Both units are on their own island, voltage wise. The commons may be at the same potential but the signal(s) are 7 volts different, because of the voltage sources. (One is 12 volt, the Bluetooth is at 5 volts)
uhh AFAIK there are 2 gnd systems on each board so total 4 gnd connections.
GND is not GND
I find it better to think of return paths of currents that the signal and power take.
GND is not GND
I find it better to think of return paths of currents that the signal and power take.
No. Both are 3.3V. The Bluetooth modules output section on the source is regulated to 3.3V by an internal voltage regulator. The input section on the power amp is regulated to 3.3V by an internal voltage regulator.
The OP said he did that, = no joy
trace the return currents of the hot signal inputs sourced from the MP3 board to amp then returning back to the MP3 again ( if you cut the signal common , the signal currents now have to return through both boards power supply returns! now the small AF signal is combined with pulsating power IR drops = bad)
Last edited:
Hi Saturnus and thanks for that once again,
I'm a total newbie in DIY audio stuff so I'll give it another try on this one, I'm going to explain what I did (about one month ago, before I started to read on how to fix Ground Loops) so you can confirm (or not) if that's the solution you are mentioning so I can try it again (I might have done something wrong eventually).
The Source (Audio Module) comes with 3 Output wires (Lout, Gout, Rout) and the Amp comes with 3 Input wires (Lin, Gin, Rin), what I did initially after "discovering" the Ground Loop was simply leave the G wire of both the Input and Output disconnected, in other words, I connected Lin to Lout and Rin to Rou and put some insulation tape around the Gin and again some insulation tape around Gout, so basically both G were not connected in anyway (that I can think of at least).
Is this what you are mentioning by not using the Signal Ground or I do need to take aditional steps? I will retry it just to be 100% sure but I'm almost certain that I did it exactly the way I described and the irritanting noise didn't disappeared or got any better.
Thanks a lot.
No. Both are 3.3V. The Bluetooth modules output section on the source is regulated to 3.3V by an internal voltage regulator. The input section on the power amp is regulated to 3.3V by an internal voltage regulator.
Perhaps but this is not the issue. Each unit has its own power supply. The Bluetooth unit immediately changes the 12 v.d.c. to 5 v.d.c. for its supply. The amp used the 12 v.d.c. direct from the battery for the amplification and may have onboard regulators for other functions. (Clock circuit, preamp section, ect..) The only thing common to both units is the ground, not the signal..
I doubt that any of the voltage conversion on either board is an isolated output type. Even though the voltages may be the same in some places on both boards, they still regulate from different sources. One direct from the battery and the other (the Bluetooth) after a 5 volt regulator..
Perhaps but this is not the issue. Each unit has its own power supply. The Bluetooth unit immediately changes the 12 v.d.c. to 5 v.d.c. for its supply. The amp used the 12 v.d.c. direct from the battery for the amplification and may have onboard regulators for other functions. (Clock circuit, preamp section, ect..) The only thing common to both units is the ground, not the signal..
I doubt that any of the voltage conversion on either board is an isolated output type. Even though the voltages may be the same in some places on both boards, they still regulate from different sources. One direct from the battery and the other (the Bluetooth) after a 5 volt regulator..
Xarlitos, yes. What you describe is precisely what will solve the ground loop problem for this specific amp used with this specific mp3 module.
If there is still noise there might be a mechanical ground loop as well or maybe you have connected a charger to the battery at the same time as playing. Chargers and low cost power supplies in general are notoriously noisy. If you're still having problems, and you can find nothing else wrong then I don't know really. Cutting the signal ground solved the issue for me... using the same amp (well, not really the exact same but the chip is the same TDA749x series) and the bluetooth chip might be a different one than mine but since it is clearly a CSR chip of the low cost series so they are pretty much identical. Mine was a CSR8640, btw.
If there is still noise there might be a mechanical ground loop as well or maybe you have connected a charger to the battery at the same time as playing. Chargers and low cost power supplies in general are notoriously noisy. If you're still having problems, and you can find nothing else wrong then I don't know really. Cutting the signal ground solved the issue for me... using the same amp (well, not really the exact same but the chip is the same TDA749x series) and the bluetooth chip might be a different one than mine but since it is clearly a CSR chip of the low cost series so they are pretty much identical. Mine was a CSR8640, btw.
Last edited:
Depending on how the designer wants it, "signal ground" node may not be same as PSU 0V with a single-ended supply.
There is probably dc-blocking capacitors on the signal lines on one or both of the PCB already. But in addition to that, try a dc-blocking cap on the "signal ground" too.
Anyway the 100ohm I mentioned earlier is to reduce the load impedance at the load (amp) side so less noise voltage is generated from EMI. If EMI isn't the problem then ignore that.
There is probably dc-blocking capacitors on the signal lines on one or both of the PCB already. But in addition to that, try a dc-blocking cap on the "signal ground" too.
