Tachometer Blues

[barrylab]

When my trusty Westerbeke water pump died and pumped coolant into the engine, I decided to have it rebuilt. That was two years ago, and I just hit the water again. Part of the delay was caused by my engineer mentality that suggests you do things in a logical order and they are much easier to do. I replaced the fuel tank while the engine was out (good thing too, the old one started to leak on the fabricator's floor). I also decided to get a new tachometer. The old one had an hour meter that was blank, it didn;t show anything because all the numbers had long since faded away.

Here's the saga of the new tachometer:
I bought one from TAD marine at the ransom price of over $300.00. I justified it by saying it was a Westerbeke, so I was maintaining some historical integrity. While the engine was getting it's rebuild, the tach somehow disappeared into the mass of stored items taken off during the removal. When it cam time to put the engine in, I wanted a tach, so, like a fool I bought a second one. I thought, well, now I have a spare ...
When I put the new tachometer into the instrument panel, I redid all the wiring from the engine to the panel, as a previous owner had replaced it with house wire of many types bolted together in places and wrapped with enough electrical tape to make a black mummy.

When I started the engine, the tach moved from the 0 position, counterclockwise to the 4000RPM line. I read the literature that came with the tach and realized you need to adjust a knob that has settings of 4,6,8,A & B
The literature doesn't say which to set it to, it says to use it as a coarse control and use the internal potentiometer (with a 2mm allen wrench) to fine tune it. (i don't care what you say you don't read the directions until it doesn't work). None of the settings caused any different behavior.

I found the "spare tach finally and tried it with identical results, so I brought one to work, where I could play with it in a controlled lab environment. I found that you need to remove power to reset the reverse slammed condition, then no signal to the tach input pin (S) results in 0RPM. With the coarse setting at 4, a 2KHz square wave produces about 500RPM, a coarse setting of 8 creates 1000RPM when 20KHz is put in. In both cases the needle moves up and down with the fine adjust and adjusting the frequency. A setting of 6 has similar behavior, so I thought they might be right about the coarse/fine definition.

Settings A & B did nothing no matter what I did with amplitude, frequency, duty cycle or fine tune settings.

Discouraged by this I did more research. I turns out the tach is made by Datcon. It is a PN 71725-00 Model 124A40EM. This is apparently a marine industrial version of a truck model they manufacture. It's available for about $125.00 from several places. They actually have a pretty good manual that explains 4 is for a 4 cylinder ignition signal, 6 is for 6 cylinder, 8 is for 8 cylinder, A is for a 10 pole alternator signal and B is for a 12 pole alternator signal. My Balmar is a 12 pole so that should work. I also found a rnat by someone saying they bought a smaller Datcon tach and had to put a capacitor in series with the tach to get it to work.
Guess what I'm trying tomorrow.

[barrylab]

After calling everyone from Perkins (whose part number Westerbeke uses) to Westerbeke (who branded the tachometer) to Datcon (manufacturer of the tachometer) to  TOAD marine (who sold me the tachometer), I called the Westerbeke local distributor: Hanson Marine and spoke with Bob Hanson. He made a few calls and told me that Balmer (the alternator manufacturer) suggested the ground connection was probably bad. I'd just ohmed out the connections and had <1 Ohm in the ground path, so I thought this was another wrong guess. Just to be sure I wired a new connection to ground - The tachometer just started working. Seems the old connection is intermittent, probably in the wire as this wiring is almost 40 years old and isn't tinned.

Hats off to Bob Hanson of Hanson Marine. He was the only person to come up with a potential solution.

[SVJourney]

Good troubleshooting Barry,  Thanks for sharing it.

Wayne

[PeteW]

It's easy to assume that simply because the alternator is bolted to the engine block that its got a good ground regardless of the size or condition of the actual ground wire. On my motor both the alternator and the starter have there own ground wires to the negative  buss. In the case of the alternator its a AWG 00. For my alternator, which is an 100 ampere Ford, I have AWG #2 for both ground and B+. The crimp lugs are also soldered on both ends.

For the tach to work there is also the return ground on the tach. This is the return for the +12V supply, the illumination bulb as well as the stator square wave signal. From a signal integrity point of view it would make sense to give the stator signal wire its own return wire going between the alternator and the tach. I might even make these two a twisted pair. This would remove electrical noise from 100 amp charging currents elsewhere from riding on the tach signal and making your reading go crazy.

While you were troubleshooting the tach issue did you happen to notice if the alternator was putting out full current?  I'm going to guess that maybe it wasn't.

In troubleshooting you know its always the last thing you look at that is the problem. Glad you figured out.

What I want to know is? Why does my tach hold its RPM reading as it was at the time I turn the ignition off ? Is this normal ?  Pete

[barrylab]

The alternator has a black 00 wire to the ground terminals of the battery, a red 00 wire to the battery isolator, and a ground 10 gauge wire to the voltage regulator. The starter has a black 0 gauge wire to the ground terminals, and a red 0 gauge wire to the battery switch.The grounds to the instruments come through the wire harness. There is an 8 gauge green wire and an 8 gauge red wire that bring ground and 12V to the instrument cluster (lights mostly). The lights work fine, The alternator charges the battery, but the ground path to the tachometer seems to have an issue. I don't get much DC drop, but the tach won't work without the additional 16 guage "signal ground" I've run from the negative battery terminals to the tachometer ground pin. I suspect corrosion wherever the green wire is connected to the engine, though I haven't traced that back yet. I just assume that if I got less than an ohm with my meter and there was just a couple of millivolts of drop when measuring DC, that the wire was OK. Somehow ground must have shifted, or the inductance of the harness caused a drop. Somehow when I measured the AC signal with the same handheld meter, I got 7.6V, from the tach ground to the tach "S" terminal, so it still doesn't make sense.

All the wiring except the engine harness are new this year, and now the engine harness is due for an upgrade.

[PeteW]

Since your problem seems undetectable by measuring ohms. You may have to get a little more aggressive and start looking for AC and DC voltage drops between battery neg terminal, alternator neg terminal and tachometer case ground with the motor running. You should be able to zero in on a noisy ground this way.

Also I've had some unbelievable situations regarding electrical connections to things aluminum. Aluminum oxide is a perfect insulator and it looks like clean aluminum. I had to grind it off. A thought since the alternator and perhaps the tach housing are aluminum.

Might be time to freshen up some of the crimp on terminal lugs by cutting them off and crimping and soldering new ones.

Adding a dedicated ground return ( 12 gauge) alongside the tach S wire to the alternator would be proper grounding technique for routing that signal wire. In fact, you may want to move the tach ground to the alternator and remove the tach ground to the neg battery. Otherwise if the alternator ground were to fail,  the tach ground wire to the battery would see 100 amps and burst into flames.