The Consequence Of Not Using High Temperature Wires

[Lyen]

It is very important to use high temperature wires for your battery. I have another incident a couple days ago with my 48V 10Ah LifePO4 pack. This is a completely different issue and is not related to the incident I had previously > http://endless-sphere.com/forums/viewtopic.php?f=14&t=18577.

As you can see from the picture, the BMS some how got too hot inside the rack bag and started to cook itself. Fortunately, the burning smell was giving me the hint to pull over and do a visual inspection. The battery wire had partially melted on top of one of the MOSFET but luckily did not touch the metal surface area or otherwise I am in deep trouble.













Now all I have to do is to replace the wires and put the MOSFETs flat on the circuit board to make it works again. I think I got lucky this time for not generating another fireball behind my butt.

 

[mwkeefer]

Well lyen if you hadn't added the extra 2 fets on the PCB and yet another stacked on top at the end? Then it wouldn't be an issue right?

Expecting transporter beam with my new controller to arrive by 10AM EDT = )_ {man that will be the day!}

-Mike

 

[justadad1957]

glad you caught it in time
sholdnt be near as costly to repair this time
i am wiring with high temp silicone
and hoping for zero errors on the pacs i am building

btw are you keeping a video cam with you at all times
i am sure all of us on the es forum would enjoy watching

 

[grindz145]

Even with high-temp wires, it probably wouldn't hurt to insulate the fet a little bit. Ouch.

 

[dogman dan]

At least don't run wires , or anthing else for that matter, across hot fets.

And don't smother it inside a bag either. Let that thing have some air.

 

[BatteryMooch]

A LOT of air!
It looks like all the FET mounting solder melted and pooled at one end of the board.

 

[wasp]

you refer to your other post about the lifep04 battery catching fire but you never answered who the maker was for that battery?
and please dont tell me it was acme...

 

[Lyen]

you refer to your other post about the lifep04 battery catching fire but you never answered who the maker was for that battery?
and please dont tell me it was acme...

Hi Wasp! The reason I did not disclose the name of the maker is because I do not want to do any harm to their business. Also, there is no real hard evident to proof if it was cause by short circuit or if it is the failure in the original construction and/or design. My only intend was to just wanted to let people know and expect the worst if something goes wrong.

 

[Lyen]

Well lyen if you hadn't added the extra 2 fets on the PCB and yet another stacked on top at the end? Then it wouldn't be an issue right?

Expecting transporter beam with my new controller to arrive by 10AM EDT = )_ {man that will be the day!}

-Mike

Hi Mike! What makes you think I added the extra 2 fets on the PCB? It already comes that way. I believe you were referred to the 24v or something.

 

[Tiberius]

Well yes, the consequences of a short in one of the balancing wires running round inside a pack can be serious. Actually a short in one of them is bound to be serious.

I'm not so keen on using silicone insulation because, while it will stand higher temperatures, its not resistant against mechanical problems. There are high spec wires with double layer high temperature insulation. They are not difficult to get or overly expensive. They are also usually thinner than single sheathed PVC wire for the same current capacity too.

The downside is that they are less flexible, the insulation is more difficult to strip and greater care is needed with fitting crimp connectors. The upside is the insulation doesn't melt when you solder the ends.

Nick

 

[John in CR]

My Konion packs never get warm enough to feel any heat with my hand, much less hot enough to melt plastic. This is just another strike against batteries that require BMS's as far as I'm concerned.

I take that back, I my pack did feel warm a couple of times, but that was just because my bike was sitting in the sun.

 

[Lyen]

I have researched for possible heatsinks to use. Here is one of the potential candidtate:

A memory module heatsink with heatpipe. What do you guys think?

 

[cell_man]

The 80A constant BMS I use doesn't get more than mildy warm when running 90A Max and 110A peak according to the CA. The 30A constant BMS I was using also never got more than a little warm running 65A max with 75A peak current. What is the BMS rated at and what were you running it at?

 

[Lyen]

The 80A constant BMS I use doesn't get more than mildy warm when running 90A Max and 110A peak according to the CA. The 30A constant BMS I was using also never got more than a little warm running 65A max with 75A peak current. What is the BMS rated at and what were you running it at?

Hi Cell_man!

