924Board.org

augidog · Joined: 03 Mar 2003 Posts: 1360 Location: New Jersey

I've read the tach side of the coil is a frequency rather than a voltage.
Or is it resistance that climbs with the rpm? If so, what voltage does it send out?
It’s raining here so I cannot take a multi- tester out.
I have some questions for someone good in designing electrical circuits.
_________________
1978 924 95 mile daily driver.
Audi TB/POR174M/High Flow Cat/2.25" exhaust
I knew that positive thinking thing wouldn't work.

numbbers · Posted: Fri Nov 05, 2004 11:21 am Post subject:

You will read a DC voltage on the "tach side" of the coil, but that is just a residule voltage from the AC (frequency) voltage. You will need an ocilliscope to actually see the AC pulses.
_________________
1980 924 Turbo

Sleykin · Posted: Fri Nov 05, 2004 11:55 am Post subject:

It is DC pulses somewhere around 6VDC peak. Unless your ballast resistor is bypassed and then it would peak near battery voltage. The PRF varies with engine RPM.
_________________
Glenn Neff
Medford, OR
87' 924S

John Brown · Joined: 07 Nov 2002 Posts: 903 Location: Leesburg VA

What is it you want to do?

Voltages are what you can measure if you like. (Amperage too, in theory, but that's harder to do and voltage will tell you anything you need)

The voltages are squarewave, or would be in a perfect world.

The frequencies involved are actually quite low: At 6000 rpm a 4 cylinder ignition system is only working at 200 Hz, or about the middle of the range of a good bass woofer. The tachometer is really then just a form of frequency readout. It counts the pulses as the ignition primary is switched off (ungrounded) and on (grounded).

BTW, somebody please tell me if I'm stupid, but I come across comments that seem to imply the Bosch ignition (on the 924) is a capacitive discharge type. For the life of me all the information and observations I can find are that all of the various incarnations used on the 924 were inductive types. Am I missing something?
_________________
John
80 931 - #931 44Cup
99 Escalade - tows track cars
gone but not forgotten: original 924.org car - 82

augidog · Joined: 03 Mar 2003 Posts: 1360 Location: New Jersey

Ok, here is my idea.
I want to make a shift light for my car using resistors at the appropriate rpm.
I want to use the opposite side of the highbeam indicator on the dash.
The same light as the high beam would be great(nothing big and stupid looking) .
I want it to blend in and this spot is not in use.( you can see the spot in my avatar left)
Any help would be greatly appreciated.
If I rev the engine can I see a rise in volts?
Any idea on how to make it?
Do I need a frequency to voltage chip?
It sounds like a fun project. Common you electrical wizards, lets figure it out.
I know it can be done
http://members.tripod.com/xexorz/schematics/analogtach.html
http://www.harlan-engineering.com/store/shiftlight/shiftlight.html
http://www.prestage.com/forums/ShowPost.aspx?PostID=74604
Mine is a little more ambitious.
A single light.
Set to go off at 3200 rpm and 3600rpm and 4000rpm
Problem is it needs to activate 3200 and then disconnect so that lights don't keep flashing when rpm drops down after shifting.
I thought maybe mercury switches on the shifter but that’s sounded a little barbaric.
Maybe a ground kill switch after it hits which can be reset at idle.
I have a basic drawing if you guys can fill in the componets.
_________________
1978 924 95 mile daily driver.
Audi TB/POR174M/High Flow Cat/2.25" exhaust
I knew that positive thinking thing wouldn't work.

Smoothie · Posted: Fri Nov 05, 2004 3:05 pm Post subject:

924RACR · Posted: Fri Nov 05, 2004 10:42 pm Post subject:

Good work, Smoothie. This may explain why even Permatune warns that you may experience misfires when swapping in their CD ignition box, if any part of the ignition system is not up to snuff...

