• State TI Thinks Resolved
  • Locked Locked
  • Replies 15 replies
  • Answers 1 answer
  • Subscribers 29 subscribers
  • Views 3140 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

DRV8811: DRV8811 Issue With Microstepping

Richard Elmquist
Mastermind 45250 points
Part Number: DRV8811
Other Parts Discussed in Thread: , DRV8846

My customer asked the following question:

Can the driver chip support switching back and forth between full step mode and quarter step mode in one running sequence?  

For example, set USM to 10 and send clock to step input pin for the first 3 steps in quarter step mode, and change USM to 00 for the next 100 steps in full step mode; and then change USM back to 10 for the last 2 steps in quarter step mode.  
The reason is to use quarter step mode to achieve resolution and full step to run fast.  Any special setting is needed in order to do so?

Is there any concern of loss step induced between mode switching?  The absolute accuracy is crucial to our application.

I explained to them that they could change the step size at any time as I saw no restriction in the data sheet. I asked them to set up a run profile to verify this. I received the following from the customer:

I set up an experiment to play with the DRV8811EVM board with our stepper motor AM1020V (made by micromo). I directed a fine laser beam to a mirror glued to a disk mounted to the motor shaft and looked at the reflection on the wall to exam the step by step motor movement.   

When used "pulse step" to run the motor in signal step mode, I found that motor seemed to rotate consistently in full step mode.  However, it was quite obvious that motor skipped every two steps in quarter step mode.  The symptom is that when clicked on "pulse step" button motor responded to two clicks and stayed un-move for the next two clicks, and then responded to the next two clicks, and then stayed un-move for the next two clicks, it went on and on like this very consistently. 

 The jumper configuration stayed the same as default; the PPS (clock rate?) was set to about 250HZ; the Accel Rate was set to 100 out of 255, the Decay control was set to Mixed mode, time base was set to 10, 100... (the above settings were changed but symptom seemed to be the same).

They sent further comments and a pdf that I have attached:

DRV8811EVM_10-18-2017.pdf

We address all positions in terms of quarter steps from home.

A typical move (excluding backlash compensation) consists of three phases, depending on the number of quarter steps to move.

Phase 1: After a 4ms pause, while direction and quarter step modes are selected, 0-3 quarter steps will taken (as needed) at 250Hz rate, to bring motor to a full step position (a multiple of 4 quarter steps)

Phase 2: Following another 4ms pause, during which time full step mode is selected, 0-n full steps (as appropriate), are taken at 250Hz rate.

Phase 3: Following another 4ms pause, during which time quart step mode is reselected, 0-3 quarter steps (as appropriate), are taken at 250Hz rate, to bring motor to final position.

Optimizations:

1. The Phase 2 and 3 pauses will only occur if the number of steps taken is > 0.

2. If the entire move does not require any full steps, the phase 1 and 3 quarter step counts will be combined into a signal move that will occur in phase 1.

We have found that 1 or 2 of the Phase 3 quarter steps are usually skipped by both our existing Allegro and TI DRV8811 driver chips.

The attached would be the test report of the stepper motor performance with DRV8811EVM board.  It describes why the stepper motor went 2 steps and skipped 2 steps in quarter step mode. 

 This pattern does not seem to be correct in looking at the Microstepping Indexer in Table 2. Do you know what might be happening from looking at the information that the customer sent.

Thanks for your help with this!

Richard Elmquist

  • 0
    TI__Guru** 114905 points
    Hi Richard,

    Which model of the stepper is used? What is the motor voltage, resistance, and current?

    Can you confirm VM = ~14V?

    Is it possible to capture the current in one of the windings? In most cases this is more helpful than the voltage.
  • 0 in reply to Rick Duncan
    TI__Mastermind 45250 points

    Rick,

    Thanks for your quick response!

    I will get the responses from the customer.

    Here is some further information from the customer:

    One more test was performed today and found the following:
    • Power on the DRV8811EVM board and re-launched the program
    • Used the same GUI parameter setup that resulted in 2 moves ad 2 skips yesterday on 10/18
    • Stepper motor was able to respond to every signal click to advance the motor (2 smaller steps followed by 2 larger steps)
    • Aout1 and Bout1 waveforms showed oscillation as shown in the test report
    The attachment was the test report with the latest amendment.  Please help to find out why it went oscillation and what was wrong with parameters setting.
    I have attached another pdf of their findings:
    I will send you the answers to your questions as soon as I receive them.
    Please let me know as soon as you have some idea of what is happening.
    Thanks for your help with this!
    Richard Elmquist
  • 0 in reply to Richard Elmquist
    TI__Guru** 114905 points
    Hi Richard,

    Voltage oscillations are expected when the device is regulating current. That is why it is important to see the current waveforms, ideally with the voltage waveforms in the same image.
  • 0 in reply to Rick Duncan
    TI__Mastermind 45250 points

    Rick,

    They stated that the supply voltage was 12V and that they were using the EVM.

    I am having them measure the current and I will reply back as soon as I hear from them.

    Thanks for your help with this!

    Richard Elmquist

  • 0 in reply to Richard Elmquist
    TI__Mastermind 45250 points

    Rick,

    Here is some further data from the customer and some comments. Can you please look at this and comment on the customer's ideas?

    I think the "missing step pulses" are caused by the half step offset between full-step mode, and all the micro-stepping modes.

    For the DRV8811, all of the microstepping modes start at a step angle of 0 degrees, while full step mode starts at 45 degrees, so both coils get energized in the home position.

