Tach troubles

[lateapex911]

In there ever expanding universe of Porshe problems....one thing leads to another...

I had ignition woes, so I replaced my points with a Crane Optical seup, and the CDI (not functioning) with a MSD 6AL I happened to have. I connected the tach via the purple/black wire to the MSD 6AL tach output and got the same bouncy and sometimes intermittent tach I've always had.

Now, the intermittent part has taken over...actually the tach is dead. Since it's bouncy, I thought, "Time to send to North hollywoodSpeedo", but as I know it will be gone for a bit, I decided to hook up a little shift light indicator box from my racecar to get by with.

I breadboarded it, set the 4 light thresholds, fired up the car, and it worked great.

So I went to install it, and when I hooked ot to the tach purple wire, I got nothing. Hmmmmmmmm

So I got the multi meter out, and put it on the diode sound setting, hooked it to the purple wire at the back and in the dash, and got.....nothing. Hmmmm...I switched the setting to ohms, and got a high 3K or so reading. Hmmmmm..

I checked my Porsche wiring diagram, for my car, a 73E originally with MFI, and see that the wire appears to be a straight run, with a stop along the way to a temp switch. (Which is gone I believe, as my car is now carbs).

So, am i missing something?
Is this a dirty connection somewhere?
A broken connection?

Me being an idiot?

Thanks!

[pookie]

On mine 76 with 81 3.0 and MSD. I had to run a new wire from the tach straight to the box. Plugged right into the MSD it works no problem. Did you already send your tach off? I'd try it with a straight line from tac to MSD before I shipped it off.

[304065]

Hi Jake.

Until Warren checks on I am as close as you are going to get to an expert.

Let's work backwards: the 1973 tach is driven, more or less, by rectangle-wave pulses that are more or less the same duration regardless of input frequency. (thanks to 914GT for that clear description!) The old '73 tach, whether it contains an IC or not, is based on a monostable multivibrator or "one-shot" in which a triggering pulse results in a little shot of power to the meter movement. Think of it as a volleyball, the more frequently you hit it, the higher it rises.

Now, inside the Bosch CDI box there's something called a "pullup resistor" connected between the 12V supply to the box and the tach input lead. When the points are closed, they complete the circuit to ground and current flows in the circuit, (exactly how much varies by the year and the value of the resistor) which triggers the SCR inside to dump the capacitor into the coil and you get a spark, whoopee!

Why I have wasted your valuable time explaining that is because it's critical to understanding WHAT SIGNAL THE TACH WANTS TO SEE. It's looking for a very high-impedance square wave type of pulse with a very low current value.

Now, the output of the MSD box is a 12V square wave with a 20% duty cycle. (Actually, it can't be a pure "square wave" because that's a 50% duty cycle like a 4-cylinder distributor generates, but that's what MSD calls it.) That's a little different than the pulse width coming from a set of points. 38 degrees of dwell per ignition pulse, three pulses per 360 degree revolution of the crank, 38*3 =114 degrees of point closure per revolution, divided by 360 degrees is about 32% duty cycle. Remember, the points still operate like they did with the old kettering ignition in which point closure actually allowed the magnetic field to build up in the coil. Why I'm telling you THIS is that it's probably the source of your ORIGINAL tach problems-- the tach expects a wider pulse width and you are sending a narrower pulse width and it's getting confused, electronicaly speaking.

How to fix it? Your MSD triggers off a white wire connected to the points. Connect the black wire with purple stripe directly to the points instead of the output terminal on the side of the MSD and see if that restores accurate tach function. Look, current HAS to flow in the MSD's trigger circuit, I don't know how much, but some must be present in order for it to trigger the MSD's multiple-spark cascade.

Now, another problem is that the Crane optical igniton foo-foo probably doesn't generate enough of a signal to properly trigger the tach. My advice to you is to revert back to good old Bosch points, set the gap to .035 (I think that's the spec off the top of my head when I was trying to help out Wayne V. with his 73RS motor at the race), and the dwell to 38 degrees and it may restore accurate tach function for you. Also, I don't know what kind of pulse width the Crane generates: it may be enough to fire the MSD on the front edge of the signal, but the pulse may not be enough to drive the tach.

Or you could stay tuned here, I'm in the middle of a tach retrofit that utilizes a modern movement and board to replace the old VDO stuff.

Good luck!

p.s. the Black wire with purple stripe runs from the points to the MFI "speed switch" which is used to shut off the fuel on overrun, and then to the "C" terminal on the tach. If the speed switch is removed because you have carbs, please make sure the old connector's not shorting out somewhere. Since you no longer have any points and the ground is controlled by the Crane optical ignition, it's being switched by a semiconductor, not a metal contact, and that could cause your tach to not work at all while the engine still runs-- if it was points and the purple wire grounds out the engine won't run at all. That's why you get 3Kohms when you measure ground to the blk/purple wire-- if it were points you would either get zero ohms (points closed) or infinite ohms (points open). Optical ignitions are usually a bandaid fix for some other problem, they suck and I replaced the one in my '66 with Bosch points and a 3-pin CDI and the car postively SCREAMS now.

