Ground loops are by far the most common problem with connecting desktop PCs to audio equipment. New posts about ground loops problems pop up constantly.
So you need to eliminate the ground loop. IMO the easiest and most reliable thing in this case is to use an insulation class II power adapter for your amp Appliance classes - Wikipedia which does not use the protection earth wire (which creates the ground loop in your case).
USB soundcard may not help, if the ground of the USB output on the motherboard is still far away from the PSU terminal and the board carries large currents in the return section used by the USB ground. It is often the case.
Any older notebook adapter will do, provided it has enough power. In my project I measured the TPA3116 distortion and powered with 20V, the distortion started rising at 2x18W/8ohm. IMO 60W for your 3ch setup will suffice, 90W would be better. I am sure if you search shops with used PC/old junk in Greece, you will find such an adapter for almost free. Otherwise buy from ebay from Europe, e.g. Genuine Original 2-Power 19.5V 4.61A 90W AC Power Supply Charger Adapter PSU | eBay
So you need to eliminate the ground loop. IMO the easiest and most reliable thing in this case is to use an insulation class II power adapter for your amp Appliance classes - Wikipedia which does not use the protection earth wire (which creates the ground loop in your case).
USB soundcard may not help, if the ground of the USB output on the motherboard is still far away from the PSU terminal and the board carries large currents in the return section used by the USB ground. It is often the case.
Any older notebook adapter will do, provided it has enough power. In my project I measured the TPA3116 distortion and powered with 20V, the distortion started rising at 2x18W/8ohm. IMO 60W for your 3ch setup will suffice, 90W would be better. I am sure if you search shops with used PC/old junk in Greece, you will find such an adapter for almost free. Otherwise buy from ebay from Europe, e.g. Genuine Original 2-Power 19.5V 4.61A 90W AC Power Supply Charger Adapter PSU | eBay
You are right, ar 40 Euros this sound card is not worth it. I got one for 12 Euros inc. P&P.
I needed something driven by a laptop with linear response and a 5 Volt microphone input + an optical output, which it does very well. For loudspeaker measurements there is not really much sense in anything sampling more than 48kHz.
Concerning the passive transformer. Ask you self what you want: You want to have no DC to keep noise away. Any audio device will have a roll off at two sides of the spectrum, if it is not DC coupled. So there will be, maybe -3dB at 20 Hz and 20 kHz. Now ask yourself again, what does your sound system deliver? Be honest, sub 40 Hz is not realistic. If you have a roll off, just push the bass up a notch and you compensated for this. What negative effects do you think, has a reversed phase? Just a tipp: The music does not play backwards.
In this forum the stupid talk is filtered out a bit, other forums are lead by the most stupid talkers...
In the end, you have to try your self. The Dynavox will work, change the sound subjectively a little, as you never will be able to do a perfectly leveled A/B comparison.
In a blind test some will like the transformer "sound" even better, just like some like tubes in the audio chain.
PS Ah, China. No I just yesterday got some stuff, perfectly in time. I mentioned that you have to plan in advance, a little, as the transfer times may vary. Like at our homes, there are quick and slow working company´s.
I needed something driven by a laptop with linear response and a 5 Volt microphone input + an optical output, which it does very well. For loudspeaker measurements there is not really much sense in anything sampling more than 48kHz.
Concerning the passive transformer. Ask you self what you want: You want to have no DC to keep noise away. Any audio device will have a roll off at two sides of the spectrum, if it is not DC coupled. So there will be, maybe -3dB at 20 Hz and 20 kHz. Now ask yourself again, what does your sound system deliver? Be honest, sub 40 Hz is not realistic. If you have a roll off, just push the bass up a notch and you compensated for this. What negative effects do you think, has a reversed phase? Just a tipp: The music does not play backwards.
In this forum the stupid talk is filtered out a bit, other forums are lead by the most stupid talkers...
In the end, you have to try your self. The Dynavox will work, change the sound subjectively a little, as you never will be able to do a perfectly leveled A/B comparison.
In a blind test some will like the transformer "sound" even better, just like some like tubes in the audio chain.
PS Ah, China. No I just yesterday got some stuff, perfectly in time. I mentioned that you have to plan in advance, a little, as the transfer times may vary. Like at our homes, there are quick and slow working company´s.
Last edited:
Thank you for your help. I decided to first check them with a Bluetooth adapter if the latency isn't bad i guess i will stay with that, otherwise i will order the dynavox isolator.
I just received one of the 2.1 boards with variable frequency control, identical to this one:
TPA3116 2.1 50Wx2+100W Amplifier Board Digital HIFI 12-24V Super Bass Speaker | eBay
It was very well packed and arrived in perfect physical condition. However, there seems to be a fault with the bass volume potentiometer. Looking at the amplifier from the front. the bass volume pot is second from left. When I rotate this pot anticlockwise to decrease the bass volume, it does not reduce the bass output to nil - it is still very audible with the bass volume pot at 0%. If I then rotate the pot clockwise, bass volume will increase as expected, until the pot is at about 50%, when the bass cuts out completely. If I continue rotating the pot clockwise, I will get bass output again at about 80% of the pot's full rotation, although the output/volume is much lower than at 50%. This output will increase until the pot is at 100% clockwise rotation.
