Re: PWM a truck alternator

On Sat, 03 Oct 2009 15:56:05 -0700, Joerg <invalid@xxxxxxxxxxxxxxx>
wrote:

Jim Thompson wrote:
On Sat, 03 Oct 2009 14:29:08 -0700, Joerg <invalid@xxxxxxxxxxxxxxx>
wrote:

Jim Thompson wrote:
On Sat, 03 Oct 2009 13:27:52 -0700, Joerg <invalid@xxxxxxxxxxxxxxx>
wrote:

Michael Robinson wrote:
What's a good frequency range for PWM of a truck alternator? Cummins 5.9
liter diesel engine on a Dodge Ram.

In the circuit I've designed, the frequency varies with duty cycle -- it
hits its highest frequency at 50% duty cycle. Just looking for a ballpark
range -- 100 to 1 kHz, 1kHz to 10 kHz, or what? I know 100 Hz is fast
enough not to see headlight flicker, but might there be some other reason
100 Hz would be too low...?

I usually go above the audio threshold but mind animals (thery hear very
much higher than we do). I don't know alternators enough to say whether
they'd like such direct PWM or whether current-sink switcher plus filter
would be better. PWM in the audio range could cause a nasty buzz on the
stereo.


A little background. I'm using PWM for a couple of reasons -- won't need a
heat sink, and it's an interesting challenge. The stock configuration in
this truck has the vehicle's computer doing linear control of the field,
holding the system voltage at a measly 14.2 volts. This truck eats up a lot
of electrical power in the winter what with the intake air heater and other
stuff I have installed. Would have to drive for many hours between each
start to get the battery back up to anything near a full charge, especially
in cold weather. Batts (truck has two) are chronically a littlle low even
in the summer. So I intend to up the voltage to about 15 this winter for
the batteries' sake.

15V might damage your battery.


I'm not going to bother trying to find out if it's even possible to reset
the voltage at the computer. Both ends of the field connect to terminals
outside the alterntor, making it dead easy to install my own DIY external
voltage regulator.

If that's so easy to get to, where does the vehicle computer sense how
high the battery voltage is? Not sure if this would be feasible here,
but what if you'd place a (big enough, if it uses its own supply to
gauge Vbat) P-Channel FET in series there? This you could be controlled
via a circuit that makes your desired 800mV drop at this FET and the
computer would now "think" that 15V is really 14.2V. Can be adjusted for
outside conditions if desired. I still believe 15V can eat your battery.
An quick read of my ancient patents will show you how to easily
temperature compensate a regulator.

What's so difficult about temperature-compensating a regulator?


Maybe it's time to again demonstrate my excellence in circuit design
?:-)

Well, if you feel overly generous you could volunteer to design one for
Michael :-)


Do modern alternators use a low-side or a high-side switch on the
field?

No idea, the ones I came across were low-side. But those aren't in cars.

Before my designs of the early '60's, the regulators were
electromechanical, and high-side. Then we went low-side to
accommodate readily available power NPN transistors.

Car companies will do weird things just to save the cost of a wire, ...


So would I. The epitome of frugality was my Citroen 2CV. Alternator on
the crankshaft, no belt costs, meaning no belt to break. In fact the car
did not have any belts. Not even safety belts but I put good ones in
there. Distributor? Nope, too expensive, coil was across two spark plug
tips so it always fired on both cylinders. That made for some
"interesting" effects when starting via a hand crank, including the
embarrassing moment when the 2nd muffler pot blew off the pipe with a
loud bang and rolled down the street.

The gen was a DC brush type and AFAIR only one fat yellow wire went
there. The regulator was a large metal pot with rheostats and huge relay
contacts so they probably did it bang-bang. The whole gen was smaller
than a coke can.

Oil pressure switch? Nope. Overtemp lamp? Nope. Other than a charge
failure light there was nothing on the dashboard. Well, there wasn't
really any dashboard.

[...]

An accident waiting for a location ?:-)

...Jim Thompson
--
| James E.Thompson, CTO | mens |
| Analog Innovations, Inc. | et |
| Analog/Mixed-Signal ASIC's and Discrete Systems | manus |
| Phoenix, Arizona 85048 Skype: Contacts Only | |
| Voice:(480)460-2350 Fax: Available upon request | Brass Rat |
| E-mail Icon at http://www.analog-innovations.com | 1962 |

I love to cook with wine. Sometimes I even put it in the food.
.

Loading