Battery Gassing?

[twosouls2gether]

Hey guy, not sure if this is the right forum or if it should be in the mechanical forum... but away. I built a small flashing led collar for my father-in-law's (someday) dog, and while it's not the brightest it will certainly let you see him at night (since he's a black lab.) However, today after I came over I noticed that the collar was blinking like crazy, not the slow once every 2 seconds or so. More like 2hz, more or less. I took it off him and unscrewed the pvc pipe to see what was going on and it was flashing just fine afterwards. Well, now we went to get our pictures taken and came back and again it was flashing quickly. I unscrewed the pipe again just a little this time and it returned to normal. Previously I had this circuit under my bed flashing for maybe 2 weeks and it always stayed the same rate. I know that the temperature can effect components but just unscrewing it shouwldn't have much to do with it. I think maybe the lithium 123 battery is out gassing as it's being used, however according to the circuit spec a small 150mah coin cell should blink for 6 months, and uses about 200uA of current so would it really be out gassing? I think it has something to do with pressure, but how would pressure affect a lithium battery? The circuit is very simply, 2 transistors, 2 caps, couple of resistors. I do need this to be completely waterproof since this dog is half beaver, loves to eat wood and if it's wet he's in it! Should I try some kind of pressure check valve? A catalyst pellet? Anywhere to find small valves or pellets? Maybe I'm WAY off? I can provide a schematic if anyone thinks that has anything to do with it, I don't have the link on me over at the inlaws. Any help would be appreciated.

 

[IcantC]

Hey guy, not sure if this is the right forum or if it should be in the mechanical forum... but away. I built a small flashing led collar for my father-in-law's (someday) dog, and while it's not the brightest it will certainly let you see him at night (since he's a black lab.) However, today after I came over I noticed that the collar was blinking like crazy, not the slow once every 2 seconds or so. More like 2hz, more or less. I took it off him and unscrewed the pvc pipe to see what was going on and it was flashing just fine afterwards. Well, now we went to get our pictures taken and came back and again it was flashing quickly. I unscrewed the pipe again just a little this time and it returned to normal. Previously I had this circuit under my bed flashing for maybe 2 weeks and it always stayed the same rate. I know that the temperature can effect components but just unscrewing it shouwldn't have much to do with it. I think maybe the lithium 123 battery is out gassing as it's being used, however according to the circuit spec a small 150mah coin cell should blink for 6 months, and uses about 200uA of current so would it really be out gassing? I think it has something to do with pressure, but how would pressure affect a lithium battery? The circuit is very simply, 2 transistors, 2 caps, couple of resistors. I do need this to be completely waterproof since this dog is half beaver, loves to eat wood and if it's wet he's in it! Should I try some kind of pressure check valve? A catalyst pellet? Anywhere to find small valves or pellets? Maybe I'm WAY off? I can provide a schematic if anyone thinks that has anything to do with it, I don't have the link on me over at the inlaws. Any help would be appreciated.

I can't help with schematics, but if you think the battery is gassing or can explode please remove device from the dog .

 

[Mr Happy]

Seems highly odd. Hard to imagine pressure having a big effect on a circuit, and hard to imagine a coin cell outgassing enough to affect the pressure.

The blink frequency would change with variation in capacitor or resistor values. Have you considered condensation? Condensed moisture could divert current and mess the circuit up, especially if it is only running on microamps to start with. I'd try putting a silica gel package inside the enclosure with the circuit, or leave it open in a warm dry place for some time before sealing it.

Edit: Or perhaps you could spray the whole circuit with clear waterproof resin (a non-conductive one designed for circuit boards) to seal it?

 

[65535]

Most cells produce little gas if any, I doubt that is your problem. Maybe some pictures or something, it's possible that weight is shifting and putting pressure on a lead shorting it out and causing the rate to change.

 

[twosouls2gether]

Seems highly odd. Hard to imagine pressure having a big effect on a circuit, and hard to imagine a coin cell outgassing enough to affect the pressure.

The blink frequency would change with variation in capacitor or resistor values. Have you considered condensation? Condensed moisture could divert current and mess the circuit up, especially if it is only running on microamps to start with. I'd try putting a silica gel package inside the enclosure with the circuit, or leave it open in a warm dry place for some time before sealing it.

Edit: Or perhaps you could spray the whole circuit with clear waterproof resin (a non-conductive one designed for circuit boards) to seal it?

