Engine RPM meter / General Science and Electronics / Forums

Forums

4hv.org :: Forums :: General Science and Electronics

« Previous topic | Next topic »

Engine RPM meter

1 2 next

Move Thread

LAN_403

Blizzara^

Thu Dec 17 2009, 04:36PM

Registered Member #2550 Joined: Thu Dec 17 2009, 04:25PM
Location:
Posts: 4

Hello all, if you just have time, could you help me with this problem?

So I am doing two-stroke engine rpm meter, tachometer with Oopic microcontroller. The thing is, I cannot get the sensor working. I have optocoupler (4n25) beetween the sensor and oopic. There is pull-up resistor in the oopic's side.

The design: 9v battery - 820ohm resistor - 4n25 - NPN transistor - ground.
And from the engine: 5 turns over the ignition wire - 10k resistor - NPN transistor base.

So it is quite simple. And it kind of works, it shows rpm on the lcd (which is connected to oopic), but that is not stable. Sometimes it shows rpm just right, but usually it hovers from 500 to 3500 when the real rpm is 2000.

If you have any clue how to improve things, please tell. Sadly I do not have oscilloscope :(

Proud Mary

Thu Dec 17 2009, 07:02PM

Registered Member #543 Joined: Tue Feb 20 2007, 04:26PM
Location: UK
Posts: 4992

Perhaps you should sharpen the leading edge of the pulses, or even reduce the gain of the first stage.

radiotech

Thu Dec 17 2009, 09:38PM

Registered Member #2463 Joined: Wed Nov 11 2009, 03:49AM
Location:
Posts: 1546

Could the issue be jitter in the signals the sensor is picking up?
I found that retroreflecive tape strips greatly improved motor shaft
readings with an optical pickup tach. When using magnetic pickups connected to a frequency/period counter coil positioning could prevent a reading of 2f. what it should have been.

klugesmith

Fri Dec 18 2009, 12:01AM

Registered Member #2099 Joined: Wed Apr 29 2009, 12:22AM
Location: Los Altos, California
Posts: 1716

Blizzara^ wrote ...
And from the engine: 5 turns over the ignition wire - 10k resistor - NPN transistor base.

If you mean the pickup wire is wrapped 5 turns around the HT wire between magneto and sparkplug,
that is the wrong configuration an inductive pickup. To first order it does not sense i or di/dt in
spark plug circuit, but it will get a common-mode signal from capactive coupling.
(please clarify the connections at both ends of your pickup coil).

I think you would get a stronger/cleaner signal with either
a) even one turn that threads the magneto core between center and one outside leg
(polarity matters; the voltage spike when current is interrupted should turn -on- the transistor
which drives your optocoupler primary).
b) thread HT ignition wire through a ferrite toroid, with your pickup coil also wrapped on that.
(for comparison, look at the clip-on pickup of an inductive timing light).

Renesis

Fri Dec 18 2009, 12:11AM

Registered Member #2028 Joined: Mon Mar 16 2009, 08:13PM
Location: Norway
Posts: 319

The easiest methode of measuring an engines RPM is to simply measure the ac frequency in the alternator (assuming your engine has an alternator). This is a popular method of measurement with older diesel engines, witch have no ignition or other engine control circuitry.

Electroholic

Fri Dec 18 2009, 02:54AM

Registered Member #191 Joined: Fri Feb 17 2006, 02:01AM
Location: Esbjerg Denmark
Posts: 720

are you polling the pin or is your program interrupt based?

Blizzara^

Fri Dec 18 2009, 05:03PM

Registered Member #2550 Joined: Thu Dec 17 2009, 04:25PM
Location:
Posts: 4

It is event based program.

