Hackaday Links: Sunday, July 21st, 2013


Adafruit tears down a set of brainwave cat ears. They’re made by Necomimi and use your brain waves to adjust a pair of plush cat ears on the headgear.

If your desktop computer is sitting on the floor you may have damaged USB dongles by hitting them with your knees. [Megacier] prevents this from happening again by building a flexible dongle link.

Can anyone help [Brian Benchoff] find a datasheet for this International Rectifier 92-O350 so he can fix up his old VT100 terminal?

Here’s a quick example of how to graph data from a Raspberry Pi on the sen.se cloud service.

Have some extra fun with your oscilloscope by displaying any image. This set of conversions starts with a picture and ends with an audio file that will draw it on the scope’s screen.

You’ve probably already heard that the Sikorsky Prize for human powered helicopter has been claimed. If you didn’t see any footage of the flight now’s your chance. [Thanks Adam]

37 thoughts on “Hackaday Links: Sunday, July 21st, 2013

    1. Also I see no reason why Mike would think it labelled as O350 when it’s clearly a zero. That is just needlessly confusing, especially if you’re asking for help to locate it’s documentation. Punctuation in that case is critical!

        1. Yeah I looked at the images, it just seemed obvious to me that it was a zero and that a capital O would seem a little out of place for it’s location in a model/part number. Sorry, didn’t mean to be as ranty as it sounded. Just thought it was important.

    2. If the photo posted in reference to the part is correct, there’s no denying that it’s an International Rectifier part. The diode-in-a-circle symbol can be easily seen.

      I assume this big diode would most likely be found in the power supply. From Archive.org’s VT100 schematic (http://archive.org/details/bitsavers_decterminachematicFeb82_6949589), page 23, the only large TO-3 diode (D36) is listed as “UES2601 PIV=50V I=30A”. Datasheets for the UES2601 are available from Microsemi and show the device as a dual-diode package with either common-anode or common-cathode arrangement (depending on model). The decapped IR part doesn’t look to have two die, but the “-0350” in the part number would be consistent for marking of a 50V/30A device.

      1. Although… I originally missed the “E” and “B” designators stamped in the case.

        The schematic also references two TO-3 transistors, Q7 (a 2N6547 NPN) and Q310 (unnamed except for “NPN 125W SI250 25”. Which reference designator corresponds to the part in question?

        I will say, if it’s not D36, that’s an awful strange naming coincidence.

        1. I think your right it’s the 2n6547. From the prints listed previously I would suspect it is the switching transistor Q7. I haven’t seen many power diodes fail on the output of a switching power supply.

      2. The “E” and “B” stamped on the bottom make it a bipolar transistor, with the can being the collector terminal. If either junction is still working, you can determine whether it’s PNP or NPN by measuring with an ohmmeter from B to E or C. (Most bipolar power transistors of that era, unless being used as a push-pull pair like in an audio output amplifier, were NPN, because these were superior to PNPs in every respect.) Failing that, you can tell from the circuit: if the collector is connected to a positive potential, it’s NPN; if negative (with respect to E), then it’s PNP. You can use the polarity of any electrolytic cap connected to the collector for that information.

        For the Vce rating, look at what is supplying the circuit – you can tell from electrolytic capacitors in the circuit what the maximum amount it can be required to withstand is, UNLESS this is in the horizontal deflection circuit of the terminal, in which case it will be breaking current in an inductor, in which case it will have to withstand a great deal more than that. In that case, just use the fattest (i.e., highest voltage/highest current) TO-3 packaged transistor you can find that’s made specifically for flyback transformer drive in TVs and terminals. For the Ic rating, you can seldom go wrong by using a transistor with a higher-than-required Ic rating, so just go for the highest available within reason.

        There’s really nothing magic about bipolar power transistors – the circuits using them are usually very tolerant of differences in hfe (beta) or ft (transition frequency), so as long as you are conservative with the voltage and current ratings, you should be safe. The best news is that transistors keep improving over time, so if it was possible to find a transistor to do the job in 1980, it should be downright easy to do the same job in 2013.

        I ran into similar problems trying to identify an RCA-made transistor in a Wanlass power supply many years ago. I even wrote to RCA (this being before email existed), and they told me it was a part custom made for Wanlass, so they did not have and/or could not disclose the specifications, and Wanlass was out of business so I couldn’t ask them. I just estimated voltage and current ratings based on the circuit, and chose a more capable transistor than these implied, and was successful.

        It’s quite likely that this was a transistor made by IR to DEC’s specifications, so finding an actual datasheet may be improbable.

        1. Be careful: the E and B markings are stamped to almost all TO-3 cases whether they contain a BJT or not. Many LM309 (predecessor to the LM78xx i think) also have E and B markings.

    1. That looks like what we call an “in-house part number”. Companies who buy large quanties of parts can have them marked any way they want. Sometimes the in-house number has it’s own unique data sheet and is not sold to the public, though you can find a replacement that is equal to or better. The best way to figure it out is contact the company that made the terminal. Wyse?

