T&T: : Tachometers was Advice on Equalizing & Tachometer?

[Dave Cooper]

<Arild wrote in part: Likewise a tach that measures the time from one pulse
to the next at one pulse per revolution. The cost of building a precision
quartz timing circuit is far more than the counters circuits now being used.
We had a similar problem in early Loran and GPS displays. If they updated
too quickly and the display numbers changed so fast it looked like
flickering; people thought the instrument was in error.>

I don't know the internal circuitry of the Navman digital tach but I do know
that the display is rock solid and I can increment the throttle under a
steady load such that it will show as little as a 5 rpm or so increase. The
pick-up for this is a 1/4" wide piece of silvered tape stuck on a 1"
diameter shaft which rotates at engine rpm.

The display never jitters and is always response to the slightest rpm change
as the engine loads and unloads to the wind/wave/load pattern. Remember this
is an old mechanical governor so it doesn't hold load rpm as well as an ECU
engine.

The VDO programmable tach which converts the digital pulse inputs via a
small processor to an analog 270 degree sweep readout is also gives a steady
reading and will show the small changes if you are staring at it closely.

I agree that in the old days counting pulses and averaging them may have
been required to "settle down" the displays. Water speed impellors,
anemometers, wind angle devices all come to mind. Averaging over 10 seconds
or so isn't unusual for these devices. These weren't getting as solid a
signal as we get from a rotating engine part with a very precise time from
the leading edge of one pulse to the next. Neither did we have cheap chip
sets that could do this. I suspect that Navman is measuring this time in
microseconds with a clock reference, doing the math and displaying the
result with firmware interaction.

However the older Loran and the GPS displays you are referring had/have to
average the results of each computational event as the input data varied
from one event to the next. Having a running average over many events was
necessary to give acceptable displays for the user. Newer GPS's with the
multiple receiver technologies allow many more "triangulation computations"
per event so allow faster updates with less location "jitter". I remember
when the first GPS's came out. They worked fine for taking "sights". Then
the next generation could be left on and would update every few minutes.
Then we got "continuous update" models. However, if you moved too fast the
GPS position was always well behind you. Aircraft units were always way more
expensive to allow faster processors to be used to keep the location in the
plane ;-) Today the cheapest GPS is hard to out drive in a car and light
planes. I can't say how they perform at mach something but the in cockpit
GPS on the Cessna 206 we were just in at 125 kts wasn't any better than my
$99 handheld.

I think the tach displays and the loran/GPS displays are separate and
unrelated issues other than the commonality of the viewers eyeball/brain,
IMHO.

On ECU engines one needs to know where you are in the revolution to make
certain engine events happen. Having a TDC reference and then lets say 100
pulses per revolution allows us to get 3.6 degree accuracy for each pulse.
This allow much better control of injector timing and valve timing of
variable timed engines.....sophisticated stuff in most of those new tier 3/3
1/2 engines. Look at the new European turbo diesels. They are totally
operated by software, get incredible power and efficiency with low
emissions. The wave of our future boat engines, I suspect.

As always YMMV...

Dave & Nancy
Swan Song
Roughwater 58