Generic Incremental Quadrature Pulse Encoders

Overview

Quadrature pulse encoders generate an A and B pulse which are 90 degrees out of phase with each other. These encoders are usually used with external counting circuitry which counts either the pulses or the edges. The resulting count increments or decrements depending upon the direction of rotation since the rotation direction determines whether the A pulse will lead or lag the B pulse. Quadrature pulse encoders have traditionally been referred to as incremental encoders. However, since some manufacturers are now using the terms incremental and absolute to mean something else entirely, in this manual we will call these encoders "Quadrature Pulse" to reduce confusion. When testing these encoders, select “Generic Incremental Encoder” from the Encoder Feedback menu of the Select Feedback window which appears after pressing the Select Feedback button (see previous sections on Basic Operation).

Many encoders currently used with servo motors are quadrature pulse encoders. Furthermore, several serial encoders also include a quadrature pulse section. Examples of this would be the Sumtak, Sanyo Denki, and Heidenhain Endat encoders. Thoroughly testing these encoders generally requires testing the serial section by choosing the correct serial model selection and testing the quadrature pulse section by selecting an incremental encoder with the correct number of counts per revolution. Please refer to Section 2.6 Manufacturer Specific Encoder Details for more information on this topic.

Besides quadrature pulse, there are encoders that are single channel or single channel with an index (i.e.: they provide only an A line or an A and Z line). These encoders are often used as tachometers to provide a pulse rate proportional to RPM. Technically these encoders cannot be called quadrature pulse because they do not have two pulses that are 90 degrees apart. This also means that they cannot indicate direction. They will increment the count in either direction. In industrial counting systems, another type of pulse called A Count B Direction may be encountered. With this type of pulse, each pulse on the A line will increment or decrement the count by 1 depending upon whether the state of the B Direction line is HI or LO respectively. Stepper motor indexers commonly utilize this type of pulse. The single channel and A count B direction signals are not supported by the current software revision of the TI-5000JX.

 

Connection

The generic incremental quadrature pulse encoder will have A, B, and Z lines (often both true and complement lines for each) which must be connected to J1 as indicated by the J1 pin configuration in Table 2.1. It will have a 0V common power supply line that must be connected to J1 pin 2. If it operates from 5 VDC power, the 5V line must be connected to J1 pin 1. If it uses something other than 5 V, it may be powered by a bench power supply (if no adapter module for the TI-5000JX is available for it). The power supply line from the encoder should be connected to the + output of the bench power supply. The – power supply line should be connected to the 0V line from the encoder at J1 pin 2 of the TI-5000JX. J1 pin 1 of the TI-5000JX must not be connected when an external power supply is used.

Generic incremental quadrature pulse encoders often have commutation lines (often both true and complement), U, V, and W (sometimes S1, S2, S3 and other designations) for providing startup commutation information to the servo motor drive. These lines should be connected to J2 pins in accordance with the J2 pin configuration.

When Hall effect devices are connected to the TI-5000JX, the U, V, and W lines should be connected to J2 pins 7, 8, and 9 respectively. The 0V common power supply line to the Hall devices should be connected to J1 pin 2. If the Hall devices can be powered by 5 VDC (which they often can), the + Hall power line should be connected to J1 pin 1. If a higher voltage is required, the + Hall power line should be connected to the + output of an external bench supply. The – output of the external supply should be connected to the Hall 0V line at J1 pin 2. J1 pin 1 must not be connected when an external supply is used or the TI-5000JX may be damaged.

 

Data Display

Data Display is selected by clicking the Data Display button to read A, B, and Z line states and the count accumulated by the transitioning of these lines in a quadrature pulse pattern. The following figure shows the Data Display for Generic Incremental Encoders.

 

Commutation

The commutation information usually appears in the commutation frame, but it really depends upon whether the encoder utilizes sine/cosine commutation. Only a few encoders employ the sine cosine commutation output. The two types of commutation currently displayed are discussed in the next two sections. Clicking on the Sine/Cosine Commutation radio button will change to display to better show the sine/cosine commutation. Clicking on the Normal Commutation radio button will change it back.

Normal Commutation

The normal commutation display automatically appears as the default when first entering Data Display, and it will display commutation in the Commutation frame.

