Start a new topic

CS4398 DAC fried by shorting output cap's

CS4398 DAC fried by shorting output cap's

The root cause for my problems turned out to be the missing earth connection.
The problem was not the earth of the PC, but a missing earth connection for the PSU for my T3S amp (the mains multi-plug extender had a internal failure).
That means that a earth connection is established, through the PC, when the USB cable is attached - this gives an initial current-spike as the primary to secondary capacitor and the two y-caps of the PSU are charged (the capacitance from USB shield to earth leg of the Power-plug is measured to 140 nF).
Somehow this current passes your DAC and fries it...
Have you ever considered a low capacitance protection device like the NXP PRTR5V0U2D on the USB input?

Thanks / Peter
* OK, so I changed the CS4938 (using a SMD soldering station and ESD wristband) - all solderings where as before.

* Removed the shorts on the output CAP's.

* Connected the POT and the T3S AMP and powered everything up.

* 5V supply is measured to 5.1V and the 3.3V is 3.28V - seems OK.

* Attached the USB cable from my PC



No sound, but smoke coming from the CS4938 area... The green LED on the input selector board is on, when the USB source is connected.



I got a minor electrical chock when I detached the USB cable (Touching the shield of the USB connector) - typical for a non earthed powersupply (the input filter in the PC-PSU divides the mains-voltage in half on the earth node - equal to 115VAC in Denmark).



* Can this be the root cause for all my problems?

* Have you experienced more components that need to be changed?

* Can you provide a more up to date schematic for the DAC?



Thanks / Breinholm



Hi, Breinholm.

There must be static electricity on your PC, so be careful, it might be dangerous to electric parts.

I also suggest to test the DAC board along when you fixed it, that will prevent to cause more problems. Use a headphone and a optical input will be better to test the DAC.

Now that's face the problem, if you can, please follow these steps .

1/ The diodes D3~D6 near power input and check if they are ok. ( power off and nothing connect)

2/ Use a multimeter to check the impedance of A5V to GND is normal or not. ( power off and nothing connect)

3/ Power on if the 1+2 points are ok and check the A5V/VD33 is normal or not. ( power on and nothing connect)

4/ Test with optical input and headphone output. ( power on )

5/ Test more.

Here is the sch file may helps.

http://forum.hifimediy.com/newreply.php?do=newreply&p=112
* OK, so I changed the CS4938 (using a SMD soldering station and ESD wristband) - all solderings where as before.
* Removed the shorts on the output CAP's.
* Connected the POT and the T3S AMP and powered everything up.
* 5V supply is measured to 5.1V and the 3.3V is 3.28V - seems OK.
* Attached the USB cable from my PC

No sound, but smoke coming from the CS4938 area... The green LED on the input selector board is on, when the USB source is connected.

I got a minor electrical chock when I detached the USB cable (Touching the shield of the USB connector) - typical for a non earthed powersupply (the input filter in the PC-PSU divides the mains-voltage in half on the earth node - equal to 115VAC in Denmark).

* Can this be the root cause for all my problems?
* Have you experienced more components that need to be changed?
* Can you provide a more up to date schematic for the DAC?

Thanks / Breinholm
yes, the DC get shorted to GND in this case. It might have survived if the pot was not as low as 10K, but a cap should be used in anyway when there's a pot.
Contact us if you need help to replace the cs4938 chip.

We'll add some warning about this on the product page.
Hi, it's bad news my friend.

To find out if we can bypass the caps or not, just check if the DC bias is same at both side of caps. The DA-X0 has 2.5V bias at the output of cs4398, and if you add a pot, the pot's bias is 0V, so it's not ok to reduce the caps.
Just bought the CS4398 DAC and a T3S Amp.

Connected the DAC output with the T3S trough a 2 x 10K POT - Everything worked fine.



Now, two CAP's in series, with a POT in between, is one CAP more than needed to prevent DC currents between the boards.

So I shorted the output CAPS of the DAC - and blew the CS4398 IC!! :o



The cause must be excessive currents flowing out of the CS4398 and into the input CAP's of the T3S, when the potentiometer is turned:

The resulting circuit is a 2.5V voltage-source(CS4398), a voltage divider (The POT) and a 2.2uF CAP.

The DC potential at the output of the POT goes from 0 - 2.5V as you turn the POT - hereby charging / discharging the T3S input CAPS - causing too high currents in the DAC IC.



Now you're warned..



/Breinholm
Login or Signup to post a comment