AD2S80_arc min meaning

Question I want to know the meaning of "Accuracy 8 arc min" in the datasheet of AD2S80A. Does it means the error is 8 arc per minute? Answer The accuracy of an ADC is measured in terms of LSB. The accuracy of an RDC is measured in terms of Arc Minutes. This is standard for all resolver to digital converters. There are 60 arc minutes in one degree. There are 21,600 arc minutes in one full revolution i.e. 360 degrees x 60. A 10-bit lsb = 21.1 arc min A 12-bit lsb = 5.3 arc min A 10-bit lsb = 1.3 arc min A 10-bit lsb = 0.3 arc min The spec on the AD2S80A datasheet is 8 arc min +1LSB. So at 10 bit resolution, for example, this means that the max accuracy of the part is 29.1 arc min. If the resolution is 16 bit the max accuracy is 8.3 arc min.

AD2S80A: Can I tie both INHIBIT and ENABLE to 0V ?

Question Can I tie both INHIBIT and ENABLE to 0V ? Answer No. The normal operation is that INHIBIT is inactive (HI) until you want to read data, when you wish to read data, you take INHIBIT LO (preventing updating of the output latches during the read) and ENABLE LO (presenting the data in the latches to the output pins). You may not tie both INHIBIT and ENABLE to 0V as this means INHIBIT is active and will prevent the transfer of the data from the up-down counter to the o/p latch. But you may tie INHIBIT HI and ENABLE LO thereby allowing continuous reading of the output data lines. By disabling the INHIBIT you do run the risk of reading invalid data during an update of the output data.

AD2S80A: Output is not linear, 2-3 degrees phase difference between sine and cosine input,

Question R/D-converter's output is not linear (0-2degrees of non-linearity): There is a small phase shift of 2-3 degrees phase difference between sine and cosine input. R/D-converter's output is not linear. If I rotate the resolver one full round, the output of the converter has 0-2 degrees of non-linearity. The resolver is checked, and it is OK (only 5 arcmin error). What could be the problem? Answer A phase difference of 2-3 degrees between the sine and cosine inputs is quite a large phase difference and is the most likely cause of the errors. This phase difference may be due to the resolver you are using or due to a mismatch in the sine and cosine cables (different cable lengths or different loads could cause this differential phase shift). For further information, refer to page 13 of the datasheet, the section on differential phase shift. This section also provides some suggestions for minimising the errors caused by differential phase shift on the input signals.

AD2S80A_high voltage interface problem

Question Now there is one customer would like to use AD2S80A to replace our older and high price parts RDC1740 and RDC1742. But she encountered the problemthat her input signals are 16V, which is higher than AD2S80A can tolerate. Fortunately, there is one application note AN-265 (in the attachment)described the solution in details. But the transformer 5S70 and 5S72 are obsolete now.The questions are:1. Whether we have any parts to replace 5S70 and 5S72?2. If these two parts are not available, whether we can say that the resistor attenuation can ensure the performance of AD2S80A? If can, whichkind of resistors and which tolerance should be selected?3. Whether there is the third way to do that? Answer Thank you for your email concerning the AD2S80A. The answers to yourquestions are as follows:1. No, we do not have any parts to replace the 5S70 and the 5S72.2. As the Apps note, AN-265, pointed out it is the matching of theresistors, R3, R4, R5 & R6, that is important and not the actual values.The tolerance should be as tight as possible. An example is given inthe Apps note which suggests that if there is a 1% error in the ratioR3/R4 : R5/R6 i.e. caused by a combination of matching and/or toleranceof the resistors this will result in a 16.7 arc min additional error through theconverter i.e. 1% = 10,000ppm, => Error = 1,000 arc seconds = 16.7 arc min.3. The only other method I can suggest may not be practical for your customer.If possible, I would recommend that they reduce the amplitude of the referencevoltage which is being applied to the resolver, which results in lower amplitudeSin and Cos outputs. The advantage of this method is that any errors introducedon the reference excitation is seen equally on both Sin and Cos signals andas a result cancelled out.However, as I mentioned, this may not be practical for your customer. Feedback: NO

