Hackaday Links: August 29, 2010

Hotel room door lock picking

Here’s further proof that you should never leave anything of value in your hotel room. We’re not worried about someone getting in while the room is occupied. But these methods of defeating the chain lock and opening the door without a keycard (YouTube login required) do show how easy it is for the bad guys to steal your stuff.

iPhone frequency generator

Need one more way to make that iPhone a useful lab tool? Why not use it as a frequency generator. Start with a free app and mix in an audio cable with test leads and you’re in business.

Drag Soldering

[Andrei] sent us a link to a video about drag soldering. This is a method of soldering fine-pitch chips using a small bit of solder and a fat solder tip. The link he sent is dead now but we found another great example of the process. We were just using this method earlier in the week to solder a TSSOP38 package for an upcoming project and it worked like a charm.

Laser etched PCB

Here’s some art in PCB form thanks to a laser. We thought this might be interesting to share after seeing those art pieces made from old circuit boards. This example is laser etched, but not directly. As you probably guessed, the copper clad board is coated with resist and the laser etches some of it away. Whatever got zapped by the laser dissolves when the board is placed in acid, leaving [Riley Porter’s] art behind.

36 thoughts on “Hackaday Links: August 29, 2010

  1. @Osgeld

    Agreed, most SOIC and the thick width leads aren’t bad to do drag soldering with. The problem comes into play when you have something like a 100pin TQFP package. Doesn’t take much to destroy those pins.

  2. Drag soldering is best for small pins. My very first SMD part was a 80 ld tqfp with .5mm pitch, and it was drag soldered.

    Some bridges come on the end pins, but dragging back again usually fixes it.

  3. @Osgeld: You obviously don’t know how to solder properly.
    I’ve drag soldered ARM μCs in 144-pin LQFP packages and have never bent any leads.
    The key is a good soldering iron, 0.30mm solder and good liquid flux.

  4. He isn’t seriously advising anyone plug their smartphone into an electronic circuit like that, is he? It is one thing to drive a speaker like he did in the example, but if you actually try and connect that up to a device you are testing or a circuit in development…

  5. @Myke:

    You are correct. The trick is to use a good temperature controlled iron, fine gauge solder(practicing to only feed *just* the right amount), and *generous* amounts of good liquid flux(but not TOO much). For extremely fine-pitched packages, you can solder a few pins at a time(however many the width of the soldering iron tip will reach) by dragging away from the pins instead of perpendicular to them. With good flux, and good precision hand control, you really only need to touch the flux, and not the pins themselves.


    Hopefully, anyone that would need to use this would construct a simple buffer/isolation circuit for use when driving circuits that will sink more current than the phone can source. Hopefully… :)

    Also, MS3FGX, what does your name mean?

  6. “@Osgeld: You obviously don’t know how to solder properly.”

    actually yes I do, grats on accomplishing 1 80 pin qfp, let me get the red carpet out and throw you a parade

    but still doesn’t negate the fact, no matter how good your soldering tools are (and I have some fine soldering tools) an extra ounce of pressure and your fiddling with trying to separate a half dozen pins soldered to 2 pads

    its not impossible to fix but by the time you screw with it, you could have soldered the entire package 1 pin at a time and been better off

    mileage may very, its not a sure fire way

  7. @thatguy:

    If I am not mistaken, piezo elements tend to be capacitive loads, rather than resistive, so you wouldn’t have to worry about excessive current draw, just a lack of performance. Speakers, on the other hand, are resistive/inductive loads, and if you use the wrong speaker(2Ω speaker on a circuit meant to drive a 32Ω speaker), you could burn out the final-stage amplifier, depending on it’s design. I do have a feeling, however, that the outputs that are coupled with series capacitors and/or resistors that would allow a full short to ground without negative effects, but I could be wrong…

  8. -outputs that are
    +outputs are

    Sorry… :) Ability to edit comments without registration, based solely on email address and IP, anyone? Yea, on second thought, that may easily be spoofed… :(

  9. @Osgeld:

    LOL, yea, you’re right, if you get too much solder in between the pins, it can be “hell” to get out with just flux, and, in trying, you would likely destroy either the chip and/or the traces, and possibly de-laminate the board and separate the vias(in a multi-layer circuit board). In those cases, I usually either remove the chip and start over, or use more flux and some solder wick, whichever is more appropriate for the situation. Vacuum based de-soldering systems, will, of course, not help in this situation…

  10. not only solder tween the pins, just a chunk o crap gets on your tip and your applying a tad too much pressure you can very easily drag pins onto each other by bending them, it only takes a second

    I have successfully soldered 68k cpu’s in qfp pagages getting into the much higher range of pitch and pins than the “examples” given before, but 1 wrong slip and it WILL “ruin a fine day”

  11. A simple idea I have used in the past for fine stuff is to flux everything up, then saturate the end of some solder wick, then “hot mop” with the “grain” of the leads

    the wick soaked in hot solder is much softer than a iron tip, and you have less of a chance of bending leads together if your in parallel with them

  12. @Noobius:

    It’s not impossible, but, yea, it’s alot more difficult without a mask. I those cases, though, you would likely be able to use reflow if you are building your own circuit(for initial placement, and not replacement)…


    Don’t “sideways-sweep” fine-pitched QFPs, pull away, like I said. You can “side-sweep”, but you must be careful to use the flux as your thermal conductor, and not actual contact with the pins. Otherwise, try pulling away from each set of pins to solder the chip(after you have tacked down the corners) using lots of flux, and if you don’t apply too much solder, you should never have bridges. I have replaced dozens of 304 pin QFPs using this method. FWIW, removing the old chips(I believe mentioned here before) for those high-density packages can be easily done with an Exacto knife with a dulled down tip, by rocking and cutting your way through the top of the pins, and then removing the legs with a soldering iron after the epoxy chip casing has been removed…

  13. Brilliant. This type of soldering was the way they did things at my last job, 7 years ago. In that time this it the first time I’ve seen a proper explanation of it and I’ve never understood it until now.