[jstobo]

If you trigger the tach off the points or crane you will the the MSD Tach adapter. The black purple wire you are trying to use may not go to the tach. On the Distributer from the points , two wires should leave, a black one going to the CDI and a Rd/Bl that chages to BL/Purple that goes to the Speed Switch(RPM Transducer). From the Speed Switch a Bl/Pu wire goes to the 14 pin connector going to the tach. You need to tie into the wire leaving the speed switch. If you still have MFI you need to send the MSD Tach output to the speed switch for the fuel cutoff circuit.

[lateapex911]

WOw...great responses, thanks! I cant hit every point now as I am running out, but,,,

the black/purple wire is, I thought, a direct shot from the dist area to the tach now. The speed switch is gone.

The odd thing is that since I have owned the car, in '92, the tach has been bouncy. That was when it had MFI.

Then after the rebuild, and the carbs were added, still bouncy.

Then the ignition saga. In an effort to get what have turned out to be carbs in need of a rebuild running correctly, the points and CDI box was removed, and the MSD and optical pickup added. The CDI was not functioning at the time so it made sense. And I hate points, LOL.

After the swap, I decided to try, just for giggles, hooking the tach output of the MSD into the black purple wire where it enters the harness near the distributor, just for giggles. Viola! Working tach! Bouncy as always.

Then one day, no tach. Next day fine. A week later no tach. ANd so on with the ratio ebbing into the no tach category. What makes me think the wire is an issue is because I can run my little light box (4 lights sequence based on RPMs set) off the MSD with a jumper, but not with the black purple wire.

John, your tach retrofit is of interest.!

More (more itelligent too, I hope) response later.
thanks guys!

[304065]

Jstobo, the MSD 8910 tach adaptor is usually required only in '69-71 cars with the old low-impedance tachs. These cars, yours included, used a silver can mounted on the engine electrical console to generate a voltage spike when the points closed that would simulate the old high-voltage spike from the coil. I don't know the cutover point when VDO changed to a low-impedance tach that would run directly off the points, but it was sometime in late 1971 or early 1972.

The 8910 is merely an inductor coil and diode that acts much like the old ignition coil-- a magnetic field builds up in the inductor and when it collapses it generates a high voltage in the circuit. This should not, however, be a problem for Jake, who has a 1973 tachometer.

[petro206]

Originally Posted by 304065

How to fix it? Your MSD triggers off a white wire connected to the points. Connect the black wire with purple stripe directly to the points instead of the output terminal on the side of the MSD and see if that restores accurate tach function. Look, current HAS to flow in the MSD's trigger circuit, I don't know how much, but some must be present in order for it to trigger the MSD's multiple-spark cascade..

Hi 304065
I resume this old post to ask about MSD ignition 6AL and Tachometer.
I have a 2.2T and i tried to connect the gray wire from MSD to purple/black wire, with no success.

After this, i tried to connect purple/black directly on the distributor (White wire from MSD) with no success.

I don't know if my tach is PRE 71 (1) or POST 71 (C) and i can't read because now i'm very far from my car

I would like to create a DIY circuit. What can i do?

I understand that MSD generate a square wave 22.5%, while i need a square wave 32%

Is there an easy way to create a circuit to enlarge the square wave from MSD output?

if i have a PRE 71 tach (1), which is the type of inductor coil i have to use to increase the voltage ?

other suggestion?

[Trackrash]

I can help with some of your questions. Here is a thread and if you look at post #12 you see a picture of what the Porsche's tach adapter or as they call it a RPM sensor as used on the 70 and 71 models. Tachometer trouble
Here are some adapters that are supposed to solve the problem. http://www.pelicanparts.com/cgi-bin/ksearch/PEL_search_2014.cgi?SUPERCAT_FLAG=Y&make=POR&Context_make=&please_wait=N&LastVisited_input=5862&Previous_Section=&forumid=&threadid=&FILTER_SVSVSI_5862=ON&description_side=tachometer&GoSearchSideFlag=&command=DWsearch&command=DWsearch&REFINE=Y&FORUM_REFER=&SUPERCAT_FLAG=Y&Previous_Section=&LastVisited_input=&make=POR&description=tachometer&x=59&y=17&FILTER_SVSVSI_5862=ON

[petro206]

Thank you Gordon,
i understand that i have to use the speed sensor or intermediate unit to increase the voltage peak.

first test
MSD tach output --> speed sensor (or intermediate unit) --> Tachometer

second test
White wire connected to distributor --> speed sensor (or intermediate unit) --> Tachometer


Yes, i know there are some adapter, but i understand that it not sure that work 8910 or 8920, you could buy bought to see works the tach.

Do you have some advice to avoid to fries the tach? Use a 0.5 A fuse on black/purple wire for example?

Is there a limit for voltage peak?

If i remember well, i connect black/purple directly to the + (or -) of the coil, but at idle i read 3000 rpm (multispark). how bigger is the voltage there?

It's very hard to find the correct way....

[RichardNew]

You should probably talk to Buz at Seattle Speedometer. It sounds as if you're in Italy though. Heres' an article I wrote about gauges recently.

Richard Newton