Ideas on how to resolve Kingsong 14" overheating

[Cloud]

I also always wondered what was the effect of the heat coming from the batteries. In the overheating situations, the battery compartment feels very warm. One doesnt feel it because of the padding on the sides, but i actually removed the padding at one point and the plastic side panel felt very hot. If the contribution if the heat generatoped by the battery is substantial, perhaps the approach to cooling should be revisited.

[esaj]

9 minutes ago, Cloud said:

I also always wondered what was the effect of the heat coming from the batteries. In the overheating situations, the battery compartment feels very warm. One doesnt feel it because of the padding on the sides, but i actually removed the padding at one point and the plastic side panel felt very hot. If the contribution if the heat generatoped by the battery is substantial, perhaps the approach to cooling should be revisited.

Guess it depends where the batteries are located, in the same compartment / next to the mainboard or in their own compartment(s). Cooling the batteries might not be as easy, as I don't think you can slap a heatsink on them (the outer casing underneath the plastic/whatever covers the cells actually is the cathode of the cells, afaik). Might need a fan or something.

[Cloud]

1 minute ago, esaj said:

Guess it depends where the batteries are located, in the same compartment / next to the mainboard or in their own compartment(s). Cooling the batteries might not be as easy, as I don't think you can slap a heatsink on them (the outer casing underneath the plastic/whatever covers the cells actually is the cathode of the cells, afaik). Might need a fan or something.

They have their own compartment but its connected with the cb co partment and its below it on the sides. 

[esaj]

On 8.4.2016 at 3:34 PM, dmethvin said:

A lot of it would depend on how much space you have to work with. I'm not familiar with the KS14 layout or control board so it's hard to say.

Can you install a plate on the *inside* of the case, facing the wheel? It could be well sealed so although it was exposed to water and dirt it wouldn't get near the board. 

Finally if you want silent cooling it would be interesting to try incorporating a Peltier to move the heat away quickly, but you'd need a 12v source and most likely want some type of thermostatic control. They're not expensive but the dimensions may not be easy to find for this application.

http://www.amazon.com/TEC1-12706-Thermoelectric-Peltier-Cooler-Volt/dp/B002UQQ3Q2

I actually have one of these TEC1 12706-peltiers (cost a couple of euros with free shipping, Aliexpress of course ), and tested it with batteries (8 x basic alkalines, can't give much more than 1-2A or so maybe). Even with that "low" current, it did become pretty cool (and of course hot on the other side). The problem I see with using such on a wheel is that the current is large (up to 6A), which might be way too much for the step-down on the mainboard, and you need a separate (probably large) heatsink for the hot-side (if it's really around 60W of power).

EDIT: About identifying peltiers based on the markings ( http://pcbheaven.com/wikipages/The_Peltier_Thermo-Element/ ):

Sometimes, the TECs have identification markings on their face, just like the following picture:

 
 

In this picture you see the ID: TEC1-12709

• The first two digits shall be always "TE"
• The next digit shall be "C" or "S". "C" stands for standard size and "S" for small sized.
• The following digit is a number and indicates the number of stages that the TEC has. In our example (and the vast majority of TECs) is a one-stage TEC
• Right next comes a dash. After the dash, the 3 first digits indicates the number of couples that the TEC has inside. In our case it has 127 couples. If the couples are 2-digit, then the number has a leading zero, for example 062 for 62 couples.
• Next comes two more numbers that indicate the rating current of operation for the Peltier. In our case this is 9 Amperes
• Some times follows a "T" and three numbers. This indicates the maximum operating temperature for the TEC. For example, "T125" is 125°C rated.

[dbfrese]

Instead of using the peltier to cool using energy off the battery, why not use the other property of peltiers, which is to generate electricity from heat? You could capture the waste heat, transforming it into electric current, and transfer that energy to the outside of the case where it would power a light bulb, or even a resistor. Outside the case, the heat would be much easier to dissipate. 

 

[Cloud]

3 minutes ago, dbfrese said:

Instead of using the peltier to cool using energy off the battery, why not use the other property of peltiers, which is to generate electricity from heat? You could capture the waste heat, transforming it into electric current, and transfer that energy to the outside of the case where it would power a light bulb, or even a resistor. Outside the case, the heat would be much easier to dissipate. 

