13
\$\begingroup\$

I still haven't found a really good solution for grabbing tiny surface mount stuff.

  1. Alligator clips are definitely way too big.
  2. Mini-grabbers like this with the spring-loaded hook are an ok solution for bigger stuff, but no good for little surface mount components. I can solder on little bits of wire and then mini-grab those, but it's tedious and time-consuming. And in my experience they tend to be broken more often than not:
    minigrabbers

  3. I got some Tektronix KlipChips and some HP 5090-4356 clips on eBay.
    Tektronix KlipChips and HP 5090-4356
    The Tektronix clips are hard to open, because the plastic head tips to the side and jams instead of sliding down when you try to squeeze it. The HPs don't have that problem. The Tek pincers are made of wire, so they splay out sideways instead of grabbing well (and get permanently bent, making them jam on each other and not close completely, so connections to thin wires are intermittent).
    Wire tips jammed together

The HP pincers are made of thin strips of metal, so they are more rigid in that direction. The wire pincers depend on the bendiness of the wire to overlap each other side-by-side (which doesn't work very well) while the HP pincers fold inside each other, with one a little shorter than the other. This seems to work better, but they get bent outward, and the tubes are plastic, so they get bent outwards by any angular stress and no longer close well.
Tip splayed open
The orientation of the pincers on the HP relative to your fingers is more natural.

These work on SOICs or bigger, but after they've become worn they don't stay on as well. They don't work for surface mount resistors (they just pop off if disturbed even a little) or smaller-pitch ICs.

And I've broken the retractable hook tip for oscilloscope probes by attaching it to a circuit and then letting go, and the weight of the bulky probe and wire put too much stress on the plastic tip and bend it.

Ideally I'd like something that can grip one side of a surface mount resistor, or a pin on a smaller pitch IC, not fall off, not produce any stress on anything, and not break easily. Any advice, ideas, tips, tricks, ingenious alternatives, cheap Chinese knock-offs of better clips?

(Previously asked on xkcd forums and adafruit forums.)

\$\endgroup\$
14
\$\begingroup\$

If you're designing your own boards, add test points. Aside from that, soldering on short bits of wire is the best solution I've found so far. I find that the amount of time it takes to strip and solder a wire is far less than what I'd otherwise waste futzing with clips or trying to hold a probe just so while also keeping an eye on a scope.

\$\endgroup\$
  • 1
    \$\begingroup\$ Soldering wire (usually 30 AWG Kynar/wire-wrap wire) leads to the board is compact, handy, and almost always do-able (if you use leadless packages you might have to scrape solder resist off a trace). However, as the OP mentions, some probes can't hold fine wire well, but at least you can abuse the other end of that wire all you like. \$\endgroup\$ – Nick T Nov 11 '10 at 6:39
  • 1
    \$\begingroup\$ Yeah, putting a loop in the wire makes it easy to grab reliably with bigger clips. \$\endgroup\$ – endolith Nov 11 '10 at 14:38
  • 1
    \$\begingroup\$ One way to quickly get a bunch of wires is to take apart a stranded cable. \$\endgroup\$ – ryantm Nov 12 '10 at 1:11
7
\$\begingroup\$

I've found these SMD test clips quite effective. They look similar to the Pomona clips. I haven't broken any yet.

enter image description here

\$\endgroup\$
  • \$\begingroup\$ And they're only $16 each! \$\endgroup\$ – endolith Nov 11 '10 at 17:42
  • 1
    \$\begingroup\$ I have some of these, they are quite nice for the price. They work on 0.5mm pitch, but not on adjacent pins. \$\endgroup\$ – starblue Nov 11 '10 at 19:56
  • \$\begingroup\$ Link seems to be broken. \$\endgroup\$ – Technophile Feb 27 at 19:04
6
\$\begingroup\$

Are you asking about debugging others' boards or your own?

If you have the ability to control the PCB, some test pads are an easy thing to add. Just a little rectangle of copper makes a world of difference in avoiding shorts, damaged traces, and the like.

Even a small pad can be poked reliably with pogo pins or an oscilloscope probe. However, if you want to be able to clip to the connection, either solder a little loop of wire onto the point, or get some of Keystone's test points[pdf]. They have SMT ones as small as .1"x.04" which are easy to put down and then remove for later use, and the through-hole versions clip and unclip in a .04" or .063" diameter hole. Just enlarging a few vias on your main communication busses in case you need to plug one of these in is an easy way to protect yourself.

If you're working with boards over which you have no control, a Chip Clip is often helpful. The plastic jaws grip the underside of the chip (between and below the leads), and there are contacts for the chip. It looks like this (8-pin SOIC clip shown):

alt text

It breaks the contacts out to .1" headers. They're available for PLCC and SOIC packages, and I think TSSOP as well. They require a chip with exposed leads and a grippable body, so that rules out QFPs, BGAs, and rectangular passive devices, but those are nearly impossible to clip even with other methods.

I recently had to reprogram several hundred SPI flash chips and didn't want to take the time or money to solder the required computer interfaces onto all those sparsely populated boards. So, I removed the chip from one board (which had the interface), ran wires from the empty pads pads to chip clips, held the micro on the new board in reset (It was the only thing connected to the SPI bus), and went to work. I went through a few clips in the process, but it was nowhere near the cost of the magnetics.

