FC3S.USD

iTrader: (2)

Hello,

I was wonering if anyone has installed the ecu upside down ? or any input. I dont think it would do anything but always good to double check.

I dont want to isntall in stock location due to heat , i stripped all the sound mateial and everything so the car gets quite hot there.

Yes i could insulate that area , but would much rather install remotly, my second best option is in glove compartment, but i dont want it that visible jsu tincase of bad luck ( someone looking for change brakes into car etc etc )

so i want to install under the glove compartment. It wont get all the heat fromt he floor bed, and is nto visible.

Any input greatly apreciated,

its a e6x if that matter any.

RETed

It's not recommended, as with any electronics - just like amplifiers.
If you've ever pulled the "guts" out of the ECU, the board is on the bottom.
The power transitors are all along the sides - this allows the heat sink to "pull" the heat away from the board - when mounted right side up.
Flipping the ECU upside-down just negates this design feature.
Most car audio amplifiers are built like this too.

If you have to do this, I would recommended installing a small fan to keep the ECU cool...


-Ted

Wankel7

Mine ended upside down...it is mounted in the glove box. After passing my harness thru the firewall and feeding the connector to the glove box I could only mount the ecu upside down. The harness was not very pliable.

So, we shall see if it causes any problems. The ecu sure does get hot.

James

FC3S.USD

iTrader: (2)

well I dont have much stuff under my dash , so i will see if theres other mounting points.

thanks for the input.

purple82

LOL, gravity has no influence on heat conduction. Moving heat from the board to the sides of the box where the heat sinks are is almost totally independent of orientation.

Mount it in any direction you want.

The thing to think about, though, is the more out in the open it is, the more air can move around the outside of the box and the cooler it will stay. Also, if it can be placed in a cooler area of the interior, like under the seats or storage bins, it will stay cooler.

Claudio RX-7

iTrader: (1)

Im with purple on this, the fact is that its not an amp, nor does it have the transformer and power demands or component weight of an audio amp. And letting air circuilta through the casing fins is more agreable.

Wankel7

Gravity doesnt have anything to do with heat...but heat rising has to do with heat.

I agree with ReTed that the ICs of the ecu are designed to have heat rise so it's heatsinks work right.

However, mine is upside down in my glove box and is gonna stay there

James

purple82

Conduction from the board to the heatsinks has nothing to do with gravity. Put a cold surface next to a hot one in a vacuum and they will come to the same temperature in the same amount of time in any orientation that you set them.

You're confusing conduction with convection, which is when heat is carried away by a fluid, in this case air. Convection IS often dependent on orientation, but convection in this case, is external to the box and whatever the box orientation, it's heat removal will be more dependent on the location of the box and it's surroundings than on it's orientation.

In other words, if the heat is trasported from a solid to solid, like the board to the heatsink, orientation doesn't matter AT ALL. If it's being carried away by a fluid, like air, orientation often does matter.

Internal to the box, there is going to be some recirculation of air, this is why I stated that orientation has ALMOST nothing to do with cooling of the box. It's not significant compared to the conduction.

purple82

By the way, an amplifier should also be independent of orientation IF the box is enclosed. If it's open on opposite sides to let air flow through but doesn't have a fan, then you would have to pay attention to orientation.

Wankel7

No fan and no air holes...yeah that does make sense... thanks!

RETed

It's stupid replies like this that should be banned.

I never claimed gravity had anything to do with it.

So what's your experience in electronics?
I'm curious...


-Ted

purple82

I've got a masters degree in mechanical engineering and over ten years in thermal analysis of computer systems. You?

Now if you want help in understanding heat transfer, ask a useful question.

O 16581 72452 5

Haha, punk'd.

I love this forum, if someone is wrong, they'll call the person that actually knows what they're talking about "Stupid" and "Retarded" because fact doesn't agree with their theory.

RETed

Done enough car audio to see amps burn up in "enclosed areas".

What you're talking about is ideal systems, but does this translate into real world?

If 100% of the heat (or enough of) is transferred into the heat sink, then maybe what you say it's true...
How about if the heat sink hits thermal capacity?
Where does all the heat go?

Masters in ME?
What specialty, if I may ask?

This is more of an EE study...
You might know more about the metallurgy, but I'd think EE would be a better field of study for this particular subject?

Hmmm...maybe you can tell me how my Crane Cams HI-6 died cause it was mounted "upside down" between the headlight and battery box?


-Ted

RETed

Yeah, why don't you keep your retarded replies to yourself.

For someone to imply gravity to affect cooling, that's pretty damn stupid.
We're not in kindergarten - maybe you are.
If you believed that, I'll call you stupid all the time.

I never called him "retarded".
I think you pretty much defined "retarded" already.


-Ted

purple82

First things first, this is a thermal issue, which is a speciality in mechanical engineering. It happens to be my speciality. EEs come to me to answer questions on how to cool off hot parts.

Now, did you even read my posts? Look above, I clearly stated that orientation hardly makes any difference, but the space around the ECU does. I don't think I can explain this any more clearly without repeating what I said in the posts above.

I honestly don't know what you mean by heatsink "thermal capacity". Please elaborate.

I love it when people who don't understand the physics of a situation pass it off as "theory" or "ideal" and assume things happen differently in reality. It's not the case folks, I could tell you within 3 degrees what every component in that ECU would be if I knew all the powers, materials and geometries.

You stated above that heat rises, right? I assumed you would connect this with gravity, but it sounds like you have some other ideas. Care to share what you think it is that makes heat rise? Also, can you explain to us why you think the stereo equiptment overheated?

purple82

I hate to break it to you, but gravity does effect cooling. In a natural convection situation it's what causes hot air to move up (because of it's lower density) so that cold air can move over to the hot surface.

