So I got a weird hobby in building old mopeds. Here in the states it's kind of un-common but go fast parts for peugeot 103's, puch's, vespa and a lot of other mopeds are all over Europe and with the internet and a few state side importers like 1977 mopeds and treats it's getting more common over here. I frequent a site called mopedarmy.com(you can see all the crazy fast builds there) and there is a common problem, besides the over use of the word rad or blaster. We are basically stuck with either a trailtech vapor/vector for an arm and a leg or the tiny tach which is a little digital display pos. Both of which induce a tach signal from a little wire wrapped around the sparkplug wire. The thing is everyone has generic 3 wire tachs for sale or already sitting in their garage. The problem is no one has figured out a way to make these very common tachs work on a simple two stroke. All moped's ignitions are very similar and are ignition timed by either magneto, ac generator(hope I'm getting this term right, permanent magnet mounted to flywheel with a coil on the stator producing ac to charge external coil, breaks primary + with points), or a simple ac cdi. Most mopeds are 6v and most don't have regulators or rectifiers. I'm an ase certified auto mechanic and even have my ase in auto electrical but can't do much aside from finding shorts and open circuits on cars. I've been searching this site for similar posts but have found most to be way over my head. I'm so lost as to how to make a square wave out of ac wave that it's pathetic. Hell, I'm not even sure that a tach really requires a perfect square wave to function or if a switching 12v or switching ground is what triggers a tach. Every idea I've had seems to fail the deeper I get into this site so I'm hoping to get some help from the people who know their stuff. Where to start? Any help would be greatly appreciated. Ryan.

 

I would start with a LM2917 tacho ic & use it to drive the meter part of the old tachos you have.

 

Well I'm looking at building something that could be used universally with any three wire tach that way I can post a schematic and a parts list for everyone. I've found that there are tach adapters for magnetos and a few other similar devices that might be what I need but they cost $100 and up which isn't something I'd like to spend. I did some more reading trying to figure out what exactly a tach needs as a input and it seems that a tach needs a wave that is above a specific voltage(what I don't know) and the bottom of the wave grounded. Anyways what I'm trying to say is that the tachs I have do not work with either a 12v open 12v open pulsing 12v or a open ground open ground pulsing ground. It needs to be a wave. Now that being said I don't think that three wire tachs require a specific dwell period as dwell isn't the same on two cars at the same rpm. So I think it just counts the how many times the voltage goes above a certain voltage but also needs a ground between them, but I don't know 100% about a tach really needing the ground between. I'm rambling. I need something that will take a ac signal(like the lm2917 does) and puts out a 12v~/ground switching signal. Does the lm2917 do that? Sorry if I'm being quite the dullard.

We thank you guys.

 

This is an old 3 wire tacho which I removed the internal circuitry & re built it around the LM2917 Ic making it mutch more versatile. It can be used on points ignition as per the diag, to run of a few turns of insulated wire around the plug lead remove the 10K resistor. It can also have a hall device to drive it. Mutch easier to do it this way. I use this tacho on all sorts of engines for checking RPM.

 

Thank you thank you thank you. But..... I'm not trying to build tachs, just some sort of adapter that everyone can use. If it helps I relegated myself to not using the spark plug wire for the signal and instead I am now looking for a way to use the 6v ac output that most bikes have(others have 12vac). Most people object to running a battery since there is no need to and most didn't come with batteries. So my problem is two fold, how to get a stable 12vdc output whether at idle or high rpm, for tach main power, out of a 6vac system and how to give the tach it's 12v-ground signal it needs. I don't think anyone is going to try to run anything but a tach from the 12vdc power created so the wattage required is only as much as the tach needs. Can I use a voltage doubler(or triple-er, etc.) to boost the voltage at idle to 12vdc and use a lm317t adjustable voltage regulator from radio shack to maintain 12vdc even as the rpm's climb and the input ac voltage(6v~) climbs?

