1067 Controller - unrecognized

Comments & issues
kayaker
Phidgetly
Posts: 35
Joined: Thu Jan 03, 2013 3:58 pm
Contact:

1067 Controller - unrecognized

Postby kayaker » Tue Feb 12, 2013 8:45 am

I've been working with 2 of the new 1067 bipolar controllers for the past few days. I oftern get an "unrecognized device" error when I plug the boards into the USB port. I never got this error message with the 1063 boards. Is there a new driver for the 1067's. Any thoughts on how I might get by this issue?

I run them on a Win7 - 64 bit Pro | Intel core(6) i7 12GB | 930 @ 2.80 GHz | duo 23" 1920 X 1080 Dell box, so it's not the machine.

Thanks in advance.
Take Care,

Jim. . .

fraser
Engineering
Posts: 264
Joined: Thu Nov 19, 2009 4:41 pm
Contact:

Re: 1067 Controller - unrecognized

Postby fraser » Tue Feb 12, 2013 9:10 am

Yep, those controllers are running a new USB stack. I'm also running 64b windows 7. I feel like the only time ive seen that happen is if you plug it in, and unplug it before the drivers get installed, Windows seems to get confused and not install it properly the second time you plug it in. This is all just off the top of my head though. I can test this issue out for you and I will let you know what i find.

Does this happen on a consistent basis? And on the same device being plugged into the same USB port?

kayaker
Phidgetly
Posts: 35
Joined: Thu Jan 03, 2013 3:58 pm
Contact:

Re: 1067 Controller - unrecognized

Postby kayaker » Tue Feb 12, 2013 5:23 pm

fraser wrote:Yep, those controllers are running a new USB stack. I'm also running 64b windows 7. I feel like the only time ive seen that happen is if you plug it in, and unplug it before the drivers get installed, Windows seems to get confused and not install it properly the second time you plug it in. This is all just off the top of my head though. I can test this issue out for you and I will let you know what i find.

Does this happen on a consistent basis? And on the same device being plugged into the same USB port?


I've had them plugged in all day and just went to launch an app and neither of the attached 1067's showed up in the app interface. Unplugged one then plugged it back in and got that attached message. Pulled both of them and plugged them in a couple more times and they the next time I launched the app they were both there.

Is there a way to force Windows 7 to reinstall the drivers?

It seems somewhat random. The same USB ports are being used as with the 1063's for the same app.
Attachments
1067 error message.JPG
1067 error message.JPG (22.94 KiB) Viewed 12638 times
Take Care,

Jim. . .

User avatar
Patrick
Lead Developer
Posts: 3109
Joined: Mon Jun 20, 2005 8:46 am
Location: Canada
Contact:

Re: 1067 Controller - unrecognized

Postby Patrick » Fri Feb 15, 2013 9:24 am

Ok,

1. Does a 1067 that was previously working and plugged in stop working without being unplugged/the computer being rebooted?
2. How often would you say the 1067 comes up as 'unrecognized'? 1/10 times? 5/10 times? 9/10 times?
3. Do you have a different system that you can test the 1067's on? Maybe a different OS?
4. What motherboard to you have? Are you attaching the 1067's directly or via a hub?

This is not an issue that we have seen here, and we aren't able to reproduce, this is why we haven't responded for a while.

The fact that 2 1067's do this point to either a computer problem, or a 1067 firmware/hardware bug. It's unlikely that you got 2 faulty 1067's. A firmware bug would be nice, as we can juts issue an update, but we can't do anything unless we can reproduce the issue.

-Patrick

kayaker
Phidgetly
Posts: 35
Joined: Thu Jan 03, 2013 3:58 pm
Contact:

Re: 1067 Controller - unrecognized

Postby kayaker » Fri Feb 15, 2013 11:13 am

Patrick wrote:Ok,

1. Does a 1067 that was previously working and plugged in stop working without being unplugged/the computer being rebooted?-Patrick

No. It is either not recognized when I reboot the computer which manifests itself as neither 1067 showing up in the app where I have the connected serial number shown on the form. It will generate the "Unrecognized error" when I unplug and plugin the cables. This never happens on this same app with a 1063 controller
Patrick wrote:2. How often would you say the 1067 comes up as 'unrecognized'? 1/10 times? 5/10 times? 9/10 times?

