New (LTE) Rotary Cell Phone [video] | Hacker News Reader

New (LTE) Rotary Cell Phone [video]

youtube.com

310 pointsxwx66 comments

Very nice.

Our home has a rotary phone in most of the rooms (they're cheap on ebay and easy to repair if they're not working). Each one is plugged into a Grandstream HT802, which gets it onto our home network. A raspberry pi runs FusionPBX, which gives each phone a number, and lets them all dial each other. The kids love it!

what a cool idea, how hard was it to set up the PBX on the pi?

Oh no, don't give me ideas to do the same for our girls.

I love it when people say "Oh, I'm not a programmer" and then continue showing the stuff they've built (and fully programmed) that is a way beyond in complexity than what most of us professional programmers get to work on, like ever... :)

I've worked a lot with quants. Most of them are 'not programmers'.

The very complexity of the stuff they build is part of the problem. They are typically very smart, but often build Rube-Goldberg machines instead of looking for a simplification.

yeah, agreed : the difference between a non-programmer who programs and a programmer is the methods by which they achieve the results.

an easy way to find work to help with on github is to go to hackaday or some other such 'hackaton/diyer' community and follow the project links -- you can spend the entire evening fixing foot-guns and lazy/naive implementations of fairly well understood early CS concepts -- that's not to say that professionals don't make the same mistake, but amateurs who are just trying to get something to work rarely hit the codebase again after the thing does what they want.

> amateurs who are just trying to get something to work rarely hit the codebase again after the thing does what they want.

Are you saying that programmers do? I mean, they should, but…

I've often been prevented from improving existing pieces of code.

Usually with a variety of reasons, but it boils down to not prioritizing keeping the code clean. Invariably that's led to a higher carrying cost and decreasing ability to meet business goals.

But you're right, it's a rare developer that's interested in doing the unsung work of keeping a codebase clean.

Same. We had some really smart SOC networking guys, who naturally start automating everything in Python.

When you interrogate, they'd say almost sheepishly 'oh I'm no programmer I just threw this together.' I don't think they realize 98% of programmers are just throwing things together and acting like a God when it actually works...

Me too. I laughed at "Oh, I'm not a programmer. I had to write all the code to drive the OLED screen . . ."

Inspiring. Love when people do this. Makers gonna make . . .

I see what you’re saying but I also think that it is a legitimate statement. I have interest in building analog synthesizer gear and efx pedals. I’ve built a few circuits on breadboards but I am not an electrical engineer or synthesizer designer. The statement is likely just a way to say excuse the mess/possibility of bad practice in my implementation.

Oh, I totally was not questioning the legitimacy of the statement - most of my work is not in service of producing software that is sold. My work is instead a mess of bad practice to just get something accomplished for others.

More generally speaking, I wish more people were like this creator - unafraid to just dive in and get stuff done. I have done a little bit of teaching introductory programming and one of the things I emphasized is that "works but is suboptimal along some professional measure" mostly rounds to "works" even in professional settings.

Part of my actual day job is working with non-professional software developers and helping take "non-professional" prototypes to "better" implementations. Having a working prototype is often a much better starting point that a poorly specified set of requirements . . .

Not a constructive comment... but this project made me smile during the whole video. This is awesome! Great hacker in my book.

This is contributing to the thread telling the project is wholesome.

I see value in sharing a positive reaction and in times like this when I click the comments before the link, this got me to actually go watch.

Very cool.

My only nit is the local/nonlocal/alt switch is unnecessary. Local & nonlocal numbers have a different number of digits. Checkdown logic can determine which it is. The phone book certainly has less than 7 digits of pages, and less than 7 digits of contacts per page. The only issue is determining if a short number is for a contact or a contact page. That can be dealt with by pressing a different button than the call button if it's a page navigation.

In some areas even local calls require an area code. There was a change not too long ago that increased the number of areas like that. https://www.usatoday.com/story/news/nation/2021/10/19/area-c...

