Carbon monoxide (?) mystery

[tonywright]

Great info Michael and others. I would have to agree that the batteries now sound like the most likely source. Seems like installing a good venting system for when the batteries are being charged is a really good idea. Those gases can be pretty explosive, so the alarms warning you are doing so for a good reason!

It would explain why the alarms are no longer sounding, since the batteries by now are probably fully charged, and so are giving off less gas. They must have had a fairly low charge when they were first installed?

Tony

[Michael]

The following is the latest update on this carbon monoxide - propane - hydrogen from batteries or ??? mystery, from one of Hali's co-owners:

"I should have sent this last night, but I did record the following in the
log book. When I arrived at Hali last night - 6:15pm - the propane alarm was
sounding off every few minutes. I opened the hatches and the alarm stopped
within 5 - 10 minutes. I checked the [carbon monoxide] detector and it showed a reading of
52 but there was no audible alarm. I had a good look around Hali to ensure
that the propane tank valve was tightly closed, and that the propane system
switch was in the off position - both affirmative. There was no smell of
propane in the cabin.
That is about all I can report other than it was a great night for a sail."

[Note by Michael.  As perhaps noted previously, the CO detector user manual indicates that it is a time-sampling device.  Apparently, the alarm will sound when certain concentrations persist for certain time periods.  I speculate that in this case a sufficient level of CO (or whatever is activating the alarm) had not persisted for long enough to trip the audible alarm.]

[Jon Schneider]

It's really sounding like the battery charging is the culprit, but just for S&Gs, try taking the propane tank off the boat to see if the same event occurs after she's sealed up again.  Also, your LPG solenoid switch is off, right?  I suppose it's possible that even with the tank valve tightly closed, there's a leak in it, and if the solenoid is routinely left on (but why would anyone do that?), then perhaps the tank is still in the running as a the culprit.  BTW, you might want to check that your charger is working properly (i.e., putting out the right voltage, adjusting to the right stage, etc.), and you might want to check to see how the fluid level is in your new batteries.  Charging enough to cause these sensors to trigger sounds pretty high to me.  Do you leave the fridge or any other power-hungry device running while you're away from the boat?  If so, the second test you might want to run (if the charger checks out) would be to turn those appliances off; in fact, turn the charger off and don't plug in.  If that solves the problem, you'll have discovered the culprit.  BTW, thanks for an incredibly detailed and interesting write-up of the problem and situation. 

[Michael]

Thank you for your suggestions, Jon.  We will try them on Hali.

According to our standard procedures, when we leave Hali, we leave shore power connected, battery charger on, all elective electrical devices (including 12 volt Adler Barbour refrigerator and propane solenoid) off, and dual house battery selector switch at off.  Therefore, the only electrical devices that will operate are the battery charger and the direct-wired bilge pump.

We are getting more water in Hali's bilge than I think proper.  This is another unsolved mystery.  Its relevance to this discussion is that the bilge pump consequently cycles on relatively frequently:  I would say about 6-12 times a day.  I suppose this regular draw on the batteries results in frequent charging of the batteries and, in our "battery gas is responsible" thesis, the regular production of hydrogen gas that activates the propane gas detector (and, it now seems possibe - see below - the CO detector).

As mentioned previously, a recent change aboard Hali was the replacement of the two old 4D batteries with two new (Interstate) 4D batteries.  I checked the fluid level in only one cell of the new batteries at the time (May 13) of installation.  Thanks to your suggestion, Jon, I will now check the fluid level in all the cells and realize I should have done that at the time of installation.

Thanks again.

AMENDMENT: Since writing the foregoing, I have learned that hydrogen gas will activate CO monitors.  At least, the fact that hydrogen gas will activate CO monitors is the basis for a patent application (http://www.patentstorm.us/patents/5959188/description.html) for a device to test CO monitors by generating activating samples of hydrogen at the CO monitor.  And the fact that hydrogen gas will activate CO monitors is mentioned in another patent application (http://www.freepatentsonline.com/6948352.html).  Both these patent applications have interesting things to say about domestic CO monitors of the type that I guess most of us use aboard our boats.  Assuming the patent applicants are right that hydrogen activates CO monitors (and I have no reason to think otherwise but also no understanding of the monitoring device or chemistry that would enable me to say one way or the other), and that, as Electro Systems Inc. has given as their "best guess", that the charging of the batteries aboard Hali is producing hydrogen that is activating the propane gas alarm, it seems we have some basis for saying that the charging of the new batteries on Hali could be causing both the propane gas and CO gas detectors on Hali to sound their alarms.  So one further useful thing to do will be to test for hydrogen gas on Hali.  I think it was Tony Wright who suggested above doing tests for specific gases.