Anyway the 100ohm I mentioned earlier is to reduce the load impedance at the load (amp) side so less noise voltage is generated from EMI. If EMI isn't the problem then ignore that.
Last edited:
Exactly. Just because it's called ground doesn't mean it's common to the power source..
Ok.. I'll go with that, being I don't have either unit here to use a VOM on. It's that easy to know for sure..
Hi,
I can do that for you, just tell me what measurements you need me to do and I'll have it done, maybe you're both right and I just have a defective IC...that would ruin my day, but it's a possibility.
Thanks
Hi again,
Just made some testing, tried all combinations between Source-Isolator-Amp, the end audio is always the same (= No Bass or lower-middles), even if I totally reverse the isolator (connect the Isolator Input to the Amp and the Isolator Output to the Source...)...This last step, the totally reversed one without any noticeable changes, seems a little odd for me, any thoughts if it's normal?
Any chance I screwed-up when wiring the L,G,R output&input to the sepparate Female and Male RCA connectors? Seemed pretty straight forward when I did it, all 4 RCA connectors had a red wire and a loose wire inside, red wire for the signal and loose wire for the ground, right? To the ground I had to Split the Common ground of the input/output to feed both RCA connectors on each side (attached)...Just want to make sure I did everything right.
Yet again, the isolator you pointed me is a 3.5mm jack, do you think it's worth the shot or the end result will be the same as my RCA Isolator?
Thanks
Attachments
Hi Saturnus,
I just tried it, no results...It's crazy because I really mean NO results, the Audio does not change one bit, not even a "pop" when I connect or disconnect the signal ground, it's like if there was an invisible wire connecting the Amp and the Source together that I simply cannot see...
Desperated as I was, I also tried to step up the voltage going to the Amp (12v to 18v) just to see how it worked out. The regulator is extremely noisy, lots of frikin "static" coming out of the speakers, and, the "bluetooth noise" is still there (which I now like to call, The Robot) BUT, with the voltage regulator the Signal Ground does, in fact, changes the Audio quality - If I disconnect the Signal Ground, the noise is louder, if I connect the singnal ground, it gets lower and if I shorten the Signal Ground wire it also reduces the noise...How come with a voltage regulator the Signal Ground interferes with the Audio and without the Voltage regulator, it doesn't? Creepy stuff for me :/
So, following your last post, what is a Mechanical Ground Loop?
DC-Blocking cap, I saw that several times during my readings but it never seemed to play an important role so I have no idea what it is because I didn't search for it, is it a standard capacitor? Any idea of how I could calculate it's value?
Thanks a lot for everyone's help.
Mechanical ground loop is just that if the amp and source and mounted on the same surface and that surface is conductive and/or magnetic there could be a ground connection that way.
Maybe it's just radiated noise from the amp output. It is switch mode after all and it could interfere with or react with signal wires. Do you use shielded or twisted cables?
Also make sure that wires to ground from the amp and mp3 card are securely connected and use an appropriate thickness of cable.
I've had similar results as well where I simply had to resort to saying: "ok, these parts simply won't work together without separate supplies", and invested in an isolated ground supply just for the mp3 module. Or using a much smaller battery for that and a big one for the amp.
Maybe it's just radiated noise from the amp output. It is switch mode after all and it could interfere with or react with signal wires. Do you use shielded or twisted cables?
Also make sure that wires to ground from the amp and mp3 card are securely connected and use an appropriate thickness of cable.
I've had similar results as well where I simply had to resort to saying: "ok, these parts simply won't work together without separate supplies", and invested in an isolated ground supply just for the mp3 module. Or using a much smaller battery for that and a big one for the amp.
Last edited:
Ok... Let's review..
"So, what's the problem? Well, when I assembled everything for testing purposes, I connected the Amplifier to the Battery and the Audio Module to a 12v Transformer, spot-on, no noise at all no matter what function I was using (USB, Bluetooth, FM or Aux) BUT, when I connected both the Amplifier and the Audio module to the same power source (either to the Battery or to the 12v Transformer) Oh boy, what a mess."
This is basically the same issue I had..
"I modified a Lepai LP-808 by adding a 12 volt to 5 volt regulator for a Bluetooth receiver. It didn't work so well when I connected the rcvr. directly to the amp. (Both being powered from a common 12 volt battery) I installed an isolation transformer network between the Bluetooth and the amp input and it works fine.."
When Xarlitos used separate power supplies, all is good. When he combined both items to a common power supply.. Well.. 😉
I would suggest connecting back to separate supplies again to make sure that nothing is damaged and start from scratch..
"So, what's the problem? Well, when I assembled everything for testing purposes, I connected the Amplifier to the Battery and the Audio Module to a 12v Transformer, spot-on, no noise at all no matter what function I was using (USB, Bluetooth, FM or Aux) BUT, when I connected both the Amplifier and the Audio module to the same power source (either to the Battery or to the 12v Transformer) Oh boy, what a mess."
