NOT another nixie clock!

[Laurence Wilkins]

Over almost the last ten years, I have been designing, producing and selling a variety of logic and PIC Micro-controlled nixie clocks, from classic 2x3 mux 74141-based designs to multiplexed and direct drive IN18 monsters, to miniature desk clocks and single digit weirdness. I feel I've "been there" now, and want to find other appropriate outlets for these glorious glowing anachronisms! (that is: nixie tubes).

I've been giving some serious thought recently to designing and building a good quality 50,000 count (4 3/4 digit) Bench Digital Multimeter with a fair few (sensible) bells and whistles, but of course with a nixie display. It would use a standard but very capable DMM "front-end" chip to do all the measurement, opto-isolated to a controller and then onto a direct drive nixie display (with Volts/Amps/Ohms/Hertz symbols!) Battery or mains, USB output (or is that input?). I am aiming for CAT IV performance by design (but not by certification - too expensive), good accuracy as afforded by the front end chip, and in a good quality case (likely to be the most expensive component). I'd like to offer it both as an assemble-able kit (so I'd presolder small SMD parts) or a fully assembled instrument. This would be a "serious" bit of kit, not something which looks like it was thrown together in a biscuit tin!  I've yet to decide whether the software would be open source, but it might be nice to let "the community" develop additional functions in software (data logging, averaging, etc.)

I aim to produce a batch of maybe 50 instruments, but I really don't have much of a clue as to the demand, out there. "It's going to depend on cost," you say. Well, based on BOM costs so far, it is not going to come in much (if anything) below about GBP £400 (USD $550). Yes, you could spend that on a new meter and get guaranteed similar specs, but it wouldn't have a nixie display which is of course the unique selling point.

It might take me around 12 months from pressing the button to a finished product. I've looked, but there appears to be nothing else out there.

So what does the group think? Are we nixie nuts a very small group? Are clocks enough? Is that price way too high? (the BOM costs means the price can’t be much lower, for small volumes!)

Regards

Laurence

www.MrNixie.com

(the one in England they call) Mr Nixie

[Dylan Distasio]

I would love one, and applaud the idea, but that price point will put it out of range for me. 

--
You received this message because you are subscribed to the Google Groups "neonixie-l" group.
To unsubscribe from this group and stop receiving emails from it, send an email to neonixie-l+unsubscribe@googlegroups.com.
To post to this group, send email to neoni...@googlegroups.com.
To view this discussion on the web, visit https://groups.google.com/d/msgid/neonixie-l/6f8b3456-3add-41b2-9106-765741d98a48%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Samuel G. Guss]

Greetings Nixiers,

Laurence, i think it's a fabulous idea. I, too, have been thinking about
non-clock nixie projects, but hadn't thought of a modern bench meter.
Great idea! To those like Dylan (and me) who are a little nervous about
the pricing, i suggest starting a NMM fund *now*. Less than $50 a month,
and when it's ready, you'll be able to put cash on the barrelhead!

In all seriousness, Laurence, i hope you go forward with this project,
and count me in for a kit.

All the best,
Sam

On 2016-10-07 15:07, Dylan Distasio wrote:
> I would love one, and applaud the idea, but that price point will put
> it out of range for me.
>
> On Oct 7, 2016 3:02 PM, "Laurence Wilkins" <in...@bcaf.net> wrote:
>
>> Over almost the last ten years, I have been designing, producing and
>> selling a variety of logic and PIC Micro-controlled nixie clocks,
>> from classic 2x3 mux 74141-based designs to multiplexed and direct
>> drive IN18 monsters, to miniature desk clocks and single digit
>> weirdness. I feel I've "been there" now, and want to find other
>> appropriate outlets for these glorious glowing anachronisms! (that
>> is: nixie tubes).
>>
>> I've been giving some serious thought recently to designing and
>> building a good quality 50,000 count (4 3/4 digit) Bench Digital
>> Multimeter with a fair few (sensible) bells and whistles, but of
>> course with a nixie display. It would use a standard but very
>> capable DMM "front-end" chip to do all the measurement,
>> opto-isolated to a controller and then onto a direct drive nixie
>> display (with Volts/Amps/Ohms/Hertz symbols!) Battery or mains, USB
>> output (or is that input?). I am aiming for CAT IV performance by
>> design (but not by certification - too expensive), good accuracy as
>> afforded by the front end chip, and in a good quality case (likely
>> to be the most expensive component). I'd like to offer it both as an
>> assemble-able kit (so I'd presolder small SMD parts) or a fully
>> assembled instrument. This would be a "serious" bit of kit, not

