if the design proves to be good.
I understand this is a very broad request, but I'd appreciate any tips and suggestions. I'm really itching to do this project 🙂
I found ! ! ! Problem in "software volume" !
When switching on web gui "software volume" off and turn on "hardware volume" is the buzz disappears.
But I need to adjust the volume rotary encoder !!!
What should I do ???
Is anybody any thoughts on this?
Helps! Help! Help! 🙂))))
Anybody...[emoji30]
When switching on web gui "software volume" off and turn on "hardware volume" is the buzz disappears.
But I need to adjust the volume rotary encoder !!!
What should I do ???
Is anybody any thoughts on this?
Helps! Help! Help! 🙂))))
Anybody...[emoji30]
Have anyone tried this board? It looks like its filterless; TPA3116D2 2 0 Stereo Sound Dual Channel HiFi Large Power Digital Amplifier | eBay