Name that Ware April 2021

The Ware for April 2021 is shown below.

Both boards are from the same machine. I really admire the construction quality of these boards! Thanks again to Don Straney for contributing these fascinating wares.

11 Responses to “Name that Ware April 2021”

  1. banjaxed says:

    The main board looks like a VXI card for data acquisition, early 90s?
    The other board could be generating the high-speed clock signal for the flash ADC.

    • willmore says:

      Is it VXE or VME? I don’t remember them that well.

      The bottom board loole like the signal starts at the upper left in that big metal can. Flows down through what looks like a mixer with some filters. Then it flows right through a chain of amplifiers. Finally it goes up the right side and out that RF connection. The upper middle of the board with the DB25 connector has 8 lines of high voltage to TTL voltage conversion with some nearby +/-12V regulators, so I’m guessing something like RS232 or maybe some parallel interface. Looks like there’s got to be at least 8 channels for data there. There’s 9 resistors and a resistor pack of the same size there, so I’d guess 9 channels. Could be 9 inputs (with those resistors serving as a voltage divider) and 8 outputs at +/-12V through the Philips chips. Might be a bidirectional buss as well as those Philips chips have an OE signal. Since I don’t see anything with any brains on it, I’m going to say not RS232, but just some control signals. Maybe to adjust gains or change the frequency of what’s in that can. Hard to tell without seeing the other side of the board.

  2. willmore says:

    The digital board is really pretty, too. It’s got a 125MHz 8 bit ADC with an 80MHz ECL can XTAL near it, so let’s assume it’s running at 80MHz. It feeds four (presumeable cascaded) 4Kx9 asyncFIFO chips through a few ECL to TTL converters. There’s a little more ECL logic nearby. Some selectors, some flipflops, etc.

    I’m going to assume that every chip with a label on it is a PAL of some sort. There are enough of them to run a few state machines to interface with the bus and manage the data flow. There aren’t enough of them to make up much of a processing system, so I’m guessing there’s no ‘brains’ on this board.

    This board looks European. I can’t put my finger on it, but the RF connectors are a type popular there and the choice of component suppliers doesn’t show any kind of regional loyalty. The RF transistors look like European part numbers, too.

    I can tell what these boards do, but I’m having a hard time figuring out who designed them.

  3. Josh M says:

    Pretty sure that’s the Bruker logo blurred out on the first card’s connector.

    Therefore, this must be some kind of high-end data acquisition hardware. (You’re all quite welcome for that startling bit of insight.)

    • jackw01 says:

      Was going to say the type and layout of the RF connectors on the first board remind me of something I’ve seen before on some MRI or NMR equipment, and it seems like Bruker is a major manufacturer of those things.

      • willmore says:

        I recognize them (or slightly larger variants) from E1 data lines used in Europe from my days designing cell phone base stations. I never looked into what kind of connector it was as I live in NA and they’re nowhere around here.

        At this point, it’s pretty clear this is a data capture board from an MRI/NMR system–thanks to Josh M. I saw the logo and tried to find similar ones, but my google failed.

        To address J. Peterson’s comments (below?) I would say that using through hole at this point in time means that this was late production for an earlier design. Industries that require approvals tend to not change designs unless they have to. So they tend to be pretty conservative in their component selection–they stick to parts with a long production life even if that costs them in other ways. I would bet you could still source parts to make this board today without much problem. The critical Datel and IDC chips are still available in those packages. The rest is pretty jelly-bean stuff, so I’m sure someone still makes it.

        Props to Bruker for their clean design.

  4. J. Peterson says:

    No guess to the origin, but the chip date codes are all reading late ’90s (97 etc.).

    What’s really strange is most manufactured items had switched to surface mount components by them (TSOP, QFN, etc), so it’s strange to see everything mounted through-hole 1980s style.

  5. kurth says:

    I work in a facility that still relies on a great deal of VME hardware like this that has been running 24/7 for 25+ years. This stuff is over-engineered and built like a tank. It keeps running when the HVAC fails and the ambient temperature hits 95F. Awesome stuff.

  6. Ratz says:

    I don’t have any startling insights, but I do rather like the way the transistors are mounted in holes in the PCB. That’s a new one on me.

Leave a Reply