AD2S80A: Noise on DIR output

Question The DIR output has a lot of noise on it. What could be the cause of his noise? Answer The DIR output is asynchronous, so in theory it can change in between LSBs even though the direction of rotation does not appear to have changed. However this would usually only be seen if there are very small changes in the direction of rotation i.e. less then 1lsb. This is unlikely to be the cause of the significant noise which you are referring to. The DIR output is derived from the velocity signal. This can be observed on the Integrator output pin. This output voltage is converted to current through the R6 resistor and then fed back into the VCO input. This is then passed through comparators to determine the actual direction. If there are noise/spikes on the VCO input this could trip the comparators but may not be enough to affect the current input to the VCO which defines the position output. You should start by checking for noise on the VCO input.

ad2s80a: Accuracy vs. temperature

Question What is the difference between the AD2S80ABD and the AD2S80ASD ? Is the ASD just more classified than the ABD according to accuracy/temperature? I mean, has the ASD and the ABD the same accuracy of 4 arcmin between -40?C and 85?C ? Or does the ASD type have a accuracy of 8arcmin over the whole range ? I will need a AD2S80 that has an accuracy of 4arcmin between -20?C and 75?C. Answer All the parts have been tested over the full temperature range. And the variation in accuracy is never bigger than 1 arc min. Therefore for a temp range between –20 and +75degC you have the BD (Industrial) and TD (military) with an accuracy of ±4 + 1 arc min (the +1 means that it can change until 1 arc min more over the range) and the UD (military) with an accuracy of ±2 + 1 arc min. The SD you mention has a ±8 + 1 arc min accuracy.

AD2S80A: Can 2 AD2S80As be used with the same inductosyn?

Question Can 2 AD2S80As be used with the same inductosyn? Answer Yes, but they must have the same reference frequency from the same excitation source. I don't see any problem in principle with connecting two AD2S80A components to the same output. Both devices have an input buffer so they shouldn't load the pre amp output. I assume that both outputs will receive the same reference frequency from the excitation source.

AD2S80A_closed loop bandwidth

Question How could I choose the closed-loop bandwidth (fBW with the AD2S80A Are there guidelines. Resolution Ratio of Reference Frequency/Bandwidth 10 2.5 : 1 12 4 : 1 14 6 : 1 16 7.5 : 1 Could you please help to give some comments on it? Thanks. Answer That section is a little confusing I agree. The data sheet states, 5. Closed-Loop Bandwidth Selection (C4, C5, R5) a. Choose the closed-loop bandwidth (fBW) required ensuring that the ratio of reference frequency to bandwidth does not exceed the following guidelines: Resolution Ratio of Reference Frequency/Bandwidth 10 2.5 : 1 12 4 : 1 14 6 : 1 16 7.5 : 1 Typical values may be 100 Hz for a 400 Hz reference frequency and 500 Hz to 1000 Hz for a 5 kHz reference frequency. This would be better interpreted as choose a bandwidth at least the ratios mentioned. For example with 12 bits resolution and a 5 kHz reference frequency the bandwidth should be 1.25kHz maximum but down to 500Hz or a 10:1 ratio should be fine. If the customer continues to have trouble determining the component values I can provide some assistance or I suggest the component selection program. It can be found at http://ez.analog.com/message/41875#41875 if you scroll down through the text.

Difference between part AD2S80ATD/B and AD2S80ATD?

Question What is a difference between part AD2S80ATD/B and AD2S80ATD? In other words could you explain what suffix /B means in the part number? Answer The accuracy spec in the datasheet for AD2S80ATD (T grade) is 4 arc minute typical at 25degC. AD2S80ATD/B is a special grade of AD2S80ATD that meets the 4 arc minute accuracy spec over the full temperature range of -55degC to +125degC and 12V +/-10% supply variation.

AD2S82A: Maximum tracking rate.

Question We need the maximum resolution of conversion in the entire range of the motor (0-1350rpm). AD2S82 datasheet says that 16bit resolution is allowed only up to 975rpm. Discarding what datasheet says I calculate the value of R6 for 1750rpm using 16bit resolution and we try to spin the motor. We were able to operate without any problems for the entire range of the motor, so the question is: can we trust to use this hardware setting or in long term there could be some problems? We have also another application where the motor is 20000rpm and we need to use at least the 12bit resolution. Answer The AD2S82A operates as a tracking resolver-to-digital converter and forms a type 2 closed loop system. The output will automatically follow the input for speeds up to the selected maximum tracking rate. If the maximum tracking rate is exceeded the signal at the output of the phase-sensitive detector (DEMOD O/P)will increase and the converter will be unable to reach a null. The factors which affect the maximum tracking rate are discussed in page 3 of the application note AN-264 Dynamic Characteristics of tracking converters.