Hello everyone,
I'll let you know as much as I can about my actual setup (it will be important to achieve a conclusion about my problem).
First, I have a ADAU1701 on a board that I designed. I'm using all ADC and DAC, one of MP pins I use for an digital external volume pot.
The ADC0 is my main input, receiving a balanced signal from my audio interface. I put two ampOps before DSP for a few reasons. First is to receive balanced signals (ofc), second to protect the DSP in cases of surge (rare cases but they happen).
The ADC1 is a generic input, for tests, phone input, etc.
The DACs, I use to send audio to a dual TPA3116 board (2.1). DAC 1 to Mid | DAC 2 to High | DAC 3 to Low.
At output, I use a active output filter like the datasheet, and another ampop to gain stage. (I prefer to do analogic gain over digital gain inside the DSP).
OK, now the power supplies
The supply is provided by a switched 24v/15A (generic chinese but with good and double core filters), and for +/-15v, I'm using two LM2596. One of then I put as inverting configuration (It does a regulator and a negative supply.. I'll show).
To make 5v, I use a LM7805 from +15v, simple and basic.
3.3v from a LM1117.
PS.: Look at the number of switched sources (maybe so much RF)
AND THE PROBLEM:
After a while (about 30mins) my output start with a little distortion, until the output is completely distorted and the audio pratically disappear after a few minutes.
Things that I already tried:
-Change the PLL_LF caps (same value, different materials)
-Another ADAU1701
-Place a 2.2uF and 100nF caps at Reset pin
-More capacitance before and after 3.3v regulator
-Another VDrive transistor (Same BC327)
-Make a new sigmastudio project from the scratch, nothing changed. A simple bypass (input->output), nothing too.
Ah, PS.: when fully problematic signal, I do a simple pertubation (touch with my fingers any V+ pin) and everything back to normal.
In Attachment, the files you need to analyze the case.
I'll let you know as much as I can about my actual setup (it will be important to achieve a conclusion about my problem).
First, I have a ADAU1701 on a board that I designed. I'm using all ADC and DAC, one of MP pins I use for an digital external volume pot.
The ADC0 is my main input, receiving a balanced signal from my audio interface. I put two ampOps before DSP for a few reasons. First is to receive balanced signals (ofc), second to protect the DSP in cases of surge (rare cases but they happen).
The ADC1 is a generic input, for tests, phone input, etc.
The DACs, I use to send audio to a dual TPA3116 board (2.1). DAC 1 to Mid | DAC 2 to High | DAC 3 to Low.
At output, I use a active output filter like the datasheet, and another ampop to gain stage. (I prefer to do analogic gain over digital gain inside the DSP).
OK, now the power supplies
The supply is provided by a switched 24v/15A (generic chinese but with good and double core filters), and for +/-15v, I'm using two LM2596. One of then I put as inverting configuration (It does a regulator and a negative supply.. I'll show).
To make 5v, I use a LM7805 from +15v, simple and basic.
3.3v from a LM1117.
PS.: Look at the number of switched sources (maybe so much RF)
AND THE PROBLEM:
After a while (about 30mins) my output start with a little distortion, until the output is completely distorted and the audio pratically disappear after a few minutes.
Things that I already tried:
-Change the PLL_LF caps (same value, different materials)
-Another ADAU1701
-Place a 2.2uF and 100nF caps at Reset pin
-More capacitance before and after 3.3v regulator
-Another VDrive transistor (Same BC327)
-Make a new sigmastudio project from the scratch, nothing changed. A simple bypass (input->output), nothing too.
Ah, PS.: when fully problematic signal, I do a simple pertubation (touch with my fingers any V+ pin) and everything back to normal.
In Attachment, the files you need to analyze the case.
Measure all the supply voltages when it is distorting, one could be going south.
Maybe because too much current is being drawn from it, and overheating.
Maybe because too much current is being drawn from it, and overheating.
Last edited:
Yes, and I did a test by modifying and placing them directly on the pads... Nothing
Yep, I thought this too... I checked many times (with my fingers), but everything is normal, nothing more than 40ºC
Member
Joined 2018
1. Check the polarity of ADC side coupling ele-caps.
2. Measure the DC voltage of ADAU1701 pin43 through 46.
2. Measure the DC voltage of ADAU1701 pin43 through 46.
Sorry for the attachments. I just saw that it hadn't loaded.
Here they are:
Here they are:
Attachments
It's like this today...
And the DC Voltage:
VOUT0 (46) - 1.527v
VOUT1 (45) - 1.529v
VOUT2 (44) - 1.527v
VOUT3 (43) - 1.531v
Attachments
I tested every VDD pin..
Everything seems normal... 3.3v has 3.317v, 1.8v has 1.772v... After 2h on.
Temp ok, nothing overheating.
Is C27 needed? I thknk theproblem is likely C28 has no path to ground, try adding a 100K to ground after C28
C27? is it useless? Do a DC block or everything else? Can I remove (bypass) it?
And after C28 is the ADC resistor... I'll test today
It may be ok to bypass C27, check for DC before C27, if non or low then try baypassing it, C28 needs a path to ground after it, been there done that, bad distortion after 20 mins or so.
Ok, I tried... Before C28 has no DC. I'll bypass. Ty.
I put a 100k resistor after the C27 and it made no difference.
I'm getting more crazy about this problem
Just a wild guess...
Make certain that you are using an 18K resistor on pin 3 (ADC_RES). I had a lower value on this accidentally in one assembly, and it caused distortion that changed over time. It probably isn't the same problem you are having, but it's worth checking to make sure...
Make certain that you are using an 18K resistor on pin 3 (ADC_RES). I had a lower value on this accidentally in one assembly, and it caused distortion that changed over time. It probably isn't the same problem you are having, but it's worth checking to make sure...
Checked. Perfect 18kohms.
I made a star ground... Think is correct, 22pF crystal caps. maybe change these? More, less... No ideas
Member
Joined 2018
Hi icarovps-San,
C39, C48, C58, C67 polarity must be wrong, but I think this is not root cause...😕