Page 1 of 2

MELF

Posted: Mon Nov 02, 2015 3:00 am
by mcs_5
I don't use the very small passives everyone else seems to use (0805 is the minimum), but I do use a lot of 0204 and 0207 MELF resistors. Does the LitePlacer come with a tool that will handle those?

And how would placing something like MSOP-10 ICs work? I try to stay with 0.65mm pin spacing or larger, but LT likes 10-pin MSOPs (0.5mm spacing) a lot for voltage regulators and references...

Re: MELF

Posted: Tue Nov 03, 2015 2:46 pm
by Spikee
I have done 0.5mm qfn and 0603 with good accuracy. No problem.
I put the chips in the pickup location , the component filter detects the center and the rotation offset.

0201 (imperial) is pretty high end. 60-80K pnp have troubles doing these reliably.

I would just try and see what happens.

Re: MELF

Posted: Tue Nov 03, 2015 3:06 pm
by mcs_5
The resistors I'm talking about are quite big. A 0204 MELF is approx. the size of a 1206, and the 0207 MELF is about a 2010. The potentially difficult part is that they are cylindrical rather than flat...

Re: MELF

Posted: Tue Nov 03, 2015 3:22 pm
by Spikee
the difficulty will be in picking them up , not placing them.
Juki and the likes have custom nozzle for this:
http://www.nozzles4smt.com/Juki-Nozzle- ... _1452.html

This is just something that you need to try out

Re: MELF

Posted: Tue Nov 03, 2015 3:55 pm
by JuKu
The needles that come with the machine are flat. If you file them, I would guess you get enough suction. I haven't tried if you actually can, but with a sharp file, I think it could work. I don't have any MELFs so I can't try it myself right away.

Re: MELF

Posted: Tue Nov 03, 2015 4:44 pm
by mrandt
Depending on part size, it might help to put a small rubber or silicone suction cup onto the needle.

Simple "vacuum IC pickers" often come with an assortment of these cups.

Examples:
http://www.ebay.com/sch/sis.html?_nkw=V ... Pen+IC+SMD

Re: MELF

Posted: Tue Nov 03, 2015 4:55 pm
by Spikee
I also put the smallest one in the set that came with the liteplacer. It creates some kind vacuum buffer that seems to help with sucking up components. The cup is not even touching the passives or whatever but it still works.
This might do Ok with MELF. But if it is possible it would be easier to change your design to the "normal" rectangular design.

Re: MELF

Posted: Tue Nov 03, 2015 6:03 pm
by JuKu
mrandt wrote:Depending on part size, it might help to put a small rubber or silicone suction cup onto the needle.

Simple "vacuum IC pickers" often come with an assortment of these cups.

Examples:
http://www.ebay.com/sch/sis.html?_nkw=V ... Pen+IC+SMD
The kit comes with one of these pens, because of the cups.

Re: MELF

Posted: Tue Nov 03, 2015 9:53 pm
by mcs_5
Spikee wrote:But if it is possible it would be easier to change your design to the "normal" rectangular design.
Low-noise resistors are not cheap - unless you use the MELFs. Bulk metal foil resistors are great - but the price difference would probably make it cheaper to buy a production-grade PnP machine ;)

Re: MELF

Posted: Wed Nov 25, 2015 11:28 am
by JuKu
I got some MELFs in the mail last week. Thank you, that made for an interesting experiment. Unfortunately, the results were disappointing:

Using a regular needle, the pickup was reliable, but the contact area to the part was so small, that it rotated during flight. Using a rubber cup didn't work either: If the cup touched the part good enough to provide friction, it also pushed the part away from an already leaky connection, making for an unreliable pickup. So, for short term, the ruling is that LitePlacer can't do MELFs.

How to solve this? A (far) more powerful pump might do it. I don't have one to try immediately. Once I get one, I'll come back to report. The true solution would be a speciality nozzle, like this http://www.smtsky.com/products_detail/& ... 18639.html or the leftmost nozzle on the bottom picture of this page: http://www.nozzles4smt.com/blog_archive_3_2011.html. Even making a V shaped tip from plastic tube (like unshrinked heat shrink tube) might do it. However, this would need a homing function for the A axis, which the software doesn't currently have.

[The A homing needs to be done anyway, an automatic nozzle change (Which I'm looking at) can't work without it. This could be done either optically (with its own complications) or with a hardware switch, such as http://fi.farnell.com/multicomp/m140t01 ... dp/1735353 and a shaft collar. That requires an additional wire, though. This needs some more thinking.]