OTOH, to be able to swap in a box with a rev-limiter would be nice for the racers...
_________________
Vaughan Scott
Webmeister
'79 924 #77 SCCA H Prod racecar
'82 931 Plat. Silver
#25 Hidari Firefly P2 sports prototype

Smoothie · Posted: Sat Nov 06, 2004 12:55 am Post subject:

Here's an adjustable shift light - no "pills" needed. Problem is it only deals with one rpm setting at a time. Then again, you could get 3 for ~$110, but then again by the time all 3 lit up, you'd be blinded. - http://store.summitracing.com/default.asp?target=esearch.asp&N=100&Ntk=PartSearch&Ntt=SUM-G2945+&x=18&y=12

Or how about a light that uses pills - figure out the resistances needed to replace the pills and give you the shift points you need, then wire the 3 resistors into a 3-position switch and mount the switch to your shifter. -Or [this one's off-the-wall and would be tricky] if there's a way to use momentary switches - mounted so they're activated in turn by the position of the shifter. - http://store.summitracing.com/default.asp?target=partdetail.asp&part=ATM%2D5330&N=110&Ntk=KeywordSearch&Ntt=shift+light
_________________
"..it's made in Germany. You know the Germans always make good stuff."

'82 924T, US version, dark green metallic, 5 speed Audi 016G gearbox

CMXXXI · Joined: 05 Nov 2002 Posts: 1939 Location: Vicksburg, MS

Just so happens I re-set my timing and idle this afternoon. My hand-held tachometer has been acting up (bad tach-dwell switch) and on a fluke I measured the voltage on the (-) side of the coil, and it was something like 5.6v, regardless of RPM. As JB says, the tach just measures the pulses, on/off/on/off/on...

On a side note, does anyone have informal records of the difference between a shop tachometer and the one that is in the dash? My hand-held says 1000rpm, but the in-dash one reads about 1200. At 2Krmp on the hand-held, the dash says about 2300. I set my idle based on the hand-held to 900 rpm, but the in-dash reads about 1050-1100. Just curious if there is any documented history as to the OEM tach's accuracy. (Is there any way to "positively" calibrate a hand-held unit?)
_________________
'79 Eurospec 931

augidog · Joined: 03 Mar 2003 Posts: 1360 Location: New Jersey

Smoothie · Posted: Sat Nov 06, 2004 11:05 am Post subject:

I think a better choice of light would be the o2 sensor light if you have one. That thing really gets your attention - it's downright annoying in fact while the highbeam light isn't all that bright. -And if what you're pondering works out for you, I'd take the voltage you get out of it (assuming it'll be something lower than 12V), use that lower voltage to switch on a relay that's in turn wired to supply 12V to light your light.

Here's something on high-pass filters - http://en.wikipedia.org/wiki/High-pass_filter
_________________
"..it's made in Germany. You know the Germans always make good stuff."

'82 924T, US version, dark green metallic, 5 speed Audi 016G gearbox

augidog · Joined: 03 Mar 2003 Posts: 1360 Location: New Jersey

John Brown · Joined: 07 Nov 2002 Posts: 903 Location: Leesburg VA

Thanks for the confirmation. Thought maybe I was going senile.

I'm no expert but CD ignition doesn't make much sense to me. Not with today's technology.

Transistor ignition was a big deal around 1965 or so I think. Remember my Dad putting 'transistor ignition' on a TR3 about then. 'Course, to the extent that transistor ignition is mainly about how the primary is switched, it would apply to either inductive or capacitive systems.
_________________
John
80 931 - #931 44Cup
99 Escalade - tows track cars
gone but not forgotten: original 924.org car - 82

Smoothie · Posted: Sat Nov 06, 2004 2:35 pm Post subject:

Sleykin · Posted: Sat Nov 06, 2004 4:38 pm Post subject:

The digital voltmeter is measuring the peak (or nearly the peak) voltage of the pulses. It is concivable to make a simple R/C time constant that would output a voltage proportianal to the frequency of the pulses. Pulse duration would also come into play here. You could then use a diferential amplifier with a pre-set voltage on the proportioning input and wire it as a schmdit trigger to fire a lamp or a relay. It would not be very precise though. A better method would be to use a DAC and set a trigger point from that. You could use micro switches on the shift lever to insert the appropriate bias for the shift points you want to set. That way one light would tell you when to shift...but it seems like a lot of bother to me

Several other ways would work but all would be more involved.
Sensors on the tach to fire a light when the needle goes by?
_________________
Glenn Neff
Medford, OR
87' 924S