That's interesting. Per the explanation on the FCC page, it's so all numbers that start with 988 go to the 988 line. I guess they want to catch misdials by callers who need the service, but are only able to catch one class of misdial (extraneous trailing digits) without requiring thousands of people to change their numbers. There were obviously some hard conversations behind that choice.

It's not just trailing digits, if you don't have a call button (like a traditional home phone, push button, or rotary) you need to know when the dialing has finished so that you can make the cell call

Yes, I would have expected to replicate the standard user interface. That's what we did, here.[1] (at 00:08). That antique phone (actually a 1970s replica) produced a dial tone if lifted, and then you could dial calls. It's running over cellular, via a commercial box that interfaces landline phones to Bluetooth, comprehends dial pulses, and can generate ring voltage.

[1] https://vimeo.com/124065314

That is true in some parts of some countries

Am I the only one who is blown away by what looks like a PCB chip assembly machine in a personal residence near the end of the video?

That was the only part of the video I didn’t understand. Was that really at home?!? How much are those machines, because it can’t be cheap. I can’t imagine that there would be enough demand for these phones to really require a purchase of that level of production equipment.

For me, my question would be how much of the production is outsourced? Hopefully there will be more videos about the production/business side.

Basic pick-and-place machines run about $2k. Reflow ovens cost about $200. Those are a must if you want to produce hundreds of populated PCBs at home. For one-off projects you can do most anything easily by hand with a high end auto-heating soldering station like a Metcal and microscope for under $500. My wife and I build and modify our own electronics for lots of random things. The equipment lives in the shed next to the laser cutter, 3D printer, and woodworking tools.

I am kind of surprised most tech professionals -don't- have basic electronics dev gear at home given how accessible it is these days.

Go learn to make a thing! You will feel more pride in it than a few more hours of doom scrolling.

My favorite feature, is that the power switch actually separates the battery from the electronics. The thing is OFF!

This is an incredible project and I really want to buy one just because it’s so unique. Really impressed with Justine’s work!

I wonder which LTE modem this is using. Some LTE modems (like the PinePhone one) run Linux on one of the processors and folks have replaced the proprietary distro on that processor with a free one. If this were using the same modem as the PinePhone one, you could run all the Arduino firmware code on the modem instead.

It's a uBlox Lara-R6 [0], which is a different one than the Quectel EG25-G of the PinePhone [1].

[0] https://www.skysedge.com/unsmartphones/RUSP/index.html [1] https://wiki.pine64.org/wiki/PinePhone_Pro#Specifications

For those with an interest, see also

> Rotary Cell Phone (Description and Build) (youtube 52 mins)

> Mar 5, 2020 My open source rotary cell phone went viral (surprisingly, to me and my husband), and people seem to be interested in it. The original description with links to notes and design files is here: [2] and a starter kit is now available here: [3]

[1] https://www.youtube.com/watch?v=0euCWf0FpOA

[2] http://www.justine-haupt.com/rotarycellphone/index.html

[3] https://skysedge.us/

Cool! I'm relatively young, so the only interaction I've ever had with a rotary phone was my dad finding one in a box at my grandparents' house. Still, this seems like fun. Cool that you can communicate with the cell modem directly.

I still have a rotary landline (well, VOIP, using a Grandstream ATA).

It is fun! But I hate people with a lot of 8's, 9's, and 0's in their numbers. They take forever to dial :)

>It is fun! But I hate people with a lot of 8's, 9's, and 0's in their numbers. They take forever to dial :)

That's when you cheat and just input the pulses with the switch hook by pressing them quickly. 8 times in rapid succession for 8, 9 for 9, 10 for 0.

We used to do that to place free calls on pay phones. I don't recall it being a particularly fast way of dialing, though it was great if you wanted to keep your dimes for hockey cards. Yes, I'm that old.

Am I the only one that is bothered that the finger stop (or whatever it is called) moves?

This is presumably a (physical) space-saving decision. The stop looks like it's right next to the 1. But dialing a digit requires moving the dial. So how can you dial 1 if the stop is adjacent and immovable? It looks like the answer for this small dial is that the stop moves only ~one tenth of a turn.

