Thursday, April 30, 2015

Current transformer project

I'm perpetually behind on projects here, which I think is a condition of living on a boat, or perhaps just a condition of modernity. So I've been taking the time, on otherwise idle days at anchor, to whittle away at the list. I'm also perpetually behind on blogging some of the major projects I've already done, such as the great 24-volt conversion project that I've been promising to write up for over a year now.

In an effort to keep the backlog from growing, I thought I'd use some of the time to write up one of the most recent, considering weather has pinned us down here for at least another day or so. As with so many projects, this one has been on the list for a long time, but has bubbled to the top because recent events on board increased the urgency. I am talking here about fixing the main AC panel ammeter that tells us how much power we are drawing from shore or generator systems.

The ammeter, in this case, is part of the Blue Seas AC/DC circuit breaker panel built into the helm console, which is original to the boat and has been there, according to the photo record, since it was originally finished under the first owner's watch. While large and swoopy-looking and "custom" labeled, this breaker panel is actually an off-the-shelf item available in the Blue Seas catalog. The labels come with it on a big sheet of commonly-used circuit names, for customization by the installer.


Main AC ammeter (left) and selector switch.

As such the ammeters are actually pre-installed and pre-wired on the panel. The DC ammeter is hard-wired and measures all the current passing through the main DC breaker on the panel, while the AC ammeter is wired, along with the AC voltmeter, through a built-in switch that allows it to be switched from one AC leg to the other.

This is all well and good, but in this configuration it has, as far as I can tell, never, ever been able to measure all the current the boat is drawing on either leg. That's because there are some large AC loads that do not actually pass through the panel on the helm.

When we got the boat, the big culprit in this regard was the washer/dryer. It's located in the engine room and is the only 240-volt appliance on the boat. When the automatic transfer switch (ATS) was added during the last owner's stewardship, the electricians split the feed to that appliance off right at the ATS, running it through a separate tiny panel with a two-pole breaker for the washer/dryer and another single-pole breaker that supplied the engine room exhaust fan. Any current being used by those items was thus not included in the totals shown on the helm ammeter.

The engine room fan was moved off this arrangement early on, because wired in this way it could only be operated on shore or generator power. We don't typically run the generator under way, which is exactly when we need the exhaust fan, and so I moved it to a circuit connected to the inverter.

Having the enormous load of the dryer invisible to the ammeter was problem enough in itself. We often tripped the shore power breaker when using the dryer, until we learned just how much (or little) other load we could have on the system at the same time. We had to go by feel, guessing at the amount the dryer was using at any given moment, subtracting that number from the 50-amp shore supply, and keeping everything else under the remainder based on the panel ammeter.

The problem grew much worse when I completed the aforementioned great 24-volt conversion project. That involved, among other things, a new inverter/charger, which was best wired with a 120/240-volt, four-wire circuit, thus loading both legs of the input power, in contrast to its predecessor which was a 120-volt-only, three-wire load and was supplied by a circuit on the main helm AC panel.


The "new" main panel, in the engine room, a 120/240-volt 8-space "main lug" QO-series panel. Two-pole breakers feed the inverter/charger and washer/dryer, while two single-pole breakers feed the two sides of the former main panel at the helm.

The helm panel has no provision for 240-volt circuits (it's arranged as two separate 120-volt panels), which is probably why the washer/dryer was not wired to it, and why I could not wire the new inverter/charger to it either. Moreover, wiring such a big load through the helm meant the power would pass right by the load in the engine room on its way to the helm, through the panel, and then all the way back again, adding cable, weight, heat, and voltage drop.

So now we have two of the largest consumers on the boat -- the dryer and the battery charger -- that do not register on the helm ammeter. Even with the higher 67-amp capacity of the generator circuit, we still occasionally trip a main breaker, and this problem is just getting worse now that we are running air conditioning more often. And if we want to deliberately load the generator right to its limit, to reduce wet-stacking, we have no good way to monitor our progress toward that end.

The simple answer to all of this is to move the input for the panel ammeter down to the engine room, immediately after the transfer switch, so that it can see the full draw of everything on the boat. Easily said, but a lot of work to actually do.

AC current is measured with a device known as a "current transformer" (CT). Unlike most transformers you might encounter in a typical electrical system, which transform one voltage to another, a current transformer actually transforms one current to another, proportional, current. In the case of our ammeter, it's a 1,000-to-one ratio: a 50-amp current passing through the main feed is transformed into a 50-milliamp current into the meter, while, say, a 12-amp current would result in a 12-milliamp meter input. The meter pegs at 50 milliamps, but the dial is marked from zero to 50 amps.

Physically a CT is a toroidal coil, with the current-carrying conductor to be measured passing through it. It picks up the magnitude of the measured current by induction. The CTs in the Blue Seas panel are very small, just big enough to go around a 6-gauge wire with a bit of room to spare, making them easy to fit into the tight space for which the panel is designed. That also made them ideal for installing in the tight space of our new main electrical panel in the engine room.


One of the two Blue Seas CTs, still installed in the old panel.

The CTs that Blue Seas uses have their pair of lead wires molded in. Blue Seas has soldered the other ends of the leads to the selector switch. I cut the leads at about their midpoints. To extend these down to the engine room I needed about 35' or so of two-pair wire, but all I had on hand was some 4-pair CAT-5 communications cable. While I would not use untinned, coarse-strand cable like this for any critical application on a boat, for the meter it would suffice. I used two pair for each CT to minimize the resistance. At least the fine twist of CAT-5 will also minimize inductive interference in the signal.


CT relocated to new main panel in the engine room. The zip tie just keeps it from sliding on the wire.

After moving the CTs and splicing the wiring at each end, we fired up the genny for a test. While I was concerned that such a long run of meter wiring might impact the accuracy, a quick check with my clamp-on meter revealed that the helm meter is spot-on. We now have an accurate reading of the full draw on both legs up to a maximium of 50 amps, which is the limit when we are on shore power.

The generator is a 16kW unit, however, which means it is theoretically capable of producing 67 amps, and in practice has a 70-amp main breaker. Consequently, when running on the generator the 50-amp ammeter can and does peg occasionally, particularly on the leg carrying the battery charger, which by itself can draw up to 30 amps. For the time being, we avoid leaving the meter connected when running above 50 amps (the selector switch has an "off" position).

I'm not too worried about running the CTs at 30% above their 50-mA rating, as these are minute amounts of current. But the meter itself could be damaged long-term, and besides, we'd like to know exactly how much we are drawing even when we go above 50 amps. So I will probably devise a shunt that can be switched in when on the generator to divide the reading in half. I'll need to experiment to determine the meter's internal resistance in order to size the shunt.

We almost never trip breakers when running on the generator, which is large enough to run almost everything on the boat simultaneously unless we are running the dryer, which we try to run mostly on shore power anyway. So the meter is most useful in managing the more limited shore power feed, and for that, the 50-amp full-scale reading is perfect.

No comments:

Post a Comment

Share your comments on this post! We currently allow anyone to comment without registering. If you choose to use the "anonymous" option, please add your name or nickname to the bottom of your comment, within the main comment box. Getting feedback signed simply "anonymous" is kind of like having strangers shout things at us on the street: a bit disconcerting. Thanks!