>> something which _looks like_ it was thrown together in a biscuit

>> tin! I've yet to decide whether the software would be open source,
>> but it might be nice to let "the community" develop additional
>> functions in software (data logging, averaging, etc.)
>>
>> I aim to produce a batch of maybe 50 instruments, but I really
>> don't have much of a clue as to the demand, out there. "It's going
>> to depend on cost," you say. Well, based on BOM costs so far, it is
>> not going to come in much (if anything) below about GBP £400 (USD
>> $550). Yes, you could spend that on a new meter and get guaranteed

>> similar specs, but _it wouldn't have a nixie display_ which is of

>> course the unique selling point.
>>
>> It might take me around 12 months from pressing the button to a
>> finished product. I've looked, but there appears to be nothing else
>> out there.
>>
>> So what does the group think? Are we nixie nuts a very small group?
>> Are clocks enough? Is that price way too high? (the BOM costs means
>> the price can’t be much lower, for small volumes!)
>>
>> Regards
>>
>> Laurence
>>

>> www.MrNixie.com [1]

>>
>> (the one in England they call) Mr Nixie
>>
>> --
>> You received this message because you are subscribed to the Google
>> Groups "neonixie-l" group.
>> To unsubscribe from this group and stop receiving emails from it,

>> send an email to neonixie-l+...@googlegroups.com.

>> To post to this group, send email to neoni...@googlegroups.com.
>> To view this discussion on the web, visit
>>
> https://groups.google.com/d/msgid/neonixie-l/6f8b3456-3add-41b2-9106-765741d98a48%40googlegroups.com

>> [2].
>> For more options, visit https://groups.google.com/d/optout [3].

>
> --
> You received this message because you are subscribed to the Google
> Groups "neonixie-l" group.
> To unsubscribe from this group and stop receiving emails from it,

> send an email to neonixie-l+...@googlegroups.com.

> To post to this group, send email to neoni...@googlegroups.com.
> To view this discussion on the web, visit

> https://groups.google.com/d/msgid/neonixie-l/CAJrqPH-E%2BpAROAy%3D4UGZfEMF60wP39qe21va1crFDOBKVoAuSw%40mail.gmail.com
> [4].
> For more options, visit https://groups.google.com/d/optout [3].
>
>
> Links:
> ------
> [1] http://www.MrNixie.com
> [2]
> https://groups.google.com/d/msgid/neonixie-l/6f8b3456-3add-41b2-9106-765741d98a48%40googlegroups.com?utm_medium=email&amp;utm_source=footer
> [3] https://groups.google.com/d/optout
> [4]
> https://groups.google.com/d/msgid/neonixie-l/CAJrqPH-E%2BpAROAy%3D4UGZfEMF60wP39qe21va1crFDOBKVoAuSw%40mail.gmail.com?utm_medium=email&utm_source=footer

[gregebert]

Yeah, that's a project I've thought about, but my conclusion was that the DMM chips were all intended for direct-drive LCD, or perhaps LED. I get stuck in an endless loop of trying to justify building something really interesting that wont get used very much vs adding features that would be unique (measure milliohms, inductance, kilovolts, and true-RMS milliwatts).  Guess what...I get nowhere.

A Raspberry Pi would likely be the best platform, as there is all sorts of software out there to enable it to run as a networked device. 

[David Forbes]

Laurence,

I like the way you think!

I might go for a new Nixie bench meter of that caliber. The Fluke 8200A, with
autoranging added, would be an excellent starting point for aesthetics. It would
want to be 4 inches deep instead of 14, of course.

