LaCrosse BC-9009 / BC-900 - The Melt-Downs Continue...

[Mr Happy]

That is also not uncommon for new, non-LSD cells to fail to terminate the first few cycles when using too-low a charge rate. From everything in your description of your incident, what you experienced had nothing to do with the charger melt-downs, and was quite normal when chagring new, non-LSD cells at too-low a current.

This is all true, but the one piece of data that doesn't fit is that "the plastic coating had greatly softened. and it burned my finger when removing it from the charging bay". I have never seen an AAA cell get that hot when being charged at 200 mA, no matter how long it is left on charge. This seems to suggest a charge current greater than 200 mA was being applied, which is not inconsistent with previously observed fault conditions in the BC-900.

 

[FlashPilot]

That is interesting. The charger was definitely set at its default setting of 200 mA at the time this happened. Three of the batteries finished within 15 minutes of each other. The overheating one went on for another hour before I became concerned and decided to remove it. That's when I noticed that it was overheating. Is it possible that the failure is intermittent with a repeat looming in the future? I now only use eneloops, but touch them frequently as I babysit the damn thing.

 

[Turbo DV8]

I have never seen an AAA cell get that hot when being charged at 200 mA, no matter how long it is left on charge. This seems to suggest a charge current greater than 200 mA was being applied, which is not inconsistent with previously observed fault conditions in the BC-900.

He stated the other three slots terminated fine. As far as I understand the failures, when the charger fails it's not going to cause only one slot to runaway overheat, while the other three are under complete control.

 

[MarioJP]

That is interesting. The charger was definitely set at its default setting of 200 mA at the time this happened. Three of the batteries finished within 15 minutes of each other. The overheating one went on for another hour before I became concerned and decided to remove it. That's when I noticed that it was overheating. Is it possible that the failure is intermittent with a repeat looming in the future? I now only use eneloops, but touch them frequently as I babysit the damn thing.

I charge my AAA cells at 700mA they all terminate without overheating. Use these cells for my 16 led flashlight. I figure AAA cells don't have much capacity anyways so at least charging them faster can be a fair trade.

 

[uk_caver]

This is all true, but the one piece of data that doesn't fit is that "the plastic coating had greatly softened. and it burned my finger when removing it from the charging bay". I have never seen an AAA cell get that hot when being charged at 200 mA, no matter how long it is left on charge. This seems to suggest a charge current greater than 200 mA was being applied, which is not inconsistent with previously observed fault conditions in the BC-900.

It would seem odd for a cell to get finger-burningly hot when only having ~300mW of energy input.

 

[uk_caver]

He stated the other three slots terminated fine. As far as I understand the failures, when the charger fails it's not going to cause only one slot to runaway overheat, while the other three are under complete control.

Depending how the failures actually happen, I guess it might be possible that charging cell->overcharging cell is more likely than charged cell->overcharging cell - a fault might stop a PWM drive on high, or simply have active FETs latching on, while not affecting cells not being charged.

If fault timing purely random, if cells were generally charged with ones in a similar charge state, there may be much more time for a fault to randomly happen in between 'start of charge' and 'first cell charged' than between 'first cell charged' and 'all cells charged'.

 

[ValVe]

Hi guys!

I've just nearly experienced a meltdown on a BC-9009 with a "IV" power supply. Third button became kinda soft to the touch ;-)

So i've dismantled the unit to look what's inside.

The tiny TSOP6 mosfets are definitely the cause, as they're heating up pretty badly, and 2nd and 3rd are almost below the buttons, which are made of crappy plastic with a low melting point.

So I've made a small aluminium heatsink just wide enough to fit between the jutting soldered parts, to help dissipate heat more evenly. It's now glued to the pcb using a couple drops of epoxy, and I've put a generous amount of thermal compound to the mosfets.

And guess what? The aluminium plate is also running hot at ~70` Celsius during a 1 Amp charge of 4 AA cells. I wonder how it even managed to work before.

Me wants some weed the engineers were smoking while designing this crap.

 

[Bones]

Hi ValVe,

Welcome to CPF, and many thanks for the details on heat-sinking your BC-9009. We eat this stuff up around here.

Considering the flaws you and others have noted in its design, it will be interesting to see whether the replacement power supply will actually bring the temperatures down enough to eliminate the melt-downs.

 

[ValVe]

Well, I've been testing the heatsinked unit extensively for several hours now, with the the PSU at 3.2 VDC.

So far, the buttons have not softened up, and the heatsink is not going above 75C. Without fans or forced airflow or anything, just sitting fully assembled on a metal sheet.

If the buttons do melt, however, I'm thinking about modding them too - maybe molding something more substantial with quicksteel or something like that to replace the lower part.

As far as experience shows, heatsinked mosfets are now very unlikely to go into thermal runaway, if they had somehow done that before.

I've also applied much more grease to the NTCs, as there was almost none before that. Something tells me that may have caused at least half of the critical termination failures.

