Tachometer black box for magnetos, Attn byacey, lark2004 (DONE?!?)

[dafox99]

Re: Tachometer black box for magnetos, Attn byacey, lark2004 (DONE?!?)

No rush. I just downloaded tina-ti, which is supposed to be a pretty cool freeware spice based sim from TI. I hit the setup.exe, it seemed to unpack something and then ... nada. Doesn't seem to be anywhere. Dang. <br /><br />Would love to get a copy of CM when you have time. Again, no rush. I've gone about 10 years without one, I can certainly wait awhile.<br /> <br />From '86 - '95 my whole job was doing circuit simulations and managing a group of guys who did the same with the Cadence Design Systems tools. Very very expensive stuff used by chip makers and system designers. I had been certified as a SPICE expert, but that was sooo loonngg ago .. just remember twiddling tnum, ITL4, LVLTIM, VNTOL and different integration methods on the super-tough circuits. It was pretty fun for awhile, but got tedious after a few years. I doubt I remember much, but I'm betting it's a lot more fun as a hobby than a profession. Thanks, Paul.

 

[dafox99]

Re: Tachometer black box for magnetos, Attn byacey, lark2004 (DONE?!?)

Hey again, Paul. Well, the tina-ti didn't load on this computer as it is from my company and just doesn't allow foreign programs. I loaded it on my home computer, and it's pretty cool! VERY intuitive .. you can throw a circuit up in no time, and do transient analysis immediately! Works great. I input the active feedback circuit I proposed, and it works in theory. Then put a resistor in series with the zener to see how you could make your reference float up with Vcc. That worked as well .. I used a 220 ohm. Of course, these free simulators don't let you introduce noise, so it's all math and theory at this point .. but at least that part works!

 

[Paul Moir]

Re: Tachometer black box for magnetos, Attn byacey, lark2004 (DONE?!?)

I only got it half-assembled here. But it's 3:00 so I'll have to give it a shot tomorrow!

 

[Paul Moir]

Re: Tachometer black box for magnetos, Attn byacey, lark2004 (DONE?!?)

Tina-Ti, I'll have to give that one a try. There's some things in Circuit Maker that get on my nerves. But then again, I'm used to the esoteric CirCAD interface...<br />Ok, got the circuit working properly with 4.7k subsituting the 5k. Turns out all my ~3v zeners were burnt out - I must have gone hunting identifying them with two little a resistance (or too high a voltage!). Once I got a working one it immediately came to life with nice, fat, square signals going to the tach.<br />It works no problem down to 10v (where the tach shuts down) and up to about 15 or so once the 220 ohm resistor got dropped in. It gets a little funky around there (the pulses shorten to the point where the tach is occasionally missing them) so I think I'll experiment with a slightly higher resistance. No suprise - I am using a 3v zener rather than a 3.3v.

 

[dafox99]

Re: Tachometer black box for magnetos, Attn byacey, lark2004 (DONE?!?)

Very cool! I think you are on target diddling with the 220 .. it was a swag. I'm thinking a little larger sounds right.<br /><br />Good luck

 

[dafox99]

Re: Tachometer black box for magnetos, Attn byacey, lark2004 (DONE?!?)

One other thought as you diddle. You may want to just look at the inputs as you bring the voltage up to 16.5 ... could be the 4.7k/1k ratio may also be the culprit on the low end. It may be moving to high and you may need to decrease the 1K or increase the 4.7K. If you can hook the scope up differentially, or look at the inputs with two channels, you should be able to tell if it is the high end (zener/220) or low end (4.7/1) that is giving you fits (or both).<br /><br />Best,

 

[Paul Moir]

Re: Tachometer black box for magnetos, Attn byacey, lark2004 (DONE?!?)