But you defintely would have my interest if you made one that felt, looked and
worked like a Fluke 79 handheld instrument. Rechargeable li-ion battery required.

I have a metric buttload of the B5870ST and B5853ST (and the B5866ST +/-) tubes,
if you need those.

On 10/7/2016 12:02 PM, Laurence Wilkins wrote:
> Over almost the last ten years, I have been designing, producing and selling a
> variety of logic and PIC Micro-controlled nixie clocks, from classic 2x3 mux
> 74141-based designs to multiplexed and direct drive IN18 monsters, to miniature
> desk clocks and single digit weirdness. I feel I've "been there" now, and want
> to find other appropriate outlets for these glorious glowing anachronisms! (that
> is: nixie tubes).
>
> I've been giving some serious thought recently to designing and building a good
> quality 50,000 count (4 3/4 digit) Bench Digital Multimeter with a fair few
> (sensible) bells and whistles, but of course with a nixie display. It would use
> a standard but very capable DMM "front-end" chip to do all the measurement,
> opto-isolated to a controller and then onto a direct drive nixie display (with
> Volts/Amps/Ohms/Hertz symbols!) Battery or mains, USB output (or is that
> input?). I am aiming for CAT IV performance by design (but not by certification
> - too expensive), good accuracy as afforded by the front end chip, and in a good
> quality case (likely to be the most expensive component). I'd like to offer it
> both as an assemble-able kit (so I'd presolder small SMD parts) or a fully
> assembled instrument. This would be a "serious" bit of kit, not something which

> /looks like/ it was thrown together in a biscuit tin! I've yet to decide

> whether the software would be open source, but it might be nice to let "the
> community" develop additional functions in software (data logging, averaging, etc.)
>
> I aim to produce a batch of maybe 50 instruments, but I really don't have much
> of a clue as to the demand, out there. "It's going to depend on cost," you say.
> Well, based on BOM costs so far, it is not going to come in much (if anything)
> below about GBP £400 (USD $550). Yes, you could spend that on a new meter and

> get guaranteed similar specs, but /it wouldn't have a nixie display/ which is of

> course the unique selling point.
>
>
> It might take me around 12 months from pressing the button to a finished
> product. I've looked, but there appears to be nothing else out there.
>
>
> So what does the group think? Are we nixie nuts a very small group? Are clocks
> enough? Is that price way too high? (the BOM costs means the price can’t be much
> lower, for small volumes!)
>
>
> Regards
>
>
> Laurence
>

> www.MrNixie.com <http://www.MrNixie.com>

>
>
> (the one in England they call) Mr Nixie
>

> --
> You received this message because you are subscribed to the Google Groups
> "neonixie-l" group.
> To unsubscribe from this group and stop receiving emails from it, send an email
> to neonixie-l+...@googlegroups.com

> <mailto:neonixie-l+...@googlegroups.com>.

> To post to this group, send email to neoni...@googlegroups.com

> <mailto:neoni...@googlegroups.com>.

> To view this discussion on the web, visit
> https://groups.google.com/d/msgid/neonixie-l/6f8b3456-3add-41b2-9106-765741d98a48%40googlegroups.com

> <https://groups.google.com/d/msgid/neonixie-l/6f8b3456-3add-41b2-9106-765741d98a48%40googlegroups.com?utm_medium=email&utm_source=footer>.
> For more options, visit https://groups.google.com/d/optout.


--
David Forbes, Tucson, AZ

[Jon D.]

I would be interested, depending on the features.

[Jonathan Peakall]

I too have as many clocks as I need, although I haven't made as many as you. I have a kitchen timer, an indoor outdoor thermometer (pixies) and a hot tub temp. display. I have an HP VOM that I have been meaning to fix.

I have been planning to make a frequency counter. In my case I would use it as part of a custom ham radio, nixies for frequency, neon bar graph tube for an S meter, magic eyes for forward and reflected power, that sort of thing. Others might use it for all sorts of purposes, like measuring RPM.