[Ron Hill]

Michael : Off of the CO topic, but you should get some Gore Dripless pack or at least go to Radio Shack and get a counter that you can wire into the bilge pump and KNOW exactly how many times the pump is cycling for sure!!   A thought. 

[Jon Schneider]

Ron, do you know if it's possible to switch to the Gore packing while the boat's in the water?  Or at least remove one or two lines of current packing and replace them with Gore while you're in the water?

[dave davis]

It looks like you are getting lots of inputs and some very good ideas.
I would not rule out the propane possibility. I had a resent CO alarm in my house and it was due to a plumber igniting his propane torch so that he could solder the copper piping in the bath room. The alarm unite was about 15 feet from his work. The alarm was immediately set off as he was lighting his torch. I don't know if the alarm is sensitive to the raw propane or to the byproduct of the flame.  By taking the unit outside, the alarm stopped immediately.
You might also examine the valves and the tubing by using a bubble/soap method to look for leaks in and around your propane tank.
As was previously suggested, I have had the gas company come out to the house and they had a snifter to search for leaks at no charge.
Another approach you might try is to use the alarm unit as a detector.  Place it in the battery well while you are charging and or in the charger compartment while you a charging. It may act as a good snifter and will help isolate the source.
Good Luck, Dave

[Michael]

Regarding the propane/CO/hydrogen/?? issue:

1.  The CO monitor was reading in the 70s at the boat yesterday afternoon (before we removed the propane can, before we ran the engine, and before we aired out the boat). No combustion engine appeared to be or to have been operating in the vicinity. The CO monitor readings dropped to nil within about 10 minutes of our ventilating the boat.
2.  After our short cruise, we took the propane can off Hali as Jon suggested.
3.  Dave, thanks for your information.  We have never smelled propane but I will not rule any possibility out yet.

Ron - Thank you for the suggestions re: stuffing box and bilge pump. At the last change, we did not use Goretex dripless stuffing but can/should re-do the packing now that we have some on hand.  I had thought vaguely about getting a counter to measure how often the bilge pump cycles on.  Your suggestion has brought the issue to the fore.

[Stu Jackson]

1.  Check the hose clamp on the bilge pump hose connection.  Mine let go a while ago and the pump ran continuously for a week until it dropped the house bank below 10.5V.

2.  Why do you leave your boat plugged in?  What needs to be charged if everything is OFF?  Only thing I can think of is the bilge pump.

3.  Much very good advice on the issues and step-by-step checks on the CO stuff.

4.  ANY stuffing box can be done in the water.  See Projects and FAQs for descriptions and discussions.

[Michael]

Stu, thank you.

1.  Your point ('why leave the boat plugged in') indicates a good next-step test (after we find out whether the alarms still go off with the propane can off the boat), namely, to unplug Hali from shore power (or at least turn the mains battery charger switch off) so that no battery charging takes place when we are not at Hali.  It seems so obvious...after you mention it.  (Is it your view that the boat should generally not be plugged into shore power?) 

2.  We changed the bilge pump line and put on new 316 ss hose clamps late last summer.  The pump seems to discharge a good deal of water overboard.  But I will check the hose clamps anyway.

3.  To second your comment about changing the stuffing when the boat is in the water, we certainly have found it easy (after reading the great posts on this site about doing it).  Nonetheless, three or four of us were on hand, armed with rags and screwed-up courage on the first occasion. 

Amendment: Since writing the foregoing, a co-owner has emailed:

"We were out on Hali tonight for a short sail and drift. The propane alarm
was sounding when D. arrived - the alarm cleared once she opened the
companionway. My brother, E., a heavy duty mechanic in his previous life,
was with us again tonight. He and I had talked about our alarm situation
when we were out last week. When I told him about the hydrogen theory, he
thought that was a distinct possibility but also thought that it might
manifest itself in a warm to hot battery (and it there was a hot one we
should get it out of there because batteries can explode). We checked the
new batteries. One cell of one of the batteries seemed a little warmer than
the others but it was almost imperceptible. No immediate danger there. We
did not think to check the starting battery. When we were leaving, I noticed
that you had removed the propane tank to eliminate that possibility - E
suggested we might disconnect the battery charger for 24 - 48 hours and
check daily to see if the alarm was still going off.

