Wednesday 9 June 2010

Comment moderation

Due to excessive spam in #gta02-core blog (mostly porn links) I'm enabling comment moderation. I'm sure you understand.


Tuesday 20 April 2010

(Kinda) Weekly News for April 20th, 2010

Welcome to this edition (sixth) of #gta02-core news, for April 20, 2010.

Why no posts since October 2009?

My fault, I'd should update this more often, but there were actually several reasons. Some changes on my personal daily life, many projects going on, work, few developments on #gta02-core project, all account for the lack of news.
I hope at least the latter will force me to update this blog, but that will depend on a lot of factors.

So, what are the latest developments ?

Well, let's try to summarize. A lot of background work has been done in fped and other not-so-visible areas of the project. Our beloved mentor (Werner) also had a twist on his life (he moved), and no one actually stepped up for his "job". Most news are actually not so good news, I'll try to explore those as we move along.

Fedora Electronic Lab

The guys at Fedora included support for kicad, fped and openocd on their Fedora-12 release, making development of Electronic Hardware Development a lot easier on their platform.

The pending ECN

A lot of ECN have been worked on: ECN0019 (buzz fix), ECN0024 (GPS ground and layout), ECN0010 (Audio amplifier) [which includes changes to power unit], ECN0015 (Calypso headset), ECN0032 (EMI/ESD protection), ECN0017 (BT to main PCB), and others like
ECN0031, ECN0033, ECN0036, ECN0037, ECN0011 (varistor cleanup), ECN0008 (mistery caps), and ECN0038.

Component status

Trying to understand what components actually Openmoko had in stock started last year (as you can read on this post).
Werner posted a list of the inventory back in January, and the the BOM process we ought to
follow on this post, followed by a list of missing items.

Then, we realised we have another big problem - some components are not easily sourceable, so we need to move to other alternatives.

To focus, or refocus

I did a lenghty post back in February about to focus on current hardware, or instead updating our design to match newest devices, like iPhone and HTC. We'll stick to our
current design.

Cashier needed

We need to figure out how to deal with all donations project is receiving, and perhaps have someone focused on dealing this this issues. If you're interested, please stand up.
There's also an discussion going on with Jon "maddog" Hall about the subject.

Researching component alternatives

Some of our components are hard to source, and there's work to be done to find where we can get those components from. For some of them we need to look up alternatives, if you find something similar and easy to source, please let us know. One late component that was found hard to source is the backlight LCD inductor. Alternatives were using a larger footprint but max. current compliant one, or to use a smaller one and live on the incertainty. We decided for the larger one. Another list of missing components came up, so if you have some time helping us locating those, please do.

Again the backup battery

ECN0040 was revisited recently, since we actually never had a concrete desision on this. We'll probably move to a large capacitor and drop the battery.


Whoa ! I should frame this, nail it to the wall, and show it to any
Openmoko veteran who comes visiting ;-)

Werner Almesberger

This news is a great Christmas gift ;)
Patryk "LeadMan" Benderz

Done - wouldn't want anything fishy going on :-)
Dave Ball

Sunday 4 October 2009

Weekly News for October 4th

Welcome to the fourth edition of #gta02-core-news, for October 4, 2009.

Last week news

As you might have noticed, I have not posted any news entry last week. There was so little done and so few messages exchanged on the mailing list I decided to wait for another week to post.

ECN progress

Dave closed some of the pending ECN for EMI filters, SPI degugging and Telit integration. He also noticed some others are still not closed, despite their changes appear on the schematics already. Most of them only need small updates, except for the mistery caps for the GPS RF. Rene gave some ideas, but I have not had the time to properly analyse the problem. Maybe this week. Rask also updated some ECN with new test points.

Backup battery - to be or not to be ?

Rask did us some update regarding the removal of backup battery. There is no conclusion yet about this matter, so unless someone comes up with a good reason to change it, it will be kept as-is.

USP SMT line

Werner posted some more pics of the SMT line at University of São Paulo, Brasil. Very well documented, with references to the individual machines. Definitely worth a look.

Quotes of the Week

(...) I'll be quite busy carting equipment around and doing a gazillion of little administrative things. (The logistics of making phones are ridiculously easy in comparison ;-)
Werner Almesberger

I can most certainly help out with any kernel related issues, and even do some minor board modifications after the fact.
Christopher Friedt

I don't think we've had problems with undefined state on power up. If we manage to power up at all, we can (and do) load our own settings anyway.
Rask Ingemann Lambertsen

Sunday 20 September 2009

Weekly News for 20th September

Welcome to the third edition of #gta02-core-news, for September 20th 2009.


Some good news from USP, Brasil!. Read below for more details. Besides that project is evolving slowly. I hope we can start the layout soon.

Finances again

Werner has posted a lenghty message about the state of project finances, where he states "I'd estimate the total budget of the project that will still need covering, with pooled and individual costs at something like EUR 15'000-25'000", not counting other costs like usage of openmoko infrastructure and the components we'll be "offered".

