Is there a reliable, cost effective way to get a parade pattern onto a home pc?

 

I've seen some 'oscilloscopes on a card' that plug into an expansion slot on a PC. The ones I saw were much cheaper than even a cheap complete oscilloscope but they were still a couple hundred bucks a few years back.

You'd have to build a voltage divider and protection network for the front end of the 'scope, and maybe trigger the sweep off an inductive pickup on the #1 wire, timing-light style. That's about as cheap as I can see, but still give pretty accurate & repeatable results.

 

thanks rustydawg I not experienced enough to acomplish that yet, but sounds like I will be when I'm done.

My next question would be how to communicate the to the pc?
Probably have to get it into serial data and right a small program to manipulate the data(also not experienced there)


Got any breif sketches of your idea?

 

After a quick Google search I came across a site where they're giving away the software that allows you to use your PC's sound card as the A/D converter to make an analog voltage signal computer-compatible. It turns your PC into an oscilloscope.

I found the software here: http://polly.phys.msu.su/~zeld/oscill.html

The nasty-looking schematic here is a starting point for an interface to connect your sound card to the vehicle's tachometer lead. The circuit provides for scaling down the large voltage spikes present at the tach lead by about 1000:1 - this is just a starting point and protects the input of the sound card (which is looking for an input voltage of about 2 volts max, I believe.)

The diodes across the output of the voltage divider (and input to the scope) are present to clamp any voltage spikes at just less than +/- 1 Volt to be safe. These should really be Schottky diodes because they're faster acting.

The timing-light style inductive pickup serves as a trigger for the other channel of the scope, allowing you to start your sweep when the #1 cylinder fires. The software allows for two channels (the left & right of the sound card's inputs) so the inductive pickup could be this second channel. An old timing light pickup would be great, but you could likely fabricate some kind of pickup coil that slips over the #1 spark plug wire before you fire up the engine.

Like I say, the resistor values of 2.2 Meg Ohm and 2200 Ohm are just starting points to ensure one doesn't drive huge voltage spikes into the PC's sound card (or motherboard in many cases where one is using on-board sound) so these values may need tweaking. This is all theoretical of course - I haven't tested any of this.

I have, however, downloaded the 16-bit Windows version of the software and will experiment with it on my old PC. I can keep you posted on my luck with it.

 

I've been experimenting with the freebie software on my old P166 and soundcard. It's okay, but the software only works with the card in 8-bit mode, so the vertical resolution could be better. It allows sampling at 11025, 22050, and 44100 samples/sec - so that's not bad.

My Dirty Bird is stored for the winter, both the other vehicles have distributorless ignition so I can't very well try interfacing the sound card inputs to a tach lead.

Nevertheless, this is a great tool for DIY EFI builders and guys tuning their EFI - this 'scope could watch injector pulse width, MAF pulses - all kinds of stuff.

Below is a screenshot of what I got, feeding a sinewave into the sound card.

 

Sounds to easy to be true!!! Thanks rustydawg, you're frigging awesome!!

However doesnt look like that program will quite do a raster or parade pattern.

 

See if you can find a copy of Jari Porhio's "revkit" for Linux.

 

unfortunately Apeiron I dont use linux.

also rustydawg, the lead you designed would be for measuring directly, I would probably use two inductive pickups measuring off the coil wire and using #1 as the trigger, then it would just be up to the program to line them up and label them.

 

I think the inductive pickup on the #1 wire would be effective as a trigger... But a voltage divider measuring voltage directly off the primary side of the coil would indicate relative secondary voltage on the secondary side of the coil where the plugs are.

This way a high secondary voltage (high impedance plug wire, missing ground electrode etc) would show as a higher amplitude spike on the scope; a shorted plug would show a lower voltage.

Using an inductive pickup on the coil wire for the display would show relative current on the secondary side of the coil instead of voltage.

So long as you know what you're looking at on the display, you're alright - but I am under the impression that shop type scopes display secondary voltage on the raster, so that's what I was trying to copy.

I think you are wise to start with inductive pickups though - they're probably much safer and friendlier to the sound card than driving coils impulses into the poor sound card.

 

Shop type ingnition scopes are expensive, like couple thousand. I would love to do this, I have funny ignition problems all the time.
From the sound of it, you could do this for cheap. What wattage are those resistors and diodes for.
Could you illustrate the secondary wiring like you did above?

 

Bens3rdGen, yes this scope should work in theory but it's untested so far. The resistor values were a safe starting point that should give some kind of trace indicating each plug firing, just how much detail will actually show up using the freebie oscilloscope software is tough to predict.

There is no further secondary wiring - the tach lead is connected to the primary side of the ignition coil, so whatever voltages occur at the plugs will show in relative scale at the tach lead where the oscilloscope is measuring / displaying.

As far as wattage goes, I would say 1/4 or 1/2 Watt resistors would be fine - and any rectifier type diodes (1N40004 or so) to limit the input voltage going in to the sound card. You may even want to double up on the diodes (put two in series, both polarities).

 

Putting them in series will increase the voltage that could arrive at the sound card.

 

Yes, adding the second diode in series increases the maximum voltage out of the divider and into the sound card.

Using the freebie 'scope software with my soundcard (Soundblaster AWE64) I see that the the oscilliscope starts to clip the waveform peaks at about +/- 2 Volts (using unity gain settings on the 'scope software).

The diodes serve to protect the input of the soundcard from large over-voltage spikes, using only one diode clamps the input peaks at less than +/- 1 volt, well before the soundcard reaches its maximum voltage limits. So you're giving away resolution by prematurely clamping the input voltage swings before the A/D converter reaches its limits.

I'd suggest two or even three diodes in series, both polarities. I'm too lazy to redo the shady schematic...

Another guy's sound card might act differently, but should be safe at two or thee diodes. Just don't run the circuit without any!

 

Pretty much, I just want to ignore the knock sensor and decide for myself if the engine is detonating, and not have to figure out if it is false knock from a huge cam or something. I can think of a thousand other things, mainly to ensure the engine is tip top before i start tuning

 

If you're playing with a sound card, you can record the knock sensor signal.