EEVblog Electronics Community Forum
Electronics => Projects, Designs, and Technical Stuff => Topic started by: Non-Abelian on November 26, 2020, 10:27:08 am
-
This might seem like an odd question, but is it possible to manufacture an LCD in this sort of configuration. I don't care about the external shape other tan it be square or round and somewhere in the vicinity of 1" x 1 1in or a diameter of 1" (give or take some). I would like radial line segments extending from the center (not unlike a watch might be). I understand square is much easier so that's fine and it only needs to be monochrome. The unusual aspect is that I would like there to be a hole in the center to accomodate a rotary encoder. The basic idea here is to provide digital control in an analog appearing way control, but which can be used for multiple functions (or not) such that when using a particular function, the data are recalled and the display changes appropriately and can be adjusted by rotating the encoder, just like a pot (by incorprating a digipot or several other means) and then stored. That way, I can use a single control with a more "natural" interface instead of menus (which I find to be annoying as I'm sure others do). It makes it easy to select a function, adjust it and store the data.
Menus are basically a pain in the ass and require too much navigation. So, far, I have sent enquiries to several manufacturers without receiving replies, so I suspect the answer is no, although I do not really understand the problem with this. All we are talking about here is a rather common basic form with lines and some numbers with a void in the center that can be drilled out. Pretty much like a watch with a hole in the middle. (Yes, I could do this with lots of small LEDs, but that would be cumbersome and look like a kludge.) Thanks to everyone in advance for whatever illumination (pun intended) you can provide.
-
I guess you'll need some kind of business reference to be taken seriously.
I'm no expert in LCD manufacturing, but I would assume it should be possible to create such a form, however the handling and sealing might require manual labor or specialized machines.
Unless you order several millions of those displays, no manufacturer will jump in, I guess.
Maybe a nice DIY project...
-
It's actually intended for a product and if the price made it viable, I'd buy several thousand. For that matter, after searching for an existing part as I described, I could see it being commercially viable itself. Dave did a video on custom LCDs and the price seemed extremely reasonable for what he specified and this seems much simpler.
-
Why not use a simple graphic on a touch screen?
The economics of scale makes that so much cheaper and the user interface seems more flexible. Its no problem to input vector or scalar quantities and the widgets come with all the software frameworks and on all platforms.
Even non-human canines can use the resultant, omnipresent "apps" to do just about anything. (See this article in Wired, for example.)
https://www.wired.com/story/the-tech-helping-dogs-learn-to-communicate-with-humans/ (https://www.wired.com/story/the-tech-helping-dogs-learn-to-communicate-with-humans/)
-
Could you simply program an area of a touch screen to do that (i.e., rotary encoding)? Like current smartphones and pads do.
-
Ordering custom LCDs is like ordering PCBs. There are manufacturers doing this, and it's not difficult (the exception being your hole in the centre, I imagine that could make trouble).
Almost all manufaturers are in the far east, but you could try your luck in Switzerland (due to the watch production network there).
Let us know how you get along.
-
A ewatch was a simple example. There are all sorts of industrial displays that use LCDs as fake meters. The issue comes down to a hole in an LCD, not anything else that isn't commonly done by the zillions.
And of course I could program a fake knob on a screen, but that sucks as an interface. In addition, if a real knob was not intuitive, no one would bother to fake one on a display. At this point, I've pretty much given up that I could get something like this done, but I would really like to know why it's such a big deal. I understand LCDs from a physicist's perspective, but not from a manufacturer's perspective, but even from what I imagine would be required, I have difficulty understanding what the manufacturing issue would be.
-
How about a real encoder and knob, and a standard bargraph above it, maybe arranged vertically? That should give the functionality you are after, would not require custom parts, and the user would not cover part of the display with his fingers while operating the knob.
-
Sure and something like that is what I've considered as a fallback and although I imagine it will come down to that, I'm still curious as to what is so difficult about putting a hole in an LCD. As a more concrete example of how much more economical it would be (interface wise), consider this rack mount studio compressor:
https://vintageking.com/ssl-stereo-bus-compressor-500-series
Now, most things in recording studios are done by ear, not by selecting exact numbers and obviously the one in the link can only be set to one specific setting and has to be readjusted each time you change something. Now, instead of that being a single instrument that could store AND show settings recalled from memory, you could select say a second intrument via a single button press with the second instrument using the same knobs (or some of the knobs or whatever). That becomes a very easy to use and intuitive instrument that doesn't require menus or other kludges that are designed to "look" that way, but suck to use. No menus or other crap to make doing what you what to do a challenge beyond the difficulty of getting the result which is the actual job at hand. You can turn a knob, store the value and when recalled, you have a visual cue as to where it's set without having to look somewhere else to see what you're doing, which is also an issue in a tight space. We could go further and imagine that the entire assembly, LCD, rotary encoder and chips are a single assembly that apart from a bus connection to a processor is more or less like a drop in replacement for a pot. How easy would that be to use? (Rhetorical question, obviously).
Mostly, I've gotten to the point that I am relly curious about why this seems so difficult or at least appears to be given the lack of responses from the manufactureres I've contacted.
-
There was a thread in here a couple years ago where somebody had a small run of custom LCDs manufactured for a vintage pocket computer, that was the only time I've ever seen that done but as I recall they were surprisingly inexpensive and they turned out quite well. I don't remember who it was or what the thread was called but I would start by contacting that company or look for something similar.
-
The issue is that an LCD display with a hole in it, although technically feasible, would require modifications to the manufacturing process, since the overwhelming majority of LCDs produced have a rectangular profile with no holes.
First the front and back glass pieces need to be made and a hole cut into them, which they may not have the machines in the factory already setup to do since all of what they usually do is cut out rectangular pieces.
