Status & next milestone for GTA02-core?

classic Classic list List threaded Threaded
13 messages Options
Reply | Threaded
Open this post in threaded view
|

Status & next milestone for GTA02-core?

Ron K. Jeffries
Am curious what the next milestone is for GTA02-core.
Have not been following the project for a while, so am
wondering when you expect to be ready to have PCBs
fabricated.

My context: starting research for possible article
about open hardware.

---
Ron K. Jeffries
http://blog.eronj.com

_______________________________________________
gta02-core mailing list
[hidden email]
https://lists.openmoko.org/mailman/listinfo/gta02-core
Reply | Threaded
Open this post in threaded view
|

Re: Status & next milestone for GTA02-core?

Dave Ball
Ron K. Jeffries wrote:
> Am curious what the next milestone is for GTA02-core.
> Have not been following the project for a while, so am
> wondering when you expect to be ready to have PCBs
> fabricated.
>  

Hi Ron,

I'll try and help, though you shouldn't count me as an official
spokesman for the project :-)

Status:  Currently we have drawn the schematics (excluding GSM), and are
in the process of reviewing those to identify errors (bugs), and working
through ECN's.

Once that's complete our next milestone will be work producing the
footprints, and then design the PCBs

If current discussions ([1][2]) turn out favourably, we _might_ be in a
position to fabricate in mid/late sept.

You've probably seen the project details at [3], the most important of
which is that this project is aimed at taking the design as far as
prototypes.  As the designs are open, my hope is that this or our next
design will be of interest to someone who has the resources to
manufacture, though we don't currently have the resources to do this
ourselves.

We've recently adopted the phrase "open design hardware", and have some
background in the "disclosure requisites" white paper Werner produced
[4][5].

Anyone that's interested would be welcome to join us, and help with both
reviewing the current schematics [6], and moving onto footprint and
board design.


Hope this helps!

Dave



[1] http://n2.nabble.com/Calypso%27s-last-dance-tp3307295ef1958.html
[2]
http://n2.nabble.com/The-University-of-S%C3%A3o-Paulo%27s-intent-to-join-Openmoko-development-tp3250060ef1958.html
[3] http://wiki.openmoko.org/wiki/Gta02-core
[4] http://people.openmoko.org/werner/oshwdr.pdf
[5]  
http://n2.nabble.com/RFC%3A-Open-Source-Hardware---Disclosure-Requisites-tp3306999ef1958.html
[6] http://people.openmoko.org/werner/gta02-core/gta02-core-all.pdf


_______________________________________________
gta02-core mailing list
[hidden email]
https://lists.openmoko.org/mailman/listinfo/gta02-core
Reply | Threaded
Open this post in threaded view
|

Re: Status & next milestone for GTA02-core?

AlvieBoy
Dave Ball wrote:

> Ron K. Jeffries wrote:
>> Am curious what the next milestone is for GTA02-core.
>> Have not been following the project for a while, so am
>> wondering when you expect to be ready to have PCBs
>> fabricated.
>>  
>
> Hi Ron,
>
> I'll try and help, though you shouldn't count me as an official
> spokesman for the project :-)

Aren't we all ? :P

> Status:  Currently we have drawn the schematics (excluding GSM), and are
> in the process of reviewing those to identify errors (bugs), and working
> through ECN's.
>
> Once that's complete our next milestone will be work producing the
> footprints, and then design the PCBs

We also need to first gather some technical requirements for PCB manufacture (we are not yet sure who it will be). USP (University of São Paulo, Brasil) got in
touch with us and might be our partner in PCB assembly, at least for prototypes and testing. (It's not official yet, Werner is "in charge" of speaking with them
and matching goals) :P

Also we are stuck ATM with some legal issues (TI Calypso) and trying to reach a fitting solution. Might take a while.

So stay tuned, and of course any help on those issues will be highly appreciated.

Best,

Álvaro


_______________________________________________
gta02-core mailing list
[hidden email]
https://lists.openmoko.org/mailman/listinfo/gta02-core
Reply | Threaded
Open this post in threaded view
|

Re: Status & next milestone for GTA02-core?