$500 is way beyond what I can spend on a meter. :-(

Jonathan

[Laurence Wilkins]

Thaks for your reply, William. Yeah, well, it costs a lot (before I consider any return for my time!). This is my concern - if the market fo rthe price I need to charge isn't there, I don't have a project, after all...

To unsubscribe from this group and stop receiving emails from it, send an email to neonixie-l+...@googlegroups.com.

[Laurence Wilkins]

Thanks for your note.

Well, you can buy "front end" DMM chips that just do the measuring with no built in display, but some serial I/O to a processor/display/keyboard combo. That's my route.

Not settled on the processor, yet. I also play with Raspberry Pis, and while I might use one for early software development, I have ruled them out for a place in the final product for a couple of reasons. The most annoying aspect of my otherwise nice Fluke 289 DMM is the ~10 second boot time. I want that voltage reading, NOW! Pi's boot in around 30 seconds, and even though the application is for  a bench meter, I would find that start-up delay frustrating.  Also, as others have found, Pi's do not take kindly to just being switched off, without a proper power-down. I need something more reliable and able to withstand "mild" abuse that that.

[Laurence Wilkins]

Yes, my envisioned Nixie DMM will include a frequency counter, but not fancy period or averaging functions, not much beyond maybe 40MHz.

[Laurence Wilkins]

Thanks for your interest, I'll keep you posted!

[gregebert]

It's a classic tradeoff regarding RasPi vs Arduino vs FPGA. Everyone has different pain-points for cost, power, boot-time, features, development effort, etc.

Maybe Aduino is a better option; I'm going that route for my next clock project. If you plan your design, you can add various shields for added features and still run the DMM and nixies (maybe thru I2C ?). My only dislike regarding Arduino is the I/O is rather slow, and there aren't a lot of pins.

An FPGA will give you the fastest boot-time (milliseconds), tons of I/Os, and probably lowest power-consumption (below 400mW for me), but it involves the most work (you better be a good Verilog or VHDL coder). I've done 3 different clock designs with FPGA's, and it was a bit more work developing the code. However, it's far, FAR better than hardwired logic such as what I did on my first nixie clock.

A RasPi will have everything you want, plus more. But it's going to cost a bit more and use more power. However, it makes code updates very simple (connect to internet & download), not to mention you can do bizarre things like logging into your device  even if it's at a customer site far, far away.

I do have the option of pulling the FPGA or Arduino aside and using something else such as a RasPi, if I wish, on most of my designs (OK, not the wristwatch...).

[David Forbes]

I think you would be bonkers to run an OS in a voltmeter. I have removed
the OSes from many of our telescope's embedded computers, because
software updates are the last thing you want to have to do in a machine.

A standard AVR 32U4 processor as used in an Arduino Leonardo is plenty
of horsepower, and the limited I/O is not a problem if the display is
driven by 74HC595 SPI shift registers and the DMM chip has an SPI or I2C
port.

That's what I used in my scope clock, which does way more work than a
Nixie tube voltmeter does.

Plus, Arduino is a rather friendly development environment for those of
us who like to make our tools do odd things.

> --
> You received this message because you are subscribed to the Google
> Groups "neonixie-l" group.
> To unsubscribe from this group and stop receiving emails from it, send
> an email to neonixie-l+...@googlegroups.com

> <mailto:neonixie-l+...@googlegroups.com>.

> To post to this group, send email to neoni...@googlegroups.com

> <mailto:neoni...@googlegroups.com>.

> To view this discussion on the web, visit

> https://groups.google.com/d/msgid/neonixie-l/1b0d46a5-1539-4f17-b799-4c265aca20fc%40googlegroups.com
> <https://groups.google.com/d/msgid/neonixie-l/1b0d46a5-1539-4f17-b799-4c265aca20fc%40googlegroups.com?utm_medium=email&utm_source=footer>.
> For more options, visit https://groups.google.com/d/optout.


--
David Forbes, Tucson AZ

[John Rehwinkel]

> It's a classic tradeoff regarding RasPi vs Arduino vs FPGA. Everyone has different pain-points for cost, power, boot-time, features, development effort, etc.

