Reply by heindekock September 7, 20092009-09-07
We have successfully done this by monitoring the input voltage. If at 11 to
12.6 the engine is off, if the motor is running it will be at 13.2 to
13.8V. We have used this on motorcycles (6 to 6.8), Petrol and diesel cars
as well as Diesel trucks (24, 27.6).

Just a few warnings, when the starter motor is kicking in the voltages go
all over the place, especially on older vehicles, so use a TVS at least.
Also beware that if the battery gets to the end of its life it acts more
like a cap, the voltage will jump up quick but drop if the motor is idling.
We used 30 samples on one second interval to overcome most of these
issues.

To overcome different readings on different hardware mostly due to
resistor values we did not go to the extent of calibrating the ADC but we
just added an ignition voltage as a soft setting. Example: my car indicates
ignition if the input voltage rises over 12.8V, in my wives car its set as
13V.

Have fun
	   
					
---------------------------------------		
This message was sent using the comp.arch.embedded web interface on
http://www.EmbeddedRelated.com
Reply by Meindert Sprang August 26, 20092009-08-26
"Hans-Bernhard Br&#4294967295;ker" <HBBroeker@t-online.de> wrote in message
news:h71g17$hnl$01$1@news.t-online.com...
> But none of that bears much relevance to the actual question. The > answer to that is a solid "Don't do that!" If you want to know if the > main engine is running, acquire that signal at its source, not at some > barely related point several massive disturbances away from it. > Guessing what the voltage at a terminal shared by an alternator, a > battery, and probably a several active loads might mean is just a > nightmare waiting to haunt you.
Jumping into this thread: The suggestion might already be made but every alternator I know of has a 'W' terminal, outputting an AC voltage directly from the windings, commonly used to connect to a revolution counter. That would be your best source to check whether the enginge is running or not. Meindert
Reply by -jg August 25, 20092009-08-25
On Aug 26, 8:02=A0am, Hans-Bernhard Br=F6ker <HBBroe...@t-online.de>
wrote:
> 1 Lucky Texan wrote: > > A. Some vehicles at idle with headlights, heater fan and other > > accessories on MAY not be charging the battery - so current direction > > would give false indication of engine OFF. > > Any remotely current car with a on-board electronics worth its salt will > up the idle revolution frequency sufficiently to avoid that. =A0Very old > ones would do it blindly if you switch on the lights, modern ones have > considerably more sophisticated battery energy management. > > But none of that bears much relevance to the actual question. =A0 The > answer to that is a solid "Don't do that!" =A0If you want to know if the > main engine is running, acquire that signal at its source, not at some > barely related point several massive disturbances away from it. > Guessing what the voltage at a terminal shared by an alternator, a > battery, and probably a several active loads might mean is just a > nightmare waiting to haunt you.
Yes and no. Sometimes allowing a simple retro-fit is worth the effort. ie We have done battery-voltage decision based systems, for coarse power saving decisions - but we did not make mission-critical decisions. . The OP should measure systems (MORE THAN ONE), and look for the DC and AC (Audio) signatures at the battery. eg A properly advanced system could get Tacho info, from the charge ripple frequency spectrum ;) Including temperature improves what it possible, as Lead Acid batteries have a clear tempco. -jg
Reply by August 25, 20092009-08-25
1 Lucky Texan wrote:

> A. Some vehicles at idle with headlights, heater fan and other > accessories on MAY not be charging the battery - so current direction > would give false indication of engine OFF.
Any remotely current car with a on-board electronics worth its salt will up the idle revolution frequency sufficiently to avoid that. Very old ones would do it blindly if you switch on the lights, modern ones have considerably more sophisticated battery energy management. But none of that bears much relevance to the actual question. The answer to that is a solid "Don't do that!" If you want to know if the main engine is running, acquire that signal at its source, not at some barely related point several massive disturbances away from it. Guessing what the voltage at a terminal shared by an alternator, a battery, and probably a several active loads might mean is just a nightmare waiting to haunt you.
Reply by 1 Lucky Texan August 25, 20092009-08-25
On Aug 24, 6:06=A0pm, "eeboarder" <jme...@emittechnologies.com> wrote:
> >"Matt" <fleck2...@yahoo.co.uk> wrote in message > >news:MPG.24f531d7f04409639896de@news.eternal-september.org... > >> In article <h6enpi$m...@news.eternal-september.org>, > >> jstew...@jkmicro.com says... > > >>> eeboarder wrote: > >>> > I'm working on a system using a PIC24FJ256GB106. One task it will b=
e
> >>> > doing > >>> > is monitoring the battery level of an industrial engine that charge=
s
> >>> > one > >>> > 12V or two 12V(24V) batteries with an alternator. These are typical > >>> > automotive batteries. The PIC also has to keep track of when the > engine > >>> > goes from on to off or off to on. > > >>> > Is there any way I can get an engine on/off signal just from the > level > >>> > of > >>> > the lead acid batteries? Otherwise, I'll have to use another > component > >>> > (oil > >>> > pressure switch) to get engine operational status. > > >>> No. =A0You can't. =A0It won't work. =A0Bye... > > >> However you could monitor the rotation of the flywheel with an Hall > >> effect sensor and that will give you a lot of information. > > >Surely if the batteries are being charged it can only be because the > voltage > >applied exceeds their internal EMF? > > >What voltage do you see when the engine is on and off? > > >I would also imagine that the voltage would vary when the alternator is > >turning (possibly a more or less rectified sine wave). =A0You could even > >(tongue in cheek) extract a spectrum and work out the engine speed. > > >Take some measurements. > > I'm very sorry this was posted three times. I submitted the original post > last Monday, but they weren't posted until this Saturday. I thought I did > something wrong posted it again. > > It won't happen again. > > Also, thank you all very much for your help.
I still think your best hope would be to detect current DIRECTION. The 2 problems is, A. Some vehicles at idle with headlights, heater fan and other accessories on MAY not be charging the battery - so current direction would give false indication of engine OFF. B. There would still be 'noise' and , even with Germanium diodes, SOME threshold issues to filter or hysterisis to work around. But, in a general sense, current direction would be a binary indicator.
Reply by eeboarder August 24, 20092009-08-24
>"Matt" <fleck2001@yahoo.co.uk> wrote in message >news:MPG.24f531d7f04409639896de@news.eternal-september.org... >> In article <h6enpi$mo$3@news.eternal-september.org>, >> jstewart@jkmicro.com says... >>> >>> eeboarder wrote: >>> > I'm working on a system using a PIC24FJ256GB106. One task it will be
>>> > doing >>> > is monitoring the battery level of an industrial engine that charges
>>> > one >>> > 12V or two 12V(24V) batteries with an alternator. These are typical >>> > automotive batteries. The PIC also has to keep track of when the
engine
>>> > goes from on to off or off to on. >>> > >>> > Is there any way I can get an engine on/off signal just from the
level
>>> > of >>> > the lead acid batteries? Otherwise, I'll have to use another
component
>>> > (oil >>> > pressure switch) to get engine operational status. >>> >>> No. You can't. It won't work. Bye... >> >> However you could monitor the rotation of the flywheel with an Hall >> effect sensor and that will give you a lot of information. > > >Surely if the batteries are being charged it can only be because the
voltage
>applied exceeds their internal EMF? > >What voltage do you see when the engine is on and off? > >I would also imagine that the voltage would vary when the alternator is >turning (possibly a more or less rectified sine wave). You could even >(tongue in cheek) extract a spectrum and work out the engine speed. > >Take some measurements. > > > >
I'm very sorry this was posted three times. I submitted the original post last Monday, but they weren't posted until this Saturday. I thought I did something wrong posted it again. It won't happen again. Also, thank you all very much for your help.
Reply by Bill Davy August 19, 20092009-08-19
"Matt" <fleck2001@yahoo.co.uk> wrote in message 
news:MPG.24f531d7f04409639896de@news.eternal-september.org...
> In article <h6enpi$mo$3@news.eternal-september.org>, > jstewart@jkmicro.com says... >> >> eeboarder wrote: >> > I'm working on a system using a PIC24FJ256GB106. One task it will be >> > doing >> > is monitoring the battery level of an industrial engine that charges >> > one >> > 12V or two 12V(24V) batteries with an alternator. These are typical >> > automotive batteries. The PIC also has to keep track of when the engine >> > goes from on to off or off to on. >> > >> > Is there any way I can get an engine on/off signal just from the level >> > of >> > the lead acid batteries? Otherwise, I'll have to use another component >> > (oil >> > pressure switch) to get engine operational status. >> >> No. You can't. It won't work. Bye... > > However you could monitor the rotation of the flywheel with an Hall > effect sensor and that will give you a lot of information.
Surely if the batteries are being charged it can only be because the voltage applied exceeds their internal EMF? What voltage do you see when the engine is on and off? I would also imagine that the voltage would vary when the alternator is turning (possibly a more or less rectified sine wave). You could even (tongue in cheek) extract a spectrum and work out the engine speed. Take some measurements.
Reply by Matt August 18, 20092009-08-18
In article <h6enpi$mo$3@news.eternal-september.org>, 
jstewart@jkmicro.com says...
> > eeboarder wrote: > > I'm working on a system using a PIC24FJ256GB106. One task it will be doing > > is monitoring the battery level of an industrial engine that charges one > > 12V or two 12V(24V) batteries with an alternator. These are typical > > automotive batteries. The PIC also has to keep track of when the engine > > goes from on to off or off to on. > > > > Is there any way I can get an engine on/off signal just from the level of > > the lead acid batteries? Otherwise, I'll have to use another component (oil > > pressure switch) to get engine operational status. > > No. You can't. It won't work. Bye...
However you could monitor the rotation of the flywheel with an Hall effect sensor and that will give you a lot of information.
Reply by Jim Stewart August 18, 20092009-08-18
eeboarder wrote:
> I'm working on a system using a PIC24FJ256GB106. One task it will be doing > is monitoring the battery level of an industrial engine that charges one > 12V or two 12V(24V) batteries with an alternator. These are typical > automotive batteries. The PIC also has to keep track of when the engine > goes from on to off or off to on. > > Is there any way I can get an engine on/off signal just from the level of > the lead acid batteries? Otherwise, I'll have to use another component (oil > pressure switch) to get engine operational status.
No. You can't. It won't work. Bye...
Reply by Grant Edwards August 18, 20092009-08-18
On 2009-08-18, eeboarder <jmeyer@emittechnologies.com> wrote:

> I'm working on a system using a PIC24FJ256GB106.
Yea, we know. We got it after the first 4 copies of the posting. -- Grant Edwards grante Yow! An INK-LING? Sure -- at TAKE one!! Did you BUY any visi.com COMMUNIST UNIFORMS??