Dave Ball
Álvaro Lopes wrote:
> Dave Ball wrote:
>> shouldn't count me as an official spokesman for the project
>
> Aren't we all ? :P

Indeed - All of us and none of us!  :-)

>
> We also need to first gather some technical requirements for PCB
> manufacture (we are not yet sure who it will be).

Yup - Werner's list at
http://lists.openmoko.org/pipermail/gta02-core/2009-July/000275.html 
looks like a good start, though as you pointed out more details still
need agreeing with whoever ends up producing the things!

> USP (University of São Paulo, Brasil) got in touch with us and might
> be our partner in PCB assembly, at least for prototypes and testing.
> (It's not official yet, Werner is "in charge" of speaking with them
> and matching goals) :P

I'll let Werner speak up on that when / if there's progress, but last I
heard sounded encouraging.  I'm keeping my fingers crossed it works out
- it certainly sounds like we and USP have similar goals  :-)

> Also we are stuck ATM with some legal issues (TI Calypso) and trying
> to reach a fitting solution. Might take a while.

Entertainingly, it seems the nabble permalinks I used in my last mail
didn't work - here they are again.  [1] is the calypso (and alternative)
discussion on the mailing list, [2] the USP discussion, and [5] the
disclosure requirements / ODH (open design hardware) discussion.

[1] http://lists.openmoko.org/pipermail/gta02-core/2009-July/000239.html
[2] http://lists.openmoko.org/pipermail/community/2009-July/051057.html
[5] http://lists.openmoko.org/pipermail/gta02-core/2009-July/000238.html

Sorry about that folks!

For anyone that's interested, the TI legal issue is that the calypso
documentation and OM's designs are under NDA, so iirc the only way for
us to use that device would be if TI relaxed the terms of the NDA. A
potential alternative to the calypso (option gtm501) has been
identified, though we don't yet know for sure that we can use it.


Cheers,
Dave



_______________________________________________
gta02-core mailing list
[hidden email]
https://lists.openmoko.org/mailman/listinfo/gta02-core
Reply | Threaded
Open this post in threaded view
|

Re: Status & next milestone for GTA02-core?

Werner Almesberger
In reply to this post by AlvieBoy
?lvaro Lopes wrote:
> Aren't we all ? :P

I'd consider it a good thing that there are several people who know
what's going on in the project :-)

My current esimate is that we'll indeed SMT mid to late September:

- two weeks of work on footprint definitions in August,
  one week of this overlapped with ...

- two weeks of layout

- one week of review,

- two weeks of PCB making,

- and then SMT.

The schedule probably needs a week or two of buffer.

The PCB should be relatively uncritical, so I'd select a site based
on where SMT will be done. I've been in contact with USP for the last
few weeks, directly and indirectly, and I'm still not sure if they're
ready to commit to running gta02-core.

In case this doesn't pan out, there may be another opportunity for
doing SMT at a site that isn't just a prototype manufacturer: a while
ago, someone from a company whose R&D group has an SMT line has
suggested they may help. But I still have to find out if they're
serious and if their line has the capabilities needed for gta02-core.

- Werner

_______________________________________________
gta02-core mailing list
[hidden email]
https://lists.openmoko.org/mailman/listinfo/gta02-core
Reply | Threaded
Open this post in threaded view
|

Re: Status & next milestone for GTA02-core?

Werner Almesberger
In reply to this post by Dave Ball
Dave Ball wrote:
> potential alternative to the calypso (option gtm501) has been
> identified, though we don't yet know for sure that we can use it.

By the way, there may be a third possibility for telephony: the
Telit GE865-QUAD GSM module:
http://www.telit.com/en/products/gsm-gprs.php?p_id=12&p_ac=show&p=11

Besides only being 2g, the main problem of that one is its thickness
of 3.0 mm, 0.7 mm more than the GTM501. I don't know if that would
fit. What's great about Telit is that all their documentation is
openly available and they have a more developed global reseller
network than Option.

- Werner

_______________________________________________
gta02-core mailing list
[hidden email]
https://lists.openmoko.org/mailman/listinfo/gta02-core
Reply | Threaded
Open this post in threaded view
|

GSM modem [was Re. Status & next milestone for GTA02-core?]