I will call the battery supplier tomorrow and ask if they have a way to
determine if a battery is giving off hydrogen gas but disconnecting the
battery charger might be worth a try. What do you think?"

And I replied:

"Great report.  Great idea to disconnect the battery charger
or leave shore power off.  Co-incidentally, Stu Jackson, who I think
is an electrical engineer, suggested doing that in a post he made on
the "Carbon monoxide (???)...." thread on the c34.org website today.
Clearly, great brains think alike!  After reading Stu's post, I was
tempted to call you to find out how things had gone tonight...but
desisted because of the late hour.  Then N. (a Blackberry addict)
came home and said, L. has sent an email about Hali.

Now that we have had alarms sounding when no propane is on the boat
(or almost no propane...I guess there might be a bit left in the
lines), it seems that we can discount the possibility that propane
aboard Hali is triggering the alarms. (I suppose we can't yet discount
the possibility that an activating gas of some sort is reaching Hali
from elsewhere.)

So I agree that as a next step we should shut off the battery charger
(or unplug the shore power)."

It will be interesting to learn what the battery manufacturer has to
say. "








 

[Jon Schneider]

Michael, I'm sure you realize this, but simply unplugging the shorepower charger won't relieve you of fixing the problem.  Someone will need that charger after a day or two out upon arrival at transient marina, but it would certainly be unsafe, despite the alarms, to plug in.  If it turns out to be the charging process, it needs to be fixed.  It can't be good for the batteries, and it sure can't be good for people sleeping aboard even if the levels are slightly below the alarm trip. 

[tonywright]

Of course if your bilge pump is cycling as much as you think it could be, then you would not want to leave the boat for too long with the charger disconnected (as someone said, more boats sink at the dock than at sea).

On Vagabond I generally have a dry bilge. I don't recall there ever being enough water in it to trigger the bilge pump (other than when I drain the HW tank). If I find any water in it, I mop it out, since it is too low for the bilge pump to suck out. As a result, when I leave the boat I make sure than all thru hulls are closed, and turn all electricals off. I only charge every couple of weeks, or after a weekend trip.

Michael, I don't recall whether you have a solar powered vent installed on Hali. If not, have you thought of putting one on? ( I have the vent, but haven't got around to installing mine yet: I'm still screwing up the courage to drill a 4" hole in the hatch!). Obviously you need to find the source of the gas and fix it. But having a vent continuously expel it in the mean time seems like a good idea?

Tony

[Stu Jackson]

Is it your view that the boat should generally not be plugged into shore power?

Michael, In all of our years with Aquavite and in earlier boats [celebrating our 25th year of sailing on SF Bay this past weekend], we never left our boats plugged in, and none of 'em sank!  The reason is basic:  we just do not know how poorly wired our neighbors' boats may be.  Stray electrical can come from a bad dock wiring design or installation [not in our marina, though] or from poorly wired other boats even to a good marina system.  We intend to have the zincs continue to do their work and not spend money on a new shaft or propeller if we can help it.

That said, I've learned a lot recently from John Nixon, a frequent contributor to this Message Board.  Once I [finally] installed our Link 2000, I learned two very important things:  1) my "energy budget" done in my head for the past ten years was pretty much right spot on (see: http://c34.org/bbs/index.php/topic,3976.0.html); 2) my estimate of input TO the house bank FROM the various charging sources was NOT very good because I hadn't taken into account the acceptance rate of the batteries and the tapering off of the charging amperage even with a smart regulator and a smart three stage Freedom 15 inverter charger.

