Add your comments about the Circuit Note found at http://www.analog.com/en/circuits-from-the-lab/CN0241/vc.html.
Hi, why is there a TI (TXB0104) Level Translator instead of an ADI one in this CN? It seems to me like eg. ADG3304 (Low Voltage 1.15 V to 5.5 V, 4 Channel, Bidirectional, Logic Level Translator) could work there. Do I not see anything important?
Thank you in advance.
You bring up a very good point. Unfortunately the ADC in the circuit (AD7920) does not have the capability to drive the ADG3304.
The ADG3304 should have an output impedance of less than or equal to 150 ohms and a minimum peak current driving capability of 36 mA.
The AD7920 can source and sink 200 µA.
I see now, thank you. I checked all ADI level translators and ADG3308 and ADG3257 seem to be specified different. Can you, please, check, why these can't be used? I know that ADG3257 doesn't allow an exat voltage level setting.
Although it does not say that the input driving requirements of the ADG3308 requires 36mA of input driving current, it has the exact same architecture as the ADG3304, so I have reason to believe it will have the same problem as the ADG3304.
The ADG3257 is an interesting choice for the application. It could work. I believe you would have to add a diode with a forward voltage drop of 0.7V in series with the 5V supply. This will allow for 5V to 3.3V level translation.
Yes, I saw the diagram with the diode at the supply pin. Thank you.
Could you, please, check this possibility with the application engineer of this CN? And maybe ask him for the reason the ADG3257 wasn't used?
And could you, please, check with the designer of the ADG3308 if it is possible there is a bit different structure in there? Or to let him know about the missing note...
I am the engineer that developed CN0241. I did not use the ADG3257 because I was unaware of its existence.
I suppose you could use this part in this application strictly as a level translator but that doesn't necessarily mean that we should use this part. If you read carefully, the ADG3257 allows for 5V signals to be input to the device, but it cannot drive 5V signals. In this sense, the part is not a true level translator. The minimum input voltage for a logic high on the AD7920 is 2.4V, so the ADG3257 could work.
This part is best suited for high speed applications between a processor and an external memory unit due to the extremely small propagation delay.
I'll check with the product line of the ADG3308 to see if there is an architectural difference between the ADG3308 and the ADG3304.
Thank you James,
that is clear now.
Retrieving data ...