Dave Ball
Werner Almesberger wrote:

> Dave Ball wrote:
>  
>> potential alternative to the calypso (option gtm501) has been
>> identified, though we don't yet know for sure that we can use it.
>>    
>
> By the way, there may be a third possibility for telephony: the
> Telit GE865-QUAD GSM module:
> http://www.telit.com/en/products/gsm-gprs.php?p_id=12&p_ac=show&p=11
>  

I keep forgetting about the Telit :-)  I prefer the option because it's
smaller (I don't like the idea of filing down plastics to get a larger
module to fit), it's 3G and newer.

Having said that, the Telit does have a UART interface (where the option
appears to be USB only) - so should be easier to fit into our existing
design.


Do we have samples of the telit available for prototypes, or if not
would it be worth investigating getting hold of some at the same time as
the Option option is explored?

Then at least we have a firm fallback.

I get the impression that the calypso isn't going anywhere, so how about
getting started with Telit based schematics while we're waiting to find
out if the Option is a goer or not?


Dave

_______________________________________________
gta02-core mailing list
[hidden email]
https://lists.openmoko.org/mailman/listinfo/gta02-core
Reply | Threaded
Open this post in threaded view
|

Re: GSM modem [was Re. Status & next milestone for GTA02-core?]

AlvieBoy
Dave Ball wrote:

> Werner Almesberger wrote:
>> Dave Ball wrote:
>>  
>>> potential alternative to the calypso (option gtm501) has been
>>> identified, though we don't yet know for sure that we can use it.
>>>    
>> By the way, there may be a third possibility for telephony: the
>> Telit GE865-QUAD GSM module:
>> http://www.telit.com/en/products/gsm-gprs.php?p_id=12&p_ac=show&p=11
>>  
>
> I keep forgetting about the Telit :-)  I prefer the option because it's
> smaller (I don't like the idea of filing down plastics to get a larger
> module to fit), it's 3G and newer.

I'll be honest here. Even at "typical" consumption the Option chip seems to draw up to 1.1A @ 3.6V. This is completely absurd for a mobile device. We draw at
most 2.5W from USB (500mA@5v). Chip consumes 3.96W. Completely absurd. I did not find anything about "idle" consumption.

> Having said that, the Telit does have a UART interface (where the option
> appears to be USB only) - so should be easier to fit into our existing
> design.

True, but USB is less error-prone, less sensitive to EMI.

> I get the impression that the calypso isn't going anywhere, so how about
> getting started with Telit based schematics while we're waiting to find
> out if the Option is a goer or not?

I'm with you Dave.

As long as we can get the schematic "non-intrusive", meaning it does not require changing the already drawn ones.

Álvaro

_______________________________________________
gta02-core mailing list
[hidden email]
https://lists.openmoko.org/mailman/listinfo/gta02-core
Reply | Threaded
Open this post in threaded view
|

Re: GSM modem [was Re. Status & next milestone for GTA02-core?]

Werner Almesberger
In reply to this post by Dave Ball
Dave Ball wrote:
> I get the impression that the calypso isn't going anywhere, so how
> about getting started with Telit based schematics while we're
> waiting to find out if the Option is a goer or not?

I like the idea ! The worst case outcome is that it would not fit
under the WLAN, in which case we'd have a WLAN XOR telephony
placement option. As far as prototypes go, not too shocking a
solution. (Besides, the AR6001 would be a certain death row
candidate for anything that follows, so not much is lost there.)

- Werner

_______________________________________________
gta02-core mailing list
[hidden email]
https://lists.openmoko.org/mailman/listinfo/gta02-core
Reply | Threaded
Open this post in threaded view
|

Re: GSM modem [was Re. Status & next milestone for GTA02-core?]

Werner Almesberger
In reply to this post by AlvieBoy
?lvaro Lopes wrote:
> I'll be honest here. Even at "typical" consumption the Option chip seems
> to draw up to 1.1A @ 3.6V.

I think this is maximum average (over some period of time) current.
E.g., when sending, you wouldn't draw more than 4 Ws within a 1 s
interval, some of which will be in the form of 3A spikes.

> True, but USB is less error-prone, less sensitive to EMI.