The boxes with headings H1, H2, H3, H4, H5, and H6 display the states of the lines connected to J2 pins 7, 8, 9, 10, 11, and 12 respectively. This display is used to display the commutation states of both Hall effect switches and encoder commutation lines. Any line that is not connected will show up as ‘H’ since the line is pulled to a HI state by internal pull-up resistors. When using Hall effect outputs, the complement lines often do not exist, so the last 3 outputs will be HI all the time. Just ignore any parts of the display that are not being used.

When complement lines are in use, it is important to verify that the true and complement lines are in the opposite states at all times. In any case, it is important that the encoder provide the commutation states in the correct pattern. You should rotate the encoder in the forward direction of the motor to verify the correct commutation pattern along with true and complement states. The Table below shows the standard manner of connecting commutation signals to the J2 connector and indicates under which heading the line state information appears. The table also indicates the standard commutation pattern that you should expect to see. You are free to connect commutation lines in other ways if you prefer. This is the way test cables from Mitchell Electronics, Inc. will be wired because this is compatible with the TI-3000JX Run Test System.

This commutation pattern will repeat as many times in one motor revolution as there are pole pairs. In other words, this pattern will repeat 4 times on an 8 pole motor.

The practical use for the commutation line states is for “setting the commutation” or aligning the encoder so that the commutation line pattern is synchronized with position of the rotor magnets relative to the stator armature windings. This can usually be easily verified applying a small DC voltage to the armature leads to lock the rotor and observing the commutation line pattern as the shaft is wiggled slightly from the lockup position. The TI-5260 PM Rotor Lockup Switch is a very convenient and organized way to provide lockup currents to the windings for checking and setting commutation alignment.

An example would be a Yaskawa motor with U, V, and W lines. If it is locked up with +U –V –W (sometimes called +U to neutral), it should show (in View 1) V HI, W LO, and U ready to toggle with a small wiggle of the rotor.

This pattern will repeat over and over. Notice that the Z pulse is independent of the A and B pulses. For practically all encoders, the Z pulse will go HI once per revolution. The portion of the angle for which it is HI, and the states of A and B during which it is HI is very dependent upon the particular encoder manufacturer.

The box labeled MECHANICAL ANGLE will display the correct mechanical angle providing the correct number of counts/revolution has been entered. The box labeled ELECTRICAL ANGLE will display the correct electrical angle providing that correct numbers have been entered for counts/revolution and number of motor poles.

Currently the TI-5000JX will show a positive count when quadrature signals are received with phase A leading phase B. Many American and European motors will use encoders which have this phase relationship for CW (clockwise) rotation. For many of these same motors, the forward armature direction is also CW. In these cases the displayed electrical angle provides a very convenient way to check and set commutation.

On the other hand, the forward armature direction for many Japanese motors is CCW. When turned in the
CCW direction the encoder phase is often B leading A which results in a negative count. This means that the displayed electrical angle will decrease instead of increasing when the motor is turned in the forward CCW direction. While this does not rule out using the electrical angle display for commutation setting, it does make the display confusing. The TI-7000 allows the user to set the positive direction of encoder rotation, and this feature may eventually be incorporated into the TI-5000JX.

 

Sine Cosine Commutation Encoders

The Normal Commutation display automatically appears as the default when first entering Data Display. If you click the Sin/Cos Commutation radio button, the display will switch to the sine/cosine display shown below. Clicking on the Normal Commutation radio button will switch the display back to the normal commutation.

The C and D channel sinusoidal amplitudes are converted to an angle which is displayed in the Commutation frame under the label MECHANICAL ANGLE. This angle should change smoothly from 0 to 359 degrees in one revolution. Each ¼ turn should cause a 90 degree change in the angle. It is very similar to verifying a resolver.

If the number of poles is set correctly, the electrical angle will be displayed under the label ELECTRICAL ANGLE. This is probably not too important since you will want to use the Normal Commutation display to check and set commutation. Also, if you have an encoder in which the C and D channel signals go through 360 degrees more than once, the mechanical and electrical angle interpretation will no longer be correct.