News from Brasil - USP

Some great news for us all. Werner went to Brasil this week, to USP (University of São Paulo) and was happy to tell us that Prof. Marcelo Knoerich Zuffo of the University of Sao Paulo (USP) has offered gta02-core the use of his SMT facility. As if this wasn't generous enough, all he asks for in return are a few of our prototypes for his team to play with.. There was much rejoice, because this opens way for quicker and cheaper production of the first prototypes. He also took a pic of the SMT facility. Many many thanks to Prof. Marcelo Zuffo, and to Jon "maddog" Hall for his help (although he declines all and any responsability for it :P). There was some discussion on the same thread about PCB manufacturers, their capabilities and cost. We have quite a lot to choose from, you can read all the gory details on the mailing list archives.

Quotes of the week

It did not take a lot of convincing.
Jon "maddog" Hall

IIRC, we still have a hole where the connector for the external GPS antenna used to be.
Rask Ingemann Lambertsen

Heh, the things we do to make it into Alvaro's quotes of the week ;-)
Werner Almesberger

Sunday 13 September 2009

Weekly News for 13th September

Welcome to the second edition of #gta02-core-news, for September 13th 2009.

About this blog

Hasn't been easy to write this on Fridays, so I've decided to do it during the weekend, when I have more spare time. Thanks for your understanding :)


Again this has been a very quiet week. Only a handfull of emails were exchanged on the mailing list, and those only cover a few aspects of the project. I guess most of us are back to work after the big holydays, meaning less time to work on #gta02-core. I expect more people to join us, now that we are moving torwards completion of the first prototypes.


There's an ongoing discussion on the mailing list about project finances and prototype cost. Werner did an initial estimation of 500-1000€ perprototype (these are prototypes so there is some amount of uncertainity about whether they will fully work, work to some extent, or just draw some smoke when we first power them on).
Getting some funding to help us making the first prototypes has been also on top of the table, but this is very time consuming and sometimes even disappointing. Nathael brought us attention to the fact some of us cannot really afford that prototype cost, but might be interested in donating some smaller amounts of money to the project, and even build some website to capture external donations. Let's see how this idea evolves on the next weeks.

The new GSM modem

As you know we dropped our old GSM modem (TI Calypso) and replaced it with a Telit one. This week Dave Ball decided to have a first stab at schematics, and raised a lengthy discussion about RF, and power issues (glitches, ON/OFF circuitry). One important aspect of the discussion is removal of external GSM antenna circuit. Addition of some more test points was also discussed, these will surely make all those hardware hackers happier for sure.

Quotes of the week

Ah, what's the trouble ? I'm good at fixing fped ;-)

Werner Almesberger, author of fped

By the responses we've had so far, I'm happy to go with the internal antenna only, which should be simple enough even for me! :-)

Dave Ball

Saturday 5 September 2009

Weekly News for 5th September

Welcome to the first edition of #gta02-core-news, for September 5th 2009.


There has been not much progress this week. Mostly discussion about shielding, debug connector FPC and audio ECN collection. One OT discussion about binary drivers was also very active. And a new schematic - wireless on-board.

Bad news from OpenMoko

Werner brought us some piece of bad news, apparently OpenMoko might not have as many parts in stock as initially thought. We'll have to wait for more information now, but this might put us on a bad position, because some parts are very hard to source.


Rask has asked a local supplier a quote for custom shields for #gta02-core. I think personally this prices are a bit high, we might have to live with our current shields if possible.

The footprint problems

Werner brought our attention to the fact that several manufacturers have different values for footprint parameters. This will require us to choose some common values ans use them, instead of using some non existent industry standard.

New on-board wlan

Your blogger finished the 1st version of the on-board WLAN module. There's an ECN pending for this module too. Some component names had to be renamed but there will exist a mapping somewhere.

SPI and debug connector

Rask found some inconsistencies in the way we use SPI on the debug connector, which might cause debug connector to interfere with WLAN.

Printed Circuit Board

There has been some discussion on board parameters, like minimum trace widths and vias. Other parameters were also discussed, like surface finishes. We're still trying to find out a good balance between our requirements and total board cost.

Quotes of the week

" At the end, i think we should consider that this will be *only* a prototype and not a usable phone, so we don't need a lot of gimmicks which are not absolutely required. "
Rene Harder

" It really helps then if you don't have to suspect every single solder joint to be a traitor to the cause ;-) "
Werner Almesberger

Thursday 3 September 2009

Goals and Objective

Hello all,

Although I am not a blogger, I think we need alternative ways to let community know what we are doing at #gta02-core project. So I decided to start this blog and hope it will help community to get in touch with latest news from our project.

For those who do not have a clue about what #gta02-core is about, take a look here

If you still have doubts, please drop me an email and I'll be glad to dissipate those. But there should be enough information there, like mailing list address and archives.

So I'll try to weekly post news about the project here. Very summarized, since I am not someone that speaks or writes too much.

I will try to post here every Friday.

Let the power of Open Design Hardware be with you.