you don't have a great need for precise motion control (relays are
slow). The key to success here is a proper relay bias and a guard
around the MOSFET. You should also put a zener on the gate if you are
using any voltage over 10V for the motors so that the gate ringing won't
overshoot the Vgs rating of the MOSFET. Also, make sure that you have
the PWM set to zero when you switch directions with the relay to avoid
arcing the contacts and the relay will last longer. It is a good low
cost solution for high current motor controls.
DLC
Shaggy wrote:
>At the rate things are going for me, I may not realistically be getting to
>this until after New Years. I will certainly try to get to it sooner than
>that, but I have a house remodeling project that I need to get done
>inbetween now and when I leave for the holidays. Oddly enough, the house
>project is related to the robot, as it involves the LAN that brain will be
>distributed across.
>
>Thanks for the link and the suggestions, that will give me a couple steps in
>the right direction for when the motors start off in motion.
>
>The part a bout the MOSFET and relay seems like the most solid solution.
>Would I be correct in assuming that you were suggesting that the relay would
>be an open switch when either the OOPic was off, or when the OOPic was
>starting up, including the 500 delay (which I have included after seeing it
>so highly recommended here by you and others). Then I would close the relay
>via the MOSFET, and leave it closed all the time that the OOPic was running.
>This could be accomplished using a logic level MOSFET, which I probably have
>lying around. Is there any particular relay you would recommend?
>
>I'm inclined to actually go in that direction first simply because it seems
>more definitive than using the resistors to bias the input. Perhaps that's
>not the best reason to choose one technique over another. The resistors
>would certainly be cheaper, but neither solutions is actually expensive, so
>I tend to favor the definitive one. Is that wrong?
>
>
>
>rtstofer wrote:
>
>
>>--- In oopic@yahoogroups.com, Shaggy <charrington@...> wrote:
>>
>>
>>>Then I will go with the Sabertooth.
>>>
>>>Thanks for the information.
>>>
>>>
>>Until I read the documentation at
>>http://www.dimensionengineering.com/Sabertooth2X5.htm I was concerned
>>that the device would only work with R/C pulses. That would seriously
>>limit resolution due to the manner in which the OOPic generates the
>>pulses. The .pdf is here:
>>http://dimensionengineering.com/datasheets/Sabertooth2x5.pdf
>>
>>But the good news is that the H-Bridge will also work in analog mode
>>and you can use the oPWM objects to control the motors.
>>
>>You may want to deal with the issue of 2.5V being the OFF condition.
>>When the OOPic first starts up, the IOLine is an input and floating.
>>The motor may run away briefly. I might put a 20k resistor between the
>>input and +5V and another 20k resistor between the input and Gnd in an
>>attempt to bias the input to 2.5V until the oPWM object takes over.
>>
>>Obviously, you want to start with the oPWM at 50%, then establish the
>>IOLine and finally set the operate property. Hopefully, you will get a
>>bumpless transition into the stop condition.
>>
>>Thinking about the OOPic.Delay = 500 as the first executable statement
>>(and it is HIGHLY recommended) the motor could run away for a very long
>>time.
>>
>>If the bias idea doesn't work, put a suitable SPST relay in series with
>>the motor power and control it via the OOPic and a trasistor of MOSFET.
>>
>>One way or the other, you need to deal with this bias situation.
>>
>>Oh, and read the note on filtering the PWM signal on page 10 of
>>the .pdf. If you use the bias resistors, the resistor at the pin needs
>>to be a lot smaller (maybe 1k or less) to allow the pin to pull the
>>signal all the way up or down. The 1k resistor is working in series
>>with the bias resistors and forms a voltage divider. Given that the
>>resistor is much smaller, the capacitor should be much larger - say 1
>>ufd (probably electrolytic).
>>
>>I suspect there will still be a little tinkering to do...
>>
>>Richard
>>
>>
>>
>>
>>
>>
>>
>>
>
>
>
--
------------------------------------------------------
Dennis Clark ooPIC Tech Support
www.oopic.com
------------------------------------------------------
[Non-text portions of this message have been removed]
Yahoo! Groups Links
<*> To visit your group on the web, go to:
http://groups.yahoo.com/group/oopic/
<*> Your email settings:
Individual Email | Traditional
<*> To change settings online go to:
http://groups.yahoo.com/group/oopic/join
(Yahoo! ID required)
<*> To change settings via email:
mailto:oopic-digest@yahoogroups.com
mailto:oopic-fullfeatured@yahoogroups.com
<*> To unsubscribe from this group, send an email to:
oopic-unsubscribe@yahoogroups.com
<*> Your use of Yahoo! Groups is subject to:
No comments:
Post a Comment