| Electronics > Projects, Designs, and Technical Stuff |
| decent but affordable 16 bit DAC |
| << < (4/7) > >> |
| SiliconWizard:
--- Quote from: nemail2 on June 07, 2019, 11:52:03 pm ---So I found this one: http://www.ti.com/lit/ds/symlink/dac8563t.pdf and was wondering if it was even possible to feed it with 5V (which would give me 5V full scale output) but use 3.3V (that's the MCU's voltage) for the SPI interface. According to the datasheet, the DAC recognizes voltage above 2.1V as logic input HIGH. According to the datasheet it should be possible even without the need of any voltage level shifters. --- End quote --- Datasheet states that logic inputs high level is 2.1V to AVDD, so you can safely use 3.3V to 5V logic if you're using a 5V AVDD. Obviously don't forget to connect the 5V supply to the VREF pin as well, otherwise it will use the internal reference (2.5V). Note that the internal reference is pretty accurate so if a 2.5V full-scale output could be adequate, I'd suggest using it instead of a generic 5V the accuracy of which may not be all that good. Of course you can always use a 5V voltage reference both for VREF and AVDD: this DAC draws little current. |
| nemail2:
kinda decided to go with the DAC8311, unless something bad comes up... (see my previous post). But thanks for your response anyway. |
| David Hess:
--- Quote from: nemail2 on June 09, 2019, 11:56:50 am ---nah, wouldn't be an issue for me. I mean, who needs 12.000V and couldn't get away with 12.010V (if that'd be the only settable voltage due to resolution)? I know, bad example - but for me, that wouldn't matter. How'd I do that, with MOSFETs? Are they the way to go? Or is it another "it depends" like always in engineering? :scared: :-DD --- End quote --- It comes down to either switching a gain stage after the DAC or switching the gain of the control loop which has the complication of affecting the frequency compensation. It could be done with MOSFETs but there are lots of ways. |
| OM222O:
Look up "universal shunt" or "ayrton shunt". Same idea applies here. You have a resistor in the feedback path of the op amp. If the shunt is disconnected, it only acts as a dummy resistor and you have unity gain. Then depending on the value of the universal shunt, you will have some gain which is 1+(feedback resistor/shunt resistor). If you want a specific impelementation for the universal shunt, I can provide a schematic using N channel fets. Also your op amps seem good for a gain of 6 (100uV max offset) but keep the OPAx188 (OPA2188 in your case) as a drop in replacement upgrade ;) |
| OM222O:
Actually I already have a universal shunt schematic with kelvin connections! See the attached picture. You can Ignore Q4-6 since they are for the sense connections and you don't need them here. The DMN3018SSD should be good enough, but if you need lower capacitance, consider the FDS6930B (again, drop in replacement). This should give you 6 orders of magnitude resistance which is more than good enough for my applications, but needless to say, you have to tune the values and the ranges of the resistors for the gain values you want. Good luck :-+ |
| Navigation |
| Message Index |
| Next page |
| Previous page |