At one place I worked, for pin sampling, we used heavy-walled borosilicate capillary tubing instead of those vacuum tubes. A little awkward at first but, when I got used to it, I liked it much better than the vac tubes. I'm not sure of the dimensions. I think it was either the 3rd or 4th one in this chart. 1.25-1.75mm ID/7-8mm OD or the 1.75-2.25mm ID/7-8mm OD. Expensive in bulk but only about $7 per 4' length. There are probably smaller lab supplies that sell the 4' pieces individually.
https://us.vwr.com/store/catalog/product.jsp?catalog_number=89003-302
To use these, we put about a 6" length of latex tubing on one end of the glass tube. The other end of the latex tubing attached to the end of a bulb-type turkey baster. You squeezed the bulb flat, plunged the other end of the glass tubing below the slag into the metal, released the bulb, and then quickly withdrew it. Like the vac tubes, this all has to be done fast or the glass will soften and start sagging. About 3" of metal was sucked into the glass tubing. The end of the glass tubing was then laid on a steel table, gently tapped to break off (as evenly as possible) the glass a little above the pin, then the glass containing the pin was broken off of the pin. The glass tubing left over was re-used until it got too short to work with (about 12"). I would guess we got about 8-10 pin samples per tube. The jagged end of the glass tubing didn't interfere with the sampling.
The awkwardness, of course, was in handling a 48" length of glass tubing. After doing it 3 or 4 times, though, it became easy.
Here are the vac pin tubes. When I buy these, I get the smallest diameter.
http://www.lmine.com/mm5/merchant.mvc?Screen=PROD&Store_Code=LMS&Product_Code=1829V&Category_Code=