Prev: ModelSim XE III error
Next: Xilinx Synthesis Tool generates clock signals from combinatorial logic
From: Nial Stewart on 13 May 2010 06:20
15 years ago a 2x5 pin 0.1" pitch through hole programming header wasn't
excessively large, these days it's a bit of a joke, worse if you use a
surface mount part.

It would be good to have a more compact 'standard' surface mount programming
header.

I've used Molex Picoblade vertical headers and connectors reasonably sucessfully
but these probably aren't robust enough for high volume operation (it's only
rated at 30 mating cycles though I've had a lot more out of it).

Has anyone any better solutions?


Come on Altera (and the rest), give us a standard.



Nial.




From: Rob Gaddi on 13 May 2010 11:36
On 5/13/2010 3:20 AM, Nial Stewart wrote:
> 15 years ago a 2x5 pin 0.1" pitch through hole programming header wasn't
> excessively large, these days it's a bit of a joke, worse if you use a
> surface mount part.
>
> It would be good to have a more compact 'standard' surface mount programming
> header.
>
> I've used Molex Picoblade vertical headers and connectors reasonably sucessfully
> but these probably aren't robust enough for high volume operation (it's only
> rated at 30 mating cycles though I've had a lot more out of it).
>
> Has anyone any better solutions?
>
>
> Come on Altera (and the rest), give us a standard.
>
>
>
> Nial.
>

Some of the newer ARM devkits we've been using lately have come with 2x5
0.05" through hole instead. 75% of your surface area back is a pretty
decent victory.

--
Rob Gaddi, Highland Technology
Email address is currently out of order
From: Thomas Entner on 13 May 2010 12:03
With our EEBlaster (http://www.entner-electronics.com/tl/index.php/
eeblaster.html), we support a 2x3 2mm pitch header which uses just
about 1/3 of the area of the 2x5 header. We think this is a good
compromise of size, price, reliability and availability. We have the
pinout made public on the mentioned link, so everyone can use it,
either together with our EEBlaster or with a self-made adapter-cable.

Best regards

Thomas Entner

www.entner-electronics.com
From: rickman on 13 May 2010 12:46
On May 13, 6:20 am, "Nial Stewart"
<nial*REMOVE_TH...(a)nialstewartdevelopments.co.uk> wrote:
> 15 years ago a 2x5 pin 0.1" pitch through hole programming header wasn't
> excessively large, these days it's a bit of a joke, worse if you use a
> surface mount part.
>
> It would be good to have a more compact 'standard' surface mount programming
> header.
>
> I've used Molex Picoblade vertical headers and connectors reasonably sucessfully
> but these probably aren't robust enough for high volume operation (it's only
> rated at 30 mating cycles though I've had a lot more out of it).
>
> Has anyone any better solutions?
>
> Come on Altera (and the rest), give us a standard.
>
> Nial.

I've only once used anything other than the standard connectors for
programming where the devices needed to be very miniature and we had
some bucks to spend on the issue. The approach we took was to use a 1
mm pitch surface mount two piece connector and created a flex cable to
bring this connector out to a standard 0.1" header. Then we used all
the standard cables for programming, debug and even used this port in
automated testing (it had some additional capabilities). A good
connector for this is also Molex, called "Conan" (not the barbarian or
the comedian). I expect you could go as low as 9 pins on this
connector. Digikey carries them. Male: 91910-21109LF, female:
91930-21109LF. The male is available in different heights with a
mated height from 4.15 mm to 7.52 mm. The receptacle has a fixed
height of 3.28 mm, which is lower than the shortest header so that is
likely what you want on the target board. The outline is about 9 x 6
mm.

One of the really nice things about these connectors is that they have
a detent "click" when mated and do not tend to come apart from
vibration. I have used them myself for daughtercard mounting. They
serve both as a mounting method and as an electrical connector. I
have seen similar connectors that are smaller in all dimensions, but
not with a positive mated feel.

Using one of these as a "standard" programming header wouldn't be a
big problem except for the expense of making the adapter. A flex
circuit would cost quite a few bucks for the NRE. I suppose a cheaper
alternative would be a paddle board PCB. That could be inexpensive
although the flex cable is much more versatile.

Rick
From: Jon Elson on 13 May 2010 15:24
Nial Stewart wrote:
> 15 years ago a 2x5 pin 0.1" pitch through hole programming header wasn't
> excessively large, these days it's a bit of a joke, worse if you use a
> surface mount part.
>
> It would be good to have a more compact 'standard' surface mount programming
> header.
I use a 2x4 pin 2mm header for JTAG programming of Xilinx CPLDs. This
seems to work quite well, but maybe for chipscope or similar testing,
you need a couple more pins.

Jon
 |  Next  |  Last
Pages: 1 2 3 4 5
Prev: ModelSim XE III error
Next: Xilinx Synthesis Tool generates clock signals from combinatorial logic