Thanks for your BMS comparison information. This is the original Signalab BMS version 1.0 designed for 20-25A operation. The MOSFET they used are the cheapest just barely get the job done. I believe the work around is to replace the BMS, replace the MOSFET with low internal resistance, and/or add heatsink for better heat dissipation.

By the way, 你的英文水平非常好！ 请问你是美国人在中国还是中国的本地人呢？

Best Regards,
Lyen

 

[SilverSurfer]

I love it when everything is goin great and your just flying up and down the avenues without a care, but secretly, I love running into problems, preferabley ones that DONT self destruct batteries. When shit goes wrong, if gives me the need to stop and figure out what happened, why it happened, and how to improve the system so it cant happen again. For some crazy reason, I get as much if not more fun out of solving the problems as I do riding!

Good luck with solving your problem! I got eneough of my own! LOL

 

[Tiberius]

...secretly, I love running into problems, preferabley ones that DONT self destruct ....

I know what you mean. Part of the art of engineering is anticipating what can go wrong and stopping it happening. Good engineers are often those with overactive imaginations, contrary to the popular image.

Nick

 

[cell_man]

By the way, 你的英文水平非常好！ 请问你是美国人在中国还是中国的本地人呢？

Best Regards,
Lyen

I'm an Englishman in China My Chinglish is coming along great, but my Chinese skills are very lacking I'm afraid

 

[jonescg]

Aw c'mon Paul, Google translator is your friend!
我真的希望他們不會阻止，但此消息 :lol:

 

[cell_man]

It's ok, my GF did the translation

I really must get back to class next semester, but my god, learning Chinese is so bloody hard...

 

[BatteryMooch]

I have researched for possible heatsinks to use. Here is one of the potential candidtate:

A memory module heatsink with heatpipe. What do you guys think?

It could work but I have a few concerns...
Typically, memory module heatsinks are designed to only remove the small amount of heat that those modules put off. Also, using a heat pipe can be a good idea for certain setups but is expensive and very delicate compared to other heat sink designs.

Is there a thermal resistance spec for that heat sink?
What were you going to use to keep the fins from bouncing around when riding?
How are you going to remove the heat from the fins? Unless the fins are very actively cooled, you essentially have no heat sink once the fins heat up.

 

[SilverSurfer]

I have researched for possible heatsinks to use. Here is one of the potential candidtate:

A memory module heatsink with heatpipe. What do you guys think?

It could work but I have a few concerns...
Typically, memory module heatsinks are designed to only remove the small amount of heat that those modules put off. Also, using a heat pipe can be a good idea for certain setups but is expensive and very delicate compared to other heat sink designs.

Is there a thermal resistance spec for that heat sink?
What were you going to use to keep the fins from bouncing around when riding?
How are you going to remove the heat from the fins? Unless the fins are very actively cooled, you essentially have no heat sink once the fins heat up.

Wow! Thats a great invention anyway! LOL Gotta be some places we can sneak a couple in our applications.

 

[Lyen]

It could work but I have a few concerns...
Typically, memory module heatsinks are designed to only remove the small amount of heat that those modules put off. Also, using a heat pipe can be a good idea for certain setups but is expensive and very delicate compared to other heat sink designs.

Is there a thermal resistance spec for that heat sink?
What were you going to use to keep the fins from bouncing around when riding?
How are you going to remove the heat from the fins? Unless the fins are very actively cooled, you essentially have no heat sink once the fins heat up.

Camlight, you are absolutely correct about insufficient heat dissipation on the tiny memory heatsink with heat pipe. I am going back to the drawing board. Thanks!

 

[SilverSurfer]

Doh! LOL

 

[ericslai]

Hi Edward,

You know how to cool the FETs on the controllers. You can apply the same principles to cool these FETs. The most important thing is to electrically isolate the FETs to the heatsink but thermally conductive. That stuff is called thermal pad. Passive heatsinks are more reliable than active ones. No moving parts to fail but the penalty is bigger size and more weight.

 

[amberwolf]

I'd use Kapton tape or similar rather than a thermal pad; those gray (or pink) pads make great electrical insulators but not so great thermal conductors. Kapton is at least much much thinner, so even though it may also not conduct heat very well, it isn't as thick to block radiation from passing thru it.