 

I'd choose a 7812 (fixed 12v regulator) over a LM317, just for
simplicity.

 

I have a variable voltage dc power supply and I noticed that voltage slightly affects the tach and how much it reads. A full car battery should be 12.6v but most cars have a running voltage of around 14.5v. I don't know what they calibrated this tach for so having a variable voltage regulator is going to be nice.

 

Oh here is my voltage multiplier on circuitlogix.

I had to edit it because the graph window was covering the load. Any ways this is giving me right around 12v even at idle and 25v at 7.3vac, high rpm. That voltage regulator has a max input of 36vdc so I think I'm done with solving for constant 12vdc to the tach. Now just to figure out the trigger...

 

I admire your can do attitude but there are a few things to consider. For one; Diode/Capacitor voltage multipliers are not noted for their ability to deliver power. What is the current requirement of the Tach? Secondly; Voltage regulators require a minimum input voltage that you can find in their data sheets. It's doubtful that, at idle, you're going to produce sufficient voltage. You can't feed a 12V regulator with 12V.

 

You could try using the 6V ac windings to triger the tacho as the frequency changes with speed. Some outboard motors use this & have seen tachos run of one AC pase of a 12V altenator on diesel engines.

 

Ok so I just tested my tach and it draws .098 amps at 12.1vdc. So about 1.2 watts. That sounds low but I double checked and even though I couldn't check it while the tach was hooked to a vehicle it does have a shift light check button that when held down the needle goes up to where it coincides to where the shift light dial is set at. Even with the needle forcing it self up the amp draw didn't change. Is this too many watts to run off a voltage multiplier? Thanks. Ryan.

 

Would something like this work or am I just being stupid.

 

Ok so I just tested my tach and it draws .098 amps at 12.1vdc. So about 1.2 watts. Is this too many watts to run off a voltage multiplier? Thanks. Ryan.

Calculate for R: R= E/I = R= 12/.098, then make the load in your spice this value. Run the simulation and see what it looks like. 98mA is a heavy load for a voltage multiplier but lets see what the simulation returns.

One note here; When you set up a sine wave signal in Tina keep in mind that if you set the generator 'Amplitude (V)' to 6V, this is not 6VRMS. It means that the gen will swing to a peak level of +6V and -6V. If the alternator on the moped measures 6V on a VOM (RMS) then its peak level is about 8.5V.

 

Would something like this work or am I just being stupid.

Are you trying to simulate a Tach (RPM) signal? Swap the battery and the load resistor's positions, they're backward. You also need a base limiting resistor. Without it you'll fry the Base-Emitter junction of the Transistor.

 

I'm having problems getting a square wave 12v to come out. It's coming out at roughly 5volts. I think this is due to the input. Any help?
I also did try with the resistance at what the tach is and also did some measurements of the ac coming out of the bike and converted the rms to peak like you said but all of that stuff isn't on this picture. Darn program sometimes crashes when I try to save making it a pain.

 

well darn. I cant get this to work the way I want so I may have to go with the LM2917 to get the signal I need.

 

The LM2917 IC driving a large meter as in my prev post draws 26Ma at nearly full scale deflection. I would try feeding the 6V ac into pin 1 without the 10K resistor. I suspect the system should work as the frequency of the generator increases with engine RPM. You posibly will get enough voltage out of the 6V ac generator if voltage doubled & fed through a 12V three term reg to power the circuit.

 

will this work?

 

My wife wants me to get the hell off this PC and do some maintenance on her vintage Wrangler. Can you measure the following while I'm doing that?

(1) Alternator voltage at idle.
(2) Alternator voltage at at cruising speed.
(3) Alternator frequency at all of the above.
(4) Is one leg of the alternator committed to chassis GND?

Answering these questions will enable me to make an educated spice.

 

The alternator frequency is going to be tough to get without a frequency counter.

Might be easier to post a photo of the alternator windings; you could count the poles and derive the ratio from that.