Hopefully then it does have a real physical stop---otherwise it'd be quite hard to dial correctly!

The thing which hit me was the (fast) speed at which the dial returns.

In a "real" rotary dial there's a governor which controls the speed of the return, thus setting the impulse rate so that the step-by-step switches in the exchange can follow.

Every phone tech back in the day would take considerable care to set the governor speed, and would instantly recognise when it was wrong.

Eh it's not wrong here though. There are always trade offs in a project like this.

Yeah that surprised me too. Is it there so that when the finger stop moves, it registers the number? I've only had a few rotary phones in my life, but I've never seen one where that doohickey (which I believe is the technical term) moves.

The moving finger stop was a feature of the Trimline dial phone.[1] It allows making the dial slightly smaller. On most dial phones, there's a large gap between the 1 and 0 holes. On Trimlines, and this new thing, there is no gap. Without that moving finger stop, you get a dial handset like this AT&T prototype: [2] That was called "the Schmoo".[3] Huge bulge in the middle of the handset.

The moving finger stop also appeared in 1960s versions of the Lineman's Test Set (the "butt set", usually hung from a tool belt), where the dial was on the back of the earpiece.

[1] https://en.wikipedia.org/wiki/Trimline_telephone

[2] https://www.bellsystemmemorial.com/images/dl/misc/schmoo.jpg

[3] https://www.bellsystemmemorial.com/telephones-trimline-artic...

etymologically, the term Schmoo or Shmoo (typically based on the characteristic shape) originates with the cartoon character popularized by Al Capp: https://en.wikipedia.org/wiki/Shmoo

A favorite example is the shape of a budding haploid yeast cell, known in the literature as a shmoo (though as that wikipedia page will indicate, there are a lot of other delightful schmoo/shmoo eponyms): https://web.archive.org/web/20071111122821/http://discoverma...

> Is it there so that when the finger stop moves, it registers the number?

No. Moving to the stop is just to wind up the mechanism up to a defined point. After being released the dial turns back at a defined speed, and creates a number short on-hook spikes, depending on far it has been wound up - the number to be dialed.

It's possible to quickly tap the hook to dial a number, too. It's easy for low numbers, but the dial helps the slow ones of us to tap exactly 9 clicks in the defined time to dial a 9.

No idea if that un-smartphone works exactly like the real thing.

If memory serves, the gap on the dial set the "interdial pause" which created a minimum gap between strings of digits, which gave sufficient time for the exchange equipment to switch to the next selector..

I'm a developer, did small Arduino project but always found it too overwhelming to create a custom board. What do you recommend to get started? I'm impressed by the custom board of this phone!

Make a shield for an Arduino, kicad has a template for the dimensions and positioning of the headers. A fun one is to make your own Arduino board, take the design for the uno and layout your own version and see if it works, there's plenty of guides on how to position stuff and it should get you introduced to the more complicated concepts in PCB design. Order the board from jlcpcb, can't beat them in price for 2 and 4 later PCBs.

Recommend checking out the rest of Justine’s YouTube channel.

I’ll die on a hill for ex-spearmint though. That’s ok. She can’t be right about everything.

That was in reference to her video about how to pronounce "experiment" [1], right?

[1]: https://www.youtube.com/watch?v=bqxeoVUtu_A

> I’ll die on a hill

Meanwhile in multi-lingual laboritories all around the world where 'international English' is the language everyone uses, or attempts to, and 'experiment' is a word used multiple times a day, I keep hearing both. Which gives me the impression there might not actually a single right way anymore, or perhaps there never was (as in: could be British vs American or older brritish vs newer), and I stopped caring.

I realize there might still be people claiming their way is correct, in which case I have to ask: is there a single authorative reference? Only if there are no other references to be found claiming something else you can in my opinion be right, when it comes to language. Or else you have to narrow down your claim and specify a region, or a certain type of usage.