This display will show the sine/cosine commutation outputs for encoders which use that type of output, such as the Heidenhain ERN1387 and RON3350 encoders used on Siemens servo motors. The TI-5010 cable is compatible with these encoders on Siemens motors. The TI-5101 1V P-P Adapter Module must be used with sine/cosine encoders in order to multiplex the C and D commutation channels into the TI 5000JX A and B channel inputs. Dipswitch #1 on the TI-5101 must be set to the closed position in order to multiplex the C and D channels properly (note: dipswitch #2 does not matter, so you can set it closed too just to make sure you have the right one). If this is not done, you will see the angle display moving through 360 degrees with a slight encoder rotation instead of a complete revolution. It should require a full revolution to move through 360 degrees although there may be some which go through 360 as many times as the number of pole pairs for the motor. 

Signal levels, interpreted as the sine and cosine of the displayed angle, are displayed in the Encoder Status frame in boxes labeled Cosine and Sine.

C and D channel signal levels are displayed as voltages in the boxes labeled C-C* and D – D*. These are the differential voltages between the true and complement lines. The software takes into account the 3.48 gain through the TI-5101, so is the actual differential output voltage of the encoder. The Heidenhain specification is for the peak differential voltage to range between 0.75V p-p and 1.2V p-p. This means that you should see amplitudes in the ranges shown in the following table. The maximum, minimum and nominal values are shown.

Remember you are checking the peak differential voltage, so you must move the encoder to the positions shown in the table above.

Some Siemens motors may use ERN1381 and RON350 encoders. These are the same as the ERN1387 and RON3350 respectively but without the C and D channels.

The models ERN1387 and RON3350 provide 8,192 counts per revolution. If the counts/rev is set to 8192, the Normal Commutation display will also show 360 per revolution in the MECHANICAL ANGLE box, and that angle will agree with the C/D channel angle. However, please note that the original TI-5000 angle reading for the C/D channels did not agree. This is due to the way the C and D channels were assigned as sine/cosine for that unit. The current TI-5000JX uses different assignments so that the C/D channel angle and the angle from the encoder count agree. The relationship between the original TI-5000 and current TI-5000JX is shown in the following table.

The Siemens alignment is based on the Z pulse, so it is more accurate to use the Normal Commutation display rather than the sine/cosine angle for alignment. Also the count angle in the Normal Commutation display can be displayed as an electrical angle. The static alignment procedure is to apply lockup voltages with the polarity –U +V +W, and the electrical angle should be very close to zero degrees at each lockup position. Remember, to get the correct electrical angle, you must enter the correct number of poles.

 

Count

The display shows the count accumulated by the A and B line transitions in the Count frame in the boxes labeled DECIMAL and HEX. Normally the user will be more interested in the decimal count than the hexadecimal representation. You should see counting activity when you turn the encoder. If you do not, you may have a problem with the A or B lines. Again, the Line Level Test or an oscilloscope may help in diagnosing the problem. When you rotate the encoder one revolution, the count should change by the number of counts per revolution specified for that encoder. If you do not have that information, it can be determined by the Count Test.

 

Rate

The Rate frame shows the encoder rate in counts/sec or RPM. The counts per second may be important when back-driving an encoder with a drive motor for determining whether it is moving too fast to perform certain tests. The RPM may be useful when back-driving a motor for checking magnet and winding integrity. In order for the RPM to read correctly, the correct number of counts/rev must have been entered into the setup.

 

Line States

The Line States frame utilizes 3 boxes with headings A, B, and Z to display the states of the A, B, and Z lines coming from the encoder. The A and B lines should change following the quadrature pulse pattern shown below when moved in the direction in which the A pulse leads the B pulse.

The A, B, and Z lines will often be available as both true and complements. In this case, it is important to make sure that both lines are changing as required. The Line Level check (to be discussed later) provides this information. The Incremental Count Test and Phase Angle Test further ensure that this pattern is correct and producing the correct counts. 

 

Index

The Encoder Status frame begins with a box labeled INDEX. This box will indicate ALARM if the encoder is
not yet indexed and OK after it has been indexed. When you first enter the test, it will indicate ALARM. After you have rotated the encoder past the index pulse, it will change to OK and the count will zero out. If this does not occur, you may have a problem with the Z pulse. You can further check it out using the Line Level Test and/or an oscilloscope.