    This is incompatible with my current firmware, which assumes that the 1/4  positions are in phase with the full step positions.

    Is there a micro-stepping driver where the starting full step position is in phase with the starting 1/4 step position?  In other words, is there a driver chip with the same starting step angle for both full and 1/4 step modes?

    Here is a document with the results of their further testing:

    DRV8811EVM_10-22-2017.doc

    Please let me know if you have any further questions for the customer.

    Thanks for your help with this!

    Richard Elmquist

  • 0 in reply to Richard Elmquist
    TI__Guru** 114905 points
    Hi Richard,

    Is there a micro-stepping driver where the starting full step position is in phase with the starting 1/4 step position? In other words, is there a driver chip with the same starting step angle for both full and 1/4 step modes?

    There is some confusion here. The device starts at the same point regardless of the microstep setting. It is referred to as the home state and is the 45 degree electrical angle. Full step, half step, quarter step and eighth step all start here. The difference is what happens one the STEP command is issued.

    If there are missing steps, there may be another reason.

    What is the VM voltage?
    What is the step rate?
    What is the state of the STEP pin when the nSLEEP is raised?
    What is the state of the STEP pin when RESETn is raised?
  • 0 in reply to Rick Duncan
    TI__Mastermind 45250 points

    Rick,

    Thanks for your response!

    The step rate is shown in the document and the voltage is 12V. The other voltages are shown in the attachment.

    I will have to check with the customer as to the state of the STEP pin when nSLEEP and the RESETn pins are raised , but the customer has gone into detail in the attachments as to the state of the device. The settings are shown and the customer has included some scope shots of the operation.

    I will respond back as soon as I hear from the customer.

    Thanks for your help with this!

    Richard Elmquist

  • 0 in reply to Richard Elmquist
    TI__Mastermind 45250 points

    Rick,

    Here are the responses to your questions (in BLUE):

    What is the VM voltage?  12V
    What is the state of the STEP pin when the nSLEEP is raised? didn't use nSLEEP but will find out
    What is the state of the STEP pin when RESETn is raised? didn't use RESETn but will find out STEP pin turned red at each click

    They also asked the following:

    I'd like to know if you or other TI engineers know good small size stepper motors that may work better with TI stepper motor drivers. 

    The main concern in our application is accuracy.  We use stepper motor to control optical device angle and position.  Therefore, backlash and loss of steps are not allowed.  Our package is pretty compact; size is also a concern.  The load of our device is not very large (don't have spec of torque to provide yet).  The one we currently use is AM1020 series stepper motor with 69.2:1 gearbox.

    Please let me know if you are able to find the issue and let me know if you might be able to offer some options for the customer.

    Thanks for your help with this!

    Richard Elmquist

  • 0 in reply to Richard Elmquist
    TI__Mastermind 45250 points
    Rick,

    Are there any updates on this issue?

    Please let me know when you might be able to respond.

    Thanks for your help with this!

    Richard Elmquist
  • 0 in reply to Richard Elmquist
    TI__Guru** 114905 points
    Hi Richard,

    Sorry, I missed this one.

    The motor should be ok, but there are some issues with the EVM settings. If the sense resistors have not been changed on the EVM, the chopping current may be set too high for the motor. This could result in damage to the motor due to excess heat.

    According to the datasheet the motor currents are either 0.25A, 0.18A, 0.09A, or 0.045A depending on the model number. The chopping current is set to 1.6A according to the docs sent. None of the motors will reach this current, but the current will be greater than the motor datasheet. In the A-.25-8, the current could be 6X the recommended current.

    Assuming the motor model used is a V-12-250, try changing the sense resistor to 7 Ohms and setting VREF to 2.5V. This should set the current at 0.045A, allowing the current to regulate at the lower settings.

    ------------------------------------------------------------ -----------------------------------------------------------------------------

    The DRV8846 may be a better fit for this application. The device requires fewer external components and has AutoTune, a feature to control the current on a cycle by cycle basis.
  • 0 in reply to Rick Duncan
    TI__Mastermind 45250 points

    Rick,

    Thanks for your response!

    I will send these recommendations to the customer.

    Do these settings that they are using imply that the would see the issue that they are reporting?

    Would an excessive current cause the motor to not step properly?

    I just want to make sure that we are addressing the issue that the customer is seeing.

    Thanks for your help with this!

    Richard Elmquist

  • 0 in reply to Richard Elmquist
    TI__Guru** 114905 points
    Hi Richard,

    Yes, this could be part of the of issue. There could be other issues hiding behind this.

    Excessive current may affect motor stepping. It can certainly affect the motor lifetime.
  • 0 in reply to Rick Duncan
    TI__Mastermind 45250 points

    Rick,

    I know that it has been awhile, but the customer has built another board and I have attached a summary of their findings.

    DRV8811 Issue With Microstepping.docx

    They still seem to be having issues with their board. Please let me know if you have any idea what might be happening?

    Would you like to see their board? Please let me know.

    Thanks for your help!

    Richard Elmquist

  • 0 in reply to Richard Elmquist
    TI__Guru** 114905 points
    Hi Richard,

    Are the outputs loaded?
    Why are they toggling between 1/4 step and full step?
    What are they expecting to see when toggling between 1/4 step and full step?
  • 0 in reply to Rick Duncan
    TI__Mastermind 45250 points
    Rick,
    I will ask them and respond back as soon as I hear from them.
    Thanks for your help with this!
    Richard Elmquist