Depends on the day, but I would say half the time when it is unplugged and plugged back in and then it may need to be plugged in a number of times before the "Unrecognized" error message goes away. Definitely enough to be and inconvenience.
Patrick wrote:
3. Do you have a different system that you can test the 1067's on? Maybe a different OS?

At this point I have only tested it on one system.
Patrick wrote:4. What motherboard to you have? Are you attaching the 1067's directly or via a hub?


I run them on a Win7 - 64 bit Pro | Intel core(6) i7 12GB | 930 @ 2.80 GHz | duo 23" 1920 X 1080 Dell box
Patrick wrote:
This is not an issue that we have seen here, and we aren't able to reproduce, this is why we haven't responded for a while.

That would be nice to hear when it is happening, so the user (me) doesn't feel like the concern is falling on deaf ears and starts shouting louder to the point where I begin to sound like a complete idiot.

Patrick wrote:The fact that 2 1067's do this point to either a computer problem, or a 1067 firmware/hardware bug. It's unlikely that you got 2 faulty 1067's. A firmware bug would be nice, as we can juts issue an update, but we can't do anything unless we can reproduce the issue.
-Patrick

Well, it never happens with the 1063 controller of which I have two. In fact, my app employs 2 bipolar controllers, a 1067 (wishing I could use both but there are other 1067 issues that prevent that) and a 1063. The 1063 always connects, but sometimes at the same time, the 1067 does not. I may not be a smart diagnostician, but if one works at the same time another doesn't, it would lead me in the direction that it is more likely a firmware/hardware issue than as user computer issue.

Other 1067 issues. . .

The 1067 runs great if you set a target position, head towards it and when you get there, you stop. However, it doesn't seem to like to change direction well. My app runs 2 bipolar controllers. I like the 1067 because it lets the first motor run faster on your 3301 motor. With the 3304 - 3.75 stepper on the second 1067 controller, it stops randomly (and doesn not restart) during the execution of the program, which has a lot of direction changes for the controller, but none that even approach its speed capability. (I can send you the app if you like). In fairness, the 1063 doesn't like directional changes on the 3.75 stepper either. However, I have a Haydon 1.8 degree captive linear actuator, which is not a consumer grade product like the 3301 and 3304 motors. My app runs the Haydon captive linear actuator PERFECTLY on the 1063 controller. It traverses in and out (changes direction through the 1063 controller) the exact number of times relative to the speed of the other bipolar motor. But, on the 1067 it just sits there and grinds, which is a real bad thing on a captive linear actuator potentially.

Any thoughts on that one?
Take Care,

Jim. . .

User avatar
Patrick
Lead Developer
Posts: 3109
Joined: Mon Jun 20, 2005 8:46 am
Location: Canada
Contact:

Re: 1067 Controller - unrecognized

Postby Patrick » Fri Feb 15, 2013 1:37 pm

It sounds like your steppers are stalling on the 1067, while on the 1063, using the same speed/acceleration/current limit settings, they run perfectly?

In this case, I'd guess that you may need to re-calibrate your chosen current limit settings. You may need to choose different values for the 1067.

The 1067 also micro-steps full time, while the 1063 switches to full stepping for the faster speeds. This should allow the 1067 to drive a same motor at a faster speed then the 1063.

If your acceleration is low-ish, the motors may be spending too much time in their resonant zone, which is the most likely speed for a stall to happen at.


Are you changing acceleration/velocity while the motors are moving, or just position. The 1067 should handle any changes to parameters while moving perfectly - but there could possibly be a bug here if you are seeing bad behaviour.

-Patrick

kayaker
Phidgetly
Posts: 35
Joined: Thu Jan 03, 2013 3:58 pm
Contact:

Re: 1067 Controller - unrecognized

Postby kayaker » Fri Feb 15, 2013 2:04 pm

Patrick wrote:It sounds like your steppers are stalling on the 1067, while on the 1063, using the same speed/acceleration/current limit settings, they run perfectly?

Yes, on the Haydon linear actuator, the 1063 runs perfectly. Not so on the 1067, just sits there and grinds.

Patrick wrote:In this case, I'd guess that you may need to re-calibrate your chosen current limit settings. You may need to choose different values for the 1067.

