reference problem on biploar DACs?

[e61_phil]

Hi,

in normal DAC operation with just one reference and unipolar output the reference drift is always scaled to the output (if we neglect any DAC influences). For example: 1ppm drift in 10V reference is also 1ppm drift if the DAC is set to 10% and outputs 1V. Therefore, everything is fine.

But now I'm using the AD5791 to generate a bipolar output. The positive ref input is directly connected to the 7V output of a LTZ1000 reference and the negative ref input is connected to the inverted 7V (using LTC2057 with VHD200 10k:10k).

The output is now calculated by
(Vref_pos + |Vref_neg|) * DAC - |Vref_neg|  = output

And now it is obvious that Vref_neg is no longer scaled by the DAC value. Which means 1ppm of Vref_neg drift (from the VHD200 for example) will result in 10ppm at the output at 10% output (or 60% DAC value).

I think this bipolar reference inputs are very nice, but perhaps one should better reverse polarity by another way.

How do you deal with this problem?

Philipp

[Andreas]

Hello,

if I had this problem I would do one of the following:
- zero adjustment (output the DAC value which is most nearest to zero and convert with a sensitive ADC as null voltmeter)
- use a LTC1043 for the inversion of the 7V
or a combination of both.

with best regards

Andreas

[Alex Nikitin]

Hi,

in normal DAC operation with just one reference and unipolar output the reference drift is always scaled to the output (if we neglect any DAC influences). For example: 1ppm drift in 10V reference is also 1ppm drift if the DAC is set to 10% and outputs 1V. Therefore, everything is fine.

But now I'm using the AD5791 to generate a bipolar output. The positive ref input is directly connected to the 7V output of a LTZ1000 reference and the negative ref input is connected to the inverted 7V (using LTC2057 with VHD200 10k:10k).

The output is now calculated by
(Vref_pos + |Vref_neg|) * DAC - |Vref_neg|  = output

And now it is obvious that Vref_neg is no longer scaled by the DAC value. Which means 1ppm of Vref_neg drift (from the VHD200 for example) will result in 10ppm at the output at 10% output (or 60% DAC value).

I think this bipolar reference inputs are very nice, but perhaps one should better reverse polarity by another way.

How do you deal with this problem?

Philipp

I think your formulae is incorrect. It should be

(Vref_pos + |Vref_neg|) * DAC - (Vref_pos + |Vref_neg|)/2  = output .

At the midscale point both Vref_pos  and Vref_neg change would influence the output in the same way. One possible way to improve the accuracy near a bipolar zero I can think of, would be to use two DACs - one for the positive side and one for the negative side. It would introduce additional problems of gain matching and adding two outputs together but would gain an additional bit of resolution.

Cheers

Alex

P.S. - incorrect bits crossed out!

[e61_phil]

if I had this problem I would do one of the following:
- zero adjustment (output the DAC value which is most nearest to zero and convert with a sensitive ADC as null voltmeter)
- use a LTC1043 for the inversion of the 7V
or a combination of both.

At calibration the "zero" is adjusted. But after a few power cycles it will shift (within 1 LSB of the DAC).

Before I made this design I was thinking about the LTC1043, but I couldn't find the right informations in the datasheet. The absolute value of -1.000000 isn't needed. I build a quick test setup (divide by 2) a while ago and the ratio wasn't exactly 0.5. Therefore, I thought I have done anything wrong or the +/- 1ppm in the datasheet will mean other things, than I expected. Furthermore, I had to create additional power rails. (+/- 15V is too much and even +/-8V is very close to the LTZ1000 output and also very close to the absolute maximum rating of the LTC1043). That led me to the decision with the temperature stabilized VHD200 with 0.1ppm/K tracking (I know it is only typical). The resistance tracking is nearly perfect. On a 6.5 digit meter only the sign changes if I measure before and after the LTC2057.

Is the LTC1043 really better?

I think your formulae is incorrect. It should be

(Vref_pos + |Vref_neg|) * DAC - (Vref_pos + |Vref_neg|)/2  = output .

This depends on the DAC configuration. In my configuration DAC value 0 will output ~vref_neg and DAC value 2^20-1 will output ~vref_pos. Therefore, my calculation should fit.

At the midscale point both Vref_pos  and Vref_neg change would influence the output in the same way. One possible way to improve the accuracy near a bipolar zero I can think of, would be to use two DACs - one for the positive side and one for the negative side. It would introduce additional problems of gain matching and adding two outputs together but would gain an additional bit of resolution.

My thoughts are more in a way of using a multilplexer after the DAC to switch between the DAC output directly (0 - 7V), the inverted DAC output (0 - -7V) and GND. The extra bit of resolution isn't that interesting, but twice the linearity is. In a configuration like this the inverting stage will be scaled again and should be fine. Switching the output to "real GND" would be nice, too.

But what kind of multiplexer should one use to not ruin the whole circuit? Is it enough to search one, which introduced less voltage drop than accepted by loading the mutliplexer with the LTC2057 (on resistance) or are there other important parameter for such sensitive circuits?