Is the potientiometer dodgy or is the fault elsewhere? I'm waiting for an answer from the seller but I don't expect too much. If it's just the pot then I'll happily swop it out.
TPA3116 2.1 50Wx2+100W Amplifier Board Digital HIFI 12-24V Super Bass Speaker | eBay
It was very well packed and arrived in perfect physical condition. However, there seems to be a fault with the bass volume potentiometer. Looking at the amplifier from the front. the bass volume pot is second from left. When I rotate this pot anticlockwise to decrease the bass volume, it does not reduce the bass output to nil - it is still very audible with the bass volume pot at 0%. If I then rotate the pot clockwise, bass volume will increase as expected, until the pot is at about 50%, when the bass cuts out completely. If I continue rotating the pot clockwise, I will get bass output again at about 80% of the pot's full rotation, although the output/volume is much lower than at 50%. This output will increase until the pot is at 100% clockwise rotation.
Is the potientiometer dodgy or is the fault elsewhere? I'm waiting for an answer from the seller but I don't expect too much. If it's just the pot then I'll happily swop it out.
I used to have this board but one channel failed on 2nd day (I took a video of the problem, full refund from seller) but I'm using 2 similar boards (one with BT, one without) with newer layout which have the big capacitor on the other side and the pot function placements are different, they are working flawlessly.
I don't remember I have the problem like what you have, and definitely not the ones I'm using. When the LR pot is at 0, I can still hear very very tiny sound from front speakers, I think probably same for the bass pot, but since I'm using passive subwoofer, don't think it's enough to drive it. Volumes increase as expected when turning clockwise.
Last edited:
As per post #11245, I pirated a similar potentiometer from one of my other amps and the new amp now works as expected.
My only concern is that some of the components are really hot - I haven't played the amp anywhere near its capacity and the six square components at the rear of the board (chokes?) are hot, with the two components nearest to the large capacitor being too hot to touch for more than a second. Ambient temp is about 23C, component temp is 61C. Is this a normal condition for these particular boards? Amp is powered by a 24v SMPS that I have used for ages on other TPA3116 boards..
Edit:Just tested with 19v laptop brick, same temps experienced.
My only concern is that some of the components are really hot - I haven't played the amp anywhere near its capacity and the six square components at the rear of the board (chokes?) are hot, with the two components nearest to the large capacitor being too hot to touch for more than a second. Ambient temp is about 23C, component temp is 61C. Is this a normal condition for these particular boards? Amp is powered by a 24v SMPS that I have used for ages on other TPA3116 boards..
Edit:Just tested with 19v laptop brick, same temps experienced.
Last edited:
![IMG_1801[1].jpg](/community/data/attachments/717/717102-041d332893d1a36e7e1c659fa3db3e6e.jpg)
Hopefully replacing the pot cured your problem but I would just note that what you are calling corrosion looks like, is likely to be, the through hole plating/pad has been pulled away from the board on desoldering. Specifically the centre rear pin. You might want to check that the new component is properly connected. No idea about heat in the output inductors but 150u/680n suggests FCO 16KHz @15R and the components look a little bit dinky for the power levels. Assuming they are not saturating then you might consider sticking an additional heatsink on top of them, thermal epoxy, for peace of mind.
Connected as per the photo. Sub is 4 ohm. The other TPA3116 and TPA3118 amps that I've used with the same inputs and speakers have never got to this temperature. I currently have the amp switched on but with no source playing. Temp on the two chokes? to the left of those shown is about 50C, ambient is 21C.
Last edited:
No load you would be looking at resistive losses due to ripple current in the windings and magnetic losses due to hysteresis losses in the cores. On load the copper losses would rise but core losses would remain about the same. In terms of size and construction then like for like the only real variable would be core material. A higher loss material need not affect nominal saturation performance but as the word higher implies the core will run hotter and the saturation level will decrease. It need not be totally squeaky bum if the sums work and naturally it saves them money. Possibly more of a reliability issue for the big capacitor.
If the pot is working as expected (as gain on the bass output), can I assume that it is properly connected? Is there any way a somewhat poor connection on one of the pins could be allied to the heat noted in the output inductors?
Everything is connected as per photo in post #11249.
I used the amp for several hours yesterday, works great but the output inductors run at 50-60C, which is a bit worrying.
Last edited:
Loosing some of the plating in that particular area need not be a problem. Low level signal stuff. As long as a track connection is made it should be good enough. Given you can run it long term and assuming there are no audible anomalies the the best I would guess is that the manufacturer has shaved production costs by using a higher loss core material. Other than being hot and a bit of a mind niggle it is not affecting performance. Glue on some heatsinks if you are really worried but otherwise kick back and listen.