Steering motor/mechanism for APRS vehicle

[Jerry Heim]

Hey guys,

I'm looking for a recommendation for a steering mechanism for our APRS vehicle (Viator).  This needs to be torquey and quick.  I'd like 'stop to stop' (~90 degrees of rotation) control in < 1sec.  We've got slicks on our vehicle with a 200+ pound payload.  I'd like to be able to turn the wheels with the vehicle at a dead stop.  We've got lot's of power available onboard.

I do have a potentiometer set up for steering position feedback, so I don't really need a servo--I can close the steering position loop in software.

For our APRS vehicle last year, I originally purchased a DC geared motor from eBay.  During testing I eventually sheared some of the teeth in the gearbox.  I tried a few other brushed DC options and then I found this beast:

https://www.allelectronics.com/item/dcm-473/3-phase-brushless-dc-motor-w/12-rpm-gear-head/1.html

The gearbox on this thing is awesome--really, really beefy for the price, but the motor is brushless.  No problem (I thought), I'd just purchase a brushless DC motor controller to drive the thing. I tried several different commercial BLDC motor controllers--all of these commercial controllers have a 'soft start' feature--this is great if you're driving the wheels on a hoverboard, but all were found to be waaaay to slow for steering control.

Unfortunately, we ran out of time and we ended up fielding our vehicle with one of these slow-responding BLDC controllers.  I attempted to tune a PID loop to work with this steering box, but it simply wasn't practical.  The result was frequent encounters with the hay bales. 

I see three approaches for next year:

1. Keep the steering gearbox and the BLDC motor and roll my own controller.  I'm comfortable with designing power electronics, but I'm not too familiar with driving BLDC motors.  Is there any chance I'd be able to control this motor with  custom controller and get a quick response?  Note that the motor does have hall-effect feedback sensors, so quick startup seems feasible to me--there are plenty of resources online that I can use to build a controller-- but I haven't found any good reference designs that demonstrate a quick and reliable response time.  I don't have time to do this design work now, but I should have time this summer to give this a try.

2. Keep the gearbox and work to attach it to a brushed DC motor.  The BLDC motor's rotor and the worm gear of the gear box are the same piece--it wouldn't be trivial to couple the worm gear to a different motor.  Probably more mechanical work than I'm comfortable doing.

2. Shop around for a good geared DC motor or servo.

Advice?  What are you guys using?

Thanks,

-Jerry

[Jon Watte]

Have you considered linear actuators? It seems like they would not shear as easily because of the way the worm gear engages with the threaded rod. Especially if you put some moly grease on it!

You can get them in any amount of force from a variety of vendors.

Sincerely,

jw

Sincerely,

Jon Watte


--
"I find that the harder I work, the more luck I seem to have." -- Thomas Jefferson

--
You received this message because you are subscribed to the Google Groups "diyrovers" group.
To unsubscribe from this group and stop receiving emails from it, send an email to diyrovers+unsubscribe@googlegroups.com.
To post to this group, send email to diyr...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/diyrovers/994fa8cc-8cec-487c-96be-5eabced388b9%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Ted Meyers]

Problem with linear actuators is that they are slow.  Of course with enough leverage, anything is fast!  I'm using a linear actuator, even though it is slow, and I did have to reduce the leverage to keep it from stalling (I could have tried a bigger actuator, but didn't due to space and cost).  Over-volting the 12V motor made it faster though, but it still doesn't turn the wheels quickly.

Sounds like you just need a bigger motor.   I'd stick with brushed.   Of course, building your own servo control with a big, powerful motor can be risky.  You might end up tearing things apart, if you get the end limits wrong; I've made that mistake.

You might also look into hydraulics.

To view this discussion on the web visit https://groups.google.com/d/msgid/diyrovers/CAJgyHGMK1bPr2YJf%2BTa-_EckBbFELSUrG2T5_pjrNvt0qeTNDw%40mail.gmail.com.

[John Leichty]

Assuming that your existing mechanical system (motor/gearbox and whatever linkage attached to it) is capable of a quick-enough response for you, you should be able to build or buy a controller to do that. That's a bit of a cop-out response, but doing some engineering to make sure the mechanical system is suitable is where I'd start. That would be figuring out what torques are required where the rubber meets the road (and at what speeds) and the inertia of the mechanism leading up to it. Then you can see if the motor/gearbox combo is appropriately specced.

If you have the budget for it, one of the Roboteq brushless motor controllers could do the job. They're just about the cheapest "professional" BLDC motor controllers available. Something like the MBL1330 ($245) or the SBL1330 ($175).

https://www.roboteq.com/index.php/component/virtuemart/298/mbl1330-detail?Itemid=971

https://www.roboteq.com/index.php/component/virtuemart/383/sbl1360-276-detail?Itemid=971

On Mon, Feb 12, 2018 at 11:41 AM, Jerry Heim <jwad...@msn.com> wrote:

[Jerry Heim]

I did consider linear actuators--I looked at 4 inch stroke actuators available on eBay for around $100.  Seems they'd do the job, but they wouldn't have a very quick response.  That would probably be my fall-back solution.  The BLDC motor & gearbox from All electronics certainly has sufficient torque and the speed that I need.  And now my vehicle mechanics are built around the all electronics gearbox.  I'm leaning hard toward attempting my own BLDC motor controller design.  Seems to me that with the hall effect sensors in the motor that I should be able to develop something without a lot of pain (from frying my fingers on overstressed MOSFETs). 

 

John, I might pick up one of the Roboteq BLDC controllers to see how well it does in my application.  Thanks for that pointer.

Thanks for the responses--I'll keep you all posted regarding any progress.  I really can't start this project until summer though...

-Jerry