Hairbrained idea for cheap timing meter/onboard display

Sooo... right into it:

1. Why not use a strip style piezo sensor permanently mounted on the #1 injector line? Basically, you could use this (with signal amplification and a touch of other electronics) to trigger an inductive (normal) timing light.

2. Moving forward, you could ALSO use a magnetic pickup mounted on the timing tab to count revolutions. This would imply that you could, via simple, live calculations on an arduino or similar, determine timing. Calculation, in theory: ((time of injection/time per revolution)/360))

3. This RPM and timing could then be output to an LCD or LED screen, along with thermocouple input and a pressure sensor for boost for in-cab monitoring.

4. Pushing the idea even further forward, you could test each injector line by including a degree offset per cylinder, allowing you to locate problematic injectors (with a significantly lower or higher "timing" since they pop at a lower/higher than desired pressure)

Phew...

As proof of concept, I'll be attaching a strip piezo sensor to the #1 line later this week, and monitoring it's output on an oscilloscope. Theoretically, at idle I should see a signal of unknown amplitude at around 4.83 hertz.(580 rpm/60 seconds = 9.66 rpm/second, divided by 2 since it's a four cycle = 4.83 injection events per cylinder, per second) I also have a small logic board that will allow me to dial in a "gain" and format the output to a digital signal above a threshold I set. IF it works as I hope, with a detectable signal at the proper timescale, there may be hope for this goofy idea.

Somethimg tells me yer math is off

4 cycle engine requires 2 revolutions to complet 4 cycles...not 4 revolutions

Maybe my memory of the 4 cycles

Suck
Squeeze
Bang
Blow

Is flawed but I tend to think you may want to divide by 2 not 4...correct me if I am incorrect.

JM7.3CW

riotwarrior said:

Somethimg tells me yer math is off

4 cycle engine requires 2 revolutions to complet 4 cycles...not 4 revolutions

Maybe my memory of the 4 cycles

Suck
Squeeze
Bang
Blow

Is flawed but I tend to think you may want to divide by 2 not 4...correct me if I am incorrect.

JM7.3CW

Click to expand...

NUMBERS FIXED FROM BRAINFART...

I did mess up and divide by two (I think you meant to correct me to four), which gave me TDC or BDC events/second, but only 1 in four gets an injection for the "bang". Numbers fixed above

i dont like popcorn

No I meant divide by 2

100 RPM in a 4 cycle engine means 50 times engine fires for 50 power strokes

Maybe I am looking at this wrong but all you are trying to do is read the timing or firing event correct?

Maybe I should just shut up and watch...

I've thought a lot about this myself. The timing meters are not magic. It's reading 2 signals and doing the math to give you the timing. It's a small micro-processor. An arduino could do it easy. The only thing I'm not sure of is if you would have to amplify the signals off the probes. It's basically measuring the time between the injection event and the timing mark on the balancer and then doing the math to display it in degrees.

You can get amplifier boards for arduino pretty cheap so it's not really an expensive endeavor. Just have to find s piezo that works for the line and a hall sensor for the balancer bracket.

Really an interesting concept. Not sure how to apply it yet but will be watching...

Probe signals shouldn't need any extra work... there are off the shelf arduino boards that'll condition/filter them and pass them back to the main processor.

The big question is "will a strip piezoelectric be sensitive enough." Looking at the actual ferret, it uses a beefy clamp to keep the sensor solidly fixed on the line, reducing harmonics induced by engine vibrations. Getting an actual read off the injector line will be the biggest challenge.

Depending on how well (if) the piezo strip works, we could have very cheap onboard or portable timing meters, and could see the advance throughout any driving condition.

BDCarrillo said:

Sooo... right into it:

1. Why not use a strip style piezo sensor permanently mounted on the #1 injector line? Basically, you could use this (with signal amplification and a touch of other electronics) to trigger an inductive (normal) timing light.

2. Moving forward, you could ALSO use a magnetic pickup mounted on the timing tab to count revolutions. This would imply that you could, via simple, live calculations on an arduino or similar, determine timing. Calculation, in theory: ((time of injection/time per revolution)/360))

3. This RPM and timing could then be output to an LCD or LED screen, along with thermocouple input and a pressure sensor for boost for in-cab monitoring.