You may wish to force the tester to re-index the encoder on the next index pulse. Click on the “Zero count on next index” button to accomplish this. The INDEX box will change to ALARM until the next index pulse occurs. You may also click the Data Display button to re-index.

 

Line Levels

The Line Levels function provides for measuring the proper incremental encoder (not used on serial) output voltages over a 0 to 14.9 volt range.

WARNING - Input voltages above 15 VDC may damage the TI-5000JX input circuitry.

This voltage range covers many of the very common incremental encoders. If precise or high speed readings are required, a digital voltmeter or an oscilloscope can be used. However, the Line Levels function can provide a good, quick, convenient indication of the state of each output line. The Line Level Test is only available for incremental encoders. The Line Level display is shown in the following figure.

 

 

As the encoder is slowly rotated, the following items can be checked:

1. Each A, B and Z line is changing states and not sticking in the same state constantly.
2. Each A*, B*, and Z* complement line (if the encoder employs them) is changing states and is not
   sticking in the same state constantly.
3. Each A*, B*, and Z* complement line is always in the opposite state from its A, B, and Z counterpart.

The Z line would normally be difficult to catch in the HI state since it is high for a very short part of the entire rotation. However, the Line Level function also provides the encoder count and will zero the count on the first Z pulse encountered. The INDEXED column will show ALARM until the encoder is rotated past the Z pulse. When this occurs, the INDEXED column shows OK. At this point, if the encoder is rotated back until the count is again approximately zero, the point at which the Z line goes to the HI state can be found.

The realtime levels should be checked to make sure they are correct voltages. The Instruction box indicates that TTL levels would be 0.8V or below for a LO level and 2.0 V or above for a HI level. If you have specifications for your encoder that show different requirements, then those specifications are what you should apply in evaluating the encoder.

 

Real Time Levels

To help ensure that each line is reaching both states, the LO state minimum and HI state maximum values are reported on the lines “LO state” and “HI state” respectively. The minimum and maximum displays are described below.

As you turn it, it will pick up each state for the A, A*, B and B* lines very quickly. Again, the Z and Z* lines may not be caught so easily since the Z pulse occurs only once per revolution. If the encoder is rotated in such a way as to make the count pass through zero a number of times, you should be able to catch the Z pulse HI.

The first line is the real-time line level display. It shows the present voltage from each encoder line and the encoder count relative to when it was first indexed. The voltages for the true lines are in the columns designated by A, B, Z while the complement line voltages shown in columns designated by A*, B*, Z*.

The example display above shows how the real time line level display looks when the Z pulse is HI. The count is at zero, and the HI Z pulse has been found. You can see how the Count display was used to get to zero to find the Z pulse HI.

 

Min Levels

The line below the real-time levels is the minimum line level display. It shows the lowest voltage recorded for each encoder line. Every line has been shown to go to a LO state of 0.1 V. What we are looking for here is a line which has not gone to the LO state. That would indicate a problem with the encoder or cable. In the case of the Z or Z* lines, it could mean that the encoder needs to be moved more slowly through the zero count in order to pick up the state of the index line.

 

Max Levels

The line below the minimum levels is the maximum line level display. It shows the highest voltage recorded for each encoder line. All lines are 2.0 V or above which are valid HI states. This means that all lines have gone HI during our test. Again, we would be looking for a line that has never gone HI as an indication of a problem. In the case of the Z pulse, try to move as slowly as possible through the zero count so that the TI-5000JX can catch the fast moving state change of the Z pulse.

 

Count Test

The count test will verify that the encoder is incrementing the correct number of counts per revolution. It works basically the same way for incremental encoders as it does for serial encoders as described in a previous section. The Count Test must be run by turning the encoder slowly by hand. The Continuous Count Test may be used while motoring incremental encoders at higher speeds.

The Count Test is performed by slowing rotating the encoder by hand through 4 or more revolutions until the box at the bottoms indicates Test Completed. The data in the results boxes will indicate how close to the exact number of counts per revolution were registered, whether any bits failed to change state during a particular revolution (stuck bits), and whether any internal errors were detected during a particular revolution. The data boxes are described in detail below.

 

Revolution

Each row on the display contains the results of the test for one revolution. The revolution box shows which revolution the test is on at any time during the test. After 4 revolutions, the test is complete.

