Well, I think I know why they used the 9028 now. I think you're going to be seeing a lot of companies use the 9028 in 2-channel applications vs the 9038, actually.
Based on some info floating around the internet, the 9038 is basically 4 of the 9028 DAC sections in parallel internally. The output current ends up being 4x that of the 9028 in any equivalent configuration. For the 9018 and 9028, if you use them as 2-channel devices, you were looking at around 15mA into the I/V converter (ignoring bias/offset current). If you use that same configuration with the 9038 you'll get 60mA out! Not easy to handle for an op-amp with a reasonably sized feedback resistor. I would be very curious to see their mono-mode application circuit. Of course, there are other ways to design the I/V converter and output circuit, but it's a bit unfriendly to the "textbook" datasheet approach.
I'm guessing the real point of the 9038 is that you can get 8-channel performance basically equivalent to the stereo performance of the 9028. I doubt anyone finds 133dB DNR insufficient.
Last edited:
https://www.facebook.com/livermore....if_t=notify_me_page¬if_id=1481052393538348
Just a temporary diversion -
-RNM
Just a temporary diversion -
-RNM
Weird. The 710 costs only 2/3 of the 860 @ Digikey. Nevertheless, I have
ordered a few.
thanks, Gerhard
Ah, that's more than enough. I have not yet clicked on "pay safely" 🙂
But the real shock is simple mechanical thingies like 1p6t switches, let alone
the knobs for turning them...
But the real shock is simple mechanical thingies like 1p6t switches, let alone
the knobs for turning them...
Same die, just selected for leakage at HV. I was told that the internal leakage was an important performance spec and they were sorting the parts for leakage and binning them.
ah thanks, I've always wondered what the actual difference is.
I have tested the 49710 at elevated supply voltages to see what happens, checking the quiescent supply current didn't occur to me. Perhaps next time. Oh, on a short term basis they seem to work fine at elevated supplies. Tested up to +/-20V.
Alan
I have tested the 49710 at elevated supply voltages to see what happens, checking the quiescent supply current didn't occur to me. Perhaps next time. Oh, on a short term basis they seem to work fine at elevated supplies. Tested up to +/-20V.
Alan
http://www.analog.com/media/en/training-seminars/design-handbooks/Basic-Linear-Design/Chapter12.pdf
Page 12.17
http://ww1.microchip.com/downloads/cn/AppNotes/cn540604.pdf
Page 12
Is it required for audio layouts I wonder, saying that its quite easy to add to the layout...
Where have I seen something similar, a resistor with some wrapping round it that has magical quantum properties (with a slight odour of male Bos taurus excrement)
Where have I seen something similar, a resistor with some wrapping round it that has magical quantum properties (with a slight odour of male Bos taurus excrement)
The total extent of this just boggles the mind. I allow for self delusion, but really?
Can I inject something technical ;-) ?
Am trying to measure the input impedance of a BUF634 unity gain open loop buffer. Spec says 8meg (or 80meg, depending on bandwidth setting) // 8 pF.
Now this chip has no connection to ground, and has floating supplies bootstrapped from the output. So, how should I consider the 8pF? Model like 2 x 4 pF to each supply? If so, and the supplies are bootstrapped, would I measure no input capacitance?
The attached measurement belies that.
Jan
Am trying to measure the input impedance of a BUF634 unity gain open loop buffer. Spec says 8meg (or 80meg, depending on bandwidth setting) // 8 pF.
Now this chip has no connection to ground, and has floating supplies bootstrapped from the output. So, how should I consider the 8pF? Model like 2 x 4 pF to each supply? If so, and the supplies are bootstrapped, would I measure no input capacitance?
The attached measurement belies that.
Jan
Attachments
Last edited:
Yes that's also a good one.
I was thinking of using a JFET opamp buffer before the BUF634 but a FET input buffer is also possible. Need to look into that, but they are obsolete.
Anyway, I can't explain the Zin curve for now. But I am still learning to use the Bode 100.
Jan
- Status
- Not open for further replies.