next, replace the 297 was Re: Was Chopper circuit - Now ver 3.2, G298
Posted by
mariss92705
on 2002-04-14 09:15:24 UTC
Hi,
I now know the difference between the L297 and the L298. It will be
called the G298. The PCB is layed out and goes to prototype next week.
In my opinion, a constant off time chopper does suck (not sure if I
used that exact word, but it will do). Here's why:
The constant off time algorithm works like this:
(1) Apply power supply voltage to the coil. Current increases at a
rate slightly less than V/L.
(2) Sense the coil current. Is it less than the reference? If yes, go
to (1), if no, go to (3).
(3) Reverse the voltage on the coil. Current will now decay at a rate
of slightly more than V/L.
(4) Wait a fixed period of time (typically 20uS).
(5) Go to (1).
The problem is, though the "off" time is fixed, the "on" time will
vary somewhat with current, typically from 20uS to 30uS. This means
the frequency will range from 20 kHz to 25 kHz.
The motor has 2 coils, so 2 choppers are needed. Each is a free-
running oscillator.
Anytime you have two free-running oscillators, they have a strong
tendency to phase lock (both run in lock-step). While they do, the
coil currents becomes divergent, so phase lock is broken after a few
cycles. The currents is restored to the correct values and phase lock
is re-established.
This making and breaking phase lock generates the annoying squealing,
hissing and growling noises you hear. It also causes the motor to
vibrate if the audible frequencies fall within the pass band of the
motor.
Mariss
I now know the difference between the L297 and the L298. It will be
called the G298. The PCB is layed out and goes to prototype next week.
In my opinion, a constant off time chopper does suck (not sure if I
used that exact word, but it will do). Here's why:
The constant off time algorithm works like this:
(1) Apply power supply voltage to the coil. Current increases at a
rate slightly less than V/L.
(2) Sense the coil current. Is it less than the reference? If yes, go
to (1), if no, go to (3).
(3) Reverse the voltage on the coil. Current will now decay at a rate
of slightly more than V/L.
(4) Wait a fixed period of time (typically 20uS).
(5) Go to (1).
The problem is, though the "off" time is fixed, the "on" time will
vary somewhat with current, typically from 20uS to 30uS. This means
the frequency will range from 20 kHz to 25 kHz.
The motor has 2 coils, so 2 choppers are needed. Each is a free-
running oscillator.
Anytime you have two free-running oscillators, they have a strong
tendency to phase lock (both run in lock-step). While they do, the
coil currents becomes divergent, so phase lock is broken after a few
cycles. The currents is restored to the correct values and phase lock
is re-established.
This making and breaking phase lock generates the annoying squealing,
hissing and growling noises you hear. It also causes the motor to
vibrate if the audible frequencies fall within the pass band of the
motor.
Mariss
--- In CAD_CAM_EDM_DRO@y..., "ballendo" <ballendo@y...> wrote:
> Alan,
>
> And while we're at it; can we get Mariss to describe in more detail
> an offhand comment he made some time ago, which has stuck with me...
>
> That being something along the lines of: (Mariss "speaking")
> Constant time off pwm sucks (he said something more like "has very
> poor performance"<G>). Doing it differently is harder to be sure,
but
> worth it, and I enjoy the challenge...
>
> (constant time off PWM is used in the L297)
>
> Ballendo
>
> --- In CAD_CAM_EDM_DRO@y..., Alan Marconett KM6VV <KM6VV@a...>
wrote:
> > Hi Mariss, Paul,
> >
> > Yes! It does sound like an interesting project. Now all we have
> to do
> > is "publish" a simple plan (and code) for a PIC or PAL
replacement
> for
> > the '97.
> >
> > Alan KM6VV
> >
> >
Discussion Thread
Alan Marconett KM6VV
2002-04-09 12:03:31 UTC
Re: Was Chopper circuit - Now ver 3.2, G298
Paul R. Hvidston
2002-04-09 13:06:51 UTC
Re: [CAD_CAM_EDM_DRO] Re: Was Chopper circuit - Now ver 3.2, G298
Alan Marconett KM6VV
2002-04-09 14:58:04 UTC
Re: Was Chopper circuit - Now ver 3.2, G298
Paul R. Hvidston
2002-04-09 15:13:44 UTC
Re: [CAD_CAM_EDM_DRO] Re: Was Chopper circuit - Now ver 3.2, G298
Tony Jeffree
2002-04-09 17:42:32 UTC
Re: Was Chopper circuit - Now ver 3.2, G298
Alan Rothenbush
2002-04-09 18:05:08 UTC
Re: Re: Re: Was Chopper circuit - Now ver 3.2, G298
Alan Marconett KM6VV
2002-04-09 19:05:27 UTC
Re: Was Chopper circuit - Now ver 3.2, G298
Larry Edington
2002-04-09 19:35:00 UTC
Re: [CAD_CAM_EDM_DRO] Re: Was Chopper circuit - Now ver 3.2, G298
Peter Homann
2002-04-09 22:38:32 UTC
RE: [CAD_CAM_EDM_DRO] Re: Was Chopper circuit - Now ver 3.2, G298
tonyjeffree
2002-04-09 23:25:48 UTC
Re: Was Chopper circuit - Now ver 3.2, G298
Paul R. Hvidston
2002-04-12 06:56:43 UTC
Re: [CAD_CAM_EDM_DRO] Re: Re: Re: Was Chopper circuit - Now ver 3.2, G298
ballendo
2002-04-14 03:51:26 UTC
next, replace the 297 was Re: Was Chopper circuit - Now ver 3.2, G298
mariss92705
2002-04-14 09:15:24 UTC
next, replace the 297 was Re: Was Chopper circuit - Now ver 3.2, G298
tonyjeffree
2002-04-14 20:12:22 UTC
Indexer
scottdbtnet
2002-04-26 15:06:36 UTC
Re: Re: Was Chopper circuit - Now ver 3.2, G298
Paul R. Hvidston
2002-04-26 16:00:05 UTC
Re: [CAD_CAM_EDM_DRO] Re: Re: Was Chopper circuit - Now ver 3.2, G298
scottdbtnet
2002-04-26 21:50:49 UTC
Re: Was Chopper circuit - Now ver 3.2, G298
scottdbtnet
2002-04-26 21:53:18 UTC
next, replace the 297 was Re: Was Chopper circuit - Now ver 3.2, G298
Alan Rothenbush
2002-04-27 09:35:45 UTC
Was Chopper circuit - Now ver 3.2, G298
Steven J. Owen
2003-05-09 14:40:04 UTC
Indexer
Keith Bowers
2003-05-09 16:40:08 UTC
Re: [CAD_CAM_EDM_DRO] Indexer
CL
2003-05-09 18:07:33 UTC
Re: [CAD_CAM_EDM_DRO] Indexer
Harvey White
2003-05-09 19:23:06 UTC
Re: [CAD_CAM_EDM_DRO] Indexer
stevenson_engineers
2003-05-10 02:06:20 UTC
Re: Indexer
David A. Frantz
2003-05-10 11:35:15 UTC
Re: [CAD_CAM_EDM_DRO] Indexer