It is a good practice to re-run the test turning the encoder in the opposite direction, but the encoder must be turned in only one direction in each individual test. Changing directions during a test will cause the test to fail. Running the test several times may uncover a problem if the problem is intermittent. Click on the Restart Test button to re-run the test.

 

Count

The actual count at the end of each revolution is displayed in the Count box. Ideally this number should increase each time by the number of counts per revolution for the encoder. 

 

Target

The Target box provides the target count which will be registered at the end of each revolution. It is against this target that the actual count is compared.

 

Error

The Error box displays the difference in the data in the Count box minus the Data in the Target box. Since the target count is what we expect and the actual count is what we got, the difference is going to be the error.

Due to differences in how that Count Test is performed for incremental encoders and serial encoders, the error is quite different. For incremental encoders, the error for a good encoder should not be more than ±2.

 

Count/Rev

The box called Counts/Rev is the difference in the count at the end of the current revolution and the count at the end of the last revolution. Ideally this should be equal to the number of counts per revolution of the encoder. For incremental encoders, it will usually be within ±2. For incremental encoders, it is useful in determining the number of counts/rev of an encoder for which you have no specifications.

 

Stuck Bits

The Stuck Bit test does not apply to incremental encoders

 

Internal Error

Error information is normally transmitted by serial encoders, so the internal error field does not apply to incremental encoders.

 

Continuous Count Test

The purpose of the Continuous Count Test, like the count test, is to ensure that the correct number of counts occur from one index pulse to the next when the encoder is rotated. Unlike the Count Test during which the encoder must be turned slowly by hand for 4 turns, this test may be run at high speed for many revolutions. The encoder may be rotated by hand, or it may be driven by the motor. Also unlike the Count Test, it is unaffected by changes in direction, so it can be run while an incremental encoder is in use on a machine. The Continuous Count Test is available only for the incremental encoder selection.

Like the Count Test, the Continuous Count Test compares the actual number of counts received at each index pulse to the target number of counts. The Revolution box shows 10 revolutions, and since this encoder is 32,768 counts per revolution, we know that the target number of counts after 10 revolutions would be 327,680 which is the number showing in the target box. The Count box shows that 327,680 counts have been received, and the Error box shows 0 which would be the number in the Count box minus the number in the Target box. This would be a perfect result after 10 revolutions.

Due to the update rate of the display, you may not see every revolution reported as the RPM increases. This is not a problem as the tester is keeping track of the revolutions even though there is not time to display them all.

The TI-5000JX Continuous Count Test can continue to function with a change in encoder rotation direction. This is very helpful in testing an encoder while it is actually running on a machine. On direction change, this test may show an error if the index pulse is very wide in terms of pulses on the A or B channel.

If the Continuous Count Test runs long enough, the maximum capacity of the tester to keep track of counts and revolutions will eventually be reached. If the test were to continue to run after reaching these maximum values, the test would begin to incorrectly report counting errors. To avoid false error reporting, the software checks for reaching these maximums. The maximums that are detected and the action taken is a follows.

 

Count Maximum

The count maximum will be reached when the absolute value of the encoder count is greater than 2,147,483,648 (2 ** 31 – 1). When this occurs the test is stopped, and the Test Status box displays “Test Stopped – Reached Max Count”. The last test values before the maximum was reached will remain on the display and will be printed on the report.

 

Target Maximum

The target maximum will be reached when the absolute value of the target count is greater than 2,147,483,648 (2 ** 31 – 1). When this occurs the test is stopped, and the Test Status box displays “Test Stopped – Reached Max Target Count”. The last test values before the maximum was reached will remain on the display and will be printed on the report.

 

Revolution Maximum

The revolution maximum will be reached when the absolute value of the number of revolutions is greater than 32,000. When this occurs the test is stopped, and the Test Status box displays “Test Stopped – Reached Max Revolutions”. The last test values before the maximum was reached will remain on the display and will be printed on the report.

 

Phase Angle Test

This section describes performing the Phase Angle Test. For an explanation into why phase angle tests should be performed and what the results tell you, please see Section 3.2.1. The Phase Angle Test is available only for the incremental encoder selection.

