EEVblog #413 - Mailbag

[NeoMorph]

Nooo Dave... don't do it. Religion has caused more wars than any other reason in history.

All the more reason for those of reason to stand up against the silly, childish, and archaic mass delusion that is organised religious belief. 
If someone brings it up, I'll set'em straight, you're welcome in advance 
Just like if you believe a $5 meter is good value... %-B

Dave.

Problem is the damned avalanche effect. I stopped trying to explain to peeps why gods can't exist and the god squad just get all uppity and end up causing more hassle than if I had just ignored them. I went anti religion after my fiance came back from a holiday to Salt Lake City... Her relatives over there had converted her to Mormonism (or whatever). She then tried to force the religion on me... That was one wedding shot down in flames, I can tell you.

Oh an a $5 meter is good value... if the shop assistant has mislabeled the Fluke 289's lol. 

[hairykiwi]

I'm surprised nobody's already done an Arduino based OSHW watch with them.
 

Not Arduino based, true, but I have been developing an OSHW watch module using one of the original Memory LCDs - for um, quite... a... long... time...


So I'm really excited to announce the recent launch of 'Open source Time Machine #2 - OTM-02' - an OSHW watch module designed using KiCad.
Here's the KiCad design files on github: https://github.com/hairykiwi/OTM-02
You'll find more technical details there, along with a BOM when its finalised - by the end of first week in Feb, all going to plan.

Barring any major setback, the goal is to have a functional prototype module ready to show around the end of Feb.



Case design files, (for a non-waterproof, DIY 3D printable case) will be posted just as soon as I've made the prototype watch module and form/fit tested it in a case - by the end of April, sooner if possible. This is how things were looking back in September - printing on a RepRap Prusa Mendel using 0.3mm layer height:


I've subsequently redeveloped the case design from the ground up, (with almost identical appearance) and printed it with 0.2 and 0.1mm layer height - with 0.2mm being best overall. At this scale, almost every detail counts in optimising a design for DIY 3D printing.

More case and dev prototype images and schematics (pdf) are here: bit.ly/mtmpublic

The project has been so long in the making simply because I've had to learn everything about MCUs, programming and 3D printing as I've progressed - and I still feel like a MCU novice.

Anyway, if you've got any suggestions for any improvements to the module itself - circuit topology, board layout or functionality, I'm keen to hear - either here or over at the thread I started on Energy Micro's Lizard Lounge Community Projects: http://forum.energymicro.com/topic/582-otm-02-open-source-time-machine-2/

Dave, if people want to discuss this more on the EEVBlog forum, I could post a new project thread - just let me know what you prefer.

If anyone is interested in contributing to, or leveraging what I've done so far, for their own derivative projects - be my guest - that's what OSHW is all about. 

Here's a little more project background, if you're still reading: Apart from a brief intro to MCUs with the 68HC11 in the mid 90's I started again more recently with TI's eZ430-F2013 and progressed (jumped) to one of the Olimex MSP430 dev boards tempted by it's built-in LCD. Big mistake - with no clue and no worked examples I could interpret, I only ever used it connected to a four line alphanumeric display. After general disillusionment with support from the MSP430 world, I discovered the world of ARM and in particular the Cortex-M3 powered STM Primer2. That brought my first big breakthrough at 'natively' driving a graphic LCD - largely thanks to another member of STM's Primer community who had made development of projects on 'bare metal' reasonably straightforward. Soon after, Energy Micro's ultra low power 32-bit ARM Cortex-M3 Gecko dev board appeared on the market and from there, things just started to come together (ok, much less slowly) but probably in no small part due to EM's enthusiastic customer support and especially as a novice - the decent range of examples in their app notes - I really can't say enough good about Energy Micro - Groovy baby!