In discussions by email with John, as a result of what I term "Fridge Follies," wherein I was doing detective work on why our fridge would not restart during the second night at anchor, I learned two things:  1)  equalize your batteries regularly to avoid diminished performance [assuming your batteries are wet cells and your manufacturer says it's OK to do so]; 2)  John provided this: "Leaving lead acid batteries on float charge 24/7 will reduce the useful life of the batteries by as much as 35 to 40% compared to cycle charging, which by definition is to charge them only when they need recharging, then leave them alone. The continuous float charge results in accelerated positive grid corrosion, which has the effect of turning the positive plates in the battery into mush: a soft, expanded material that produces increased internal resistance and decreased deep cycle capacity. I only use cycle charging on my boats now, but I will leave them overnight or maybe for a day or 2 with the charger on just to make sure that they get fully recharged." 

I noted that in order to do so, I'd start a program of plugging into shorepower to top off the batteries once I got back to the slip before I left.  He added: "I doubt that an extra 30 minutes with the charger on when you get back to the dock is going to get your batteries fully recharged: that last 5% or so takes a long time, even with the very low internal resistance of AGM batteries like I have on our boat. If you are adverse to leaving the charger on all the time when you are away from the boat ( which I support completely, but not for the reason you might think...), a good simple way to give the charger time to do its work is to put it on a cheap/inexpensive 24 hr time switch with the ON peg pulled out. Just set it to run for 24 hours when you leave the boat: manually start it with the timer set just after the OFF pin, and it will dutifully turn itself off after a good 24 hour charge. In most cases, that will generally make sure that you have reached that 100% charge that the batteries need to not begin loosing capacity due to sulfation. 

So, for many reasons, it's important to both keep your batteries fully charged, avoid keeping them plugged in 24/7 on float, and take care of them.  I am constantly amazed at the number of plugged in boats that NEVER GET USED.  Whenever the owner comes down to the boat, there are going to be a bunch of things that don't work, most importantly their electrical systems.

The fridge works much better now after the equalization, which I intend to do more regularly now that I know.  The voltage used to drop to 12V whenever the fridge started running.  After the equalization the voltage remains high and only slowly tapers off after fridge run time and during the off cycle of the fridge, the voltage actually rises a bit.

Live and learn.

[Kyle Ewing]

Regarding partially discharged batteries and leaving the battery charger on when the boat is unattended, how do people handle leaving the boat after discharging the battery (i.e. after an overnight cruise with little motoring) then closing up the boat for the weekend? 

I dislike leaving partially discharged batteries and dislike more leaving power to the boat, especially if I might be away for a week or two.  I leave the charger on while unloading and closing up but know the batteries need more than an hour of charging.

[Michael]

1.  Tony, like you, I would be apprehensive about cutting a hole in the hatch for ventilation.  In any event, I think we are agreed that the root cause of the alarms' being triggered needs to be found and eliminated.

2.   The first step in the current plan is to continue to leave the propane tank off Hali, blow the propane lines clear, vent then button up Hali, and leave the battery charger on.  If the alarms still sound, we can discount the theory that onboard propane gas is triggering Hali's propane and CO alarms.

3.    The second step will be to shut off the battery charger, check the alarms for a couple of days, then turn the battery charger on.  If the alarms don't sound when the battery charger is off but do sound soon after it is re-started, we can (I think) be reasonably confident that gas produced during battery charging is causing the alarms to sound.

4.    From Stu and Ron's posts, I take that a cause may lie deeper: e.g., the packing in the stuffing box is worn out; too much water is getting through the stuffing box; the excess water from the stuffing box is causing the bilge pump to cycle on frequently, which draws down the batteries; the drawn down batteries are requiring frequent charging; the frequent charging produces hydrogen gas; the hydrogen gas triggers the propane and CO alarms.  If some such chain of cause and effect is at play, it will, interestingly, have been an alarm about a gas leak that led to our dealing with a water leak.  (Of course, we should have dealt with the water leak in its own right.  Also, I suppose that if the water-leak-triggers-alarms theory proves true, just because we fix the water leak won't mean we have solved the gas-from-charging-batteries issue as the batteries could need charging for other reasons at other times and would then still produce hydrogen gas at potentially dangerous levels.)

5.   Stu, I am afraid that the details of your good long post on batteries and charging will be over my head until I read a book on batteries (again!).  But I suspect the post will be of interest generally.  Would you post it as the start of a new thread so that it is not wasted in this carbon monoxide desert? If you do that, I would be pleased to post some follow up questions to you about what you have posted.

Sorry about subjecting readers of this thread to my rather tortuous "thinking aloud" about Hali's alarms problem...but your input in response is really helpful.