If I set the current limit higher, the motors get way hot!. As an example, I set the 3301 at .5 amps. and it runs fine so long as you do not run a pattern of change direction a lot. If so, sooner of later it stalls.

Patrick wrote:The 1067 also micro-steps full time, while the 1063 switches to full stepping for the faster speeds. This should allow the 1067 to drive a same motor at a faster speed then the 1063.

Yes, I understand that, that is why I bought the 1067's as 600 rpm on a 1.8 degree stepper is a lot slower than I would like for my winder.

Patrick wrote:If your acceleration is low-ish, the motors may be spending too much time in their resonant zone, which is the most likely speed for a stall to happen at.

I set the acceleration high (1,000,000) towards the upper limit so there will not be wire build up at the end of the traverse on the coil being would. I have tried 918000, 900000, 768000, 512000 and all the same. Remember, on a 3.75 stepper (3304), is where it stalls and I assume on a turn. With the Haydon 1.8 linear actuator, the 1067 just sits there and grinds, regardless of the acceleration parameter setting.

Patrick wrote:Are you changing acceleration/velocity while the motors are moving, or just position. The 1067 should handle any changes to parameters while moving perfectly - but there could possibly be a bug here if you are seeing bad behaviour.

No, I am not changing the velocity or acceleration during the operation cycle, only the target position.

Please, try the app in your lab. Maybe the way I coded the traversing of the linear actuator function isn't liked by a 1067.

Edit..................................... Just tried yet another motor on the 1067.....

I just received a larger Haydon linear actuator this afternoon. I put it on the 1067 controller that is running the actuator function of the app. In a cycle of 10,000 winds of the winder motor, the linear actuator is set to traverse 18 times. In those 18 direction changes, twice the controller did not change the direction of the rotation. It decelerated like it was going to, but continued on in the same direction. This happened once in each direction during the full operation.

When I put the new linear actuator on the 1063 and set the identical parameters for a 10,000 wind operation, the 1063 functioned flawlessly. So, if it is my coding, the 1063 is OK with it and the 1067 is not.

Here is the Sub code snippet for the direction change for the linear traverser controller/motor.

Private Sub Traverse()

LinearStepper.steppers(0).CurrentPosition = 0
StepNext = StepNext * (-1)
LinearStepper.steppers(0).TargetPosition = StepNext
TraverseCounter += 1

End Sub

My money is on either firmware/hardware issue with the 1067. It travels great in a straight line, but ask it to turn a bunch of times and it cannot process the commands property (in my uninformed and non technical opinion). Or maybe it processes faster than the 32 ms feedback timing from the controller on the position change.
Last edited by kayaker on Fri Feb 15, 2013 3:25 pm, edited 1 time in total.
Take Care,

Jim. . .

User avatar
Patrick
Lead Developer
Posts: 3109
Joined: Mon Jun 20, 2005 8:46 am
Location: Canada
Contact:

Re: 1067 Controller - unrecognized

Postby Patrick » Fri Feb 15, 2013 3:17 pm

You can send me the code and I'll have a look.

patrick@phidgets.com

kayaker
Phidgetly
Posts: 35
Joined: Thu Jan 03, 2013 3:58 pm
Contact:

Re: 1067 Controller - unrecognized

Postby kayaker » Fri Feb 15, 2013 3:49 pm

Patrick wrote:You can send me the code and I'll have a look.

patrick@phidgets.com


Done.

You should run the "Set Winder rpm:" parameter at ~1,100 rpm. and "No. of Turns" to 10,000 to get a good feel for effect direction change has on the 1067.

Thanks.

edit.................................

Here is a link to a YouTube video last week with the winder on a 1067 and the traverser (linear actuator) on a 1063. You can tell as the current.change is only available on the 1063.

http://www.youtube.com/watch?v=Yu3pyv0B ... e=youtu.be

This is how it is intended to function.
Take Care,

Jim. . .

rdrr
Fresh meat
Posts: 1
Joined: Mon Feb 03, 2014 9:20 am
Contact:

Re: 1067 Controller - unrecognized

Postby rdrr » Tue Aug 12, 2014 1:09 pm

Just curious as to whether or not this issue was resolved. About to start coding a similar project and wonder if this is something I should be looking out for.

Thanks


Return to “General”

Who is online

Users browsing this forum: No registered users and 2 guests