 

How effective are these? Do they actively capture and remove the heat or just let the element get warm? In the latter case the passive heat transfer probably wont be sufficient for cooling such a big space...i think their application are more geared towards electricity generation tather than space cooling

[dmethvin]

Peltiers can transfer the heat from one side to the other, but they do require a pretty high amount of current to do that and you have to get the heat away from the hot side or they'll burn up. If you've ever seen one of those car refrigerators they typically use Peltiers rather than refrigerant and a compressor. Good enough for keeping a 6-pack of beer cool at least.

As a generator it looks like a Peltier can generate 1.5 volts or more with a reasonably good heat source but as I understand it the temperature differential is crucial to the voltage so you'd need to get the heat away from the cold side with a good sized heat sink. This guy did a blog post. http://www.x2jiggy.com/blog/2013/4/3/thermoelectric-generator-lamp.html

[esaj]

1 hour ago, dbfrese said:

Instead of using the peltier to cool using energy off the battery, why not use the other property of peltiers, which is to generate electricity from heat? You could capture the waste heat, transforming it into electric current, and transfer that energy to the outside of the case where it would power a light bulb, or even a resistor. Outside the case, the heat would be much easier to dissipate. 

 

 

47 minutes ago, dmethvin said:

Peltiers can transfer the heat from one side to the other, but they do require a pretty high amount of current to do that and you have to get the heat away from the hot side or they'll burn up. If you've ever seen one of those car refrigerators they typically use Peltiers rather than refrigerant and a compressor. Good enough for keeping a 6-pack of beer cool at least.

As a generator it looks like a Peltier can generate 1.5 volts or more with a reasonably good heat source but as I understand it the temperature differential is crucial to the voltage so you'd need to get the heat away from the cold side with a good sized heat sink. This guy did a blog post. http://www.x2jiggy.com/blog/2013/4/3/thermoelectric-generator-lamp.html

I briefly read about "thermoelectric generator" (basically peltier acting as a generator), and the efficiency is really poor, something like < 10%. So I doubt it would work very well to first turn the heat into electricity and then burn if off in a resistor or such outside the case, as 90% or over would still remain inside. In general (although I don't really know that much about the subject), I doubt peltiers are the way to go with cooling EUCs. But who knows, maybe someone will prove me wrong

[Hunka Hunka Burning Love]

I wonder if small automated sliding vents could be incorporated that when a certain temperature is reached that they open up allowing air to pass through the control box.  That way under colder conditions the vents remain closed.  For rain, maybe an APP over-ride or a humidity sensor (or rain sensor like those automated car wipers) could be used to keep the vents closed.   Some cars do this with their front grills for aerodynamics and cooling I believe.  Or duct the intake so that it's lower with a gooseneck above and then downspout to enter the box and have a shielded exit port to minimize water ingress.   Either that or.... wire up a can of this:

http://www.amazon.com/MG-Chemicals-Super-Spray-Aerosol/dp/B0047Y9DKS

with the nozzle to automatically spray over the motherboard to cool whenever the redline is hit!  Paint the can blue with an orange NOS label and impress people!

Really though the designers need to step up their game with either more efficient components, larger or better planned heat sinks, etc rather than have consumers do the work for them.

[DS]

What about just stick an ice pack close to motherboard when needed

 

 

[Jason McNeil]

Got some data on my 7.5mile journey to the office today: there was quite a bit of headwind , temperature was pretty mild, & was riding on the limiter about 90% of the time. The temperature probe is secured directly next to the bank of MOSFETS.

Although there's a magnet RPM sensor on the outside rim of the tire, I have not been able to get it to work reliably, which is a shame, since it would be a useful piece of data; however, it's not the end of the world, since I'll be acquiring the GPS module later this week  

Conditions:

• Rider: 75kg
• Ambient temperature: 27.8°C
• Humidity: 44%
• Winds: 25.7kph, mostly headwind
• Terrain: completely flat! 
• Data capture rate: 100Hz

Setup of the EagleTree eLogger

Profile of the ride into work (there's a bug in this output graph, where the minutes are being represented 4x faster than they really are...):

Analysis: until reaching 'steady-state' the rate of temperature increase is roughly directly proportional to power output. In the first highlighted area, the avg power consumption is around the 535W mark, as this drops off to 350W (on account of tail wind instead of headwind), the control-board temperature decreases by 3°C. 

