From: Chuck Simmons
Organization: You jest.
X-Mailer: Mozilla 4.61 [en] (X11; U; Linux 2.0.33 i586)
Subject: Re: Speed control for small d.c. motors
Date: Fri, 20 Sep 2002 14:03:55 GMT
NNTP-Posting-Date: Fri, 20 Sep 2002 07:03:55 PDT
John Woodgate wrote:
> I have two vinyl disc players with Matsushita d.c. motors in which the
> AN6652 speed control chip has departed. They are known to be not very
> reliable and are now unobtainable AIUI.
> I want to replace with a discrete speed controller. I have an old
> Philips cassette recorder with a controller that looks sophisticated but
> originally didn't work because the AC127 germanium npn leaked like a
> sieve. But I can't see some of how it works anyway, which is annoying.
> The current sense resistor seems to be very low in value; only 17 mV
> across it. OK, all of and change in it gets to the emitter of the npn,
> through the diode level-shifters, but the gm of the Ge npn is not all
> that high at Ie = 4.5 mA. I don't see very much loop gain. This is what
> it looks like (use Courier font):
> +------1 kohm----------+
> | e BC328 |
> + 7.5 V (batt) o--------+----\_/---+--6.5 ohm--+---+------+
> 9,0 V (mains) | | | |
> +-||-+ 4.7 nF BA315 V |
> | BA315 V |
> +-100 ohm-+ | |
> | | | Motor
> 360 ohm | | |
> | |/ AC127 | |
> Speed adj. 100 ohm pot---|\e Ge npn | |
> | | | |
> 620 ohm +----------+ |
> | 560 ohm |
> 0 V o--------------+--------------------+------+
> The motor takes 26 mA at 3.7 V off load, which is almost the same as I
> want for the turntables, but I have a 14 V supply available. The BA 315
> diodes look like ordinary double-plug diodes, nothing special.
> With a 'good' AC127, it holds the motor current and voltage within close
> limits from 6 V to 12 V applied, but I can't get it to work with a
> silicon npn and a suitably adjusted 360/100/20 pot chain to allow for
> the larger Vbe; that is the major problem. I'm not surprised; Philips
> didn't put a Ge device in there for no reason. I suppose I'm missing
> something obvious.
The circuit is sufficiently obscure for such a simple circuit that it
seems best just to look at the physics of what is to be accomplished.
The voltage across a permanent magnet motor is Em=I*R+w*Ke where I is
current, R is internal resistance, w is angular velocity and Ke is the
generator constant of the motor. Because T=I*Kt where T is torque and Kt
is the torque constant of the motor, I can write Em=T*R/Kt+w*Ke. It
happens that Kt and Ke are directly proportional (equal in MKS units) so
that the right hand term, w*Ke, dominates in low friction situations
whenever Kt is sufficiently large. So for constant speed, without
tachometer feedback, the best you can do is a constant voltage source.
Perhaps looking at it that way will help.
... The times have been,
That, when the brains were out,
the man would die. ... Macbeth
Chuck Simmons firstname.lastname@example.org