Espritmodel.com Telemetry Radio
Reply
Thread Tools
Old Aug 13, 2004, 06:30 AM
Registered User
Aio_1's Avatar
Ireland, Kildare, Leixlip
Joined May 2002
6,441 Posts
Quote:
Originally Posted by tjcooper
Flyers,
....I have found that the metal tubes around the cells can really help control the fireball when a pack goes thermal. I hope many of you will consider using this method to transport and charge your packs. The firesafe is great for a few packs, but I can assure you that if one pack goes, they will all go.....one after the other. More than 5 packs in a firesafe could possibly cause the fireball to leak out of the safe. Only testing will tell for sure......
I'm not enthusiastic about storing cells in a container that could conceivably short them. Isn't this a danger with these tubes and the safes or do you have them lined with an insulator? One other concern would be the danger of confining a catastrophic failure. The failure mode of Lipos has one advantage over Ni** cells in that they don't explode. Is it possible that confining a flame-out in a closed box could cause the box to explode if the air inside was heated rapidly enough?

Aidan
Aio_1 is offline Find More Posts by Aio_1
Reply With Quote
Sign up now
to remove ads between posts
Old Aug 13, 2004, 09:22 AM
Tight is Right
Darren Hintze's Avatar
Lehi, Utah, United States
Joined Dec 2001
7,663 Posts
Aidan,

Good questions. About the shorting, I'm not sure about the tube but the Brinks box has a coating on it. Most of us use some kind of foam too. It does seem to me that a tube would be even better for storage because the contacts would be isolated from the surface.

In the tests done on the box I've read about, Lipos do not flame with enough intensity to rupture the box. I don't think most are is using a completely sealed box to charge anyway, wires must exit the box somehow (to the pack or to the charger). I don't think most have sealed this point.

It is important the "vent" areas be positioned to plan for flame exit.

Darren
Darren Hintze is offline Find More Posts by Darren Hintze
Reply With Quote
Old Aug 14, 2004, 12:59 AM
Registered User
Illinois
Joined Mar 2004
372 Posts
I think this thread should be required reading for Lipo users.

Keep up the good discussion.

I use a ceramic coated ammo can with foam inserts with slots for transport and storage. After that the Lipo I am gong to use is put in a pocket of my flight bag for the walk to the flight pits.
VBSuperMaxx is offline Find More Posts by VBSuperMaxx
Reply With Quote
Old Aug 14, 2004, 02:09 AM
Got shenpa?
flieslikeabeagle's Avatar
Los Angeles
Joined May 2004
10,940 Posts
It was interesting to see those pics of extremely puffed-up cells, and I agree, it looks as though LiPo cells might in fact puff out to three times their volume, and consequently float. Wouldn't have thunk it...

To get back to the question of burning under water, we're probably all aware that certain types of welding torches will happily burn under water; oxy-hydrogen, I believe, is one gas combination that will do this. So a hot enough flame will survive under water, since the water cannot conduct away heat fast enough to put out the flame. The remaining questions are, how hot does a burning LiPo cell get? And what *starts* the thermal runaway - after all, the innards of the cell are not exposed to atmospheric oxygen when the thermal runaway starts, so why shouldn't this happen under water as well as in air?

I just got my first LiPo's a couple of weeks ago, and am consequently more than a little interested in case-histories of LiPo failures.

One last thing, anyone have a link to a thread on the parallel charging/series discharging technique? In particular, what kind of connector are people using that has enough contacts to allow this flexibility? It would seem you'd need four contacts for a 2S, six for a 3S, eight for a 4S...or am I missing a clever circuit layout that does this with less than two independent pins per cell?

-Flieslikeabeagle
flieslikeabeagle is offline Find More Posts by flieslikeabeagle
Reply With Quote
Old Aug 14, 2004, 03:11 AM
Registered User
Joined Jan 2002
6,587 Posts
There's no demonstrated way yet to charge in parallel without disconnecting the series of cells, i.e., two wires are required for each cell or parallel array. Any polarized four-pole plug appropriate to the amperage will do for packs discharged 2s. For 3s packs, luc has found the Multiplex 6-pole plug to be good for 30 amps bursts in a 3s-discharge setup.

I have hopes that Quacker's dedicated 4.2 volt, variable amperage chargers, one for each cell or parallel array of cells, may be isolate the charging circuit enough that packs could be charged in parallel through center taps, without disconnecting the series arrays. We'll see.

- RD
RD Blakeslee is offline Find More Posts by RD Blakeslee
Reply With Quote
Old Aug 14, 2004, 10:15 AM
Registered User
Walled Lake, MI, USA
Joined Feb 2000
11,182 Posts
Quote:
Originally Posted by RD Blakeslee
I have hopes that Quacker's dedicated 4.2 volt, variable amperage chargers, one for each cell or parallel array of cells, may be isolate the charging circuit enough that packs could be charged in parallel through center taps, without disconnecting the series arrays. We'll see.
This has the potential to be the safest charging system to date. If it turns out to be so, RD, you deserve a lot of credit for your continuing efforts to raise awareness of this concept.
Dave Hederich is offline Find More Posts by Dave Hederich
Reply With Quote
Old Aug 14, 2004, 11:49 AM
Registered User
Joined Jan 2002
6,587 Posts
Thanks, Dave.

