Diesel Engine Activity Sensor

I need to build a portable, simple, and easy-to-attach engine monitor for logging (each minute) the average engine speed of a large diesel engine (e.g. 4 - 8 cylinders). My device needs to be very simple to attach to the engine, and cannot require manipulating existing engine wiring, fuel lines, etc. No, I can't install a tachometer. I can 'clamp' some sensor(s) onto the engine, via spring-clamp, or magnet mount, not nuts and bolts. If this were a gasoline engine, I'd clamp something around a spark-plug wire and count inductive pulses, but diesels don't have spark-plugs of course. This device will be placed by guys in the field who probably don't want to fool with it. With that picture of reality, does any means of sensing engine RPM come to mind?

If that's too easy for all of you who are much smarter than me, in the same non-invasive way, has anyone any ideas for monitoring fuel flow? Ultimately that's what I'm trying to get to through knowing the RPM. Thanks!

Does the car in question has an ODB-II interface? - Arduino CAN-Bus OBDII demo - YouTube -

Thinking out loud: add a microphone and do an FFT on the sound signal to determine the frequencies? (never tried)...

If ODB is possible this is a nice site - http://www.obd-codes.com/ -

This is cool, and although you've given me an idea for another project, the diesels in question are driving huge water pumps. Thanks!

Rob, I just saw your idea about a microphone ... I think that's worth a try, and perhaps vibration sensing as well. There is certainly a huge difference between sound / vibration at the high-rpm. I may look at exhaust manifold temperature as well by way of a magnetically mounted temperature sensor.

Why not use a hall effect sensor to measure the RPMS?

robfmcc:
Why not use a hall effect sensor to measure the RPMS?

Ditto. Usually a water pump, alternator, etc. is exposed enough to mount a sensor nearby. Another option is to hook a sensor wire to the alternator output before it hits the rectifying diodes. Just be sure not to block airflow and to use a very-temp-rated wire. It gets hot in there!

Hall sensor. Now there's an idea. But just to be certain I'm in line with your thinking, will you elaborate in a sentence or two about what the hall-effect sensor is picking up if a) mounted near a rotating water pump and /or b) near an alternator output (pre-rectifier)?

Pardon my previous post if I appear lazy, but I really wish to be certain I'm not missing anything. I assume in one instance, the Hall Effect sensor is reading the pulses produced by a passing magnet (that is a magnet mounted on a rotating element). I found this link which demonstrates this.

http://profmason.com/?p=703

In another instance, it's sensing the effect of a changing electric field (e.g. Alternating Current), say in monitoring the alternator wire discussed above.

It occurs to me that the rotating element is a consideration, as some rotating elements don't rotate at the same speed as the crankshaft. So that would have to be factored in.

Hi there,

The hall effect sensor may not require a magnet as long as it's placed near a rotating metal piece that is notched in some way (gear teeth, holes in plates, etc.) If it's sensitive enough and mounted close by, it'll note those changes. Another approach with holes is using an LED and LED receiver.

As for the alternator, I'd consider using a simple voltage detector network with an input into the ADC. Another approach could be the use of a Schmitt trigger / comparator approach where a pulse is sent to the Arduino every time the raw (i.e. pre-rectifier) output of the alternator exceeds a certain threshold. Since you're interfacing with potentially huge potentials and currents, going the alternator approach is less desirable in my view. I'd really look into anything that rotates on the engine that is accessible.

Speaking of which, if a belt is accessible, consider an IR/light approach using the belt. Belts are usually black and engines typically don't enjoy sunlight under operation. Paint a small section of the flat (outside) part with a white stripe and then use an IR sensor to capture the changing color as it whizzes by. Should work. Just be sure to use a color/paint that will not damage the belt. Belts are usually accessible and you may even have a convenient mounting point nearby. I have used a Fairchild QRB1134 IR photoreflector for a similar project, capturing the fastest wheel/indicator on a gas meter in my basement. The application is similar, i.e. a white background with a black dial interrupting the 'view' from time to time.

All who have contributed, thank you. What a great forum this is ... I love the hall sensor idea, and I'm thinking now of one sensor to read the engine speed and one for the pump (which can during operation be disconnected from the rotating engine via a Power Take Off and belt-driven pulley). I think I can easily mount a sensor-magnet on exposed parts of the PTO shaft, as well as another on the pump shaft. I'll have to see how to mount the sensor, but I think that's not a huge issue.

In this application, I think I like the Hall better than the IR, though the IR light's reflection off of a brightly painted strip on a black belt is cool. As is the idea of an IR LED 'shining through a hole' to a receiver. These engines are largely open to the elements (skid mounted) including possibly ambient sunlight.

I also like the idea of sensing alternating current via the hall sensor as well (the alternator pre-diode approach as well). I'll have to experiment. This is a new frontier for me, the Hall Sensor that is. I'm planning on trying Sparkfun's http://www.sparkfun.com/products/9312 which at first glance appears to be the same as Hall effect sensor [US5881LUA] : ID 158 : $2.00 : Adafruit Industries, Unique & fun DIY electronics and kits unless someone thinks those are not robust enough. I have access to a machine shop, so I can fabricate a housing for the sensor.

Just to re-iterate: The Hall effect sensor would not be used for current sensing inside the alternator. I apologize for any misunderstanding.

To use the electrical circuits inside the alternator as a frequency source, you'd have to measure the rise and fall of voltages post alternator but pre-diode. Most likely with a voltage divider network. But given the voltages and currents involved, I would use this as a last resort. Even then, I'd go with an opto-isolator and similar equipment to keep the stuff in contact with the on-board power supply well-isolated from the arduino. As I recall, there can be some really nasty transients on the alternator as the regulator switches it on and off and/or the engine stops/starts. Never mind the ampacity issues. Like I said, this approach is a last resort, IMO. Stay away from inside the alternator, if you can.

Your hall-effect approach on a moving/rotating part of the engine sounds much more robust and isolates the Arduino from undue voltage potentials. Find a suitable spot with easy access to a mounting bracket, and you're done. I like this approach much better, just as the optical one I mentioned earlier, which may even work in outdoor conditions.

Thanks, Constantin, for your clarification. I do agree with your assessment and your concerns about the alternator. There are issues (it's very invasive) in addition to those you've cited. Hall effect as described is certainly what I'll move forward with and report back!

Just a thought, how about a airflow sensor that goes on the exhaust of the dieselengine. More exhausts are emitted at higher rpms...

Goos luck with the project!

has anyone any ideas for monitoring fuel flow?

I don't know how to do this even in an invasive way because you have to subtract the return fuel and I think because it's hot you can't just put another sensor in the return line as it gives invalid readings (don't know about that but that's what I've been told)

I did hear of one system that had a holding tank for return fuel and measured how much was pumped out of that to keep it at the same level.


Rob

Depends on the technology also. Many engines now use a single fuel supply line with no return, as I recall with a variable-speed fuel pump. One benefit is that the fuel temperature in the tank doesn't fluctuate up and down as much, causing condensation, promoting algae growth, and then leading to unhappiness. However, you're right, older diesels return the fuel.

I'm still a big fan of using a hall effect sensor to note a sprocket wheel / water pump / etc. rotating nearby. Pretty straightforward to implement, non-contact, and very accurate.