Quite true.

> Maybe Aduino is a better option; I'm going that route for my next clock project. If you plan your design, you can add various shields for added features and still run the DMM and nixies (maybe thru I2C ?). My only dislike regarding Arduino is the I/O is rather slow, and there aren't a lot of pins.

The Arduino IDE is extensible, and many companies (including Arduino) have run with this, offering the Arduino development environment for much more powerful chips. I have a personal fondness for the AdaFruit "feather" series and the PJRC "Teensy" ones, both of which offer some very nice hardware (32-bit ARM CPUs at good clock rates and lots of I/O) at very attractive prices and you can plug them right into a breadboard if you like. I've ended up building these into a wide variety of projects.

> An FPGA will give you the fastest boot-time (milliseconds), tons of I/Os, and probably lowest power-consumption (below 400mW for me),

The lowest power consumption solution I've personally worked with is TI's MSP430 line. Those chips (especially the FRAM ones) are amazingly power efficient and reasonably powerful as well.

> but it involves the most work (you better be a good Verilog or VHDL coder). I've done 3 different clock designs with FPGA's, and it was a bit more work developing the code. However, it's far, FAR better than hardwired logic such as what I did on my first nixie clock.

I haven't gotten much into FPGAs myself (I worked with the earlier PAL and GAL technology some). Note that the environments are only available for Linux and windows.

> A RasPi will have everything you want, plus more. But it's going to cost a bit more and use more power. However, it makes code updates very simple (connect to internet & download), not to mention you can do bizarre things like logging into your device even if it's at a customer site far, far away.

A Pi Zero is $5, so it's cheaper that almost everything else out there. However, I only use something like that when I can deal with the "real OS" issues (boot time, varying I/O timing, etc.).

I have no patience for clocks, so I'm always looking for other nixie projects to build.

- John

[Nick]

Upvote here for the TI MSP43x series.

The new ones, the 432s, run at 13 microamp per MHz. Very easy to programming (C++) using free IDE. Development boards (aka "launchpads") cost just a few USD. Lovely chips...

Nick

[John Rehwinkel]

> Upvote here for the TI MSP43x series.
>
> The new ones, the 432s, run at 13 microamp per MHz. Very easy to programming (C++) using free IDE. Development boards (aka "launchpads") cost just a few USD. Lovely chips...

The "Spy-bi-wire" programming only uses two signals (and pins) and ability to run without a crystal makes it possible to breadboard the DIP ones easily. For people who prefer the Arduino environment, there's an MSP430 port called "Energia". And the Launchpad development boards Nick mentioned come with an on-board programming interface you can use to program other chips. The programming/debug interface runs through a block of jumpers where you can tap it off for your own uses. There's a family of add-on daughterboards as well.

- John

[NeonJohn]

I don't understand why these TI chips and their Launchpads aren't
getting more maker attention. Based on your recommendation, John, I
just ordered a Launchpad for this processor from Mouser. $12. I spent
last night reading the spec sheet. Wow! IT'll do just about everything
I need for my induction heaters and then some. Lots of I/O pins. fast
enough to implement in software some stuff I've been doing in hardware.

I've previously been evaluating the TI C2000 LaunchPad ($17). A fine
processor but deficient in I/O pins. Still more than adequate for most
projects.

Oh, did I mention that the GUI dev environment is available natively for
Linux? We're a Linux shop and it really ripped me to have to keep a
windoze VM to run Atmel's environment.

About 500 milliseconds after MicroChip digested Atmel, Atmel EOL'd the
chip I use in my main induction heater so I'm having to find a
replacement. This chip is a couple of bux more but well worth the money.

