This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

Stange breakdown of PCM-converters and output impedance values

Other Parts Discussed in Thread: PCM1792, PCM1796

Hi,

 

I need to konw, p to which voltage levels the DAC-output pins of PCM1792/1794A and the PCM1796/1795 are safe?

My problem is:

I use a grounded base stage (or common base stage) as I/V-converter with a JFET-current sink to the negative supply rail of the I/V-stage.

So the Emitter of the I/V-transistor is connected to the DAC-output, its collector going through a resistor (that converts the i to v) towards the positive supply rail.

The base is held at the Vbe-potential  by the base of  a second transistor whioch is diode connected and whose emitter is gnd-referenced. So the I/V-transistors emitter is held at gnd-potential.

The bases of the transistors are going via a small cap (1.5nF) to gnd (this cap helps keeping the input impedance of teh common base stage linear up to the MHz-range).

By trimming the current of the current sinking JFET one can easily tune the voltage offset at the DAC-outputs to less than 10mV. However when switching-On the power the DAC-output line falls immideately  to approx. -500mV and back to 0V (short pulse). On switch off the level jumps  from 0V to approx. +500mV and sinks quite slowly towards 0V.

I understand, that the PCM all feature protection diodes in their outputs. Could it be that theses Diodes may be distroyed at switch-On or -Off?

So, which values of  Voltage are allowed at the DAC-outputrs (beeing CMOS-devices the voltage compliance should be 1.0V to -1.25V).

Could it be a measure to add fast protection diodes with higher current capabilty than 10mA (the limit of all DASC-pins according to the Datasheet) toward gnd?

Or are there any other methods of protection if I can´t avoid voltage pulses on the DAC-output lines?

The second Q regards the value of output impedance. I already asked that Q, but never got a reply. At the time I´m calculating with 2.2kOhms paralleld with 22pF or the PCM1792/1794A.  I don´t have any figures regarding the PCM1796/1795 though.

Thanks

purist

 

  • Hi,

     

    just one thing. At least one of the DAC-chips that broke down, now shos a tiny 4Ohms input impedance at its 5V analog reference input pin (Vdd-analog).

     

  • Hi, purist,

    It seems like you're clamping the output protection diodes, and you must be damaging those. You might be able to protect them by placing low VF diodes (schottky) which would turn on prior to our diodes turning on.

    Sorry, we don't have the value of the output impedance available.

    -d2

  • Hi,

    so one can assume that the protection diodes in the outputs are not of the Schottky-type, hence they feature higher Vf -Values than schottkys?

    I now had the possibility to control one of the ´broken´ DACs and it seems that instead of a 3.3V regulator for Vdd the reg put out just below 3.0V. So at the moment it seems that the voltage variation at the DACs outputs had not been the problematic issue. Since we don´t need any 3.0V but 3.3V items I assume a wrong voltage-rated chip had been send by the supplier.

    Unfortunately TI doesn´t explain the codes printed on the chip in their Datasheets. So I was unable to even figure out of which voltage-value the TPAS730xxDBV chip should have been, though I´m quite sure that the measured 2.945V had been the result of a of a pretty well functioning 3.0V-version and not the result of a misfunctional 3.3V-version.

    In fact You can´t even estimate by the imprinted code what kind of chip is placed on the board, a regulator, or something else. After my fashion the datasheet is clearly missing on information here.  Linear Technology for example explain the code imprinted on their LT17xx series so that You can easily recognize which is which.

    I still would like to know what the output impedance values of the PCM1792, 1796 and 1795 are. Come on, You are not telling us here that nobody at TI knows the output impedance values of their own chips???? Beeing analog outputs it should go without saying that such data is regarded a must in any datasheet. having checked other threads I saw that I´m not the only one who likes to know this data.

    regards

    purist

     

     

     

  • purist,

    Yes, the diodes in our process have a higher Vf than an external diode.

    -d2