Re: [ARTIQ] uTCA backplane driver choices

[Grzegorz Kasprowicz]

Hi
The SATA type connector was installed as an option to route GTX normally
routed to AMC. Selection is by coupling capacitors placement.
The AMC-SFP adapter is another, better option. As I wrote before, my
colleague designs such thing for another project and we can use it.
Greg

On 5 July 2016 at 18:17, Sébastien Bourdeauducq  wrote:

> On Tuesday, June 28, 2016 04:00 PM, Grzegorz Kasprowicz wrote:
>
>> For Sayma board to not waste precious GTX we will add PHY with
>> 1000Base-X. We can also route it to SATA connector and provide dedicated
>> SATA-SFP cables. Most of these PHYs apart from RGMII to 1000base-X offer
>> 1000base-T, so we can also put low profile or vertical RJ45 connector
>> for quick access. Full size SFP won't fit:)
>>
>
> If it is not possible to fit a SFP, a direct transceiver connection e.g.
> SATA connector, would be the second-best option. Ethernet PHYs add (a bit
> of) complexity to the board and do not support White Rabbit-style time
> transfer. But do we have a spare GTX for such a connector anyway? Won't we
> need to keep using the backplane connector (with an SFP adapter) for the
> transceivers when the Sayma board is stand-alone?
>
> Sébastien
>
___
ARTIQ mailing list
https://ssl.serverraum.org/lists/listinfo/artiq

Re: [ARTIQ] uTCA backplane driver choices

[Sébastien Bourdeauducq]

On Tuesday, June 28, 2016 04:00 PM, Grzegorz Kasprowicz wrote:

For Sayma board to not waste precious GTX we will add PHY with
1000Base-X. We can also route it to SATA connector and provide dedicated
SATA-SFP cables. Most of these PHYs apart from RGMII to 1000base-X offer
1000base-T, so we can also put low profile or vertical RJ45 connector
for quick access. Full size SFP won't fit:)


If it is not possible to fit a SFP, a direct transceiver connection e.g. 
SATA connector, would be the second-best option. Ethernet PHYs add (a 
bit of) complexity to the board and do not support White Rabbit-style 
time transfer. But do we have a spare GTX for such a connector anyway? 
Won't we need to keep using the backplane connector (with an SFP 
adapter) for the transceivers when the Sayma board is stand-alone?


Sébastien
___
ARTIQ mailing list
https://ssl.serverraum.org/lists/listinfo/artiq

Re: [ARTIQ] uTCA backplane driver choices

[Grzegorz Kasprowicz]

we can use Si instead of CDCM61004RHBT
Greg

On 29 June 2016 at 04:56, Sébastien Bourdeauducq  wrote:

> On Tuesday, June 28, 2016 04:02 PM, Grzegorz Kasprowicz wrote:
>
>> For synchronisation over fibre we can use existing White Rabbit core.
>> The card requires only 2 VCXO oscillators and FPGA logic. The WR core
>> consumes 50% of small Spartan 45T. It ensures 1ns timing accuracy.
>>
>
> We probably won't use White Rabbit as-is, but the basic principle will be
> the same. It can be a good idea to include those VCXOs on boards that may
> be synchronized over fiber (Kasli/Metlino), though can't we use instead a
> Si5324 for jitter cleanup and FPGA PLLs for DDMTD?
>
> Sébastien
>
___
ARTIQ mailing list
https://ssl.serverraum.org/lists/listinfo/artiq

Re: [ARTIQ] uTCA backplane driver choices

[Sébastien Bourdeauducq]

On Tuesday, June 28, 2016 04:02 PM, Grzegorz Kasprowicz wrote:

For synchronisation over fibre we can use existing White Rabbit core.
The card requires only 2 VCXO oscillators and FPGA logic. The WR core
consumes 50% of small Spartan 45T. It ensures 1ns timing accuracy.


We probably won't use White Rabbit as-is, but the basic principle will 
be the same. It can be a good idea to include those VCXOs on boards that 
may be synchronized over fiber (Kasli/Metlino), though can't we use 
instead a Si5324 for jitter cleanup and FPGA PLLs for DDMTD?


Sébastien
___
ARTIQ mailing list
https://ssl.serverraum.org/lists/listinfo/artiq