Yesterday we wrapped up Alumafiesta in Tucson, and so now we’re in the “recovery” phase, just trying to get back in sync with life and unwind after a hectic week. For me, it’s time to get a few Airstream projects done.

In the first week of January I decided to do something I’ve never done before: take a friend out in the Airstream for a week-long trip, instead of my family. It might seem odd that I’ve never done that, but those of you who own travel trailers or motorhomes can testify that they become very personal. I always associate our Airstream with our family. I rarely even take the Airstream out alone.

But my buddy Nick was a pretty safe bet. He’s low-maintenance, easy-going, and a decent cook. We work on our old Mercedes cars together on weekends, and have spent many a Saturday morning digging around the junkyard, so we’ve bonded over greasy parts and underneath diesel engines. At one point Nick mentioned that he and his wife would like to own a small travel trailer someday, so I figured he was ready to get introduced to the world of Airstream.

We took the Airstream out to the southern California desert, where we met up with a bunch of other desert rats/technomads, went hiking, and ate apple pie from Julian. Then we hopped over to Quartzsite to see the spectacle of cheapskate boondockers and endless flea markets, and we wrapped up the trip with a hike up Picacho Peak in Picacho AZ. It was all brilliant except that the Airstream kept giving us little problems, signs of advancing age and frequent hard use.

I already wrote about the battery charging issue, and that will get addressed later this month when the new charger comes in. During the California trip, I found a few other glitches, most of which I fixed on the spot.  One interesting problem was the sudden failure of the P/T valve on the water heater.  This valve is supposed to relieve excess pressure in the water heater (it’s a safety device) but mine decided after ten years it was done holding back, and so it began to gush water.  One trip to the local hardware store later, Nick and I had it swapped out for a new one.

All of the repairs were small stuff like that, hardly worth breaking out the tool kit for, until the aluminum bracket that holds up the dump valves broke loose.

Even this wasn’t a major problem.  The bracket is riveted into the belly pan, which is thin aluminum, and the rivets had finally torn out after 100,000+ miles of bumping along North American roadways. The dump valves hung a little lower than usual, but everything still worked.  My only concern was that eventually, without the bracket, the connections on the plastic pipes might eventually start to leak.

Normally I’d fix this with bigger rivets—which I carry around at all times, as well as a rivet tool—but in this case there wasn’t enough intact metal in the belly pan left to make a bond I could trust.  It needed a reinforcing sheet of aluminum.  I browsed around the junk piles at Quartzsite and found an expired California license plate that was the perfect size to serve as a reinforcing plate.

I was feeling rather resourceful until it became obvious that the entire dump valve assembly would have to be removed in order to get in there with the drill and rivet tool to rig up my field repair.  This was more of a job than I wanted to do in a Quartzsite campground (we had opted for full hookups rather than boondocking again), and it seemed like an opportunity to order a new set of dump valves to swap in at the same time.  So after getting home, I placed an order for the new valve set and went off to Alumafiesta for a week.

Alas, the day before Alumafiesta I heard a strange hissing noise from the refrigerator, one I’ve never heard before.  The refrigerator used in RVs is normally silent. There wasn’t any obvious ammonia smell (a definite indicator of a major failure), nor any sign of coolant leakage, and the fridge was still working so there wasn’t anything I could so at that point.  A few days later, the fridge stopped cooling and a tell-tale puddle of greenish-yellow coolant oozed out.  RIP refrigerator #2.  It lasted just six and a half years, one of the 900,000+ victims of Dometic’s unfortunate refrigerator manufacturing fiasco from June 1, 2003 to September 30, 2006.

The immediate solution was to find another refrigerator to store my food for a few days until the event was over.  The long-term solution was to order a new cooling unit for the refrigerator.  The cooling unit is the guts of the fridge, and it’s entirely replaceable.  We could get a new refrigerator, but we really like the Dometic NDR-1062 that we have, and it has been discontinued.  It’s the only model we’ve found that yields 10 cubic feet in the space of a typical 8-cubic foot refrigerator box.