Bah, if we can't get UART right over a few cm ... ;-) One problem
with USB is that it has periodic activity, even if "idle". So you
need elaborate fallback schemes to make it really stop doing
things.

> As long as we can get the schematic "non-intrusive", meaning it does
> not require changing the already drawn ones.

Telit should be a fairly direct replacement for the Calypso.
Option would be worse, because it means that BT and modem need
to be rotated, since we only have a single USB bus. Also, I've
never seen our BT module being used with UART.

- Werner

_______________________________________________
gta02-core mailing list
[hidden email]
https://lists.openmoko.org/mailman/listinfo/gta02-core
Reply | Threaded
Open this post in threaded view
|

Re: GSM modem

Rask Ingemann Lambertsen-2
On Mon, Aug 03, 2009 at 11:29:20AM -0300, Werner Almesberger wrote:

> Telit should be a fairly direct replacement for the Calypso.
> Option would be worse, because it means that BT and modem need
> to be rotated, since we only have a single USB bus. Also, I've
> never seen our BT module being used with UART.

   Regardless of Telit/UART or Option/USB, we need to answer a question:
How do we wake up from suspend on incoming call/SMS? Remember that the
Calypso couldn't do that out of the box, it needed a firmware modification.

   The lack of USB busses could perhaps be solved by including a USB hub. We
ought to have some room to spare with the removal of the Glamo, NOR flash,
audio amplifier. If we can live with the power consumption of a USB hub, it
would solve the Option 3G vs. Bluetooth conflict and still have two USB
ports that could be useful for hardware hacks.

--
Rask Ingemann Lambertsen
Danish law requires addresses in e-mail to be logged and stored for a year

_______________________________________________
gta02-core mailing list
[hidden email]
https://lists.openmoko.org/mailman/listinfo/gta02-core
Reply | Threaded
Open this post in threaded view
|

Re: GSM modem [was Re. Status & next milestone for GTA02-core?]

Rúben Leote Mendes
In reply to this post by Werner Almesberger
Em Seg, Agosto 3, 2009 15:24, Werner Almesberger escreveu:

> Dave Ball wrote:
>> I get the impression that the calypso isn't going anywhere, so how
>> about getting started with Telit based schematics while we're
>> waiting to find out if the Option is a goer or not?
>
> I like the idea ! The worst case outcome is that it would not fit
> under the WLAN, in which case we'd have a WLAN XOR telephony
> placement option. As far as prototypes go, not too shocking a
> solution. (Besides, the AR6001 would be a certain death row
> candidate for anything that follows, so not much is lost there.)

Crazy idea #1: Remove GSM and even maybe GPS. Build an external bluetooth
enabled GSM+GPS radio and find an off-the-shelf plastic case (with belt
clip) big enough to fit the prefered (price, license, ...) modules.
Problem: can the processor wake up from bluetooth when a call comes in?

Crazy idea #2: Build the phone to fit another case. Not viable for mass
production but ok for a prototype. Some cheap cases:

http://www.dealextreme.com/search.dx/search.housing%20case

Crazy idea #3: do both of the above. ;-)

Ruben



_______________________________________________
gta02-core mailing list
[hidden email]
https://lists.openmoko.org/mailman/listinfo/gta02-core
Reply | Threaded
Open this post in threaded view
|

Re: GSM modem

Werner Almesberger
In reply to this post by Rask Ingemann Lambertsen-2
Rask Ingemann Lambertsen wrote:
>    Regardless of Telit/UART or Option/USB, we need to answer a question:
> How do we wake up from suspend on incoming call/SMS?

That's a very good question, and one for which I haven't been able to
find a good answer in the Telit documentation so far. There is a
low-power mode (CFUN=5), but it appears to be entirely host-controlled,
i.e., without letting the modem speak unless first spoken to.

There are things like ring indication that may or may not be suitable
for waking the CPU.

> If we can live with the power consumption of a USB hub,

Hmm, I'm sceptical about that. Anyway, we don't need to have that
degree of flexibility in gta02-core. If a future device will need a
lot more USB, it will go through its own set of prototypes anyway, so
there will be room for experiments.

- Werner

_______________________________________________
gta02-core mailing list
[hidden email]
https://lists.openmoko.org/mailman/listinfo/gta02-core