 

Realtime

The 93 in the Readings box indicates that 93 phase measurements have been made so far during the test. That gives you some idea as to how representative the readings are. The longer the test is running, the better representation you have of the phase angles in terms of the min/max readings. The Realtime line reports the phase and symmetry angles in real-time as they happen. The realtime data will be constantly updated as the test proceeds. However the speed at which it is updated is dependent upon the speed at which the encoder turns. The data on the Realtime line will probably be changing constantly because the phase angles tend to vary somewhat as the encoder is rotated. The better quality encoders will exhibit more constant phase and symmetry angles. To get the most accurate readings, it is essential to rotate the encoder at a constant speed (RPM).

High resolution encoders often use interpolation to divide their sine wave output into smaller pulses. The interpolation process generally causes phase angles to be jumpier. If the TI-5000JX is reading phase angles at a high rate (controlled in part by the encoder rotational speed), it can be hard for the operator to interpret the readings from jumpy data. Clicking the Take Sample button will take a sample of the real time data and hold it on the display. This makes it much easer to read. This can be done as many times as desired. The Sample line shows a sample that was saved from a realtime reading when the Take Sample button was clicked. An initial sample of the first real-time reading is taken automatically when the Phase Test is first started.

The Phase B-A box reports the phase angle between the rising edge of B and the rising edge of A with A leading B. It would ideally be 90 degrees when the encoder is moving in the direction of A leading B, or 270 degrees in the direction of B leading A. See Section 3.2.1 for a description of these phase angles.

The A Symmetry box reports the A Symmetry angle, which is the angle from the rising edge of A to the falling edge of A, and it would ideally be 180 degrees. Likewise, the B Symmetry box reports the B Symmetry, which is the angle from the rising edge of B to the falling edge of B, and it too would ideally be 180 degrees. These readings provide a measure of the symmetry of the pulses. A perfectly symmetrical pulse would be HI for the same amount of time that it is LO, and that would result in an angle of 180 degrees. Specifications for these angles vary with encoder manufacturer, but a tolerance of ±22 degrees from the ideal is a fairly typical specification.

The lowest rate for phase angle measurements is 25 counts per second. For rates slower than 25 CPS, the RT line will not be updated.

Similarly, for pulse rates above 440,000 CPS, there is not enough resolution in the internal timer to resolve phase angles to 1 degree. Therefore, above about 400,000 CPS, a TOO FAST prompt appears on the RT line in place of the phase data.

You may wish to calculate the maximum and minimum RPM values corresponding to the max and min count rates. The encoder RPM may be calculated from the number of counts per second and the number of lines per revolution as follows:

RPM = (60 * RATE) / (4 * LINE)

For example, using a 500 line per revolution encoder, 20 CPS and 400,000 CPS would convert to 0.6 RPM and 12,000 RPM respectively. This gives you the allowable RPM range for reading the phase angle for this particular encoder. You can see from this that as the encoder resolution goes up, the allowable RPM for phase readings will come down.

Substituting in 400,000 CPS for the maximum rate and 25 CPS for the minimum rate gives us the following two equations for the max and min RPM:

MAXRPM = 6,000,000 / LINE
MINRPM = 375 / LINE

Rotating the encoder while keeping the speed as constant as possible in the allowable RPM range, should result in good phase angle measurements.

 

Max Val

The Max Val line shows the maximum values of the phase and symmetry angles since the test started or the last max/min reset.

To further help in evaluating jumpy data, the maximum and minimum values are recorded and displayed along with the realtime values. The max and min values may be reset by clicking the Reset Max/Min Values button. The reset process will set the max and min values to the values in the current real-time data reading. The values are automatically reset when the Phase test is first started. If the encoder is stopped during a test, it may cause a wild max or min value to be recorded. This would be a situation in which you would definitely want to perform a reset. Of course the test can be exited and restarted to reset the max/min values as well. The max value readings are most useful when the encoder is turned at a constant speed with a motor. The max values will normally show a lot of variation when the encoder is rotated by hand and therefore may not be that useful. The speed differences caused by rotating by hand will cause variation in the readings, and it usually does not tell you much about the quality of the encoder. Therefore, max values should be ignored if the encoder rotational speed is not constant.

 

Min Val

The Min Val line records the minimum values in the same manner as described above for the maximum values.