|
Home > Archive > Electrical Engineering > March 2007 > MOSFET smd - six pins?!
You are viewing an archived Text-only version of the thread.
To view this thread in it's original format and/or if you want to reply to
this thread please [click here]
| Author |
MOSFET smd - six pins?!
|
|
|
| RTQ035N03 in TSMT-6 smd 6-pin package:
<http://www.rohm.com.cn/products/dat...f/rtq035n03.pdf>
One pin each for source and gate. Four for drain.
Is the number of drain connections simply for convenience of
mounting/orientation? One wouldn't need four drains with only one source...
Or am I missing something (which is my default presumption)?
Thanks,
--
DaveC
me@bogusdomain.net
This is an invalid return address
Please reply in the news group
| |
| Palindrome 2007-03-13, 9:25 am |
| DaveC wrote:
> RTQ035N03 in TSMT-6 smd 6-pin package:
>
> <http://www.rohm.com.cn/products/dat...f/rtq035n03.pdf>
>
> One pin each for source and gate. Four for drain.
>
> Is the number of drain connections simply for convenience of
> mounting/orientation? One wouldn't need four drains with only one source...
>
> Or am I missing something (which is my default presumption)?
Thermal dissipation.
--
Sue
| |
| Tim Wescott 2007-03-13, 1:25 pm |
| Palindrome wrote:
> DaveC wrote:
>
>
>
> Thermal dissipation.
>
To unpack:
The leads carry heat as well as electricity, and the lead frame is
designed to maximize the thermal transfer -- some or all are being made
out of copper instead of kovar to maximize heat transfer. To really
deal well with the heat you often need to solder the chip down to a
copper pour on the board -- the data sheet will tell you how, or will
point you to an app note.
--
Tim Wescott
Wescott Design Services
http://www.wescottdesign.com
Posting from Google? See http://cfaj.freeshell.org/google/
"Applied Control Theory for Embedded Systems" came out in April.
See details at http://www.wescottdesign.com/actfes/actfes.html
| |
|
| > The leads carry heat as well as electricity, and the lead frame is
> designed to maximize the thermal transfer -- some or all are being made
> out of copper instead of kovar to maximize heat transfer. To really
> deal well with the heat you often need to solder the chip down to a
> copper pour on the board -- the data sheet will tell you how, or will
> point you to an app note.
Oh, you mean like this:
"Total power dissipation: 1.25w*
*Mounted on a ceramic board"
I can find no other references to how to use this chip.
Hmm... how to mount on a ceramic board. Is it difficult to achieve that 1.25w
in real-world terms. How many designers use ceramic boards in their designs?
Thanks,
--
DaveC
me@bogusdomain.net
This is an invalid return address
Please reply in the news group
| |
| Genome 2007-03-13, 1:25 pm |
|
"DaveC" <me@privacy.net> wrote in message
news:0001HW.C21BB677016A5362F05096C8@news.sf.sbcglobal.net...
> RTQ035N03 in TSMT-6 smd 6-pin package:
>
> <http://www.rohm.com.cn/products/dat...f/rtq035n03.pdf>
>
> One pin each for source and gate. Four for drain.
>
> Is the number of drain connections simply for convenience of
> mounting/orientation? One wouldn't need four drains with only one
> source...
>
> Or am I missing something (which is my default presumption)?
>
> Thanks,
> --
> DaveC
> me@bogusdomain.net
> This is an invalid return address
> Please reply in the news group
>
To, not, clarify. Most of the power dissipated in your mosfet is dissipated
in the drain.
The chip is put into the package with the drain bonded to the main leadframe
and the main leadframe has most of the leads connected to it so you can get
the heat out.
How they think you do that is up to them and the marketing department, who
will win.
How you do it is up to you.
I've got problems with spelling dissapated dissipated but, fortunately,
that's not your problem. It's rather similar to your subsequent
question(s)......
DNA
| |
| Robert Baer 2007-03-13, 5:25 pm |
| DaveC wrote:
>
>
> Oh, you mean like this:
> "Total power dissipation: 1.25w*
> *Mounted on a ceramic board"
>
> I can find no other references to how to use this chip.
>
> Hmm... how to mount on a ceramic board. Is it difficult to achieve that 1.25w
> in real-world terms. How many designers use ceramic boards in their designs?
>
> Thanks,
If you were crazy to $pend the money, then you would be psycho-cramic
crack-pot.
| |
| Dave Platt 2007-03-13, 5:25 pm |
| > Oh, you mean like this:
> "Total power dissipation: 1.25w*
> *Mounted on a ceramic board"
>
> I can find no other references to how to use this chip.
>
> Hmm... how to mount on a ceramic board. Is it difficult to achieve that 1.25w
> in real-world terms. How many designers use ceramic boards in their designs?
You use ceramic boards in your design when you really need high power
densities! For designs where you can run your parts at a lower
dissipation per square, you can get away with less exotic substrates
like FR4.