    I can attest to the fact that it works very well on fine-lead packages and it’s amazingly fast. They used a Weller cup tip and brush-applied liquid resin. The engineer swept perpendicular to the pins but never bent anything – probably because you’re not meant to press down on them as you go.

    I believe most of the solder involved in the joint was just the pre-tinned PCB pads being wicked up.

  14. About three months ago I started working for a company that re-manufactures industrial controls.

    When I was shown this technique the first time I thought it was madness…until all the connections “resolved”.
    It’s pretty damn neat!
    Fast too once you get it down.

  15. If you drag the solder tip along the pins, not across the row, in small strokes, from centre of the chip towards the outer side, you can’t bend them. It’s a mix of drag soldering and regular soldering, you can’t really solder the entire row with one elegant move, it is a slower method but it seems to work for me.

  16. Those hotel tips are useless in most hotels now. when you flip the deadbolt lock it disables the handle from turning. That trick is only useful for breaking into a room with very old door locks. the last 8 hotels I stayed in during the past 10 months made you turn the knob 90 degrees to open the door, that ricg is incapable of turning the door handle more than 20 degrees.

  17. Not worried about your room when it’s occupied? My wife and I had a hotel room for a weekend of a wedding. In the middle of the night someone opened the door. We sat up, bleary-eyed as the intruder apologized and left. Still don’t know if it was an honest mistake by a guest or staff (bad / dupe keys) or a potential thief. However, when I’ve got a room, I now lock every last bolt, chain, etc. Especially those solid brass arms on the jamb that catch the knob on the door.

  18. I used to do this on 208 TQFP parts, but instead I used a conical hook tip on a Metcal and ran the rounded edge along the joint. Regardless of what you use, the secret to success is an iron with fast recovery and lots of flux.

  19. @Osgeld:
    “actually yes I do, grats on accomplishing 1 80 pin qfp, let me get the red carpet out and throw you a parade”

    Ofcourse you do. That also explains why you have trouble with bent leads.
    And when I said μCs, I meant microcontrollerS in plural. Countless ARM7s we used for development purposes at our company. And guess what: I’ve always drag soldered them and never ever did I have to fix a bent lead.

    By the way. You don’t really need to touch the leads. You just drag your soldering iron at the ends of the leads, over the pads. The solder will cover the pads and come in contact with the leads of the chip. It will immediately heat them up and flow around them. When you get to the end of the row, you just wick up the excess solder, if necessary.

  20. @hitek146: I’ve got an old Weller temp controlled iron at home, and have used some Hakko stuff, but nothing I’ve ever used heats up or bounces back as fast as the Metcals.

  21. I thought google checked if a video was ‘flagged’, so how the hell do they judge that lock-hack video ‘not suited for children’?
    Is that some sort of joke or just a gimmick youtube uses to try to get people to sign up? Or a gimmick the maker of the vid uses to gather more data on viewers? Or just to get them to watch by making them curious.

  22. I’ve been soldering ICs like this for years and it definitely works. Didn’t realize that not everyone was doing it this way. I have to agree with other posters that you can damage the pins if you aren’t careful. I find the easiest way is to focus on moving the solder across the pins more than applying pressure to the pins themselves. Think of it as using the iron tip to “roll” the solder ball over the pins where they meet the pads. You can actually keep the iron tip from right above the pins if you have a good clean tip and a nice temperature controlled station. Add flux and move quickly.

  23. Android has a similar sig gen app, I used it to set up an interferometer in an optics lab a few weeks ago:
    Infa red interferometer (http://en.wikipedia.org/wiki/Michelson_interferometer) with output feeding into photo-diode -> oscilloscope. Then do a fast Fourier transform, to see component frequencies (of mirror vibrations etc).
    It was easy to place the phone on the optics bench and see the oscilloscope FFT trace respond with a peak at the input freq.

  24. @Osgeld

    You must not be very good, then. I’ve never heard it called “Drag Soldering”, but I’ve been using it for about 10 years to solder very fine pitch packages, with a 100% success rate. Try using flux, it works wonders! I can solder a 48TSSOP in ~30 seconds with no bridges, all good joints.

  25. @Osgeld

    Also, if you *somehow* get solder “in between” the leads (I assume you mean a bridge), and aren’t proficient enough to remove it by hand, use some high quality solder wick. It’s gone in a snap.

Leave a Reply

Please be kind and respectful to help make the comments section excellent. (Comment Policy)

This site uses Akismet to reduce spam. Learn how your comment data is processed.