Good stuff
John

--
John DeArmond
Tellico Plains, Occupied TN
http://www.tnduction.com <-- THE source for induction heaters
http://www.neon-john.com <-- email from here
http://www.johndearmond.com <-- Best damned Blog on the net
PGP key: wwwkeys.pgp.net: BCB68D77

[Nick]

I've been using MSP43xs for a few years now - they are simply lovely. The Launchpads are great, fully functional, boards and as has been mentioned, they have a complete programming system built in. MSP430s area have a 16-bit ultra-low-power proprietary core; MSP432s are low-power, faster, industry-standard 32-bit ARM Cortex-M based. An example of the madly low-power FRAM-based MSP430s is the MSP430FR5969 - the Launchpad for that is at http://www.ti.com/tool/msp-exp430fr5969 costing a whopping USD 16 all-in.

For further, IoT, fun, they newer boards also have upower measurement capability - you can measure the power used (with tiny tiny granularity in fractons of a milli-joule) for individual functions - all supported by the IDE - either use TI's IDE or Energia as mentioned. TI call it "EnergyTrace" - use it to estimate battery life etc.

It's also worth mentioning that the TI E2E support forums for the MSPs (https://e2e.ti.com/support/microcontrollers/msp430/f/) are manned by experienced MSP engineers, both software and hardware. The designers/maintainers of the MSPWare HAL library (also written in C++) also hang out there.

There is also a great user (non-TI) forum at http://43oh.com/ - this is directly equivalent to AVRFreaks etc.

If you want JTAG, that's included but you can use the MSP-FET (http://www.ti.com/tool/msp-fet), a USB JTAG tool that also provides a UART back-channel so running a terminal emulator on your host you can talk directly to the MSP dev board application.

I love these chips - they are almost exclusively used by the trade, but there are basic ones in DIP packages up to a whole world of SMDs.

HTH

Nick

[gregebert]

I gave the MSP432 (Launchpad)  a try when it first came out; must have been about 3 years ago. I was unable to get reliable communication (characters randomly got dropped) thru the USB port on a Linux platform, so I put it in the junkbox. The $12 price was irresistible.

If any of you are currently using the launchapd with Linux and see reliable operation, please post and I'll buy another one with my next online order and give it another try. I'm assuming I have an earlier version that was not fully debugged on Linux; BTW, running on WIndoze is not an option I would ever consider.

[Nick]

I think you mean the MSP430, not the 432 - the 432 is very new.

I've never had any issues with serial comms on any launchpad and I've been using them since day 1 - maybe it's just your system or a duff launchpad, though I've never heard of one of those failing either.

I use mainly Windows, but have used the serial comms to all sorts of devices, from microprocessors to Cisco routers.

Nick

[NeonJohn]

The C2000 LaunchPad works fine under Linux. Since it looks like the
MPS432 uses the same interface, I should think it would work as well. I
use gtkterm as my terminal program.

Like you, here the W word is a profanity, never to be uttered out loud.

John

[threeneurons]

I'll add my 2 cents to what I want in a meter. A true RMS meter, that goes into the RF range. Linear Technologies use to make a chip that helped do that:

http://www.linear.com/product/LT1088

Unfortunately, its been discontinued, but the concept still applies.

How you implement the greater functionality, is up to you.

[Paul]

I think that is a wonderful idea. I have thought about doing this in the past but do not have enough knowledge to design it. I would be very interested.

[Steven Donaldson]

I agree what what has been said for the MSP430's.  I've mentioned them on this group before.  I've been using them for battery operated equipment at work and plan on using a launchpad on my next project (when I find time for that).  They are great chips but don't seem to get much love on the hobby side.

--

You received this message because you are subscribed to the Google Groups "neonixie-l" group.

To unsubscribe from this group and stop receiving emails from it, send an email to neonixie-l+unsubscribe@googlegroups.com.
To post to this group, send email to neoni...@googlegroups.com.
To view this discussion on the web, visit https://groups.google.com/d/msgid/neonixie-l/3808a728-0411-4ad8-ba50-004d24c895fc%40googlegroups.com.

[Dylan Distasio]

For those looking for a quick, fun build using these chips, I've built one of these. All of the code is freely available:

https://www.tindie.com/products/RobG/msp430-nixie-clock-kit/

To view this discussion on the web, visit https://groups.google.com/d/msgid/neonixie-l/CAOXC3DFVuzoXgGifVcvR--3nS_HrxeTxAEZ_ZHbjCJC2CeEGpQ%40mail.gmail.com.