Consider the size of a 2-watt through-lead resistor (which is probably
what you'd want to use when actually dissipating 1.25 watts), and the
space needed to mount it up a bit above the board to allow for
adequate air circulation. Consider how warm it's going to get.
Now, consider all of that heat coming out of a little SMD package,
with no free-air under-chip circulation, being carried away mostly by
radiation and by conduction into the PC board traces. That chip's
going to get hot!
I've seen some of the less-expensive PC-board material be singed
pretty badly from the heat of a TO-220 7805 linear regulator
which was screwed down (with its heatsink) directly to the board.
Even with a couple of square inches of heatsink sticking up into the
air, its several watts of heat dissipation burned the board. The
heat-per-square density of a 1.25-watt SMD is likely to be even
higher, and thus mandates the use of a high-temperature-tolerant board
substrate.
As I understand it, it's not necessarily any harder to use a ceramic
board than a glass/epoxy board. You just tell your fab house which
substrate to use, and use a somewhat larger wheelbarrow to ship them
the money :-)
--
Dave Platt <dplatt@radagast.org> AE6EO
Friends of Jade Warrior home page: http://www.radagast.org/jade-warrior
I do _not_ wish to receive unsolicited commercial email, and I will
boycott any company which has the gall to send me such ads!
| |
| Lostgallifreyan 2007-03-14, 3:25 am |
| dplatt@radagast.org (Dave Platt) wrote in
news:12ve5gp78bukn90@corp.supernews.com:
> As I understand it, it's not necessarily any harder to use a ceramic
> board than a glass/epoxy board. You just tell your fab house which
> substrate to use, and use a somewhat larger wheelbarrow to ship them
> the money :-)
>
Off topic to FET's but how much, roughly, for small quantities of maybe 2"
x 3", single sided? And can thickness be had at 1/8"? This could be a cost-
effective method for mounting electronics and optics on a single board for
laser designs where it is critical that thermal stability is high, and
expansion is low. I've seen it used in the Coherent C315M laser, and it
looks like a good way to do it. It seems to allow high quality for making
small numbers of devices by hand, and an easy way to set and reset
alignments at will, so even if it costs a bit, it might save far more than
it costs.
| |
| Lostgallifreyan 2007-03-14, 3:25 am |
| "Genome" <genomerics@ieee.org> wrote in news:GKBJh.943$Lz4.146@newsfe7-
gui.ntli.net:
> I've got problems with spelling dissapated dissipated but, fortunately,
> that's not your problem.
Good, cos 'dissipated' is the way to go... Can we enjoy enough dissipation?
I think not. Cut me some SLACK. >
| |
| Lostgallifreyan 2007-03-14, 3:25 am |
| "Genome" <genomerics@ieee.org> wrote in
news:GKBJh.943$Lz4.146@newsfe7-gui.ntli.net:
> The chip is put into the package with the drain bonded to the main
> leadframe and the main leadframe has most of the leads connected to it
> so you can get the heat out.
>
> How they think you do that is up to them and the marketing department,
> who will win.
>
> How you do it is up to you.
>
I haven't seen this thing, so I don't know if my suggestion has any merit,
but just possibly... TEC. The small ones often found in modest quantites on
eBay. Small high pump density devices metallised both sides are often
custom made, so surplus turns up at times. If you could remove the tinning,
and somehow protect the body of the device in peelable rubber, you could
etch tracks and such onto the top side for mounting stuff. That gets you a
cheap ceramic heatsink AND a pump. And an easy way to mount the parts,
if you have a temp-controlled iron.
| |
| Robert Baer 2007-03-14, 3:25 am |
| Dave Platt wrote:
>
>
> You use ceramic boards in your design when you really need high power
> densities! For designs where you can run your parts at a lower
> dissipation per square, you can get away with less exotic substrates
> like FR4.
>
> Consider the size of a 2-watt through-lead resistor (which is probably
> what you'd want to use when actually dissipating 1.25 watts), and the
> space needed to mount it up a bit above the board to allow for
> adequate air circulation. Consider how warm it's going to get.
>
> Now, consider all of that heat coming out of a little SMD package,
> with no free-air under-chip circulation, being carried away mostly by
> radiation and by conduction into the PC board traces. That chip's
> going to get hot!
>
> I've seen some of the less-expensive PC-board material be singed
> pretty badly from the heat of a TO-220 7805 linear regulator
> which was screwed down (with its heatsink) directly to the board.
> Even with a couple of square inches of heatsink sticking up into the
> air, its several watts of heat dissipation burned the board. The
> heat-per-square density of a 1.25-watt SMD is likely to be even
> higher, and thus mandates the use of a high-temperature-tolerant board
> substrate.
>
> As I understand it, it's not necessarily any harder to use a ceramic
> board than a glass/epoxy board. You just tell your fab house which
> substrate to use, and use a somewhat larger wheelbarrow to ship them
> the money :-)
>
I find that Megtron 6 holds up rather well at high temperatures and
it is a heck of a lot less expensive than ceramic and certainly not brittle!
|
|
|
|
|