When the CB temperature reaches around 45-48°C, we see it leveling off into a 'steady-state' condition, where between the airflow to the wheel housing & heat conduction through the shell body, the control-board no longer increases in a linear fashion. At 116, the power consumption diminishes, & a new lower operating temperature equilibrium is achieved.

Next Steps:

1. GoPro video combined with data feed, try to get Dashware working....
2. Integrate RPM/speed vectors
3. Replicate conditions to cause CB overheating, such as rapid acceleration/braking
4. Evaluate the rate at which the CB is able to shed heat at maximum temperature when powered off 
5. Experiment with air inlets while riding
6. Create formula to predict expected overheating conditions

[Donald1800]

Getting basic heat profile data is good and will help in determining the sources of excess heat, their percentage of contribution, and effectiveness of corrective action, but it is essential to understand that whatever combination of 'fixes' are incorporated, waterproofing the Control Circuit Board is an essential first step, as any effective removal of rapidly rising/generated heat will require air movement from the Control Board to the ambient atmosphere.

One company I worked for was on contract to provide digital circuit boards that could be exposed to excessive high humidity condensation or rain through ventilation holes/slots, and we waterproofed all of them (after cleaning with any form of 91% alcohol) with multiple coats (typically three) of Krylon Clear with each coat drying for 24 hrs. (see my comments in a post about "Waterproofing").

Once you no longer need to have concerns about water on the Control Board, any and all methods of excessive heat dissipation and removal become available for use in the most efficient and economical method.

[Donald1800]

CAUTION!

Remember - ALL MOSFETS have their Drains (with potential High Voltage) connected to their mounting tabs which are also mounted through thin insulators to the Heat-Sink. This area (the mounting tabs, thin Insulators & mounting area of the Heat-Sink) must ALSO be cleaned with any type of 91% alcohol before re-mounting the Control Board to the Heat-Sink, then treated with several coats of Krylon Clear AFTER re-mounting the Control Board to the Heat-Sink to prevent potential shorting of the High Voltage through water conduction.

[UKJ]

Hello,

I also now suffer with almost daily over heat beeps and tilt back.

I am 75 kilos, tire inflated to 3.6 bar, KS14C 800w, 680 w/hr, 3 months old

I am very concerned because it is not even hot outside yet, like only 16 c, 60 f, how much worse will it be when summer heat sets in?

By idling dancing in my small living room it will tilt back after about 8 minutes of as easy control as I can input to keep balanced.

Outside on tennis court using lines as guides going forward and backwards it will do about 10 minutes or so.

Having done tons of research here trying to buy an EUC for SAFETY, speed, riding time, off road and freestyle my beloved ks14c does all but the latter very well.

Thanks to all posting here as I must find an solution soon.

Kingsong France has posted on the French forum that they have a quick and simple fix which consists of adding a fan, I am waiting for a reply as to which fan, where and how to mount, and how to power.

Would someone PLEASE post close up photos, video of the upper sides of the case to see inside better.

Even better a video showing not just tire change as I have already studied but complete disassembly.

I have not opened up my unit yet as still waiting and reading but here are a few ideas I am pondering.

@Donald1800

VERY good idea to water proof mother board FIRST!

Maybe just opening up the top of the case just under the handle is enough to let it breath enough to cool off?

Many small holes, a couple of square openings sized to fit a fan,s later?

Water, dust proofing could be done using gortex or other fine mesh and velcro.

I have seem many small fans of many sizes that should run from power at the usb port as we know that is 5 v, better 12 v I would think but where to find on the mother board?

I am thinking that one maybe two fans, 35mm X 7mm fan mounted exterior under the handle, one fore and one aft, one input other exit should allow enough air flow.

As I ride for sport my waterproofing is not that important so far my EUC has never seen ANY H20 and hopefully will not!

As it seems that the heat sink is small and mounted above the wheel is there any way to make it bigger keeping it inside the top of the case, and, or add fins being very sure to not touch the wheel itself.

What about fins attached to the wheel rim to act as a fan also?

As for the battery compartments maybe also just opening them up while trying to keep out H20 and dust would also help?

Off to ride, any and all input ideas more than welcome, thanks!

UKJ