Ok, stick a fork in this one 'cause I think we're done:<br /> http://tinypic.com/fnrzpv.gif <br /><br />The ratio was indeed the problem (which made more sense once I thought about it) and so that had to be bumped up a bit. Also, I was running into some odd signals occasionally so I had to narrow the hysteresis quite a bit. But that's no problem. Now the circuit seems quite bulletproof; it works perfectly 10-16v. It doesn't care if the spark gap is wide, short, or shorted out. It's nearly impervious to noise. It produces 4ms pulses at ~1000 rpm. They shorten to about 2ms at 2000 RPM but don't get any shorter after that which works fine with my Faria I/O tach.<br />I've only tested up to 3100 rpms because that's about the fastest I can run my 4hp in a barrel of water. I'll try to get an on the water test in this weekend but I'm not sure it'll be possible. Anyway, I don't see any reason why it won't work to 5000 RPM. If there's any problems I'll report them.<br /><br />Any final changes?

 

[dafox99]

Re: Tachometer black box for magnetos, Attn byacey, lark2004 (DONE?!?)

Congrats Paul!<br /><br />Nothing comes to mind other than where you place your device might matter in an electrically noisy environment .. but, I'm betting you are safe and sound. I did a quick spice run on the inverting input voltages with your current values as Vcc varies. It shows the wave will run from 3.3v - 3.6 with Vcc=10.5v and from 4v - 5.5v with Vcc = 16.5v. I don't know what your mag input is looking like, but, frankly, all that doesn't matter if the darn thing works!

 

[dafox99]

Re: Tachometer black box for magnetos, Attn byacey, lark2004 (DONE?!?)

Hey Paul .. if you have a moment, you might look a couple down at "need help ID'ing an engine". Did someone replace an I/O motor with an outboard powerhead?

 

[Paul Moir]

Re: Tachometer black box for magnetos, Attn byacey, lark2004 (DONE?!?)

Sorry, I keep meaning to post my 'scope output but then I keep deleting it. I have the trigger points nicely centered on one of those unusual ignition events, which are "smaller" then regular igntion events. <br />I've been testing in the noisiest environment I can, with the sparks just inches away from the circuit and occasionally prodding it with fingers, etc. I hope this might be an adequate comparison to keying the mic on the VHF. <br /><br />What I like about this circuit is it's simple, doesn't have many parts, and is really resistant to noise. And I think it would work for anyone who whips one up - it isn't very particular on values and didn't need much fiddly work to make it happy.<br /><br />That simulation seems pretty close. Here's my measured values (Bias is the DC voltage at C1):<br />10.5v:<br />Bias - 4.3v<br />Snap in - 3.4v<br />Snap out - 4.1v<br /><br />16v:<br />Bias - 7.0v<br />Snap in - 4.3v<br />Snap out - 5.6v<br /><br /><br />The first 'scope picture above gives you an idea of what is at C1. But the odd ignition event pulses look just like that, but don't go as far down. Perhaps 2v up above the regular pulses, which are a good volt above ground with Vcc=16v. Now that's only really visible when the voltage gets cranked up, and so it's no problem at lower Vcc. <br /><br />I'll have a look at that post. I wonder if he has a Sea Drive or something.

 

[dafox99]

Re: Tachometer black box for magnetos, Attn byacey, lark2004 (DONE?!?)

Sounds good, Paul. Noise rejection and tolerance to, uh, tolerances is key. <br /><br />If there were supposed to be scope pics, though, I didn't see them in the post.<br /><br />I saw the nutty sea drive monster you posted .. never seen anything like it.

 

[Paul Moir]

Re: Tachometer black box for magnetos, Attn byacey, lark2004 (DONE?!?)

I ment the traces on the Nov 3 post. But never mind - it only took a few minutes to capture what we're talking about.<br />

<br />The top trace is at 16v and the bottom one is taken at 10.5v. The red lines are 0v (top one got cut short). The trace on the left are two "ignition" pulses from C1 or U3 pin 6. The later one is weaker for some reason. The following trace is from U3 pin 5. My capturing equpitment & software is cobbled together so I can't display them like a dual-trace. But you get the idea.