It is a primary 123 cell, not a coin cell. The whole circuit was done "free" style, or point-to-point and covered in hot glue (other option was silicone) and it certainly didn't seem to affect it in the least. If I unscrew it slightly then it works. It is a little snug in there, but I count that as a good thing so it doesn't rattle around. I unscrewed it and let is sit a bit, still has that PVC cement smell inside, perhaps that is having some effect? If I remember correct battery voltage does influence the frequency, but it would have to go way over 3v to produce that kind of quick flashing. You could say consensation, but then how would slightly unscrewing it do anything? I could put some silica in there I suppose, wouldn't hurt anything. Also I open it, close it right back up tight, nice slow flashing. Maybe the dog is screwing with me??? As for freq and temp, well when I open and close it, it's outside at the time. I have everything sealed pretty well. Maybe I'm doing something to the capacitors, as I would think they are the basic timing mechanism.:shrug:

 

[twosouls2gether]

Most cells produce little gas if any, I doubt that is your problem. Maybe some pictures or something, it's possible that weight is shifting and putting pressure on a lead shorting it out and causing the rate to change.

Well, that's what I thought, but it is encased in hot glue, and it's a tight squeeze in there so there should be no settling or shifting really. I brought it into the house to play with it, and opened it and poked and pushed and all that fun stuff, but no change. When I glued it I made sure it was working, that I saw nothing touching and good clearance, and then slowwly added glue and let it flow in so I shouldn't have changed much in there during that process. I am sure hot glue isn't the ideal insulator, but if I have to I can make another one for something like $5 and some time. I heard that some companies place catalyst pellets (such as pretzl) in water tight enclosures to reduce the pressure/hydrogen inside, thoought maybe something was occuring that would change or shift the battery voltage. I'm kind of stumped.

 

[Mr Happy]

So you soldered the circuit together, dropped it inside a plastic tube and filled the tube with hot glue? That would seem to protect it from pressure, gases and moisture. In fact, it sounds almost bomb-proof.

I don't know anything about potting circuits in hot glue, but I understand it is typically some thermosetting plastic like polyethylene with additives, so I imagine it would be a good insulator. Did you use low temperature glue? High temperature glue could have fried something, perhaps?

Really you have a mystery. Unless you can reproduce the change of flashing rate while the device is in your hands, I don't know what you could do.

 

[twosouls2gether]

So you soldered the circuit together, dropped it inside a plastic tube and filled the tube with hot glue? That would seem to protect it from pressure, gases and moisture. In fact, it sounds almost bomb-proof.

I don't know anything about potting circuits in hot glue, but I understand it is typically some thermosetting plastic like polyethylene with additives, so I imagine it would be a good insulator. Did you use low temperature glue? High temperature glue could have fried something, perhaps?

Really you have a mystery. Unless you can reproduce the change of flashing rate while the device is in your hands, I don't know what you could do.

Well, actually, it's some PVC plumbing parts from Home Depot. I ground the little flanges off one end to stick the pipe cap on and cemented it on, then wrapped teflon tap on the threads and put the other pipe cap on. Zip tied it to a collar and then placed the leds on it, running the wires under the collar with hotglue every so often, hot glued the led solder sections and all. I took the small CR123 in the battery holder, soldered all the components (8 if you don't count the 4 led's on the collar and the battery) to the battery holder, then hot glued a nice big blob all over the soldered jumble on the end. Bascily the only thing exposed is the battery (in case it runs down quicker then 3 years by my calculations). It is high temp glue as that's all I had. Now reading the package it says temps at the nozzle may reach 300 some odd degrees (doh!). However it all worked here at home just fine... and I had the damn thing running under my bed for a couple days.

Bomb proof is what I am looking for, this dog has been hit by a car, swallowed a fish hook (still there!) and caught in a snare for several days (hence trying to make a silly blinking collar to help find him). He routinly is running through the woods, chews wood and loves water like a beaver. Very active dog, needs to be tough.

BTW, just monkeyed with the circuit in spice and it seems that one capacitor is responsible for the time delay between current pulses.Hot glue problem?:shrug:

 

[65535]

Could be a faulty Cap. It is not at all uncommon to find faulty electronics components on the end user market.

 

[DM51]

Bomb proof is what I am looking for, this dog has been hit by a car, swallowed a fish hook (still there!) and caught in a snare for several days (hence trying to make a silly blinking collar to help find him). He routinly is running through the woods, chews wood and loves water like a beaver. Very active dog, needs to be tough.

