Larry,
I agree about motors, in and of themselves, not being particularly sensitive to
fast risetime needs. Yes, the driving waveform could be slowed down by
appropriate filters, but I do urge caution in one regard: what are the power
dissipation specs of the drivers used in the controller? Slower risetimes =
greater power dissipation in the driving semiconductors. If you can
successfully filter the motor drive lines and not cook the drivers, fine. I
would say that you have been most fortunate with your previous motor noise
fixes.
Remember, there is a lot of "give and take" in design of power electronics.
EMC guys want waveforms as slow as possible and driver device guys want fast
risetimes and short duty cycles. That's the way it goes.
73, Dale
WA9ENA
-----Original Message-----
>From: Larry Benko <xxw0qe@comcast.net>
>Sent: Aug 16, 2012 12:52 PM
>To: Dale <svetanoff@earthlink.net>
>Cc: rfi@contesting.com
>Subject: Re: [RFI] Motors/controllers and something to think about
>
>Dale,
>
>I agree in some instances but the discussion here has been for motors of
>significant size in appliances. Rise times when driving something such
>as a motor with a large mass is not that important. I have done
>considerable work driving fairly large stepper motors and assuming a
>drive frequency of 5kHz (period of 200us) I could never measure any
>performance differences when the rise time was .1us, 1us or, 5us. Yet a
>5us rise time produced much less RFI. However a rise time of 50us would
>seriously produce a different waveform and probably affect the
>performance of the motor.
>
>Every large appliance I have tamed was not fixable with clamp on ferrite
>cores (even with multiple turns through the core) which only affect
>common mode currents. In all cases I needed to insert (cut into wiring)
>and put in a differential filter. The differential filter slowed the
>motor drive rise times and the RFI was eliminated or greatly reduced.
>
>My $.02
>
>Larry, W0QE
>
>On 8/16/2012 9:13 AM, Dale wrote:
>> Larry,
>>
>> The issue of rise time may also be inflexible for some systems. While I
>> have often encountered design situations in which rise time was considerably
>> faster than required (the digit guys wanted "reserve speed"), there are
>> other cases in which the interfacing devices (at either end of a signal
>> feed) require very fast risetimes and the resulting steep "fronts" they
>> present. You then have an RFI/EMC issue to resolve.
>>
>> 73, Dale
>> WA9ENA
>>
>>
>> -----Original Message-----
>>> From: Larry Benko <xxw0qe@comcast.net>
>>> Sent: Aug 16, 2012 9:48 AM
>>> To: rfi@contesting.com
>>> Subject: Re: [RFI] Motors/controllers and something to think about
>>>
>>> Jim and others,
>>>
>>> I have taken several EMC compliance classes over the years and they
>>> always stress that there are 3 main issues that contribute to RFI. They
>>> are in no particular order:
>>>
>>> 1.) Signal rise time
>>> 2.) Signal current
>>> 3.) Loop area of wire carrying the signal
>>>
>>> Signal current is usually the least flexible parameter but rise time and
>>> loop area are quite easy to alter. Components can affect the signal
>>> rise time and good construction can reduce loop area. Loop area
>>> problems can be both printed circuit board issues and external wiring.
>>> No earth shaking news here.
>>>
>>> 73,
>>> Larry, W0QE
>>>
>>>
>>> On 8/15/2012 11:40 PM, Jim Brown wrote:
>>>> On 8/15/2012 10:30 PM, Roger (K8RI) wrote:
>>>>> The issue has dealt with specific makes and models with and without RFI,
>>>>> but as PWM is by its nature an RFI generator we need to stop and think
>>>>> about how companies (any company) acquires the motors and speed
>>>>> controllers.
>>>> It is critically important to realize that it is EASY to build a PWM
>>>> system without RFI if you simply pay attention to the fundamental laws
>>>> of physics -- keep track of where the current flows, ALL of the
>>>> current, including DC, baseband AC, and RF -- and design your product so
>>>> that it flows in loops that have small cross sectional areas, and over
>>>> wires that don't make good antennas. It's really as simple as that.
>>>> The most recent edition of Henry Ott's classic text on EMC even has
>>>> chapters devoted to switching power supplies and PWM motor controllers.
>>>>
>>>> 73, Jim Brown K9YC
>>>> _______________________________________________
>>>> RFI mailing list
>>>> RFI@contesting.com
>>>> http://lists.contesting.com/mailman/listinfo/rfi
>>>>
>>> _______________________________________________
>>> RFI mailing list
>>> RFI@contesting.com
>>> http://lists.contesting.com/mailman/listinfo/rfi
>>
>
_______________________________________________
RFI mailing list
RFI@contesting.com
http://lists.contesting.com/mailman/listinfo/rfi
|