 

[dafox99]

Re: Tachometer black box for magnetos, Attn byacey, lark2004 (DONE?!?)

Thanks, Paul. Not sure about the weaker second wave. I guess the mag has two coils, and one might be weaker? Looks like the thresholds are on the money. Nice work. I'd keep the circuit as-is.

 

[dafox99]

Re: Tachometer black box for magnetos, Attn byacey, lark2004 (DONE?!?)

Paul, now that the product is complete, you think this is something I should put together and drive a new tach, or go with an old airguide from ebay? Today, I have nothing.

 

[Paul Moir]

Re: Tachometer black box for magnetos, Attn byacey, lark2004 (DONE?!?)

I was in the same boat, so to speak. There was a couple reasons I went with the new tach. First, if you look at the airguide tachs on ebay, almost all of them don't come with a sending unit. To my understanding, this is a necessary part of the Airguide tachometers which would interface the gauge with a particulair ignition system: battery, magneto (various brands here), and CDI. Right now the only tach I see up there with one is for battery ignition J/E V-4s. <br />The second reason was that I wanted to pick out a particular style of tach. I wasn't after the vintage look and my personal preferences are for one that isn't much like the Airguides. The Faria is the "Euro" style one with black lettering on a white background.<br />Finally, there seems to be more outboards than tachs out there by a long shot. So someone was going to have to do something sometime. Might as well be us, here. <br /><br />I think if you were patient and did find one on ebay, it would be less expensive than going with the all-new route. Certainly less trouble once you had it anyway. But that circuit is pretty simple and I'm sure it'll work for a variety of tachs & ignitions, though I haven't done much testing. I was too busy to do an on-the-water test last weekend, and that may have been the last chance for the season. And I would also like to test it with some Sierra brand ignition coils, though I doubt that will be much different.<br /><br />But I'm confident it's something anyone could assemble & it would work just fine.<br /><br />EDIT: Looks like some Airguides don't need the "Transmitter". But which models I'm not sure.

 

[dafox99]

Re: Tachometer black box for magnetos, Attn byacey, lark2004 (DONE?!?)

Understood. I think the airguide route is clearly a crapshoot, and by the time I purchase and receive one, I could well find it doesn't work, and may never with my motor. May have to visit RadioShack and pick out a new one. I probably have some time. It was 80 in Austin today! whoah.<br /><br />Thanks!

 

[Paul Moir]

Re: Tachometer black box for magnetos, Attn byacey, lark2004 (DONE?!?)

Aw man, the reason I thought it was the end of the season was because I was on icebreaking duty last weekend in the salt inlet. <br /><br />You would be able to find all the parts you need out of a junked monitor or toploader VCR. LM324 is a subsitute if that helps.

 

[byacey]

Re: Tachometer black box for magnetos, Attn byacey, lark2004 (DONE?!?)

Hi Paul, regarding the amplitude difference in your last scope shots - the difference is probably the spark plug gap and / or leakage paths on the electrode insulator. Wider gap means higher voltage to create the arc. This would account for the difference.

 

[Paul Moir]

Re: Tachometer black box for magnetos, Attn byacey, lark2004 (DONE?!?)

I think you're 100% right there byacey. In fact I know it because I was testing with various gaps (including open & short) and the wide gaps had longer & more intense primary pulses.<br /><br />But those are actually a strange "oddity" signal I was getting perhaps 1/200th of the time and are far worse than the wide gap pulses. I never figured out what caused it - maybe just some crud on one set of points or something like that.<br /><br />Anyway, once I've got this thing properly tested out, I'll give you a shout if you don't mind. I just don't want to waste your time until I'm sure. I want to do up a shareaproject.com article too rather than relying on tinypic.com.<br />Thanks again!

 

[bb_ed]

Re: Tachometer black box for magnetos, Attn byacey, lark2004 (DONE?!?)

Dang, maybe after a couple cervesas you guys could work on solving that cold fusion problem! Any of you fellas know a guy named Nemo?