Your father-in-law's dog sounds like a real survivor, maybe a bit on the dumb side, but a good ol' dog.

I assume you want to stay on good terms with your father-in-law. I'm not sure strapping a CR123A-size firework round his dog's neck is the best way to guarantee that, lol.

 

[twosouls2gether]

Your father-in-law's dog sounds like a real survivor, maybe a bit on the dumb side, but a good ol' dog.

I assume you want to stay on good terms with your father-in-law. I'm not sure strapping a CR123A-size firework round his dog's neck is the best way to guarantee that, lol.

Yeah, that's what he said! But hey, what a way to go! I don't think I've ever met a REAL smart dog, and I don't think he's all that stupid. Just being a dog, out in the woods wandering around. Got snared up by some dumb A$$'s trap. He finally found him when he parked the four wheeler at a gate and walked in a bit calling his name. He perked up and happy ending. Sad thing is that he said he saw some tire tracks I think, recently too... so someone came down that way and saw the dog but left him to die.

Anyway, from what I understand lithium primaries don't have all that a good of a chance to just explode. We actually found Shadows "sister" in the woods recently after being missing for 4 days. 15 yo shitzu, not all that far from the house. Not in good shape and had to put her down. Had she had a collar that lit up the night might have found her. Also they live right on a highway, so I think the chances of the dog dying by lithium aren't as good as other ways he could go. That being said, if anyone knows of batteries that will survive the cold (-20F sometimes) here in minnesota and keep giving power for a long time let me know.

 

[DM51]

I have the perfect answer for you - what you need is a Glo-Toob. It uses a single CR123A lithium primary (best for very cold weather) is tough as hell, totally waterproof, attachment ring for dog-collar, has a strobe capability, etc etc.

You will get BIG-TIME approval ratings from your father-in-law if you fit his dog with one of these.

 

[twosouls2gether]

I have the perfect answer for you - what you need is a Glo-Toob. It uses a single CR123A lithium primary (best for very cold weather) is tough as hell, totally waterproof, attachment ring for dog-collar, has a strobe capability, etc etc.

You will get BIG-TIME approval ratings from your father-in-law if you fit his dog with one of these.

Sorry DM, already have one on him. I'm looking for one that you don't have to worry about turning on and off, and that will last a long time. The glo tube advertises only 100 hours of run time, and you never know when the dog will wander off. Also it's a $40 dog light! Though he did have it mounted on his back with some kind of a harness. The dogs head went through and then it wrapped about his back and then another loop around his chest. Similar to something I guess you'd see for seeing-eye dogs. Zip-tied to the back it is very visible. I've already given him a Sam's HID and a Gerber Infinity, so wouldn't be keen on another $40. And with my luck the dumb dog would wander off and die, and then I'd be out a glo tube!!!! This other one would be so nice and cheap, and work well I think, but that damn weird flash rate change. I will probably have to make a new one and probably mount it on his back like the glo tube, seems that is the best spot so far. Only concern is if he's laying down then it's not visible from his stomach side. (Then again I suppose short of putting him in a giant lighted hampster ball there will always be draw backs/blind spots).

 

[DM51]

LOL!!

Does the Glo-Toob only last 100 hrs on the slow strobe setting? I think you'll find it is much longer than that, although I haven't seen any review where they've timed it. It has slow and fast strobe. Strobe would be easier to find than constant on, as the flashing is more noticeable – like the unit you have built yourself.

Going back to your unit, do you think it could be heat, rather than gas, that is affecting the frequency? If you have it totally contained in plastic parts, the plastic will act as insulation and the circuit and the LED are going to heat up. This heat could be affecting the electronics of the flasher circuit. Was the cell warm or hot when you took it out? I hope not actually hot, because that would be dangerous with a Li cell, but a bit of extra heat might be enough to affect the flasher circuit.

Let's have a pic of this dog!

 

[twosouls2gether]

LOL!!

Does the Glo-Toob only last 100 hrs on the slow strobe setting? I think you'll find it is much longer than that, although I haven't seen any review where they've timed it. It has slow and fast strobe. Strobe would be easier to find than constant on, as the flashing is more noticeable – like the unit you have built yourself.

Going back to your unit, do you think it could be heat, rather than gas, that is affecting the frequency? If you have it totally contained in plastic parts, the plastic will act as insulation and the circuit and the LED are going to heat up. This heat could be affecting the electronics of the flasher circuit. Was the cell warm or hot when you took it out? I hope not actually hot, because that would be dangerous with a Li cell, but a bit of extra heat might be enough to affect the flasher circuit.

