BINA-VIEW: A Fascinating Mechanical Interference Display

[Fran Blanche] tears down this fascinating display in a video teardown, embedded below.

These displays can support up to 64 characters of the buyer’s choosing which is controlled by 6 bits, surprisingly only requiring 128 mW per bit to control; pretty power-light for its day and age. Aside from alphanumeric combinations the display also supported “color plates” which we found quite fascinating. The fully decked model would only cost you $1,206 US dollars per unit in today’s money or five rolls of toilet paper at latest street price. And that’s just one digit.

If you dig through the documents linked here, and watch her video you can get an idea of how this display works. There are six solenoids attached to rods at the rear of the device. A lamp shines through a lens onto the back of a plate assembly. Each plate is a strategically perforated grid. When the solenoids activate the selected plates tilt interfering with a stationary grid. This causes the light to be blocked in some regions only.

It seems clear why this never took off. Aligning these seems like a production nightmare compared to things like flip displays and Nixie tubes. Still, the characters have quite a lot of charm to them. We wouldn’t mind seeing a 3D printable/laser cut version of this display type. Get working!

24 thoughts on “BINA-VIEW: A Fascinating Mechanical Interference Display

      1. Let’s see your channel then, master mechanical logician lol! Don’t get me wrong, Fran sometimes goes down after red herrings before finding the correct solution but her channel isn’t supposed to be a “direct finding a solution the first try, exact instruction and profit” sort of channel. Sure it’s fair to say there are skills she can improve on, but it’s by no means painful watching another person learn unless you happen to be an impatient nit-picker or a run of the mill troll.

      2. I have conversed with Fran directly on technical topics in the past, and you are mistaken, at least in general. She is quite brilliant, and seems to understand the things she works on quite well.

      3. Fran has a unique way of describing things. She certainly does understand how the device works quite well and she did explain how it worked. You just have to know her way of describing things, and have a little intelligence :)

  1. Fran’s videos on the BINA(ry) View(er) are indeed fascinating, it would be interesting if, for a third installment, hook up an Arduino or some such to drive the display at its rated speed (the msec timings mentioned in the brochure).

    Making a repop would involve opening the device again and caliper measuring plates, pivot distances and so on. I’m guessing the original masks were photoetched which could be done easily enough from thin brass sheet. Scale modellers have been using these photoetched parts sheets for a long time now, and a pretty common.

    I posted on her comments there are a couple other equally interesting display indicators from the era, the Raytheon DataStrobe and the Union Switch and Signal Division Readall Readout, both on archive.org. I’m hoping she might luck onto examples of these someday :)

    1. These weren’t ever really common, and mostly seemed to be used in places where you needed a few characters to be displayed at any time. A google search shows that there were several neurology experiments in the 60s and 70s dealing with memory persistence, reflexes, etc, that used these. Because you could wire these up with minimal outside components, you could get a few characters displayed with a large amount of consistency and repeatability in the amount of time they’re displayed, the delay in action, etc without a need for a lot of even more expensive support components.

  2. Using modern mathematics, you should be able to significantly reduce the number of plates for a minimal reduction in quality.

    Just perform ICA (independent component analysis) on the full alphabet. This should give you a set of component plates which you should can mix and match to reconstruct any letter, number, or glyph.

    Then, you can replace the 6-bit+6-solenoid input system with a coding disk hand-turned by a knob. Holes spreading radially outward from the center of the disk will offset the plates up and down (based on the combination required to reconstruct that glyph). You would have a secondary lever that offsets all the plates off of the disk to prevent cycling through each glyph on your way to the next one.

      1. Yeah only if you live in our universe where even incandescent, florescent and LEDs are common.

        This would work better then a single signal fire from Gondor. Remember 1 is a mistake, 2 is coincidence but 3 is intent.

  3. This was a fascinating exercise from Fran. I began my career in 1969, but never heard of this display. The only BinaView I knew of were cheap Japanese binoculars! After a bit of searching I found what looks like an older info sheet https://ia800803.us.archive.org/25/items/TNM_Bina-View_Binary_Input_Self-Decoding_Readout_Display_Unit_20170623_0513/TNM_Bina-View_Binary_Input_Self-Decoding_Readout_Display_Unit_20170623_0513.pdf which includes data on multi-unit displays, and some mentions in scientific papers, including this https://link.springer.com/content/pdf/10.3758%2FBF03343215.pdf from 1965. The interesting bit is in the 3rd paragraph were it says that the displays were driven by a paper tape reader. Could that be the origin of this display? It could be used to display changing information from a paper tape reader with no additional equipment. This was a time when most computer peripherals were electro-mechanical so this would fit right in.

  4. The BinaView obviously needs to be driven by a computer. Its introduction is coincident with the HP2100 series which is interesting because the 2100 became the embedded controller of its day, often appearing built into a console for a single dedicated function controlling industrial and medical machinery. Since this stuff was fabulously expensive (A base model HP2100 with no peripherals was $40,000 in 1974 when my father’s lab got its unit) the cost of the BinaView would not have been extravagant for something that needed a more versatile real-time status display. I suspect the one Fran inherited was acquired for research toward supporting it in such a role, possibly for a project which either didn’t pan out or which shipped and was then forgotten.

  5. I managed to find one of these buried deep within my father’s junk collection. Apparently it was used as a readout at the gasoline distribution terminal where his dad worked as a truck driver. I powered it up and it seems to work exactly as designed even after all these years. Mine seems to be a special-order piece, it only contains the numerals 0-9 and few additional two-letter codes.
    I’ve pondered trying to produce some additional slides for it, but I’ve had a difficult time thinking of a process that would yield satisfactory results. I took some scans of one the slides that were in it if anyone would like pursue them.
    http://jacksonsmith.name/PannelScan.jpg
    I think that I’ll probably turn it into a clock (showing one digit at a time) to show off at my makerspace. I’ll have it up on my desk for the next few weeks so if anyone would like additional pictures/measurements feel free to shot me a reply.

Leave a Reply to RW ver 0.0.3Cancel 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.