Jump to content

Max Operating Temperature


dbfrese

Recommended Posts

This is not a concern yet, but I want to make preparations for the inevitable. I live in Scottsdale, Arizona located in the middle of the Sonoran desert, and summer temperatures average 107°F/42°C and frequently reach 117°F/47°C for days at a time. Seeing as how the specs for the Ninebot On E+show a max operating temperature of 113°F/45°C, what should I expect as far as performance?

How will this affect battery life, and even more importantly, will I experience sudden shutdowns due to exceeding operating temperature? Is there an overtemperature warning beep on the E+?

And... is there anything I can do to upgrade the motherboard/chip by adding to or replacing the heatsink to gain more effective heat dissipation as operating temperature approaches ambient air temperatures?

Any thoughts?

Link to comment
Share on other sites

  • 4 weeks later...

I was also contemplating this the other day.  I was wondering if there is a more robust Mosfet that could be used to replace the originals?  Why don't the 1500-3000 Watt motors have this same issue?  Must be larger Mosfets, or better heat sink?  I'd like to compare MB's to see what's different...  I'm sure they only saved a couple dollars when building the board, and I'd be happy to spend the couple dollars now to beef it up.

Link to comment
Share on other sites

From what I could see the 9B1 boards had pretty decent heatsinks compared to the Firewheel, but the Firewheel has the advantage of mounting its heatsink to a metal plate exposed to the outside (well, inside the wheel shell). So the movement of the wheel does a bit of cooling for me there. Since the 9B1 board is in a nearly closed compartment maybe the heat isn't escaping? If there was a way to get some air flow through the heat sink area that might help but it seems like it would be hard to find a place for the air to enter/exit. Maybe you could cut out some of the plastic behind the board, replace with a metal plate, and attach the plate to the board/heatsink combo? That would expose the metal to the inside of the wheel where airflow would cool it.

563528800ee6a_IMG_1695(2).JPG.b6184618b2

U2jo7pB.jpg

Link to comment
Share on other sites

We actually have a metal plate exposed to the wheel area, so there is air hitting it.  Same idea as what you have.  I was debating replacing the Mosfets with more robust versions (Something Like These).  Not sure if that would cause a failure somewhere else on the MB though.

Link to comment
Share on other sites

Maybe there's a way to attach some copper tubing to the heat sink, and take the copper tubing out to where it is exposed to air. I'm just thinking out loud without any clear plan.

 

Link to comment
Share on other sites

27 minutes ago, dbfrese said:

Maybe there's a way to attach some copper tubing to the heat sink, and take the copper tubing out to where it is exposed to air. I'm just thinking out loud without any clear plan.

 

I spent a BUNCH of time working on a plan, and I think it's good.  As an added bonus, it's a nice seat.

56d9f13022cbc_NineBotOneCooler.jpg.3009f

Link to comment
Share on other sites

2 hours ago, dmethvin said:

From what I could see the 9B1 boards had pretty decent heatsinks compared to the Firewheel, but the Firewheel has the advantage of mounting its heatsink to a metal plate exposed to the outside (well, inside the wheel shell). So the movement of the wheel does a bit of cooling for me there. Since the 9B1 board is in a nearly closed compartment maybe the heat isn't escaping? If there was a way to get some air flow through the heat sink area that might help but it seems like it would be hard to find a place for the air to enter/exit. Maybe you could cut out some of the plastic behind the board, replace with a metal plate, and attach the plate to the board/heatsink combo? That would expose the metal to the inside of the wheel where airflow would cool it.

563528800ee6a_IMG_1695(2).JPG.b6184618b2

U2jo7pB.jpg

The Hopemotion board in the Firewheel uses SMD MOSFET. They sit close to each other and not in a line like in the example of the Ninebot. Maybe heat can't transported as good. Maybe a heat pipe design like used on graphic cards or a bigger cooler with active fan could lower temp. But this doesn't address the root cause: Why do the heat goes up in this way. 

Link to comment
Share on other sites

1 hour ago, SuperSport said:

I spent a BUNCH of time working on a plan, and I think it's good.  As an added bonus, it's a nice seat.

