in need of plastic IC packager

Sep 12, 2014 30 Replies

If it more than doubles the yield, it may reduce the price overall, even if you're only bonding out a few. This would apply particularly if the high aspect ratio causes latent issues (can a die be partially cracked, like a PCB?).

Do you want to spend your time sorting out which chips are good and which are bad, and which problems are design issues and which are cracked chips? Or do you want to spend your time turning the design into money?

Tim Wescott Wescott Design Services http://www.wescottdesign.com

If he is doing MOSIS, he is probably not doing more than a handful. It's not really that bad, i have seen LCD controller close to that ratio (8:3). Once they are mounted on glass or PCB, they are rock solid.

There are cheap way and expensive way to "custom bond" this special die. Cut and modify some existing wire frames for the cheap way (3K$ to 5K$). Or custom mold new wire frames for the expensive way (perhaps 10K$ to 15K$).

How would increasing the die size increase the yield?

Rick

If wider doesn't crack?? ...Jim Thompson

| James E.Thompson | mens | | Analog Innovations | et | | Analog/Mixed-Signal ASIC's and Discrete Systems | manus | | San Tan Valley, AZ 85142 Skype: skypeanalog | | | Voice:(480)460-2350 Fax: Available upon request | Brass Rat | | E-mail Icon at http://www.analog-innovations.com | 1962 | I love to cook with wine. Sometimes I even put it in the food.

We are planning on making 120 pieces on this run. It will never go into "production", but we might want more in the future. It is a custom design for a physics experiment.

There are now open-cavity plastic packages which are similar to what you describe. They can then be filled with something a lot thinner than the traditional epoxy. That is a possibility, but I don't think MOSIS offers them with enough pins.

Jon

be doing ever, or is it to prove the design so that you can then go to prod uction?

"production", but we might want more in the future. It is a custom design for a physics experiment.

OK, a bit more than a handful. So, probably need custom rectangular wire-f rames.

We did custom ASIC packing 1,000 chips before. But they insist on making 1

0,000 wire frames. So, we took 5,000 and exchange 5,000 for beers with a m etal recycler. Took 1,000 back to the US and kept 3,000 in China. Due to d esign changes, we are not going to use the rest anyway. But it was enough to get our first prototype out.

Anyway, it would be same price for 120 or 1000.

I'm not at all familiar with chip packaging. Why would the chip crack because of the aspect ratio? Where are the forces coming from and why are they... what, maybe "unbalanced" more for the "un"square chip?

Rick

The die bonding process (usually) involves heat... thus there are rules about aspect ratio.

In this particular case... assuming not too much chip dissipation, and a small "study" sample size, maybe just carefully epoxy them down.

However the original query was about wirebond interference angles. ...Jim Thompson

| James E.Thompson | mens | | Analog Innovations | et | | Analog/Mixed-Signal ASIC's and Discrete Systems | manus | | San Tan Valley, AZ 85142 Skype: skypeanalog | | | Voice:(480)460-2350 Fax: Available upon request | Brass Rat | | E-mail Icon at http://www.analog-innovations.com | 1962 | I love to cook with wine. Sometimes I even put it in the food.

pure guessing, but I'd assume that wider simply means stronger so that when you stay square-ish the forces and the strength balance

-Lasse

ecause of the aspect ratio? Where are the forces coming from and why are t hey... what, maybe "unbalanced" more for the "un"square chip?

bout aspect ratio.

I don't believe that's too much of a problem

small "study" sample size, maybe just carefully epoxy them down.

Yes, just glue it down well.

That would be a problem if you try to fit an rectangle in a square. Common wire-frames are mostly square, but rectangular frames are possible. Howev er, there are so many possible combinations of X & Y, it is difficult to ma ke it general stock item. So, the industry simply force everybody to be sq uare. But if you got the rectangular green papers, shape is not a problem.

I didn't want to get into it, but there are all sorts of ways to relieve stress on a chip. Look at the image sensor line arrays. They have an extreme aspect ratio but are still manufacturable.

Unless you need ohmic contact to the back of the die, epoxy can be used. If you need thermal performance, there is glass bead.

Solve your bonding issue first. ;-)

Packaging is a pain since you are usually dealing 3rd party. Your packaging engineer deals with somebody in ROK (as opposed to the DPRK).

Most of the time, packaging is done in a stamped lead frame. For low volume, you can get a custom etched lead frame, or use some "standard" etched lead frame if the vendor will supply it. [Standard as in the photo work exists for it, but the package is not really standard.)

Join the Discussion

Have something to add? Share your thoughts — no account required.

Didn't find your answer?

Ask the community — no account required