I wish we could get the involvement of a LiPo systems manufacturer. As you know, several of our leading LiPo suppliers deny the value of taps.

- RD
RD Blakeslee is offline Find More Posts by RD Blakeslee
Reply With Quote
Old Aug 14, 2004, 12:14 PM
Space Coast USA
hoppy's Avatar
Space Coast
Joined Oct 2000
20,832 Posts
Are any of the manufacturers who are offering taps denying there value? That would get my attention. I would fully expect any manufacturer who doesn't offer them to downplay there value but I thought I read where all the manufacturers were bringing out tapped packs....no?

One thing is for sure, if the packs don't have taps, nobody will know if they have a balance problem or not. Balanced packs, if treated conservatively during charge and discharge cycles, may remain in balance for their life but I'm not so sure about packs that are treated with wild abandonment, thus the advantage of a tapped pack..
hoppy is offline Find More Posts by hoppy
Last edited by hoppy; Aug 14, 2004 at 12:19 PM.
Reply With Quote
Old Aug 14, 2004, 02:46 PM
Got shenpa?
flieslikeabeagle's Avatar
Los Angeles
Joined May 2004
10,940 Posts
message deleted - duplicate post
flieslikeabeagle is offline Find More Posts by flieslikeabeagle
Last edited by flieslikeabeagl; Aug 14, 2004 at 03:00 PM.
Reply With Quote
Old Aug 14, 2004, 02:52 PM
Got shenpa?
flieslikeabeagle's Avatar
Los Angeles
Joined May 2004
10,940 Posts
RD Blakeslee wrote:
Quote:
I have hopes that Quacker's dedicated 4.2 volt, variable amperage chargers, one for each cell or parallel array of cells, may be isolate the charging circuit enough that packs could be charged in parallel through center taps, without disconnecting the series arrays. We'll see.
This is what I am having trouble seeing. Take a look at the attached schematic showing such a charger, charging several series-connected cells, but with a separate charging tap for each cell. Current I1 is flowing through the first cell, I2 through the second, and I3 through the third. The problem is that the current in the return wire d-c equals (I1 - I2), and there is no way to tell the individual values of I1 and I2 from this alone. Similarly, the current in wire f-e equals (I2-I3), and you cannot tell how much is I2 and how much is I3 from this alone. If the charger was smart enough to measure the currents in a-b, c-d, and e-f, all independently, and then solve Kirchoffs equations, then it could in fact figure out the individual charge currents to each cell. It would then have to independently adjust each of its three internal voltage sources to balance the currents.

This would make for a complex charger, as it would have to contain three independent variable voltage souces, measure three floating (not ground-referenced) currents, and then do some number crunching to figure out what to change.

Maybe the Suzanne LiPo balancer is a simpler way to achieve the same result.

Am I misunderstanding something, or do you see it the same way?

-Flieslikeabeagle
flieslikeabeagle is offline Find More Posts by flieslikeabeagle
Last edited by flieslikeabeagl; Aug 14, 2004 at 03:03 PM.
Reply With Quote
Old Aug 14, 2004, 02:59 PM
Got shenpa?
flieslikeabeagle's Avatar
Los Angeles
Joined May 2004
10,940 Posts
RD Blakeslee, anyone else, anybody have a link to either a suitable 4-pin connector, or the Multiplex 6-pin connector mentioned above?

Currently my models are all in the Speed 380 or smaller size, so my current draws are probably below 10A.

I just spent some money buying rather pricy Deans Ultra connectors for my charger and planes to match the ones that came on my Kokam LiPo's. Now it looks as though I'll be spending more on buying multi-pole connectors...and getting some new LiPo's while I'm at it!

-Flieslikeabeagle
flieslikeabeagle is offline Find More Posts by flieslikeabeagle
Reply With Quote
Old Aug 16, 2004, 01:31 AM
Registered User
Sunnyvale Ca
Joined Jun 2002
1,006 Posts
results from underWater test: not good

LiPoly users:

first some apologies: I looked at my old notes on salts and current in water. The actual number was 0.35 ohms but it was for a specialized salt used to coat metals, it was not NaCl and so the values are much higher because that is a much more conductive salt. The conductivity of NaCl and KCl at saturation is around 110-150 micromohs/cm. So when you do the calculation for the 3 inch square patch, it takes millimeters of separation to get to 1-2 ohms. Sorry for the bad "rememberence" numbers. The bottom line is that saturated salt water in a tank with LiPoly cells is like a 1K ohm resistor. Enough to slowly short out the cells, but not enough to cause any thermal runaway internal to the cell.