@SuperSport Almost perfect. I see you are going for subtlety in design. Might I suggest a radiator for more efficiency?

 

.9botrad.jpg

Link to comment
Share on other sites

1 hour ago, OliverH said:

The Hopemotion board in the Firewheel uses SMD MOSFET. They sit close to each other and not in a line like in the example of the Ninebot. Maybe heat can't transported as good. Maybe a heat pipe design like used on graphic cards or a bigger cooler with active fan could lower temp. But this doesn't address the root cause: Why do the heat goes up in this way. 

Never had actual heat problems with the Firewheel last summer (although we had a cold summer), at some point I was worried about it, as I don't know if Firewheel actually has any overheat warning. Someone in the french forum had one of their mosfets blown ( http://trottinetteselectriques.heberg-forum.fr/ftopic326-340.html ):

mosfet-hs-low.jpg

Don't know if it was overheating or overcurrent (meaning like short circuit/bridge shoot-through or driving the motor way too hard; the translated text says " the person has completely lost the control following it has certainly due to excessive speed "). The mosfets there are these:  http://www.irf.com/product-info/datasheets/data/irf7759l2pbf.pdf

One thing that caught my eye, is that the Rds(on) is about 5 times lower than in the 75NF75's used by older Ninebot-mainboards (1.8-2.3 milliohms in IRF7759's vs. 9.5-11 milliohms in 75NF75's), which should translate to much less heat being generated.

Link to comment
Share on other sites

40 minutes ago, esaj said:

One thing that caught my eye, is that the Rds(on) is about 5 times lower than in the 75NF75's used by older Ninebot-mainboards (1.8-2.3 milliohms in IRF7759's vs. 9.5-11 milliohms in 75NF75's), which should translate to much less heat being generated.

Oh good catch! That would definitely make a difference in heat, and explain why the smaller heatsink could still be enough. I added some compound between the plate and the heatsink anyway to improve heat transfer, but that surface is so far from the MOSFETs that it may not matter.

The article on MOSFET failure that I read a while back said that one way they fail is by getting so hot that the solder on the internal connections reflows, which can result in either a short or an open. Roll the dice. :wacko:

Link to comment
Share on other sites

I had a bad mobo right out of the box with my 9bot with no bluetooth so when Banggood sent me a new one I noticed the old heatsink had a big fat blob of thermal paste just mashed on there.  Anyone who's changed out CPU's on a PC knows that more paste doesn't equal better so when I replaced the board I got out some of my good quality CPU thermal paste and applied just a very thin even layer on the heatsink.  I can't compare the temps from the old and new one because my bluetooth didn't work but I'm sure it must have helped out a bit with temps.  The amount of paste on the original board seems like way too much, plus who knows what kind of quality the paste was that they used.

Link to comment
Share on other sites

Better be careful if you apply thermal paste or such between the heatsink and the mosfet: the mosfet "body" (the metallic part on the "backside") must be electrically insulated from the heatsink, because it (at least usually) acts as the drain:

_scaled_mosfet_symbol.jpg.pagespeed.ce.b

On high-sides of the half-bridges, I don't think this matters much, as the high-side drain is (more or less) directly attached to the battery + terminal anyway, but on the low-side, if the drain of the mosfet is attached directly to battery, it will cause a short circuit when it starts conducting (think of a switch directly between battery + and - terminals).

Link to comment
Share on other sites

1 hour ago, esaj said:

Better be careful if you apply thermal paste or such between the heatsink and the mosfet: the mosfet "body" (the metallic part on the "backside") must be electrically insulated from the heatsink, because it (at least usually) acts as the drain

Yeah, I wouldn't use any conductive paste in there.  There are Non-Conductive versions available that will do a decent job.  It sounds like @SlowMo is familiar with using the paste though, but for those who don't understand it, I'd say get a Non-Conductive just to avoid issues.  If too thick and it squeezes out later and drops on there, that would be enough to cause a short.  Again, @SlowMo mentioned he added a thin layer, so he's probably safe.

Just imagine if this was conductive paste...

1byf626.jpeg

Link to comment
Share on other sites

Archived

This topic is now archived and is closed to further replies.

×
×
  • Create New...