Thanks for your answers!
Philipp

[Alex Nikitin]

I think your formulae is incorrect. It should be

(Vref_pos + |Vref_neg|) * DAC - (Vref_pos + |Vref_neg|)/2  = output .

This depends on the DAC configuration. In my configuration DAC value 0 will output ~vref_neg and DAC value 2^20-1 will output ~vref_pos. Therefore, my calculation should fit.

At the midscale point both Vref_pos  and Vref_neg change would influence the output in the same way. One possible way to improve the accuracy near a bipolar zero I can think of, would be to use two DACs - one for the positive side and one for the negative side. It would introduce additional problems of gain matching and adding two outputs together but would gain an additional bit of resolution.

My thoughts are more in a way of using a multilplexer after the DAC to switch between the DAC output directly (0 - 7V), the inverted DAC output (0 - -7V) and GND. The extra bit of resolution isn't that interesting, but twice the linearity is. In a configuration like this the inverting stage will be scaled again and should be fine. Switching the output to "real GND" would be nice, too.

But what kind of multiplexer should one use to not ruin the whole circuit? Is it enough to search one, which introduced less voltage drop than accepted by loading the mutliplexer with the LTC2057 (on resistance) or are there other important parameter for such sensitive circuits?

Thanks for your answers!
Philipp

Hi Philipp,

1) Your formulae is not correct - look (for example) at the midscale. At that point the DAC works as a divider by 2 and it does not matter which reference deviates, the effect will be the same. If both references deviate exactly the same voltage in opposite directions the output will not shift at midscale.

2) Yes, you can switch the polarity by a switch, that would give you the same effect (and same problems  ). My guess is that even plain old 405x CMOS switch should do just fine.

Cheers

Alex

[e61_phil]

Hi Alex,

1) Your formulae is not correct - look (for example) at the midscale. At that point the DAC works as a divider by 2 and it does not matter which reference deviates, the effect will be the same. If both references deviate exactly the same voltage in opposite directions the output will not shift at midscale.

I guess you're right with the zero point. But that doesn't really affect the problem. In the end there are some unscaled terms in the formular. By the way.. Even Analog Devices describes the output only that way (see attachment. It is from the AD5791 datasheet).

2) Yes, you can switch the polarity by a switch, that would give you the same effect (and same problems  ). My guess is that even plain old 405x CMOS switch should do just fine.

Of course, one can do it. But will it work proper? Do you have any experiences with these multiplexer in sub-ppm stable setups?

[Alex Nikitin]

Hi Alex,

1) Your formulae is not correct - look (for example) at the midscale. At that point the DAC works as a divider by 2 and it does not matter which reference deviates, the effect will be the same. If both references deviate exactly the same voltage in opposite directions the output will not shift at midscale.

I guess you're right with the zero point. But that doesn't really affect the problem. In the end there are some unscaled terms in the formular. By the way.. Even Analog Devices describes the output only that way (see attachment. It is from the AD5791 datasheet).

2) Yes, you can switch the polarity by a switch, that would give you the same effect (and same problems  ). My guess is that even plain old 405x CMOS switch should do just fine.

Of course, one can do it. But will it work proper? Do you have any experiences with these multiplexer in sub-ppm stable setups?

I've looked more closely and the formulae you've used (= same as in the datasheet) is correct (and what I've written is wrong  ). Sorry! I was right about the equal influence of both references though. Obviously my logic was better than my math today .

Cheers

Alex

[Andreas]

Is the LTC1043 really better?

Hello,

at least in the 2:1 divider configuration I usually see only the offset drift of the buffer amplifier (over a 30 deg C temperature range).
So the LTC1043 divider is rather stable.
I have typically a gain that is 4-5 ppm (RTI) or 9 ppm (RTO) too low.

As multiplexer with low leakage current and low charge injection I would try a MAX4053A.
(but also limited to  +/-8V)

With best regards

Andreas

[Alex Nikitin]

One more possible suggestion:

The midpoint stability essentially depends on the tracking between the internal resistors and the external divider used to invert the reference. The AD5791 contains two internal 6K8 resistors that can be used to produce the negative reference. That pair might provide a better tracking with the R-2R ladder and as a result a better stability at midpoint.

Cheers

Alex

[e61_phil]

One more possible suggestion:

The midpoint stability essentially depends on the tracking between the internal resistors and the external divider used to invert the reference. The AD5791 contains two internal 6K8 resistors that can be used to produce the negative reference. That pair might provide a better tracking with the R-2R ladder and as a result a better stability at midpoint.

Cheers

Alex

Nice idea! At the moment I use these resistors for bias compensation of the buffer OPamp.

Thanks for your feedback. If I ever do a redesign (the drift measurement works great and this topic here is only a cosmetic problem) I think I will go with switching behind the DAC and give the MAX4053A a try.

With best regards
Philipp