BTW, no free replacement PSU's here in Russia, alas :-(

 

[Mr Happy]

So I've made a small aluminium heatsink just wide enough to fit between the jutting soldered parts, to help dissipate heat more evenly. It's now glued to the pcb using a couple drops of epoxy, and I've put a generous amount of thermal compound to the mosfets.

If you can post a photo of this modification the details would be nice to see.

 

[MarioJP]

Any of you guys have an idea what is going to happen next with this charger. Or no more La crosse??

 

[FlashPilot]

Hi guys!

I've just nearly experienced a meltdown on a BC-9009 with a "IV" power supply. Third button became kinda soft to the touch ;-)

So i've dismantled the unit to look what's inside.

The tiny TSOP6 mosfets are definitely the cause, as they're heating up pretty badly, and 2nd and 3rd are almost below the buttons, which are made of crappy plastic with a low melting point.

So I've made a small aluminium heatsink just wide enough to fit between the jutting soldered parts, to help dissipate heat more evenly. It's now glued to the pcb using a couple drops of epoxy, and I've put a generous amount of thermal compound to the mosfets.

And guess what? The aluminium plate is also running hot at ~70` Celsius during a 1 Amp charge of 4 AA cells. I wonder how it even managed to work before.

Me wants some weed the engineers were smoking while designing this crap.

Can you please post a few pics of your heatsinks? I may do the same thing with mine.

Agreed... this thing must have been engineered by complete morons.

Thanks.

 

[ValVe]

Here it is...

(The grease is not conductive)

The four studs on case' bottom part happened to be just where they are needed to press this thing to PCB.

Maybe they actually were planning to fit something like that in here? But... uh... forgot? ;-)

 

[Conan]

Any of you guys have an idea what is going to happen next with this charger. Or no more La crosse??

Why don't you email La Crosse instead of asking this question repeatedly to forum members who have no control over the future of La Crosse?

 

[SilverFox]

Hello FlashPilot,

Welcome to CPF.

I don't think complete morons were involved in the engineering... I think it may have been really cheap morons... :devil:

Tom

 

[FlashPilot]

Here it is...

(The grease is not conductive)

The four studs on case' bottom part happened to be just where they are needed to press this thing to PCB.

Maybe they actually were planning to fit something like that in here? But... uh... forgot? ;-)

Good job Valve! :thumbsup:

First off, let me begin by apologizing for the long post.

While babysitting my potential fire starting little charger, I was online looking at all the heat sinking options available for cooling small electronic packages. For about $3 per unit production costs, La Crosse could have protected the mosfets and the clusters of chips just below them from heat buildup.

In trying to determine what gets hot and what stays cool, I removed the back cover of the charger and proceeded to recharge 4 AA's at 1000 mA. After about 3 minutes of charging, using my pinky finger, I carefully touched each component to find out where all the heat was generated. I was surprised to find that most of the heat was coming from the clusters of chips just below the mosfets (they are typically cluster of 6 or so, just below each of the 3 mosfets). While the mosfets were warm, they were not as warm as the clusters below them.

Valve, after looking at your pic and how you devised the shape of your heatsink, I was curious as to why you didn't cover all the small chips below the mosfets. Maybe it was a quick build to ease some of the thermal load. Either way, I commend you for your efforts. I returned to my pinky finger tests and checked all the smaller chips to see if there was a difference in heat (especially the ones you left unprotected) and they were all about the same heat in the cluster. After messing around with a saw, dremel and some 3/16 aircraft aluminum, I came to the conclusion that there was no way to use a one piece heat sync to do the job sufficiently. On my charger example, all the chips were of a slightly uneven height. Because of this, there would be to large of uneven gaps of heat sync paste between some of the chips and the aluminum heat sync for my liking. For the paste to work correctly, there should be very little used between the two mating surfaces. Not wanting to create more of a problem, I decided to rethink this. If I was smart, I would have tossed the whole mess into the garbage... where it belongs. Heh!

In the mean time, I tested at 700 mA and 500 mA to see if less heat would be generated at lower currents. At 700 mA, there was slightly less heat but it was still uncomfortable to touch. At 500 mA (where I normally charge AA's) it was warm but not uncomfortable to touch. I then returned to 1000 mA and flipped the charger right side up and left for 5 minutes. When I returned, I could smell plastic melting so I quickly confirmed proper charging voltage and then immediately terminated the charge. Feeling around the board, The mosfets were very hot and the smaller clusters of chips just below them were really very hot! I couldn't find any soft or melted plastic so maybe the smell was from just breaking in from that higher temperature... again, I never charge at 1000 mA, and that was a first. Its also interesting to note that until this time, all the other tests had been conducted with the charger on its back with the cover off, and with the mosfets and smaller chips pointing skyward (heat rises... hmm). Messing around a bit more, I found out that charging with the unit upside down has a huge impact on heat containment and absorption, even with the bottom case removed. No wonder this damn thing gets so hot! The heated components are mounted to the bottom of the board, and with very little ventilation to allow the heat to escape through the top. While there are vents on the side and bottom, they don't seem to be very effective in evacuating the rising heat. It just sits there and cooks itself. :sick2:

A few thoughts for better cooling:

1) Get a sledge hammer and pound this POS into oblivion, then mail it back to the morons that ignored all the problems its customer complained about with this ill fated design.