The other issue comes down to sealing the display, which again the manufacturing equipment is probably not set up to handle sealing around the edges in the middle of a display.
It all comes down to the fact that special accommodations would need to be made to accomodate the hole in the manufacturing process, and the manufacturers are probably not willing to do this for a fairly small production run (even a few thousands is nothing really).
What's wrong with a ring of LEDs? Simple, effective, easy. Combined with a custom designed light pipe diffuser you could have this show up as lighted radials rather than points of light.
Another trick which is more of an electro-mechanical approach, would be a mechanical pointer driven by a servomotor, similar to the type used in automotive instrument gauges.
-
There are plenty of circular LCDs out there.
(https://rh6stzxdcl1wf9gj1fkj14uc-wpengine.netdna-ssl.com/wp-content/uploads/2018/05/TMA-086A.jpg)
Even some pretty inexpensive ones https://maclight.en.made-in-china.com/product/kscEVzRrHiWh/China-China-1-22-Inch-Circular-TFT-LCD-Module-with-240rgbx204-Resolution.html (https://maclight.en.made-in-china.com/product/kscEVzRrHiWh/China-China-1-22-Inch-Circular-TFT-LCD-Module-with-240rgbx204-Resolution.html)
Could you make the encoder to accommodate the LCD? IOW, the LCD fits into a slot on the encoder frame - something like that?
This guy made his own LCD - pretty neat I think.
https://www.youtube.com/watch?v=_zoeeR3geTA&feature=emb_logo (https://www.youtube.com/watch?v=_zoeeR3geTA&feature=emb_logo)
-
The issue is that an LCD display with a hole in it, although technically feasible, would require modifications to the manufacturing process, since the overwhelming majority of LCDs produced have a rectangular profile with no holes.
First the front and back glass pieces need to be made and a hole cut into them, which they may not have the machines in the factory already setup to do since all of what they usually do is cut out rectangular pieces.
The other issue comes down to sealing the display, which again the manufacturing equipment is probably not set up to handle sealing around the edges in the middle of a display.
It all comes down to the fact that special accommodations would need to be made to accomodate the hole in the manufacturing process, and the manufacturers are probably not willing to do this for a fairly small production run (even a few thousands is nothing really).
What's wrong with a ring of LEDs? Simple, effective, easy. Combined with a custom designed light pipe diffuser you could have this show up as lighted radials rather than points of light.
Another trick which is more of an electro-mechanical approach, would be a mechanical pointer driven by a servomotor, similar to the type used in automotive instrument gauges.
Excellent comprehensive answer. While technically possible its far enough away from standard methods that you'd probably need to fund the research/failures yourself by partnering with a manufacturer and paying them upfront for developing the processes.
I know of 2 different companies who "solved" rotary control knobs floating on LCD display, but neither made it to mass production. Prices are still unpleasant compared to the LED rings that are used in reality.
Could you make the encoder to accommodate the LCD?
Yes, there are examples of external rotary controls with a small display in the middle of them (see example below).
-
Yes, there are examples of external rotary controls with a small display in the middle of them (see example below).
Neat! May be a bit pricey though; samples go for 3000 TWD (about 100 USD):
https://www.shanpu.com.tw/en/product/series/SLB_Rotary_encoder_TFT-display (https://www.shanpu.com.tw/en/product/series/SLB_Rotary_encoder_TFT-display)
-
Wow! I would like to thank everyone for the suggestions and especially, @TheMG for clarifying that it's really a manufacturing issue as a custom part. I did consider the ring of LEDs, prior to seeing Dave's video on how inexpensive a custom LCD can be and all things conidered, if the hole in the LCD were fesible, it would have been simpler, more cost effective and be aesthetically nicer, so I wanted to try that route (although actually mounting it for abuse might have been an issue, but without any info from a manufacturer, hard to say). I think the ring of LEDs may still be a consideration if the LED/diffuser/encoder/chips are cost effective and aesthetically attractive. ( if it were a personal project, I would not care much, but people spending money need reasons to spend it.)
However, I really like the idea of the circular LCD inset in the knob that mechnically turns the encoder, especially because in addition to displaying the current setting, it could display the current function for the control. The motor control is really a non-starter. Too slow and also expensive.
The main point here is to use enough controls to make a trade off between having enough controls to not constantly be switching functions and parameter settings for just a few knobs while still allowing for a single set of knobs to adjust parameters for just a few functions so that it's not single rotary encoder that has to constantly be switched, yet doesn't go so far overboard as to be cost prohibitive to manufacture and (hopefully sell). It would also be nice if the control itself were inexpensive enough to be a product on it's own as a side benefit if it works out that way. (After working in a lab as a graduate student whee every fitting, o-ring groove, etc. was a one-off done for short term expediency and redoing it all to make everything fit without being a jigsaw puzzle, I came to appreciate modularity and thinking how things should be designed before starting to make parts. :)
This might seem odd to attach a lot of importance to, but I f&%*ing hate kludgy interfaces that didn't give a lot of thought to "How is a non-technical person who just wants to use this going to use it easily? or "Why is it so difficult to do something that should be simple?" Again, I woud like to thank everyone who responded and I think the responses have pretty much pared down what is cost effective without losing the basic idea to just a couple of what look like fairly inexpensive options, most notably the ring of LEDs depending on how that looks and the round LCD inset into the external knob, the latter of which is probably the easiest with respect to mechanical mounting and some potential extra functionality that is useful. Thank you all again!
-
It would also be nice if the control itself were inexpensive enough to be a product on it's own
Four wide widescreen touch panels arranged so that short side connects a long side of the other and leaving square hole in the middle.
Face plate covering the center and edges so that four wide corners are open and can then select a function, something like a Rigol power supply selector.
Probably not very inexpensive.