The dealer quote was $1,560 to replace the cooling unit.  I ordered a new cooling unit for $524, to be delivered by truck freight to my door with a 6-year warranty, and I will replace it myself sometime next week. I’ve never done this job before, but it doesn’t look terribly difficult, and I’ve got friends who can help. As a few people have said to me, “If you can do your own work on that old diesel Mercedes, you can do this.”

Meanwhile today, I got under the Airstream and swapped out that dump valve set, and riveted that bracket up. It was a good warm-up for the work yet to come this month.

At times it seems like I’m constantly working on one Airstream or another. This can be frustrating at times because there are other things in life. On the other hand, each repair is an opportunity to get to know the Airstream better, learn new skills, and improve things beyond the original factory spec. There’s something very satisfying in doing it yourself.  And I can assure you that those dump valves will never drop low again, thanks to ten huge rivets and a California license plate.

Next week I’m going to camp in the desert in California, and so I’m getting the Safari ready now. I’ve learned that anytime the Airstream has been sitting for a while, it’s best to start checking all the systems at least two weeks in advance.  That way the little problems that sometimes crop up during storage can be resolved without a last-minute panic.

I figured I’d find something that needing doing, but was completely surprised by what turned up.  The Tri-Metric battery monitor was reporting the batteries were at 73%. Since the Airstream has been continuously plugged into power since late August, this was clearly suspicious. The batteries should have been at 100%.

The Tri-Metric 2020 (by Bogart Engineering) is one of several amp-hour meters you can install in place of the existing battery monitor that came with your travel trailer. I recommend this upgrade to everyone, for reasons I’ve outlined previously. It’s about $200 plus installation, and well worth it for anyone who ever camps off-grid, has solar panels, or just wants to know what’s really going on with their batteries.

The Tri-Metric is highly accurate. It “counts” every bit of power (in amps) that goes in or out of the batteries, so when it reports 73% charge, it’s pretty darned close, like within 1-2%. We’ve had that Tri-Metric running in the Airstream for nine years and it has always been reliable.

So the first thing I checked was that the Airstream was in fact receiving power.  That was simply a matter of looking at another meter in my case, but if you didn’t have one, turning on an AC-powered appliance would verify power as well. Just plug in a lamp or something.

The second thing I checked was that the power converter/charger was doing its job.  You might recall that earlier this year I switched from the factory-installed converter/charger to an Intellipower 9260 with Charge Wizard. This was in order to get better battery charging when we were plugged in. The factory put in a 2-stage charger, and the Intellipower has three stages, plus somewhat more “brain” so it doesn’t overcharge the battery, and the option of manual overrides using the Charge Wizard.

The Tri-Metric answered this question too. It was showing that the batteries had a tiny rate of discharge, about -0.05 amps. Turning on additional DC power consumers (lights, fans, water pump) revealed that the rate of discharge never changed.  That’s because the Intellipower was doing at least part of its job, stepping up the power input as needed to compensate for DC power draws. If the Intellipower wasn’t working at all, the Tri-Metric would have shown a dramatic increase of discharge.

Now, to understand what’s coming next, you need know something about the way batteries charge. A fully charged “12 volt” battery really runs about 12.7 volts.  (This varies by the type of battery chemistry used, but here I’m referring to the typical “wet cell” lead-acid batteries that come with your Airstream.)

Think of volts as electrical pressure. In order to get 12.7 volts into the battery, you have to “push” power into the battery a little harder than 12.7 volts. The harder you push, the faster the power goes in.  But there’s a limit to how hard you can safely push, so for this typical sort of battery the manufacturers usually recommend about 13.6 volt for a normal charge. When the battery is really empty you can push a little harder (meaning more volts), and when it is nearly full you have to back off and push more gently (less volts).

The Intellipower, like many other RV converter/chargers, has pre-set levels at which it charges the batteries. If the battery is full or nearly full, it charges at “storage mode” rate of 13.2 volts.  This keeps the battery topped off, compensating for a little “self-discharge” that naturally occurs with lead-acid batteries.

If the battery is somewhat discharged, the Intellipower steps up to 13.6 volts.  This is the “normal mode” of charging.

If the battery is really discharged and needs a bulk charge quickly, the Intellipower goes for broke and pushes hard at 14.4 volts. It will only do this for a little while before returning to the normal mode of 13.6 volts.