TEST ON 3S1P ThunderPower cell: I was given a previously minor swollen pack by a friend. We wanted to have the CELL destroyed before it destroyed HIM (I wish a lot more people had this attitude). It was a 3S1P ThunderPower pack (I think 1500 maH but I could see no markings). It had one of those PCBs on the top to help connect the three cell leads. I hooked it up to my constant current/constant voltage source and proceeded to put 1.58 amps into it. After 1 hour 30 minutes, it had reached 42 deg C and all three cells had swelled to roughly 4 times their original size (this is a rather low temperature for so much swelling). I was prepared for a big "vent and flames" and was about to dump the pack at the first signs of smoke into my fish bowl with saturated salt water.

The pack had swollen so much, and the packs pushed against each other so strongly, that the leads connecting the cells to the PCB broke. The temperature declined immediately and nothing else happened. I have included a picture to show the swelling. This occurred at dusk so the camera light was low and therefore the picture is grainy. But it gives you a good idea of the swelling. I had no other packs to "sacrifice" so I will have to wait on this experiment until I have some other "donations" from our flying group. I really did want to see if this thing would "burn" under water.

COATING ON TUBES: there has been conversations about the danger of shorting out packs by putting them into metal tubes. I do not worry about this because all the packs are wrapped and all the connections are carefully shielded. They have to be or otherwise shorts would occur inside the airplane with the ESC, Rx, antenna wire, etc. But on the final metal tubes, I was going to have them sprayed with black enamel to keep them from rusting. THis coating would stop almost all shorting unless very high heats were involved. I really don't see this as a significant issue.

TEMPERATURE PROBE: I have friends trying to make up a "4 pack" of temperature probes such that if any one probe goes beyond a temperature limit, it shuts down a relay which can power 115 VAC at 10 amps. I think this would be a great "second line of defense" in stopping thermal runaway. If this is cheap enough, then the "sand bag" technique or metal tube technique would be a really proven way to charge with safety.

I will keep you posted on any more testing that is done. If anyone else has experienced the "under water flames" I would greatly appreciate hearing about it with all the surrounding details. I will have to wait until I get give some more packs before I can finish the tests....and it may take 8-9 tries before I can get one to go "thermal" underwater.
Crazy Ted -- pushing the limits on LiPoly technology...and hoping not to get burned doing it!
tjcooper is offline Find More Posts by tjcooper
Reply With Quote
Old Aug 16, 2004, 07:58 AM
Registered User
Joined Jan 2002
6,587 Posts
F.l.a.b.,

4-pole A 10 amps:
http://www.atlantahobby.com/shopexd.asp?id=515

Multiplex 6-pole:
http://www.aircraft-world.com/shopdi...and+CONNECTORS
RD Blakeslee is offline Find More Posts by RD Blakeslee
Reply With Quote
Old Aug 16, 2004, 08:05 AM
Registered User
Joined Jan 2002
6,587 Posts
Quote:
Originally Posted by flieslikeabeagl
RD Blakeslee wrote:

This is what I am having trouble seeing. Take a look at the attached schematic showing such a charger, charging several series-connected cells, but with a separate charging tap for each cell. Current I1 is flowing through the first cell, I2 through the second, and I3 through the third. The problem is that the current in the return wire d-c equals (I1 - I2), and there is no way to tell the individual values of I1 and I2 from this alone. Similarly, the current in wire f-e equals (I2-I3), and you cannot tell how much is I2 and how much is I3 from this alone. If the charger was smart enough to measure the currents in a-b, c-d, and e-f, all independently, and then solve Kirchoffs equations, then it could in fact figure out the individual charge currents to each cell. It would then have to independently adjust each of its three internal voltage sources to balance the currents.

This would make for a complex charger, as it would have to contain three independent variable voltage souces, measure three floating (not ground-referenced) currents, and then do some number crunching to figure out what to change.

Maybe the Suzanne LiPo balancer is a simpler way to achieve the same result.

Am I misunderstanding something, or do you see it the same way?

-Flieslikeabeagle
F.l.a.b., Thanks for the analysis - I don't know enough circuit theory to say anything but that my hopes aren't necessarily realizable.

Thanks again.

- RD
RD Blakeslee is offline Find More Posts by RD Blakeslee
Reply With Quote
Old Aug 16, 2004, 08:13 AM
Registered User
Aio_1's Avatar
Ireland, Kildare, Leixlip
Joined May 2002
6,441 Posts
The multiplex connectors are commonly used on <20A power systems. To do this they are usually soldered up with 3 pins for - and 3 for +. If we wanted to use them for individual cell connections on 3s packs we would be limited to about 6A - 7A per cell. They'd be perfect for these applications though.

Aidan
Aio_1 is offline Find More Posts by Aio_1
Reply With Quote
Reply


Thread Tools

Similar Threads
Category Thread Thread Starter Forum Replies Last Post
Discussion Lipo loss caused by spark 2lakea Batteries and Chargers 8 Apr 29, 2006 05:16 PM