I do not have alternator as I am using racing ignition.
I would really like to use inductive sensing, because this rpm meter goes to my dynamometer. It would be kind of lame to cut everyone's wires coming from magneto just to meter rpm :D If you have suggestions how to accomplish this better way, I want to hear! :)

(please clarify the connections at both ends of your pickup coil): If I understand right, my own circuit's coil is open at one end and goes to the resistor&transistor on the other end.

klugesmith

Fri Dec 18 2009, 05:58PM

Registered Member #2099 Joined: Wed Apr 29 2009, 12:22AM
Location: Los Altos, California
Posts: 1716

Are you into Kart racing? You must show us some pictures.

And tell more about your dynamometer. I know some kids who towed a car along a level street, as a load for kart dyno measurements.

Back to the ignition time sensor (if that's where your trouble lies):
If you have a single wire from your electronics, with the last few inches wound tightly around an ignition wire, that's a capacitive pickup. Electrically, the "coil" would behave just like a similar area of aluminum or copper foil tape; the number of turns is irrelevant. More significantly, the return path is poorly defined -- how and where is the primary side of your optoisolator circuit connected to the engine metal? (OK, we could call it "ground").

I would be glad to make and post oscilloscope pictures of a similar pickup on a lawnmower engine.
First you would need to describe the relevant details a little better, perhaps aided by a picture.

-Rich
[edit] If this is for a stationary dynamometer, on which you will be testing more than one engine, why not just take the clip-on pickup from an inductive timing light? Or, uh, an inductive tachometer? With knowledge of its output voltage range, it could safely be interfaced to your OOPIC without optical isolation & primary side battery.

Blizzara^

Sat Dec 19 2009, 03:51PM

Registered Member #2550 Joined: Thu Dec 17 2009, 04:25PM
Location:
Posts: 4

Actually I tune scooters, there are three scooters in my garage right now, of which two are waiting to be tuned and one in parts :D

The ground is from transistor pin to the motor, last time it was connected with alligator clip to my cylinder cooling fins.
I attached schematic, hope it will clear things a bit. Also some pics of the setup.

The dynamometer, I have pics of it here:

under "Dynamometri",
and a blog about it here:

it is in Finnish, but you can use Google Translate. If you look at the first link, there are also pictures of my scooters and engines.

I do not know about inductive timing light, how it works.. About the ready tach, they are quite expensive. Of course if you find a cheap one, I am interested : )

klugesmith

Mon Dec 21 2009, 03:51AM

Registered Member #2099 Joined: Wed Apr 29 2009, 12:22AM
Location: Los Altos, California
Posts: 1716

Thank you for showing us your web site and blog. Nice work! I admire the dynamometer.

Today I learned some ignition details about two single-cylinder, four-stroke, gasoline engines with magnetos. (On a lawnmower and a chipper-shredder).

1. The main HV pulse to spark plug is Negative polarity on both engines. Detected that by holding a plain neon glow lamp (NE-2 style) in parallel with the spark plug. (Would probably work in series, and that would not kill the engine). Other engines may have positive pulses; in fact wasted-spark automobile ignitions use both polarities.

2. I made a pickup like you described, and looked at its output with an oscilloscope (1 megohm input impedance). The pulses were Negative, as expected, with very little ringing. Peak amplitude was more than 50 volts, and the pulse width was a few tens of microseconds. This would not turn on the NPN transistor in your circuit, in fact it would probably exceed the reverse breakdown voltage of the base-emitter junction.

3. Conclusion: for engines like mine, your optoisolator primary circuit would want a PNP transistor. Could be configured with both kinds of transistor, so pulses of either polarity would energize the optoisolator.

As for inductive timing lights, used ones are dirt-cheap since distributor adjustment on cars is becoming a forgotten art. See ebay 290383606664. The clip-on detector is a ferrite-cored current transformer. If you're still interested, I can try to 'scope the pickup waveform from mine. With an intact timing light, your OOPIC input could be just a photodiode.

1 2 next

Moderator(s): Chris Russell, Noelle, Alex, Tesladownunder, Dave Marshall, Dave Billington, Bjørn, Steve Conner, Wolfram, Kizmo, Mads Barnkob