NixieStat Thermostat is closer to reality

[Joe Croft]

Happy New Year Yall!!

To open this new years, I have been continuing on my saga with the NixieStat Thermostat. I have a hackaday project with pictures and more here:

https://hackaday.io/project/4452-nixiestat

I want to make this a kit, either both a Nixie tube version as well as  7 segment  LED version. Unfortunately I don't have the cash on hand to do this in any quantity to start so I was thinking of turning to Kickstarter. The question is, what is the quantity of kits I should aim for? If I do 50 or each kit the approximate prices would be $115.00 for the LED version and $155.00 for the Nixie tube version (darned in17 are expensive!).

So the question is: Is one hundred kits to many or too few?

Looking forward to hear from yall. Thanks!

   Joe

[Dylan Distasio]

I would be interested in one.

--
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/CAE%2BVk6MmC9FZphTFzJGJ-5jaZVrsdaBvA%2BB7QTPTMxXpb%3DZV3g%40mail.gmail.com.
For more options, visit https://groups.google.com/d/optout.

[David Forbes]

Joe,

When deciding how many of a new thing to make for sale, I look for the
knee of the curve of cost-per-unit vs quantity, and aim about there. Aim
towards the lower end of quantity if uncertain or funds are short.

On 1/2/16 7:38 PM, Joe Croft wrote:
> Happy New Year Yall!!
>
> To open this new years, I have been continuing on my saga with the
> NixieStat Thermostat. I have a hackaday project with pictures and more here:
>
>

> So the question is: Is one hundred kits to many or too few?
>
> Looking forward to hear from yall. Thanks!
>
> Joe
>

--
David Forbes, Tucson AZ

[Terry Kennedy]

On Saturday, January 2, 2016 at 9:38:57 PM UTC-5, joenixie wrote:

I want to make this a kit, either both a Nixie tube version as well as  7 segment  LED version. Unfortunately I don't have the cash on hand to do this in any quantity to start so I was thinking of turning to Kickstarter. The question is, what is the quantity of kits I should aim for? If I do 50 or each kit the approximate prices would be $115.00 for the LED version and $155.00 for the Nixie tube version (darned in17 are expensive!).

So the question is: Is one hundred kits to many or too few? 

I'd suggest reading through the QR Clock saga - http://ch00ftech.com/category/qr-clock/ - while a lot of it deals with the issues of outsourcing manufacturing to China (for only 50 units), a lot of it covers crowdfunding (on Tindie, but a lot of the info applies to pretty much any crowdfunding site), the process of fulfilling the orders, etc. 

[John Rehwinkel]

I want to make this a kit, either both a Nixie tube version as well as  7 segment  LED version. Unfortunately I don't have the cash on hand to do this in any quantity to start so I was thinking of turning to Kickstarter. The question is, what is the quantity of kits I should aim for? If I do 50 or each kit the approximate prices would be $115.00 for the LED version and $155.00 for the Nixie tube version (darned in17 are expensive!).

So the question is: Is one hundred kits to many or too few?

There's an excellent write-up here from a kickstarter veteran (she's a muscian, but the concepts are broadly applicable) with a lot of good things to consider and how to run your numbers.  Long, but worth a read if you're new to kickstarter and/or want to minimize the chance of disaster.

http://mariancall.com/kickstarter-math-is-weird/

- John

[Keith Moore]

Perhaps open source it? If you want to make money, perhaps work out a license for the software. 

It would seem to be very difficult to estimate uptake on the kits and it would be highly likely that you'd not be able to match parts purchase to the quantity of buyers. 

I build almost exclusively from kits and designs (I know very little about electronics) and enjoy sourcing the parts and buying/making from a BOM and a PCB (or eagle file). 

This is just one guy's perspective but I hope it helps. 

- Keith  

On Saturday, January 2, 2016 at 9:38:57 PM UTC-5, joenixie wrote:

[NeonJohn]

Here's my opinion as Chief Engineer and majority owner of a company with
several products on the market.

*) Lose the solid state relays. They will surely be failed in short
order the first time a service call is initiated. HVAC guys like to
short low voltage wiring to look for power. Faster than getting out a
meter.

*) Lose the LED version. The Nixies are the hook that justifies the
high price. LEDs won't.

*) Not a good KickStarter project. They (the doners) tend to like big
projects. Maybe one of the other crowsourced sites that cater to small
projects.

*) I suggest using a quick flip etch and stuff house to assemble say, 5
A/T systems and get them out there to beta testers. You MUST beta test.
Our beta test policy is that the tester buys a machine at half price
and when the test is over, he gets a production machine at no extra
charge. Another approach is to put them out at cost. Either way you
recover your costs.

*) Consider putting it with SparkFun. They have three ways of selling
your projects.

https://learn.sparkfun.com/tutorials/how-to-sell-your-widget-on-sparkfun

I have several pieces of ancillary bits and pieces that I'm about to
offer. The first is a variable speed fan controller. Stick the
stamp-sized board to a heat sink, connect 9 to 30 volts to the input,
connect a 15 volt or higher fan to the output and viola. No noise
unless it's necessary to keep the sink at the setpoint. We use this
controller in our Roy induction heater so the design is well-proven.
Therefore I'll probably go with option 1.