After getting all ham-fisted with multimeter probes and blowing up my first first Gecko STK development continued with a Tiny Gecko STK connected to the smallest commercially available LCD I could find (but still MOQ of 100 ex Taiwan). I owe a big debt of thanks to Thomas Tuxen at Review Displays UK for his time and kind support during this phase. Then, the month after I first got that LCD working I discovered silver PNLC Memory LCDs - the same model mikeselectricstuff featured in his video - were available from Mouser in unit quantities. Woohooo! That was April 2011 and what an awesome breakthrough it was in terms of the availability of cool tech for the masses. After house renovations, life, work and other distractions edged their way back up the priority list, it took the rest of 2011 to get to the stage you see in the very first image, above. Finally, during a period of flailing-economy-driven me-time, I started on the circuit design, component hunting and case integration around April last year. At the end of July I began to realise just how much case is needed around an LCD, (coincidentally I discovered the MetaWatch Strata on Kickstarter - and just how big it is on an average-sized wrist when I took delivery of one in Dec. Really well done MetaWatch, but for daily wear it's just a bit too big.) And so I switched to the latest Memory LCD technology in September 2012, (the same one that Andlier featured in his video) when those units became available as samples - many thanks to  AvnetEmbedded UK for their generous support. Since then it's been a slow but steady process of PCB design tweaks followed by case design tweaks, sometimes by inspiration, but mostly by needing a break from one aspect of the design process.

Right now, I'm moderately confident the PCB is ready to prototype -  feature creep is probably the single biggest risk to tangible progress right now.

Regarding cases - Initially I'm targeting and working on an optimising case design for the DIY 3D printing scene, with printed metal case design(s) to follow. I'm thinking along the lines of supplying a basic case design as OSHW that people can modify and/or embellish with their own unique design details. Right now 'fine detail' and 'DIY 3D printing' don't really go hand-in-hand, but with various on and off-line 3D metal printing bureaus, and the on-going improvements in the DIY 3D printing world, the future looks quite exciting for small batch high-tech manufacturing.

Looking forward to hearing your thoughts.
- Hamish

[mikeselectricstuff]

Dave, if people want to discuss this more on the EEVBlog forum, I could post a new project thread - just let me know what you prefer.

Yes - probably best to split to a new thread in projects

many thanks to  AvnetEmbedded UK for their generous support.

Avnet must have changed a lot since I last dealt with them many years ago, or you got lucky - they initially refused to take a £3000 order as they "don't do proformas" and couldn't set up an account due to insufficient credit references....

Looking forward to hearing your thoughts.

My thinking on a memlcd watch was to make it as small as possible, using a  coin cell to minimise battery size as it should be possible to get a decent lifetime.
I sort of stalled at the stage of how to do the casing though, as that's the key to making something useful rather than a toy.
Is 3d print really durable enough? One idea I had was to make a case with a sandwich of stacked PCB material that was sanded smooth.

Incidentally in my experiments with  memlcd I've found that there is no advantage  to having the frame frequency faster than about 1-2Hz - at some extreme angles you can just see a very slight flicker but nothing of concern. As the frame update is one of the main static power draws, keeping it low helps a lot.

[chickenHeadKnob]

I don't want to add to the topic drift but I have a question for you watch implementers (mike and kiwi). Have you folks considered a carabiner watch style packaging? I find after going two years without a wrist watch I only miss it occasionally. I was very hard on wristwatches breaking them in various ways during construction and other outdoor sport activities.

I live in the outdoor playground that is coastal British Columbia and what I have really wanted is a completely waterproof  and rugged  watch/altimeter that I can clip on my belt or belt loop. I  don't need GPS as I always know where I am  . The cheap carabiner watches I have looked at don't appeal to me. The altimeter should have a cumulative gain-loss for hiking in the mountains.

I know the packaging aspect is perhaps the most difficult part and figure a larger format carabiner -say about 10 cm but still pocket sized, would actually ease the task.

anyway good luck with your endeavours

[Alexei.Polkhanov]

Just a followup - ITO glass has arrived. It really does have relatively low resistance - see attached photo. Etching it will be next step I guess 

[mikeselectricstuff]

For the HV I think I just used a CCFL backlight inverter and a needle

[hairykiwi]

Thanks Mike, I'll start another thread for my project soon.

Sorry this reply was a little slow coming...

I probably did get lucky at Avnet Embedded. Nonetheless, I really appreciate their support. And they seemed happy enough to take my money by proforma invoice + bank transfer yesterday.

Initially, I went down the coin cell battery route too. But things have really moved on in the last 12 - 18 months in miniature Li-Po battery land.

