When you click on links to various merchants on this site and make a purchase, this can result in this site earning a commission. Affiliate programs and affiliations include, but are not limited to, the eBay Partner Network.
Ok, I give up. Time to buy a good tach filter. Anyone know of a good one?
I did some searching and found this filter on the Arduino site and built it:
And here's the results I got on the AEM:
Note that the RPM is not only twice what it should be, which I could change in the pulse/rev setup, but it spikes badly. So, as said, I'm ready to buy something that is said to work. Will call AEM on Monday, but in the interim I'm asking for suggestions. In the interim I think I'll do something else.
How about an inductive pickup for one of the plug wires?
Looks like a time for the Schmitt trigger circuit to be made. Not as simple as the zener clamp but will give a good clean pulse out from a noisy input.
Just for giggles, put in a damper diode across the coil. may be that will stub out the ringing. This is one picky data logger you have!
I keep thinking I'm SOOOOOOO close on the filter off the tach circuit, and one part of me says to pursue it and the other part says forget it. But, for a few minutes let's explore the "pursue it" path.
Yes, it is a picky datalogger. I'm thinking about using a 555 timer. If you remember, I got it to what's shown below with a very sharp rise, a slight glitch, and then a long exponential decay. Unfortunately, the capacitance required to do that means the decay takes about 15 ms. However, at 6000 RPM the time between cylinders firing is only 2.5 ms. (6000/60 = 100 revs/second, or 10 ms/revolution. With 4 firings/revolution that's 2.5 ms between cylinders firing.) Obviously then there's not enough time between firings to allow that much decay. So I'm thinking I could use a 555 to give a digital pulse with no "junk" on the waveform. But, typically a 555 fires off the negative edge of a signal, and I really need it to trigger off the positive edge . So I'm researching that.
I used to be a 555 guru, and have used them in many, many things including fancy battery chargers that run the battery up to a pre-set value like 14.1 volts and then letting it "coast" down to 13.0 and then going again, flashing the headlight on a motorcyle in such that it appears "on" at all times, and plenty of others. But, getting one to trigger off a positive-going input is new to me. This site says this is supposed to be the way to do it:
POSITIVE TRIGGERING
The only way to trigger a 555 with a positive pulse is to have the output sitting HIGH. A positive pulse on pin 6 will change the output to LOW. The voltage on pin 6 must be greater than 2/3 supply voltage. When the 555 detects the trigger pulse, the output goes low and a transistor inside the 555 takes pin 7 to the 0v rail. This starts to discharge the 10u via the 1M resistor and when pin2 detects a LOW, the output goes HIGH again. See Positive and negative triggering on P1 of this discussion
Gary, how about using a optocoupler? That will rid the ringing as the LED will inside the optocoupler during the dips in the pulse.
Eddie - Please say that again as there appears to be a word or phrase missing. But, I suspect you were saying something like "won't light". Might have promise.
Eddie - Thanks for that link. Please don't take me as being ungrateful, but that may be the hardest schematic to get my head around I've ever seen. So, I re-drew it, but I need y'all to look it over and see if I did it correctly. PLEASE!
Here's their schematic:
And here's my schematic. It is supposed to be the same electronically, but laid out so's up is power and down is ground.
So, what's good about this circuit? I assume that one of the advantages is that it takes some power to turn the LED on, so that ringing isn't likely to do it. Right?
Yes that circuit will use some power from the tach pulse to light up the optocoupler's LED As long as the voltages over the 5.1V zener diode will cause the LED inside the optocoupler to light up and then makes the photo transistor ground out the output to the tach.
From the scope waveforms, looks like the ringing voltage level doesn't drop below the 5.1V threshold, so it should give a nice square wave pulse as a output.
Thanks. I've been around a schematic or two in my time. Was a ham in high school in the early 60's and tutored some friends in college. I'll never forget walking up to one of them who was cramming for a test outside the classroom. Looked over his shoulder and said "That's a 5-tube super het". His jaw dropped and he asked how I could tell. Duh! I wanted to tell him he was in the wrong field, but didn't.
Ok, I'll see about ordering a bunch of things to let me build that circuit as well as the one using the 555. Something will work. However, it doesn't have to be perfect if I can tell what is happening with the R's as it's the AFR that's the real question.
Dylan - Make sure you read the part in the other thread about the distance from the ports the O2 sensor should be. Basically that is 18" - 36", but on a six that is going to be difficult to do given the length of the engine.
However, David/1986F150six will have his truck here in a few weeks and we'll be using this one on his so will have experience.
Gary, luck at the top of the linked thread and you will see what the exhaust manifold looks like on my truck and the placement of the factory O2 sensor. Close, huh?
Yes, that would be the one. Now you know why it is referred to as a "log" manifold.
The bottom one is from the 1987 and up engines with fuel injection. They are an upgrade since the flow is much better. But, provision has to be made to heat the underside of the intake manifold since they are not joined to the intake.