4. Pushing the idea even further forward, you could test each injector line by including a degree offset per cylinder, allowing you to locate problematic injectors (with a significantly lower or higher "timing" since they pop at a lower/higher than desired pressure)

Click to expand...

Your idea isn't hair brained at all. What you described above is basically the following tool. Piezo mic clamped to #1 line that triggers an ordinary timing light.

http://www.tooldiscounter.com/ItemDisplay.cfm?lookup=FERV765-01

I had a similar idea(as have others here) of using an arduino and an lcd as a tach/boost/temp display in cab. I talked with one of the hardware guys at work and he said the components to achieve this would be fairly straight forward. It would just take some time with the o-scope to get the circuits and trimming right. I'm not a h/w guy, I'm one of those ****** s/w devs, so I know very little about circuit design.

In fact, it seems like someone here already built some type of temp or boost guage using an arduino. It wasn't too long ago.



http://www.oilburners.net/forums/showthread.php?74423-Arduino&highlight=arduino


http://www.oilburners.net/forums/sh...meter-WVO-Controller-System&highlight=arduino

Be pretty neat to watch the timing while driving. The strip piezo would be pretty cool for price, but Tinytach has clamp on transducer that are not terribly priced.

What about using the one that is already one #1, that's what they time with when the engine is hot tested. IIRC there is two contacts on it, don't know whats inside of it. Just a thought.

Now you've gotten me thinking about this again. I believe this will plug into the arduino. So extend it and use it to read TDC off the balancer.

https://www.parallax.com/product/605-00005

Then you can baseline off of this and calculate the time difference between the injector and TDC on the trailing event using a hysterisis function. This should be, more or less, what the kent-moore timing device is doing. At least in my limited understanding of how it works.

Of course, you'll probably have to beef up the sensor some for external use. Maybe push it down a metal tube and fill with potting. The sensor may require a magnet, the link is broken. If so, it may not work. putting a magnet on one side of the balancer would probable not be a good idea.

So after much thought...

I see we are zpeeking green and red apples

All you want is the firing pusle...that pulse occurs every 2 revolutions of engine. So the timing tab pickup has to read every 2 revolutions...correct?

You speak of the 4 cycles but not the 2 revolutions per 4 cycles.

Those cycles take two full ups n downs of the piston which equates to 2 revolutions as the cycles are 1/2 of the equation per full revolution.

Again...as per usual I am likely talking out me **** but ....

Suck....intake...piston goes down
Squeeze..compression...piston now goes up
= 1 revolution
Bang...ignition....piston travels down
Blow...exhaust...piston going back up for 2nd time

= 2nd revolution....

Now maybe me **** spews out diatribe but something tels me thats a complete procedure in a four stroke engine.

Somebody help me see where my math is incorrect cause the timing meters read 2 revs to 1 Pulse on the injector line unless my mind is NFG....its why cam turns half speed of crank as does IP...

2 stroke is suckbangblow every turn of crank...

JM7.3CW out to lunch

Riot warrior: We're on the same page... I'm dividing rpm by 2 to guesstimate how many injector pulses per second to look for on the O-scope. I started at dividing by 2, brainfarted to 4, and hopped back to 2.

OLDBULL8 said:

What about using the one that is already one #1, that's what they time with when the engine is hot tested. IIRC there is two contacts on it, don't know whats inside of it. Just a thought.

Click to expand...

Hmm... since the service manual says to hook up the Ford timing adaptor (don't remember part #) to that, I assumed it wasn't capable of output on its own. I can probe it and see...


Needle nose: I'll have to fiddle with a spare balancer to see what's in there that triggers the kent-moore style pickup. The hole for that sensor is several degrees away from the 0 mark for a timing light... There are also inductive coils (similar to our tach sensor) that ought be able to pick up the milled "zero line" built into the balancer.

BDCarrillo said:

.. I'll have to fiddle with a spare balancer to see what's in there that triggers the kent-moore style pickup. The hole for that sensor is several degrees away from the 0 mark for a timing light... There are also inductive coils (similar to our tach sensor) that ought be able to pick up the milled "zero line" built into the balancer.

Click to expand...

The timing tab has A ZERO mark...short hole...aligned with slot...longer hole for timing probe is 20 deg offset...hence need to adjust J33300 to a 20 deg offset.

The gap from the TDC slot is what triggers the mag pick up.

BTW glad yer math is back on track...

JM7.3CW