OR

2) Get a small sealed metal equipment box (4" x 4" x 6" should do nicely). Leave the bottom cover of the charger off and use it as a template to cut a hole in the top of the equipment box. We are going to cut that hole the same size as the bottom charger plate. Look how the case fits in a groove on the bottom of the charger plate and it will become apparent what you will need to do. If it's the right size, we can drop the charger onto the top of the equipment box hole we just cut and screw it in place using the existing hardware that came with the charger. Just drill the holes in the equipment box correctly and take your time lining things up and trimming metal where it doesn't fit. This box will now become the new bottom for the charger. Now get a small PC style fan and cut a hole in one side of the equipment box to mount the fan. We want the air flow from the fan to push into the box for forced positive pressure. In effect, we just created a positive pressure forced air sealed plenum for this trashy little charger. Looking at the board inside the charger, it seems reasonable that the forced air flow should do a fairly good job of cooling both sides of the board as well as some spill for the batteries. This seems like a far better option than simply keeping a house fan blowing on the unit while its charging, because a house fan only cools the batteries and the top plastic housing... not what's inside.

There are probably many other things to consider. I read in other threads that this charger probably needs to sense the batteries getting warm at the end of the charge cycle to terminate properly. Because of this, a fan blowing from across the charger was discouraged because it would keep the batteries to cool. The heat sensors are mounted on small towers that sense the heat of the batteries while charging. Also, if the batteries get to hot, the unit is supposed to suspend charging until the batteries cool before it resumes charging. These sensors were bedded with a very small amount of thermal paste to facilitate heat transfer. If there were air moving past the sensors, this might defeat their effectiveness in being able to detect an overheating condition. From the way these sensors are mounted, it looks possible to shield them with a small dose of thermal grease where they make contact to the battery contact, and then coat the rest of the tower stanchion with electrolytic grease to shield it from the cooling breeze blowing from the fan. Maybe a piece of heat shrink tubing would accomplish the same goal. The key to this entire mod is to keep everything reversible (because most other mods that others have tried on this piece of junk have failed miserably) and this unit is still under warranty - so we need to anticipate returning it to La Crosse after if blows up like a hand grenade.

If after the addition of this fancy new cooling system, this miserable POS still keeps overheating, its possible to use flat copper wire bent in tiny "U" shaped pieces, and attach them to each chip with minimal thermal glue to help assist in heat dissipation. These little pieces of wire would be directly in the plenums air stream and should prove most effective.

Maybe all this doesn't mean two hoots! I've also read that because of the inferior quality and inadequate sizing of the mosfets, they just might be like tiny obama surprise packages, destined to fail at any moment and send a runaway 3 volts with full wall-wart current directly to your unsuspecting batteries. This, regardless of heat or intended charging current.

Maha, will be saying "Mahahahaha!, you should have bought me!" There's another thread out there where a Maha engineer explains the pitfalls of the La Crosse design. Reading into it, calling it garbage would be putting it nicely... but what the heck.

If you read this far, Thank You for reading. I know its all very boring and hardly worth the effort to mod... but hey, its another fun project! Right? Its what we all live for. Uh-huh...

Pray to the "Gods of Chapter 11" that La Crosse becomes a bad memory in the very near future.

Comments on my ducted plenum cooling concoction would be appreciated.

 

[FlashPilot]

Hello FlashPilot,

Welcome to CPF.

I don't think complete morons were involved in the engineering... I think it may have been really cheap morons... :devil:

Tom

Thanks SilverFox,

My animosity towards La Crosse is due to their gross lack of support and denial of a potentially serious problem. May they burn down their own facility with one of these pieces of junk.

BTW... lovecpf
With flashlight in hand, its a great time to be alive!

 

[guggie]

FlashPilot:

Is your charger using the IV (recalled) wall wart power supply? What is its output voltage?

 

[VegasF6]

I just joined this thread so I am not up on it yet, I need to see which power supply is which. What a shame I love this charger:

Valve, neat job on your heatsink.

Flashpilot, are you talking about the smd resistors? They should be quite warm, they are resistors after all. If they are cool enough that you can keep your finger on them they should be fine.

 

[ValVe]

My measurements (with a pyrometer, not a pinky) show that the resistor clusters do not need additional heatsinks, never going above 55C in any mode.

But you can cover them too, if you want. Not that they need it badly.

However, they sit lower on the pcb than the mosfets, so you'll need to work something out, and they have conductive parts on top, unlike the mosfets, so it is unsafe to put a metal heatsink tightly onto them. You may short something out easily.

So far I've cycled 4 sets of AAs at 1A and 2 sets of AAAs at 0.7A on my unit, and no melting buttons/burning smells/black screens or anything.