  1. The cat ear teardown totally misses that Neurosky is in the business of marketing these low-end EEG sensors for various purposes. It’s not a “general purpose ASIC,” this is Neurosky’s business and they probably built the whole daughterboard.

    1. Probably the same Neurosky board that is in their other products, with the 4 pins being a data connection. Seeing that it is 4 wire, rs485 maybe? Might be a clue to what the Neurosky chips are capable of, since they seem reluctant to share that data with hackers.

  2. Flexible usb dongle.

    I think the term he’s looking for is “USB EXTENSION CABLE” – you can buy them everyplace in various lengths for just a few bucks/pesos/euros/lumps of silver.

    Plug them into the BACK of your computer, and route the cable up to your desktop. Use a banker clip to hold the cable in place for easy plugging and unplugging of your usb devices.

    I was going to make a Youtube video explaining my oh so clever hack, but I loaned my “Captain Obvious” hat to friend.

  3. @Brian Benchoff since the International Rectifier 92-O350 is probably a horizontal output transistor used to drive the flyback transformer it should be possible to use a transistor from a tv of same or slightly bigger tv.

    make sure to have spare fuses ready since there may be more wrong or if it is different enough specs.

    if you get a brighter or dimmer picture you can compensate by turning the screen up or down.

    since it is a vt100 terminal it probably will be all 1 color the same green you see the text of hackaday’s in case you missed it column and find hacks by date calendar it would not have a screen or focus adjustment on the flyback transformer since the volktage is less than 20000.

    above is assuming it is from the video display board

      1. Well, if it’s going to +5V or +12V, it’s NPN; if it’s -12V or -23V, it’s PNP. Either that, or if as Igrunenberg says, they stamp ‘E’ and ‘B’ on devices even if they aren’t bipolar transistors, it could be most anything. If it is a dual diode as several have suggested, you will find the ‘E’ and ‘B’ leads connected directly (or almost so) to power transformer pins. If that’s the case, if this feeds a positive supply, it will be a common-anode pair, otherwise common-cathode. Again, using modern devices, just use the highest-current substitute you can find in a TO-3 package, with at least a 100V peak inverse voltage rating. Everything that is connected to this is a clue to its function and specs, but you’re really not giving us much to work with.

    1. Following that link, if you click on the Adobe PDF icon, you will be presented with a page that says they will (attempt?) to locate the datasheet for you for that part.

  4. Did anyone else think that if you combined the brainwave sensor and motors from the cat ears and a desk top missile launcher you could come up with a cool little “Predator” shoulder gun. And as for the flexible dongle, try Viagra.

    1. Not sure if you’re trolling – if you are, consider me caught!

      ‘International Rectifier’ is a semicinductor manufacturer that makes a very wide variety of devices – rectifiers, BJT’s, MOSFET’s, SCR’s, Triacs, etc.

    1. Heavy rain in the midwest,my free internet is down.
      The best hack, moving my cookie can lid antenna around where I never tried and finding as good a signal as was normal when dry. So good as to waste time watching silly cat videos!

  5. [Mike]: it might be less cumbersome in the future, if you would use a separate post for each topic, so we’re not reading a mixture of comments on six widely-divergent subjects.

  6. Since no one else has commented on it, I enjoyed seeing the human powered quadcopter. The successful attempt was news to me. I always heard that large rotors turning slowly were most efficient, but wow, those are *really* large and slow.

    1. And in one of the articles I read about this, they were claiming that helicopters are inherently less efficient than fixed-wing aircraft. Which I question, because a) the most efficient wings have very high aspect ratios – as these rotors do – and b) in a fixed-wing aircraft you have a generally small propeller (which is not very efficient, having its own considerable drag) just getting the fixed wing up to speed, and THEN you have the drag of the fixed wing. Here, ALL of the power is going into moving some very lightly-loaded, high aspect ratio wings through the air slowly.

      Now I’m waiting for somebody to cover quad rotors with very very thin photovoltaic cells. (Sorry – don’t have the resources myself.) I don’t think I’ll ever be fit enough to lift myself with a quad, but it would be quite a thrill to “ride the sun” for hours on one.

      In any case, that film was a joy to watch – it’s been a long time since the Gossamer Condor made ITS mark in history.

      1. Much of the theory is confusing to me, but I think it boils down to this. Any time a wing (or rotor) encounters the vortices created by another wing, efficiency goes down. Less of an issue for an aircraft rotor, since the rotor is moving forward fast enough that no blade passes through the same air twice. More of an issue for a biplane (the second wing contributes relatively little as a result). And a big issue for a hovering helicopter. I’m guessing due to the incredibly slow speed of this human powered heli it’s not an issue here.

        It would be cool to see it covered with thin film solar. Maybe even have the rotors filled with helium!

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.