[GastonP]

Well, regarding to the love the MSP430s get, I suspect that as soon as Microchip starts discontinuing the AVRs, the Arduino herd will have to migrate to greener pastures, and I think that's where the MSP430's are.
I'm evaluating to port the Marlin 3D-Printer firmware, which works on the ATMega platform to the MSP432 one. It has no ETA, but I think it's a worthwhile move.

On Monday, October 10, 2016 at 9:34:06 PM UTC-3, MrStevenUND wrote:

I agree what what has been said for the MSP430's.  I've mentioned them on this group before.  I've been using them for battery operated equipment at work and plan on using a launchpad on my next project (when I find time for that).  They are great chips but don't seem to get much love on the hobby side.

On Mon, Oct 10, 2016 at 5:45 PM, 'threeneurons' via neonixie-l <neoni...@googlegroups.com> wrote:

I'll add my 2 cents to what I want in a meter. A true RMS meter, that goes into the RF range. Linear Technologies use to make a chip that helped do that:

http://www.linear.com/product/LT1088

Unfortunately, its been discontinued, but the concept still applies.

How you implement the greater functionality, is up to you.

--
You received this message because you are subscribed to the Google Groups "neonixie-l" group.

To unsubscribe from this group and stop receiving emails from it, send an email to neonixie-l+...@googlegroups.com.

To post to this group, send email to neoni...@googlegroups.com.

[John Rehwinkel]

Well, regarding to the love the MSP430s get, I suspect that as soon as Microchip starts discontinuing the AVRs, the Arduino herd will have to migrate to greener pastures, and I think that's where the MSP430's are.

You're probably right.  There's already the Energia port of the Arduino IDE to the MSP43x architecture.  And it's impossible to put the genie back in the bottle.  The IDE already supports plug-ins for different architectures, and there are dozens available and more coming.  Lots of ARM based ones as well (including official Arduino branded offerings).  There is PIC support, for people who like Microchips more primitive offerings, and even X86 ones, for people who like Intel's more primitive offerings.

Back when I had outgrown the PIC architecture (and lost patience with some of the things that go with it), I had looked at the MSP430 as a possible next step, as TI has just introduced it and was offering the launchpads and watch kit at an eponymous $4.30 price.  Unfortunately, at that time, it was DOS-only, which was a dealbreaker for me (now they support modern operating systems).  I went to AVR at the time and it has served me well, but I've since embraced MSP430 and ARM too.

I'm evaluating to port the Marlin 3D-Printer firmware, which works on the ATMega platform to the MSP432 one. It has no ETA, but I think it's a worthwhile move.

If it's not in assembler, shouldn't be too tough a port (I've ported several projects from AVR to MSP430, and it's generally been straightforward).  The FRAM based parts have a nice hidden bonus as well: you can use the FRAM as RAM, gaining a bunch more memory and it's non-volatile.  It's not quite as fast as the RAM, so you have a choice of either running it with a wait state or running at 8MHz (I went with 8MHz, which was fast enough for my needs, and saved battery as well).  It's tricky, as the RAM and FRAM memory is non-contiguous, so you have to be careful with how you lay things out.  Additionally, some of the FRAM is addressed higher than (IIRC) 64k, and earlier compilers don't support that very well ("large" memory model).  I ended up using low FRAM as RAM and high FRAM as flash, which was good enough for my project.  These days, I'd hope the compiler issues have been ironed out.

