+4 V/10 mA VREF buffer amp for ADS1601 ADC?

[hanakp]

I suppose I could find one by myself, but this way, others will learn about this ADS1601 quirk:

Short version: is there a rail-to-rail opamp that would reliably work as +4 V/10 mA DC voltage buffer from +5 V or +-5 V supply? By "reliably" I mean with a reasonable margin and at least in -20 to +70 deg. C temperature range. Preferably, it shouldn't be too expensive (<4 USD) and needs to be okay with large capacitive loads. And of course, it shouldn't have too big DC offset and drift.

Long version: I built a prototype with ADS1601 ADC, because I needed sigma-delta, 16 bits and >1 MSps rate. It requires 3 reference voltages (VREFP, VMID and VREFN), but their on-chip sources are too inaccurate for my purposes (+-8% tolerance, >50 ppm/deg. C drift). The ADC allows for external references, their voltages should be 1.0, 2.5 and 4.0 V in that case. They even provide recommended buffer circuit in figure 42 in the datasheet, but there is a tiiiiiny liiiiitle problem with it: the 4.0 V input draws 10 mA. That happens because internally, there is 300 ohm impedance between the VREFP and VREFN inputs, as they show in figure 41. And since there there is 3 V difference between the inputs, it draws 10 mA. They recommend OPA2822 as the buffer, but that's not even rail-to-rail type. There is no evaluation board for ADS1601, but I checked boards for similar TI ADCs and they typically use +-7 V supply to provide enough headroom for the amps. I'd like to avoid that and use +5 V or +-5 V instead. I already tried my favorite TLC2272, but it can supply only 3.5 mA so close to the rails (it works fine for 1.0 and 2.5 V, though).

Thanks!

[MasterT]

Any 5V OPA would be o'k, there only requirements I see is stability with capacitive loads.
DS states that opa2822 characterized outputs 4.5V with load, so no issue  with not to be R2R .
Regarding OPA costs, ADC is outdated, noise is much higher than new mcp3562, so exceptionally good  2 nV noise level does not have any meanings for it. Sure, mcp6022 with 10 MHz GBW and ~8 nV would be perfect candidate.

[hanakp]

Those MCP6022 indeed require only 0.2 V headroom at 10 mA load (how is that even possible?), but their DC offset could drift up to +-2.5 mV over the temperature range. That alone could cause about 0.1% full scale ADC error. Also, all precision Microchip amps are not too great with capacitive loads. Still, they're generally cheaper than Analog and TI counterparts, so I plan to test MCP6072 along with TLV2186, OPA2378 or AD8602.

OPA2822 is unusable, the datasheet says "both inputs and outputs can swing to within 1.2V of either supply rail" on page 15.

MCP3562 goes only to 150 kSa/s, I need over 1 MSa/s. ADS1601 currently seems to be the only sigma-delta ADC that combines 16-bit resolution with such high sample rate.

[iMo]

OPA388/387 ??

[MasterT]

It's not clear why 1 msps, but my experience with SD ADC tells me that converter running at just 20 MHz clock could get you 8-bits noise free at the best. Usually makers implement oversampling / decimation filters right into the converter, not to push 8-9 noise bits out of 16 over data buss

 Though, low price 16 bits SAR easily outperform SD at 100 ksps and up regarding noise & linearity.
ADS8691, for example.

[shapirus]

I suppose I could find one by myself, but this way, others will learn about this ADS1601 quirk:

Short version: is there a rail-to-rail opamp that would reliably work as +4 V/10 mA DC voltage buffer from +5 V or +-5 V supply? By "reliably" I mean with a reasonable margin and at least in -20 to +70 deg. C temperature range. Preferably, it shouldn't be too expensive (<4 USD) and needs to be okay with large capacitive loads. And of course, it shouldn't have too big DC offset and drift.

Take a look at AD860x.

[Kleinstein]

The difficulty part is finding an amplifier that is Ok with a huge capacitance (10 µF range) at the output.  So this would be not so much a normal OP-amp, more like a voltage regulator or special reference buffer.
Given that many moder SAR and SD ADCs have similar requirements it is somewhat strange not to have extra chips for that task. Some OP-amps may be OK, but it is not that clear which as it is outside the normal load case.
There is not need for a rail to rail chips the levels of some 1 and 4 V are reasonably away from the rails. It may need separate chips for the positive and negative side - many RR amplifiers are quite poor in a  range like Vcc-1 V to VCC. So I would more like avoid a RR chip here for the 4 V.

[hanakp]

I finally got opamp samples and I tested them as ADS1601 bufers. All of them had negligible offset at 2.5 V (which draws only a small current), but it usually got worse at 1.0 and 4.0 V which drew over 6 mA in my case (but could be as bad as 10 mA). I also tested offset's temperature drift, I measured them at 20 and 80 deg. C. I used Agilent 34401A in the high input impedance (>10 Gohm) mode. All the amps were powered from non-symmetric +5 V, except TLC2272 and TLV2186 which can work with +-5V.  The results from the best to worst:

-MCP6072 was the best, despite it's the cheapest of the lot. The initial offset was in 0.02 mV range and worsened only to 0.1 mV at 80 deg.C. It kept such good parameters regardless of current load.

-TLV2186 had slighly higher initial offset (still 0.04 mV range), but was rock stable at higher temperature, the changes were practically indistinguishable with 34401A. Again, it was insensitive to load currents.

-AD8602 was rather disappointing, its offset worsened dramatically with load, it was over 1 mV at 1.0 V and 2 mV at 4.0 V. But like TLV2186, they didn't change with temperature.

-OPA378 has different package (SOT-23), so I tested 2 samples at 4.0 V only. They fared pretty poorly, their initial offsets were 2.1 and 4.7 mV and they worsened to 3.7 and 7.1 mV at 80 deg. C. Obviously, they don't like the load. Originally, I wanted to test the dual version OPA2378 in SO-8, but they're sold out at the moment. I wonder if they'd fare as badly...

-TLC2272 didn't work at 4.0 V (that's why I started this thread in the first place), but the offset was 0.1 mV at 2.5 V and 2.5 mV at 1.0 V. The latter was suprisignly stable with temperature, it worsened only to 2.8 mV at 80 deg. C. Not bad, considering it's not a precision amp.

Of course, there were only 2 samples of each type, so the results are not particularly reliable, but at least I could reject the obviously problematic ones. I plan to buy more MCP6072s from a different vendor to see whether their "win" today was just a fluke.

I also tested stability of all the amps, becuase they have to work with a large capacitive load. But all of them worked fine, there was no overshoot on ringing upon Vref ramp-up. It's because their feedback in figure 42 compensates for the parasitic pole.

[hanakp]

I got another MCP6072s from a different production lot. One amp performed worse than the last time, it had 0.056 mV offset and it also changed with temperature. Overally, the second batch performed roughly the same as TLV2186, but MCP6072s are still cheaper.