RETed

Ah great, you actually caught your own contridiction.

Guess it makes you look smart that you beat me to it.

You're right.
Without gravity, the term "convection" would have no meaning.

So I dunno where you pull this out from?

Or you just like talking in circles just to confuse the lay folk?

"Thermal capacity"?
Dunno what you fancy ME guys call it, but a mass of metal (in this case, the heat sink "case") can only conduct so much heat before it can't anymore.
So what happens to the generated heat?

I need to remind you that this is the Haltech section.
If you don't know this, but the early E11V1's were prone to "too much heat" (official explanation from Haltech themselves), so why aggrevate the problem?

Heat rises - like you said.
Why put the heatsink under the IC board?
Ideally, the heatsink case would pull all the heat away from the circuits (i.e. power transitors) and allow convection cooling to keep it within safe operating parameters.
But what if the design was flawed?
Ever touched the case of a Haltech running on the track?
I gets hot enough to burn your hands - easily over 100F.
Heat and electronics don't mix well.
So if we can avoid aggrevating the heat "problem", why would you recommend ignoring it?


-Ted

RETed

Thermal capacity...specific heat capacity...specific heat?
http://en.wikipedia.org/wiki/Heat_capacity

Thermal capacity = heat capacity
http://education.yahoo.com/reference...entry/heatcapa


-Ted

purple82

OK.

1) Convection and conduction are two different things. As I said, conduction is independent of orientation, free convection is not. A few simplified definitions for you,
conduction - heat transport through solid bodies
convection - heat transport between a solid and a fluid
There is NO contradiction in anything that I said and I'm not talking in circles if you understand the words that I'm using. Re-read the quotes you posted of what I said.

2) Again, re-read what I said. Within the box: convection is a small player, most of the heat moves from the board to the heatsinks through conduction (and some radiation, but we won't get into that until you get the conduction part). There's no orientation dependency.

3) Outside of the box it's almost all convection, hence the necessity to pay attention to where the box is placed within the car. AS I SAID, make sure there's space around the ECU where it's mounted. This allows the air to take the heat of the box away.

4) "thermal capacity" as you describe it: It's complicated, but basically, temperature rises with power. The only real limit to the amount of power a heatsink can handle is because the component it's cooling has a temperature limit. There is no amount of power that a hunk of metal can't conduct, but there is a temperature increase with every amount of power increase. Until it melts or turns to vapor, I suppose, but then other things start to happen.

5) I NEVER said heat rises. That's a rediculous statement. Air picks up heat and the warm air rises. That's due to the density difference between the warm and cool air. Heat only moves from high temperature to low, it doesn't rise or fall or move sideways if the temperature difference isn't in that direction.

purple82

Specific heat is a measure of a material's ability to hold heat compared to other materials. It does decrease with increasing temperature, but realistically, it doesn't change much and definitely doesn't change noticeably in the temperature range that we're talking about here.

J-Rat

iTrader: (2)

*gets popcorn*

RETed

You're right.
I apologize.
For some reason my eyes saw "convection", but you did type "conduction" (conductivity?).


Although the majority of the heat is produced by the power transitors, there are still other electrical components inside (on the IC board).
They might radiate significant amounts of heat or they might not?
This is going to take a lab under controlled conditions to get solid numbers, so it's a moot point.

In car audio, we are told to try and mount heat sinks so convection currents can flow upward.
Thus, heat sinks should be oriented vertically; this is the most efficient mounting method to take advantage of convection (cooling).
If we extrapolate from that, it can be implied that mounting with the heatsink underneath the IC board should radiate heat back into the IC board?
Radiated heat (from the heatsink) is not necessarily directional, and heat radiates back inside the ECU case itself?
I know it can be argued that this might not be significant, but why take the risk?
Like I said before, the E11V1's had a known heat issue.
It might be okay to mount the ECU upside-down, but why not avoid doing that if possible?

I know from my experience that the ECU (heatsinks) chassis gets very hot - enough to almost burn your hand at touch.
If I can, I'd mount the ECU right-side up, and try to avoid mounting it upside-down.
Is there something fundamentally wrong with that?


-Ted

purple82

Alright, now we're getting somewhere!

It's true, for external convection of the unit, it is better to mount it vertically if the heatsinks are along the sides, as the warm air has an easier time rising off the longer vertical surfaces then it does off longer horizontal surfaces. You're abosolutely right.

There are only two sides of this box that are used to sink heat off the ECU, the other four don't do much for heat dissipation. They are sheet metal and one is full of connectors. If the ECU is mounted with the sinks at the top and bottom it's worse for heat rejection than if the box was set on the larger flat surfaces. So there is an order to the best orientations of the box.

You have to understand, however, that this kind of convection is VERY dependent on the amount of free air space around the box. This is the major driver, much moreso than orientation. You would be far better off with a wide open space in the worst orientation than enclosed in the best orientation.

Having said that, the original question was if the box could be mounted upside down vs. right side up because it didn't sound like he was planning on any locations where it would fit vertically. There will be almost no difference in the temperature of the components between these two orientations if everything else remains the same.

purple82

I missed this comment this morning. You've made the leap that's leading to your biggest confusion.

The process of heat transfer between the IC board and the heatsink doesn't have much to do with convection. It's dominated by conduction.

The heatsink being cooled by convection is on the outside of the box, the printed circuit board (PCB) is inside the box, there's no heat exchange in the way that you're describing. Try and think of this as two different things going on. What's going on inside the box and then what's going on outside the box.