*) Deal with and mock up/3d print the case first and get the approval
of several women of various backgrounds. No only does the lady of the
house control what goes on the furniture in many instances, they almost
always control the thermostat.

Figure out how you're going to manufacture your enclosure if you go the
custom route. I'm very unimpressed with the appearance of 3d printed
parts, even from industrial machines. I can tell you all about getting
enclosures manufactured in China, for that's what we had to do to handle
the cost.

*) Do a "design for manufacturing" and cost reduction pass across the
thing. Those Dallas temperature sensors are nice but a forward biased
diode will do the same job for a tiny fraction of the cost. You can
specify 4 wire telephone indoor plant wiring for the hookup and do a
Kelvin measurement to eliminate lead resistance effects.

Up until a couple of years ago I was seeing thermistors in extremely
cheap consumer goods. Now I see diodes.

*) Get hold of an ASHRAE handbook and learn how to properly implement
the outdoor temperature modulated algorithm. Hint: Torrent :-)

*) Also study the ASHRAE standard comfort map. This maps temperature
vs humidity vs comfort. The aim of your thermostat is not to control
temperature but to control comfort.

*) To that end, you need a humidity sensor so you can properly
implement the comfort algorithm.

*) You need outputs for de- and humidifiers for the same reason.
Particularly central humidifiers in the winter. One can maintain
comfort at a much lower temperature if the humidity is kept high which
will save the owner a lot in energy costs.

Here's a concrete example. Right now to keep my honey comfortable, the
house is at 79 degrees and 20% RH. I'm about to re-install the
water-line-connected humidifier in my propane furnace. (I take it out
during the summer to prevent lint build-up). I know that with 50%
humidity, she's comfortable at 69 deg. Major savings in fuel.

*) contemplate the liability you're taking on. A malfunction could
start a fire or let a house freeze up and burst pipes. At least get a
personal liability rider on your homeowner's policy. A megabuck's worth
of coverage is very cheap.

*) Consider your engineering approach, particularly your software
approach. Are you building self-checking and redundancy and other high
reliability techniques into your code? Using the Arduino black box
library doesn't give me fuzzy feelings in this area.

*) If you're going to make these in commercial quantities, consider
getting it listed. We work with ETL (I have a grudge against UL and
their nasty politics). Figure at least $5 large.

Well, that turned into an epistle! Just some things to think about, though.

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
https://www.etsy.com/shop/BarbraJoanOriginals <-- Affordable Fine Art
Originals
PGP key: wwwkeys.pgp.net: BCB68D77

[NeonJohn]

On 01/03/2016 10:37 AM, Keith Moore wrote:
> Perhaps open source it? If you want to make money, perhaps work out a
> license for the software.

Oh absolutely open source it. Hardware and software.

Open source built Fluxeon (my old induction heater company) and now it's
building Tnduction (my new one). We offered almost all our products as
open source. Bare boards, kits and A/T boards.

I think we sold all of 5 bare boards. The kits were a disaster. People
CANNOT solder, nor can they read resistor color codes. Or follow
instructions, for that matter. And after they tried to build a board
using a Weller soldering GUN, they'd want free support.

We sold quite a number of A/T board kits. Most of those sales resulted
in the sale of a Roy, our standard-bearer product. Open source got us
started and it continues to contribute to sales today.

John

[Mich...@aol.com]

Heh.

And I thought it was free lifetime support.

 

I just hate it when you sell or even give a friend/customer a computer and it's automatically expected that you provide all known software and lifetime support.  Uhhhg.

 

Michail

[A. Nonamus]

Heh, how does that firearms safety catchphrase go? "Never point a gun at any person or any thing, unless you want to see it killed or destroyed."

[Joe Croft]

Hi Yall,

Lots of food for thought here. As for kits, I'm not too afraid of kit builders. I've sold over 100 of my Nixieneon clock kits with very few issues. Granted, I never package two values of any given part into the same bag. I truly believe that saved me a lot of grief. Also, making the steps very small and explicate helped too.

I am now starting to have reservations about calling this a thermostat and also will be contemplating making a statement dissuading users from connecting it to their house. We'll see. Maybe a subtle change to make it a universal data logger with the ability to turn stuff on and off. I will take the time to look into the personal rider on the insurance. It will probably be a good idea in any case if I am selling kits of any sort.

I never thought of this as being a production item. The intention of this was always to be a kit. It was designed using either through hole components or pre-made boards such as the keyboard controller from sparkfun. Once the boards and components are gone (or the kits), it is done. I just need to recoup the cost of the parts and hopefully help fund the starting of the next project (not sure what it will be yet).

The idea of making an LED version was to pull in more hobbyists.

As for a case, I will be honest, as with my clock I find may people created their own cases and they all surpassed anything I am able to do! I would be very happy if I didn't have to supply a case at all ;)

-joe

--
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/568954E2.5080004%40neon-john.com.