In my opinion, coin cells have a couple of detractions:

• They need changing and that seems wasteful if recharging is at all possible.
• mounted directly to a PCB, they take up a valuable PCB real estate.

My experience with designing and making DIY 3D printed parts with a RepRap printer using extruded Poly Lactic Acid (PLA) - from watch cases to cycle accessories to snorkelling accessories to battery holders - is it's plenty durable enough. And if it does break, you print another one. It has its aesthetic limitations, sure, but up to a point those can also be features. I really see DIY 3D printing as it currently stands not as the best thing since sliced bread, but rather as an enabling technology - a step in the right direction for micro manufacturing.

I believe some of the high-end Swiss watch manufacturers are already using 3D printing for making metal cases - especially for printing parts in precious metals.

I agree with your thoughts on low frame rate contributing favourably to efficiency - though I never needed more than 1 or 2Hz refresh with my dev prototype watch anyway - it's either 1/60 Hz with seconds (main time) turned off, 1 Hz with seconds refresh, or 2Hz with chronograph running + seconds refreshing asynchronously.

I really like your laminated PCB material case idea. You could possibly incorporate capacitive sense buttons by creating a paralleled bank of plates commonned together through a series of vertically aligned pads that were connecting during assembly. A flex-rigid PCB might be another (expensive option) for integrating touch pads. Who said art had to be cheap?

---

chickenHeadKnob -
Nice Idea - and a carabiner style case should be reasonably straight forward as a DIY 3D printing project.  I'd be tempted to screw a metal fixture into a printed plastic case rather than relying on a printed carabiner hook, but both styles would be worth experimenting with.

An tap/double tap accelerometer based, or capacitive-touch based user interface would help with waterproofing.

Do you know of any waterproof micro-sized absolute pressure sensors? A friend of mine was working on a dive computer project and he found it impossible to buy a micro-sized pressure sensor. I found some mini-sized sensors at a trade show recently - OK for diving applications perhaps, but still too big for a watch.

Can anyone recommend any sources for micro-sized / micro-power absolute pressure gauges?


- Hamish

[chickenHeadKnob]

chickenHeadKnob -
Nice Idea - and a carabiner style case should be reasonably straight forward as a DIY 3D printing project.  I'd be tempted to screw a metal fixture into a printed plastic case rather than relying on a printed carabiner hook, but both styles would be worth experimenting with.

An tap/double tap accelerometer based, or capacitive-touch based user interface would help with waterproofing.

Do you know of any waterproof micro-sized absolute pressure sensors? A friend of mine was working on a dive computer project and he found it impossible to buy a micro-sized pressure sensor. I found some mini-sized sensors at a trade show recently - OK for diving applications perhaps, but still too big for a watch.

Can anyone recommend any sources for micro-sized / micro-power absolute pressure gauges?


- Hamish

I didn't totally think through all the waterproofing implications . The cheap and omnipresent Bosch BMP085 pressure sensor has this to say in its spec sheet:

- for the device housing appropriate venting needs to be provided in case the ambient pressure shall be measured;
  if waterproof packaging is needed, venting can be accomplished by a vent element with a membrane like Gore-   Tex™

- liquids shall not come in contact with the device

- the clearance above the metal lid shall be .1 mm minimum

- During operation the sensor is sensitive to light; therefore the hole in the to lid shall not be exposed to direct light during operation

--end quote
Bosch's stupid pdf was locked so I hand transcribed the above. hope that helps, I wouldn't have thought of Gore-Tex. However I would think  it would turn an absolute sensor into something of a variometer with delayed response.

I am very interested in your project even if it results in something I have to mod for my own needs, Proceed!

[hairykiwi]

chickenHeadKnob - Thanks very much for the lead on Bosch's range of sensors! 

The BMP085 is a great starting point, but at 5 x 5 x 1.2 mm, it really is in the XXXL department for watch components.   

Some of Bosch's other pressure sensing components look very promising though:
The BMP180 (3.6 x 3.8 x 0.93 mm) might just fit with a small tweak to the current board, while the BMP280 (2.0 x 2.5 x 0.95 mm) would almost certainly fit. The BMP280 is a recent release though - 092012 on the flyer I downloaded - so availability could be a problem.

Their accelerometers are worth looking at too.

Thanks again,
Hamish