For the curious, here's a picture of a project I built using an MSP430FR5969 as the CPU:

http://www.vitriol.com/images/tech/equipment/homer-bare-800.jpg

- John

[Laurence Wilkins]

Thansk for your thoughts, Greg. The final choice of processor is probably going to be influnced more by software development (and learning curve) that the hardware costs, given that I doubt I'll make even 50 of the finished NDMM (as it will henceforth be known). I can crash about a bit in Python, but the interpretted nature of that language gives me concerns over speed.  The boot-time and seemingly easy ability to "trash the SD card" are worries for me in an embedded product. I know nothing of FPGAs, so the learning curve for me could easily be the most "expensive" part (even if it was the neatest solution). Arduinos are everywhare, and I can probably leverage some of my experience with PIC processors (some C, but mostly assembler) in that direction. Dont actually need that much IO, just 3-4 serial channels and a few spare lines for keyboar scanning. What do the group make of TEENSY processors?

Laurence

[John Rehwinkel]

Thansk for your thoughts, Greg. The final choice of processor is probably going to be influnced more by software development (and learning curve) that the hardware costs, given that I doubt I'll make even 50 of the finished NDMM (as it will henceforth be known). I can crash about a bit in Python, but the interpretted nature of that language gives me concerns over speed.  The boot-time and seemingly easy ability to "trash the SD card" are worries for me in an embedded product. I know nothing of FPGAs, so the learning curve for me could easily be the most "expensive" part (even if it was the neatest solution). Arduinos are everywhare, and I can probably leverage some of my experience with PIC processors (some C, but mostly assembler) in that direction. Dont actually need that much IO, just 3-4 serial channels and a few spare lines for keyboar scanning. What do the group make of TEENSY processors?

The PJRC Teensy line of processors is very nice, I've used them in several projects.  The 3.x line of ARM CPUs is very capable and the size and price are attractive.  It would be tricky to find 3-4 serial channels on an AVR or Arduino (unless you used a lot of software serial), but several of the MSP430 chips offer several UARTs, as do the Teensy 3.0 parts.  They aren't as tightly integrated into the Arduino IDE, but I don't find that a large hindrance.

- John

[Dan Hollis]

I suspect arduino herd will migrate to esp82xx, not msp430.

-Dan

> http://www.vitriol.com/images/tech/equipment/homer-bare-800.jpg <http://www.vitriol.com/images/tech/equipment/homer-bare-800.jpg>
>
> - John

>
> --
> You received this message because you are subscribed to the Google Groups "neonixie-l" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to neonixie-l+...@googlegroups.com.

> To post to this group, send an email to neoni...@googlegroups.com.
> To view this discussion on the web, visit https://groups.google.com/d/msgid/neonixie-l/97DC78EF-ACEE-4386-BC31-2D9BB834B03D%40mac.com.

[Dan Hollis]

https://www.youtube.com/watch?v=Zy-scT8uGQ0

See around 0:51, looks like IN-18

Around 2:03, looks like maybe IN-14

-Dan

[Gene Segal]

Very nice find!  I concur with IN-18 in the first shot, but I think the second one might be IN-12 with its "upside down 5")

And don't y'all forget - in Russia, arcade game plays YOU!))

Dan Hollis

October 23, 2016 at 1:03 PM

--

Sent from Postbox

[Alex]

That is brilliant, I love the more mechatronic approach used on these machines! Thanks for the link!

[Konstantin]

Hi Dan,

Thanks for sharing this video. I've been there in August while my wife and
daughter spent their time in Hermitage ;-)

It was great time... and Yes, Nixie tubes I saw there are IN-12, IN-14 and
IN-18.
A few pictures are attached.


Konstantin
www.kosbo.com

[SWISSNIXIE - Jonathan F.]

Oh very nice to see real arcade machines with nixies!

P.S. Im currently working on a slotmachine with nixies! :) Like these spinning wheel things where you have to match 3 items, but instead with nixies.

[Quixotic Nixotic]

This rare pinball machine was made in 1935 by US company Pacific Amusement, PamCo. Neontact spells out "Two", "Five", "Ten or "Twenty" in neon letters in the backbox. This example has been restored by the AgraGames CoinOpMuseum in Holland.

John S

[Dan Hollis]

Looks like someone is offering edgelit modules for sale:

http://www.nixiepipe.com/

I was going to design my own PCB for something like this, looks like
someone saved me the hassle. Fortunately I have access to a laser cutter
so I can make my own digits.

-Dan