Those are the “three stages” that I was referring to earlier, and it works just great for conventional batteries.

With the trailer plugged in, the Tri-Metric was telling me that the battery voltage was steady at 13.2 volts.  That’s not the actual voltage of the battery, because it was receiving some input from the Intellipower. To get the true voltage, I disconnected the AC power and waited for the battery to have a chance to “settle”.

Ideally I should have let it settle for 24 hours with no charge or discharge (e.g., disconnected), and then measured at 77 degrees, but I was impatient and didn’t want to disconnect the battery at that time, plus it was cold outside. So I waited six hours with a very small load on the battery (from the refrigerator’s circuit board and a few other small “parasitic” drains), and checked the voltage again.  It was 12.7 volts, which in a conventional battery would indicate that it was about full.

If this had been the end of the story I would have concluded that the Tri-Metric had somehow lost calibration and wasn’t counting the amps correctly. But that just didn’t sit well with me.  The Tri-Metric seemed to be acting normally.  After six hours of the trailer being unplugged, the Tri-Metric was reporting a 70% charge, which seemed about right. Something else had to be wrong … but what?

The answer surprised me.  Long ago we replaced the original Airstream batteries with an Absorbed Glass Mat (AGM) battery. These are sold under various brand names, such as Optima. Ours happens to be a Lifeline 4D model. I looked up the charging requirements for this battery and discovered that it has entirely different voltage requirements, as follows:

Absorption Charge voltages (“normal mode”): 14.2- 14.6
Float Charge voltages (“storage mode”) 13.1 – 13.4

Although the Intellipower charger was supplying power to the battery, it just wasn’t enough. When the battery wanted 14.2 to 14.6 volts, the charger gave it 13.6 volts. Sitting in storage, the charger gave it only 13.2 volts, which was fine for a while, but not enough to maintain the AGM’s rated “full” level of 12.9 volts. The battery gradually lost power.

The upshot for you is that there’s a dirty little secret about most power converters: they aren’t optimized for charging AGMs, at least not the Lifeline ones.  In our case, the Intellipower documentation doesn’t address this, and factory voltage output settings can’t be changed. I checked a few other popular brands and found they are exactly the same. Only a few brands, like Xantrex, have the built-in capability to push the correct voltages needed for AGM batteries. If you have switched to AGMs and haven’t upgraded your converter/charger to the right brand, your battery is going to have reduced capacity as well.

The really peculiar thing about this is that it took eight months for my problem to crop up.  Why didn’t I notice a charging problem before?

Because we have solar panels, and a separate solar charge controller (a Blue Sky Solar Boost 2000e). The Blue Sky charger can be programmed to output a range of voltage, so you can optimize it for your batteries. The factory default on that device is 14.0 volts (compared to 13.6 volts on the Intellipower), and that makes a huge difference. So when we were parked outside, our batteries were getting their last 27% of capacity courtesy of the sun and the Blue Sky—and I didn’t realize it until now.

At home, our Airstream lives under a carport, so the solar panels don’t produce any power. And, in colder temperatures, it takes a little more power to charge the batteries—about 0.5 volts more. (This is just a weird battery chemistry thing.) So after four months of sitting in the carport with slowly declining temperatures and inadequate voltage from the Intellipower, the battery slowly lost power and the solar panels weren’t there to save the day.

It’s possible the battery still is underperforming. I’m going to test it this next week when I take the trailer out of the carport and go camping for a week.  If I’m right, a full charge should be possible in the sunshine, and then I can “equalize” the battery using the solar charge controller (which goes to 15.2 volts in equalization mode), and exercise it through a few charge/discharge cycles.

I may also adjust the BlueSky charger for slightly more output voltage. I’ll have to do that after the battery has reached full charge. It may already be at an optimal setting, but since I don’t know, it will be a good exercise to check it once we have full sun.

If the battery is fine and it comes to a full charge next week using solar power, I’ll have to start looking for a better converter/charger. It’s a bummer to have to replace that unit again, but with the right unit in place the battery should charge faster when plugged into AC power—and most importantly maintain its state of charge all winter long.