PCM1702 passive output, I/V stage ???

[jbator]

Hello,

Does PCM1702 can have passive output stage and I/V vbased on resistors only?

I have CD with parallel PCM1702 per chanel and I want to change somehow it's output stage.

 

[wa2ise]

I took a look at the data sheetTI and Burr Brown web site and it looks like that this DAC will work quite well into an I/V resistor. See my web site of modified CD players with tubes and look at the circuits I did with the PCM61 and PCM56. These DAC chips have the same kind of internal circuitry as yours, so my I/V circuits should work for you.

 

[jbator]

What are the rules of chosing value of this ground resistor?

 

[wa2ise]

For the I/V resistor, most everyone selects a resistor such that the peak current yields a voltage around 200mV. This to avoid any clipping from any ESD (Electro Static Discharge) protection diodes in the DAC chip. I've gotten away with higher voltages, around 1V without any problems from any ESD diodes. It's hard to say what the chip maker did, as they expected everyone to feed the analog output signal into an op-amp virtual ground.

 

[kevinkr]

I generally aim for a max of about 20 - 30mV as good linearity (both DNL & INL I suspect) is not guaranteed with a lot of these current output dacs if there is significant voltage present on their outputs. (They were designed to run into the virtual earth of a transimpedance amplifier or I/V converter as they are more commonly known.) Analog Devices actually specified an allowable voltage at the I out on some of their older dacs before degradation occurs. (AD1860 series iirc 20mV) I think BB may have provided similar information for the PCM63 and others as well.

I haven't seen that information on newer dac chip spec sheets, but assume in an R2R ladder or newer variants of this topology that this may still be somewhat of a concern.

It goes without saying that at these output levels voltage gain is required, I have used tubes in the past and currently favor transformers either from Sowter or Llundahl for this purpose.

 

[jolintsai]

Hi guys,

Apologies to hijack this thread. I need some recommendations from you guys. I tried having a 750 Ohm resistor on the Iout to ground, and removed the OPAMP from the output stage. What I got was some distorted sound that sounded like noise but it tracks the music. When music stops, it is totally silent.

I followed what Mr.Lukasz did exactly, as shown on picture below.

Did I do anything wrong? Usually what I saw online is a 100-200 Ohm resistor. Is 750 Ohm too high?

 

[miro1360]

what is input impedance of tested device?

 

[georgehifi]

The PCM1702-4 is not a good candidate for passive I/V as it only has 1.2mA output, and by the time the passive i/v resistor has knocked that down, you end up with nothing but noise. Then you'll have to amplify the hell out of it to get some gain along with the noise.

Cheers George

 

[abraxalito]

I agree passive I/V isn't a good solution - not just from the point of view of output current but also its output impedance is rather low. Use a common gate (or base) stage to provide a low impedance to the DAC.

 

[jolintsai]

hi Miro1360,

output range +/- 1.2mA
output impedance 1K Ohm

I'm thinking that the article posted by Lampizator is misleading. 😕 I should have used something like 50-100R to begin with. 750R may be too high.

hi guys,

if 50R-100R still does not work, will try a grounded grid tube stage instead and see how it goes. thanks!

 

[Extreme_Boky]

Those DAC chips are outstanding, but they do need to see close to 0 dynamic impedance at their output. The only way to achieve this is the AD811 current feedback OP. This one will give the required impedance, fantastic speed and the required gain all in one heavily class-A-biased amplification stage.

 

[jolintsai]

Those DAC chips are outstanding, but they do need to see close to 0 dynamic impedance at their output. The only way to achieve this is the AD811 current feedback OP. This one will give the required impedance, fantastic speed and the required gain all in one heavily class-A-biased amplification stage.

Will look into AD811. TQ!

 

[analog_sa]

if 50R-100R still does not work, will try a grounded grid tube stage instead and see how it goes. thanks!

Is this plan of action based on logic? Please inform us what tube has input impedance below 50R.

 

[jolintsai]

It is the circuit. Read here. Grounded Grid Tube Gain Stage