This is basically the same issue I had..
"I modified a Lepai LP-808 by adding a 12 volt to 5 volt regulator for a Bluetooth receiver. It didn't work so well when I connected the rcvr. directly to the amp. (Both being powered from a common 12 volt battery) I installed an isolation transformer network between the Bluetooth and the amp input and it works fine.."
When Xarlitos used separate power supplies, all is good. When he combined both items to a common power supply.. Well.. 😉
I would suggest connecting back to separate supplies again to make sure that nothing is damaged and start from scratch..
Problem here is that a ground loop insulator trafo will not do anything. When disconnecting ground altogether doesn't improve or actually appear to do anything at all then obviously a ground loop insulator won't do anything either. It can't be much more isolated than not connected at all, right?
It's more than isolating the shield or ground only. Being the voltage regulator are probably not an isolated output type on either item, all commons (or grounds) are, or should be, at the same potential through the common power supply. That doesn't guarantee the audio signal connections are at the same common potential. The signal lines need isolation also..
"IF" the output signal shield of the Bluetooth and the input signal shield of the amp are connected to the common ground on both units, they "should" be at the same potential through the common connection to the power supply..
In my situation, the 5 volt non isolated regulator and the amp use the same common or ground.(Same Battery Also) The shield of the input of the amp and the common of the output of the 5 volt regulator are at the same potential. BUT, between the output shield of the Bluetooth receiver and the battery common or ground, there is is a 3 volt difference in potential. (Measured from the shield output connection of the Bluetooth receiver to the common or ground of the battery) This tells me that the whole audio output section of the Bluetooth is "floating" above ground. Why would I not think that the actual audio signal lines wouldn't be floating also?
The amp input signal section is based on the common or ground of the battery, being it operates directly from the battery, using the battery ground as its reference. I already have a 3 volt difference in the shield so I'll be safe and assume a difference in reference in the signal lines also. Isolation transformers breaks the D.C path connection in both, the shield and the signal lines between the two units..
I basically did the same thing as Xarlitos did. I ran the amp from a 12 volt gel cell and the Bluetooth receiver from a 5 volt bank battery and all was well. When I connected the Bluetooth receiver to the 5 volt regulator and to the input of the amp, all went to crap. I installed an isolation transformer and all is good..
I would suggest to disconnect all audio connections between the amp and the Bluetooth unit, power them both up, and do some voltage readings between the shield connections of both units and the L/R signal connections. One lead on the shield or signal ground connection of one unit and the other lead to the other device at the same point. Do both signal lines (L/R) between the two also. If you see any voltage reading, you need to isolate..
"IF" the output signal shield of the Bluetooth and the input signal shield of the amp are connected to the common ground on both units, they "should" be at the same potential through the common connection to the power supply..
In my situation, the 5 volt non isolated regulator and the amp use the same common or ground.(Same Battery Also) The shield of the input of the amp and the common of the output of the 5 volt regulator are at the same potential. BUT, between the output shield of the Bluetooth receiver and the battery common or ground, there is is a 3 volt difference in potential. (Measured from the shield output connection of the Bluetooth receiver to the common or ground of the battery) This tells me that the whole audio output section of the Bluetooth is "floating" above ground. Why would I not think that the actual audio signal lines wouldn't be floating also?
The amp input signal section is based on the common or ground of the battery, being it operates directly from the battery, using the battery ground as its reference. I already have a 3 volt difference in the shield so I'll be safe and assume a difference in reference in the signal lines also. Isolation transformers breaks the D.C path connection in both, the shield and the signal lines between the two units..
I basically did the same thing as Xarlitos did. I ran the amp from a 12 volt gel cell and the Bluetooth receiver from a 5 volt bank battery and all was well. When I connected the Bluetooth receiver to the 5 volt regulator and to the input of the amp, all went to crap. I installed an isolation transformer and all is good..
I would suggest to disconnect all audio connections between the amp and the Bluetooth unit, power them both up, and do some voltage readings between the shield connections of both units and the L/R signal connections. One lead on the shield or signal ground connection of one unit and the other lead to the other device at the same point. Do both signal lines (L/R) between the two also. If you see any voltage reading, you need to isolate..
As I wrote previously. I also had cases where there was a voltage differential. In one case an isolation transformer fixed it but in most it doesn't. Only an isolated ground supply or having separate supplied fixed it. In this case it should not be necessary though as the parts used are pretty much identical to ones I have tried.
The fault here is clearly not in the amp at all. There is something in the mp3 module that causes it. What exactly is difficult to say without having it in hand and start measuring but most of these cheap mp3 modules are really just barely worth the $2 price tag.
The fault here is clearly not in the amp at all. There is something in the mp3 module that causes it. What exactly is difficult to say without having it in hand and start measuring but most of these cheap mp3 modules are really just barely worth the $2 price tag.
- Status
- Not open for further replies.