Member Since: May 11, 2006

Country: United States

  • Why the hell is there no pinout on the main connector?

    There is a top AND bottom silkscreen. There is NO good reason for not having the pinout of the 2X5 header, at least on the bottom of the board.

  • Unless I'm missing something, the schematic here is still for the old version.

    Please update the schematic file.

  • For what it's worth, there may be significant part-to-part variation in the HV supply levels, since the transformer used in the supply is not designed for flyback operation.

    I may have received one which runs unusually high.

    However, this does not discount the fact that it should be regulated, and adding regulation is extremely trivial.

    I strongly doubt that any are producing the correct voltage, however. The real question is the extent of the over-voltage, not whether it is happening.

    Also, it is possible that the tube begins to ionize on it's own, and therefore prevents the HV from increasing to the levels I saw when testing the HV supply open-circuit.

    Considering that the tubes are not supposed to be used this way, and the fact that LND confirms that over-volting them is indeed bad for them, this does not excuse the failure of technique in testing the original design, and the apparent apathy to the reports of issues with excessive tube supply voltages that seem to be legion in this device's history.

  • There are part-numbers on the schematic (e.g. Q1, Q2, R1, etc) but they're not on the PCB.

    What the hell?

    I want to bodge together a custom adapter that does TTL level-shifting (5v-3.3v), and I'm stuck following traces to figure out what is what.

    There is enough space on the board for part deliminators, and you have double-sided silkscreen (there is a logo on the bottom. and nothing else). Would it kill you to label stuff?

  • Absolutely nothing. Spark-fun removed the voltage regulator for no reason I can discern.

    The circuit as drawn above actually produces about 1-1.5KV. The method he is using to measure the voltage is drawing the voltage down to ~500V.

  • At the risk of feeding a troll… Fake-Name, after your posts here do you really think they would voluntarily work with you? Seriously, do you really think that?

    I dunno, is this a popularity contest, or a business interested in producing good products?

    For the sake of honesty, no, I really expect they wont. It felt good to ask, though.

    For all I know, you might be best thing to happen to electronics since silicon doping. However, with the attitude displayed in your post I sure wouldn’t work with you unless heavily compensated or were somehow forced into it.

    I assure you, when I havn't just sunk $150 into what is effectively a useless piece of junk, I'm a lot nicer.

    One can be completely, even brutally, honest without being assine. For example, there are ways to clearly and even empathically point-out design flaws you may have identified without doing any of the


    Question whether or not anyone at the company ever reads datasheets

    Considering they have released boards for sale that violate the products specs in the datasheet I can't help but assume they don't.

    Question if Sparkfun does any quality assurance on their designs

    See previous comment.

    Imply that the only working part of the design had to >be obtained (or “cribbed” as you put it) from an >outside source

    Oh, that wasn't an implication, it was a factual comment. See here or here, modified to use a choke. This circuit is topologically identical and has the exact same values, though apparently someone decided they could remove the voltage regulator (may not have been sparkfun, whoever did it was an idiot. They could have cribbed from a version of the schematic horribly abused by someone else. It's still a travesty).

    There are hundreds of high-voltage transistors on the market (MPSA42), and tens of thousands of generic PNP devices (MMBT4403/2N4403). The fact that all the linked circuits use exactly the same devices, together with exactly the same resistor and capacitor values, constitutes something pretty damn close to definitive proof that the spark-fun circuit was in fact copied form somewhere else.

    Suggest, even in jest, that “Anyone who thought this >was ready for sale should be hit over the head with a giant stick, preferably until they move to alaska and promise to never look at a transistor again”

    Considering the actual function of the circuit, and how it massively overdrives the GM tube, I thought it an appropriate suggestion.

    Declare that Sparkfun has a history of sloppy designs

    Uh.... Hello? Have you read the comments on their products on a regular basis?

    If this were a "Hey, take a look at this schematic/project of mine, and tell me if it has problems" situation, I would be (and generally am) nice about things.

    However, this is a "hey, we're selling this fundamentally broken product. No, we didn't test it. Yes, it kinda-almost-works as is, even though it will likely damage the GM tube in the long-run, and has potential stability issues with the USB interface, neither of which we are apparently knowledgeable to notice or actually inform the person dropping $150 on this thing before they purchase it"

    So yeah, I'm a bit incensed that they're selling a broken product, with no warning that it is a broken product.

    I will readily admit I'm inclined towards hyperbolic prose. To a certain extent, it was in jest. Furthermore, this was in (large) part me venting, after spending a number of hours trying to determine exactly what the circuit was doing. Anyways, take of it what you will.

  • It's self-oscillating. Q3 and Q1 form an astable oscillator.

    Basically R2 charges C2 until the voltage at the emitter of Q3 rises above Q3's base by it's threshold voltage. At that point, the transistor begins to conduct, which turns on Q1.

    Q1 turning on pulls the voltage at the R1-R2 node to ground + D1's drop. Since R3 and R4 are no longer forming a voltage divider, this pulls the base of Q3 down, turning on Q3 even more.

    At this point, C2 discharges through R2 and through Q3 (into the base of Q1) until the voltage across C2 is no longer high enough to turn on Q3 (and through Q3, Q1). At this point, the collector of Q1 goes high again, and the process repeats.

    It's a pretty clever oscillator, even if it is drawn in a rather confusing manner. It has the advantage that since Q1 is a high-voltage transistor, when Q1 turns off, Diode D1 is reverse biased, so effectively both windings are put in series, which allows the circuit to produce a higher voltage.

  • The FT232 is missing a cap on the 3.3v internal regulator pin.

    From the FT232 datasheet: "+3.3V output from integrated LDO regulator. This pin should be decoupled to ground using a 100nF capacitor. "

    Do you guys actually bother reading the datasheets, or have any QA? it may work without the cap, but it will likely have reliability issues. The FT232's internal regulator runs the whole usb transciever.

  • I don't think there is any way to use a current transformer without some sort of preamp.

  • Stainless has absolutely crap thermal conductivity. It would not work very well at all.

No public wish lists :(