\$\endgroup\$
  • \$\begingroup\$ I don't really have control over the PCBs. I suppose I could request that they add test points, but I don't know how successful I would be.... I've seen those test point loops on eval boards. They work well. I have a Chip Clip, and did exactly the same thing to program maybe a hundred EEPROMs by removing the EEPROM from the eval board and connecting the chip clip in place of it. :D \$\endgroup\$ – endolith Nov 11 '10 at 19:09
3
\$\begingroup\$

I haven't tried it myself, but I've seen insulin syringes used. I think the idea is that they're sharp enough to catch on pads and components.

http://travis.frob.us/~travis/public/blog/images/syringe/probe_un.jpg

\$\endgroup\$
  • 1
    \$\begingroup\$ I do something very similar using a dental-pick. They're sharp, have a nice right angle which makes them easy to hold, and are generally stainless. You just clip a test lead to the bottom, and poke away. \$\endgroup\$ – Connor Wolf Nov 11 '10 at 10:54
  • \$\begingroup\$ Also, they're great when doing SMT soldering. \$\endgroup\$ – Connor Wolf Nov 11 '10 at 10:55
  • \$\begingroup\$ I'm not sure insulin comes in that "form factor" any more. My wife is diabetic and uses these type of pens: commons.wikimedia.org/wiki/File:Insulin_pen.JPG \$\endgroup\$ – Amos Nov 11 '10 at 22:25
  • \$\begingroup\$ @Amos I bet you could also get free syringes from a "needle exchange" for heroin users :-) \$\endgroup\$ – Toby Jaffey Nov 11 '10 at 22:30
  • \$\begingroup\$ Sewing pins aren't good? \$\endgroup\$ – endolith Nov 12 '10 at 5:50
3
\$\begingroup\$

The options i've seen are:

  • Solder a blue wire onto the pin you need to look at (what i usually do)

  • Both agilent and tektronics make probes with solder-able, replaceable tips, expensive

  • Use a 'third arm' device, tektronics and agilent make ones specific for scope probes but they are expensive. However you can get ones designed for chemistry work that are just as effective in my opinion and are much cheaper. Examples here and here. This holds an advantage for testing signals in potentially noisy areas of the circuit by limiting both the length of wire from test point to probe and keeping the probe/probe wire as far from the circuit as reasonably possible.

The best solution is to include test points, or at least a via big enough for a blue wire on a signal you know you'll need to test. Some times this obviously isn't possible. I always at least include a spot where i can hook to a decent ground.

\$\endgroup\$
2
\$\begingroup\$

Those Pomona grabbers starblue posted are great and work very well but they break very easily and you would have to support most of the weight of the grabber or it will come of the pin. I'd say they are not worth it for home use unless you are extremely careful and use them sparingly. They do work very well once you get one on properly. Unfortunately I don't know of anything else except for soldering or holding a probe on the pin as an alternative.

\$\endgroup\$
  • \$\begingroup\$ Yeah, everything I've seen is very expensive and very fragile. :/ \$\endgroup\$ – endolith Nov 11 '10 at 14:38
2
\$\begingroup\$

For passive components there are small spring-loaded plier-like probes with two 0.6mm contacts from Hirschmann.

enter image description here

For IC pins there are some really small ones from Pomona 6351/6352/6353, though the price is inversely related to the size (or maybe the square of the size?), so these are really expensive. I've used two of these on adjacent pins with 0.5mm pitch (unfortunately they weren't mine).

A good search term is "grabber".

\$\endgroup\$
2
\$\begingroup\$
  1. Get a header of 0.1" pins.

  2. Attach this to your prototype board with hot glue or double-sided tape, securely enough that you can attach test probes without it pulling off. Or glue on a suitable piece of protoboard and solder the header to that.

  3. Get some small gauge enameled magnet wire and connect each node you are interested in to the header.

  4. Get a paper label of appropriate size and label all of the debug connections.

Obviously, this is no good for RF or precision analog or very high-speed digital circuits. For those things a very short ground path is needed.

I have lately switched to Digikey part S5493-ND (picture follows). This mates with the Open Workbench LA. Or I can plug a normal pin header in for oscilloscope connections or whatever.

enter image description here

\$\endgroup\$
  • 1
    \$\begingroup\$ Wire-wrapping wire is easier to work with than enameled wire, I think \$\endgroup\$ – endolith Sep 8 '11 at 16:27
  • 1
    \$\begingroup\$ No matter what stripper I use, wire wrapping wire seems to eventually break where the strippers nick the wire. This has cause frustration. Enameled wire is "stripped" with heat from the iron, so no nick to concentrate stress. And enameled wire is available in a much wider range of gauges. The fine gauges can be soldered (with the aid of a microscope) to dodgy things like 0.006" traces and QFN pads. (There's never a test point when you need one.) Even if you get wire-wrap wire soldered on a skinny trace, bending it can transfer enough force to damage the trace. \$\endgroup\$ – markrages Sep 8 '11 at 16:33

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.