Bluetooth noise problem.

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Hi guys. I build these single speaker class D amps out of those 5L beer kegs. A couple years ago I made a new model with bluetooth. It's been great for a couple years, so I'm trying to make more for my friends. The bluetooth module that I used in mine is no longer available, and the ones i've tried are producing a lot of bluetooth signal noise. I've been assuming it's the cheap modules that is the problem, but after my most recent purchase of a more expensive module that is supposed to be the upgraded version of the module in my original amp, I have to accept that maybe i'm doing something wrong. So i'm here to ask for help from folks who actually know stuff. Below is a description of the problem and my setup & components.

The issue:
When the amp is powered up there is what I describe as signal noise for 5-10 seconds, and then it goes quiet. When a device connects (my phone) the signal noise comes back. If no audio is played the noise again goes away in 5-10 seconds. When audio is played the noise comes back and stays until the audio is paused or stopped. The signal noise scales with volume. Mostly you can't hear it over the music, but I can detect it in softer songs and audio books.


The basic setup:
Amp%20Plan.jpg


Specifics:
Power supply: 6x Panasonic Lithium 18650 wired 3S2P

Amp: TPA3118 PBTL Mono Digital Amplifier Board1X60W 8-24V

Bluetooth Module: Bluetooth 4.2 CSR64215

As of right now the bluetooth boards have not been modified to accept the necessary external antenna.

Thanks for any advice you can offer!
 
Thanks guys. The grounds all connect together. I'm not running a ground line from device to device, but rather they all connect to a central point. I think so anyway.

When it comes to any wires that carry sound (bluetooth to pot to amp to speaker) i'm using shielded wire. I'm using the outside bare shield wire as the negative and the inner coated wire as the positive.

I should have some electrolytic capacitors around. What volt & mdf would you recommend? I have a 16v 100uF, and a 16v 220uF. Would either of those work? I also have some others if not.

Thanks!!
 
Any RF injection might be reduced with 100pF to 1nF ceramic capacitors across signal<->ground, and power<->ground. RF on the audio signal can get rectified in the amp's input stage, polluting the audio with RF envelope waveform.


Keep antenna well away from amp, add the ceramic caps at the connections to the amp board right at the terminals, preferably soldered direct to the terminals.


Does the amp pick up GSM packet bursts too?
 
Mark, thanks! I'm going to try the capacitors.

Currently, with this new board just in testing, I haven't modified the bluetooth board for the antenna. I have the cut the on-board antenna and the solder the lead wire. Anyway, currently the bluetooth module is sitting well away from the other components.

I had to google GSM. Without any experience, i'm not certain. To me the noise in question sounds like it could be this. It's not a steady hum, but rather an irregular jumble of noises that vary in pitch and volume. If the capacitor idea doesn't solve the problem, i'll record a sample of the sound. Thanks again!
 
Ok, so the capacitor thing didn't work out. So, i decided to remove everything and start as simply as possible. I'm only using the battery, the amp, the bluetooth, and the speaker. Along the way I found something on accident that might point to the problem. Or it may be perfectly normal... but it doesn't seem right to me.

First, i connected the amp to the battery and speaker. It powered up, and there were no unexpected noises in the speaker.

Then I connected the audio out from the unpowered bluetooth to the audio input of the amp. No problems there, but again, there's no power to the bluetooth module yet.

Lastly, i started connecting power to the bluetooth. I connected the positive lead first, and suddenly the bluetooth powered up and started making the interference noise through the speaker. I had not yet connected the negative power supply lead! This made no sense to me. The only thing i could think of is that the audio connection between the amp to the bluetooth was completing the circuit. So, i unplugged the 1/4" phone jack from the audio out of the bluetooth board... and it powered down, confirming my suspicion.

Is this par for the course? Should the audio connection between the boards be able to complete the bluetooth's power circuit?

Thanks guys.
 
Yeah it's a well known problem, give your bluetooth module it's very own battery and listen to it become silent.