For example recently someone pointed out I'm wrong not using a comma after 'e.g.' or 'i.e.'. Which sounded strange to me, because I keep on reading that, including in redacted texts. That person then came up with one single textbook as reference, claiming it's the authority. I went looking around and easily found other texts, which definitely seemed to be reference-worthy material according to where they were used and referenced in turn. I never got a good answer to my question why I should believe that one book is 'the' authorative one and the others aren't so as far as I'm aware both forms are ok.

Part of the issue are professional style guides for written English, and they vary. The biggest ones off the top of my head are AP, Chicago, and MLA. I love the Oxford comma as shown in the last sentence, but in the AP guide they do not use it and it is wrong to use it for their style.

English is the best and worst language we have currently on the planet because it’s such an amalgamation and continuing to evolve and branch in ways that are hard to keep up with sometimes. It’s hard not to end up chugy don’t ya know.

Part of the issue are professional style guides for written English, and they vary.

Probably, but what I don't get is: how could one be so blind to claim their style is 'right'? It's style ffs.

For me it’s like the debate about which way to set the hang on your toilet paper roll. There is no right answer and the fun is asserting that yours is correct.

There IS a correct way. Check the manual that the manufacturer provided for the roll holding device. Using it incorrectly voids the warranty.

Hmm, that seems more like an engineering problem. Like: what's obectively the easiest way to be able to grab the paper? Surely only one answer :)

Love the quirkiness and that it dose one thing well. For myself I would have loved to make the switches a little less fiddly. A phone I can turn on and dial in my pocket would have been my target.

However I can actually see it being quite attractive for accessibility reasons. Where it always being in one mode ready for dialing the same way every time has its advantages.

The front page of HN is littered with silly/pointless stuff this evening and I just love it so much.

This is really cool. Is there any information on what kind of (standard) hardware/software this is using to communicate with the mobile phone network?

Congratulations on seeing this product all the way through production, a lot of people would have given up with all the headwinds of the past two years

I can’t wait for the cad files to get posted. I want to see how the mechanics of the new rotary dial work.

It's a really nice toy, but where I live I sadly can't afford it.

Yeah $390 + shipping + 25% VAT is a lot for this. It can't be more than $50 in parts.

But I guess this is the problem when you turn a hobby project into a niche business. You suddenly have to amortize the development costs over a small number of sales. Otherwise there's no point making it your sole means of income.

And from the initial 3D printed model to a custom made production run takes many times the amount of time it will have taken to build the prototype.

In this case there's the pick & place machine to consider too. I'm surprised she didn't outsource this. These things are expensive and I doubt she has enough parts to run it 24/7 to get her money's worth.

But no way I'll consider buying one at this price. Great project though!!

Also to add to that parts are usually only part of a project. Labor is usually a decent amount of the cost. Plus some profit (usually 2x whatever your parts cost are is a good rule of thumb, until it isnt). You also have to value your own time at this. You will sell at cost if you value your time as 0 or being nice to someone. A 3d print of something is also not something you can necessarily just take to someone and say 'make me 300 of these'. You usually have to work with them a bit and the machinery to make that can cost a bit. There is a weird spot in making things where making a dozen does not cost much other than time and parts. But you want to make 1000 and suddenly the cost goes pear shaped and does not calm down until MR=MC. And you may end up with one custom part you had to make to make the other thing that costs a bunch. Linus tech tips recently had a good summary of what they had to do just to make a screwdriver. You in essence have to build a machine to build your thing. That machine can cost a decent amount.

> Also to add to that parts are usually only part of a project. Labor is usually a decent amount of the cost.

Of course, but I didn't include this here because it's sold in kit form :)

But I'm not questioning the price. Just lamenting that the price puts it out of my reach.

I must say, I never would have thought of that!

This is why I am on the internet.

Right? I didn't even know I needed this thing, now I do.

Cool. Consider embedding speech recognition features too!

That is a really complex feature. XKCD 1425

Reply on news.ycombinator.com