Let's have a pic of this dog!

Well, the ADVERTISED runtime is from 24-100 hrs depending on mode. Yes, there is a nice double beacon strobe that works wonders. As to heat, the whole circuit isn't much and I doubt that there is much heat. Basicly from what I gave when the capacitor that is trickle charging hits the right voltage it triggers a transistor that basicly connects the led's to ground for all of 2ms. Advertised average current draw is 30uA. Although I have left out the current limiting resistor to the led's since the circuit was designed for one single red led, not 4 white ones in parallel. It also advertises approx. 6 months of operation from a 150mA-HR cell. So a CR123A with what, 900-1000mAH or more should, maybe, last longer then the dog! Not to mention for the heat that it's pretty cold out at night, below freezing. Also the led's are NOT part of the waterproof unit. The led's are attached to the collar each at 90 degree intervals with wires running to them.

I will have to get a picture of Shadow, however the gf's mom is kind of moody for the last few months so neither of us is really all that excited to go out there.

 

[DM51]

Well as long as it doesn't overheat or blow up it should be OK, although a rapid blink rate might give the dog an epileptic seizure, lol. Have you tried running it off the dog? Could his death-defying activities be making the thing go haywire somehow?

Even at a fast blink rate, the cell should last quite a long time, which was what made me think of the Glo-Toob's slow strobe setting. You could try timing both devices.

My betting is they'll both last quite a long time, even your home-made one with the strobe running fast. Powernoodle has had his HDS B60 running in locator beacon mode for over a year now, and the smart money is on the cell dying of old age before they run out of juice.

 

[DM51]

I've just re-read my post #16 above, and there's room for clarification on one point.

When I asked:
"Have you tried running it off the dog?"

what I meant by that was:
"Have you tried running it when it was not actually attached to the dog?"

and not:
"Have you tried using dog-power, instead of a CR123A cell, to run it?"

Glad to be able to clear that one up; I hope it didn't cause any confusion, lol.

 

[twosouls2gether]

I've just re-read my post #16 above, and there's room for clarification on one point.

When I asked:
"Have you tried running it off the dog?"

what I meant by that was:
"Have you tried running it when it was not actually attached to the dog?"

and not:
"Have you tried using dog-power, instead of a CR123A cell, to run it?"

Glad to be able to clear that one up; I hope it didn't cause any confusion, lol.

Hmmm... a dog powered blinking light, something like a cow bell idea? I know they have those watches that charge up due to the persons wrist moving. Not sure the actual mechanics of it, something like a shake light maybe?

 

[twosouls2gether]

Well as long as it doesn't overheat or blow up it should be OK, although a rapid blink rate might give the dog an epileptic seizure, lol. Have you tried running it off the dog? Could his death-defying activities be making the thing go haywire somehow?

Even at a fast blink rate, the cell should last quite a long time, which was what made me think of the Glo-Toob's slow strobe setting. You could try timing both devices.

My betting is they'll both last quite a long time, even your home-made one with the strobe running fast. Powernoodle has had his HDS B60 running in locator beacon mode for over a year now, and the smart money is on the cell dying of old age before they run out of juice.

From what I understand about epileptic seizures is you have to be responsive to blinking lights and then you have to have just the right combo to set one off. You think maybe he has some kind of ghosts or etheral energy about him? Like a ghost whisperer kind of thing? Hmmm...

As for the blink rate speeding up, well this was after only 2 days, and then later after I slowed it down it came back up again as I was posting the first message here, so an hour? So in a year it will be going nuclear. It's just soooo weird. I might be making another one to mount on his back, seems the FIL seemed to like that location better when the glo-tube was mounted there. If I do I'm going to try mounting all the components on a small pcb and maybe use silicone to seal it, nice and cool, no hot temps. Is silicon caulking conductive? :shrug:

 

[DM51]

If you rule out heat and gas, although in my experience you unfortunately can't rule out either of those things with a dog, especially gas lol, then it must be the blink mechanism.

I don't know anything about electronics, but if that thing just keeps going faster and faster, maybe you have a faulty transistor, or the voltage/current is unsuitable for it, or something like that.

If the components are cheap, building another one might be the answer, using slightly different components. Silicone caulking (I assume you mean the stuff you use to seal round a bathtub or shower tray) should be an electrical insulator, not a conductor, so it should be OK.