Wow, really? So i need to have one set of batteries for the bluetooth and a different set of batteries for everything else?? How would that even work? I mean, all the batteries are rechargeable and contained in the unit. I guess i would have to have 2 different charging ports and BMS boards, and charge them separately? That's nuts!

What are my options? Are there bluetooth boards out there that don't suffer from this problem? Is there a way to isolate the power between the two modules somehow, like with diodes or something?

Thanks for the response!
 
You don't have to have two batteries, but it is the way to prove that there's some funky loops being introduced when you share a battery.
You can try a dc isolater a or a dc-dc converter, sometimes that works sometimes it doesn't, I have to say I have given up and now only use fully engineered bluetooth amps, because they are really cheap and work really well.
 
What engineered bluetooth amp do you use? I've never been able to get away with using an amp with a built in bluetooth chip because of my single speaker setup. All the ones i've found are stereo, and i've been told you can't or shouldn't sum the left and right channel post amplification.

I've never heard of a dc isolator, and don't understand what a dc-dc converter would be (what's it converting to?), but i'll go do some research. Thanks again for taking the time to respond!
 
The CSR bluetooth chips are configurable, there are numerous parameters that can be changed. The CSRA64215 config is stored on the local spi flash chip.

It's possible that the CSR64215 onboard battery charger is enabled by default and that is inducing switching noise into the audio path. Is the battery charger enabled or disabled ? who knows, you will need to read the config to determine the state.

Buying a fully engineered product will already have the bluetooth chipset to have a optimised configuration and not one that is probably set with some default options.
 
Disabled Account
Joined 2010
The problem of these BT devices is pulsating current consumption combined with common gnd return lines. What can you do?
1. In case there is a symmetrical output, use this and feed into an symmetrical input.

2. Otherwise isolate pos supply input with a regulator or some resistance and put a low ESR blocking elco of several 100uF parallel to + and - terminal of the BT device.
 
Last edited:
Thanks everyone! After reading your replies I did a search on dc-dc isolating converters. I found a video on youtube of a guy demonstrating the exact noise that i'm getting with this weird ground loop, and then adding a dc-dc converter to fix it. I'm going to give that a try, but I have a couple questions:

1) I'm running everything off 6x 18650 lithium batteries, in 3S2P. The voltage varies from between around 12.6v at full charge, down to something like 8.7v when the BMS shuts things down to prevent undercharging. I also use the amp while charging, and my AC-DC wall wart is 13v.

So, with that in mind, do I still use a 12v DC-DC isolating converter? Or, do i need to find one that can handle a range of input voltage from 8v-13v?

2) When shopping for these i'm finding that they are rated not just for voltage, but also output power and current. Do you guys have any idea how much power & current a bluetooth board like mine requires?

Thanks again everyone, i really appreciate the help!
 
So, with that in mind, do I still use a 12v DC-DC isolating converter? Or, do i need to find one that can handle a range of input voltage from 8v-13v?



2) When shopping for these i'm finding that they are rated not just for voltage, but also output power and current. Do you guys have any idea how much power & current a bluetooth board like mine requires?



Thanks again everyone, i really appreciate the help!

1) I don't think the small modules come for a range of voltages, but they might. If the module will take 5v, step it down to 5v then use an isolation module for 5v

2) easiest way to know for sure is measure it. Most would be less than 5W though.
 
Thanks Uber. I'm pretty sure the bluetooth module will take 5v but i'm checking. I'm trying to achieve the best possible battery life. In the past i've done similar things with a transistor, but i think a fair amount of energy gets wasted in the way of heat. Is there a more efficient way of doing it?
 
Thanks Uber. I'm pretty sure the bluetooth module will take 5v but i'm checking. I'm trying to achieve the best possible battery life. In the past i've done similar things with a transistor, but i think a fair amount of energy gets wasted in the way of heat. Is there a more efficient way of doing it?
A switching "buck" converter is the way. Modules can be had for next to nothing. They also make eg. 12v-5v isolating power supplies.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.