TE0725 Information

[uncle_bob]

Hi

Looking at the schematic for the TE0725 module and the pictures of the same module, there are some differences.

The schematic shows the 2.5 V regulator present and the jumper resistor missing.

The picture shows the 2.5V regulator missing and the jumper resistor present.

That makes some sense. The fiber interface that the LVDS lines feed is missing both on the schematic and on the pictures. The reason for my question:

I have an application for a lot of LVDS i/o. The 725 board would do great. It would be even better if the 2.5V regulator was present on the board ( == it's like the schematic).

I'm guessing the pictures are correct and the schematic is wrong.

Bob

[Antti Lukats]

The pics are taken most likely from early assembly version.

The assembly option was changed from not-mounted to LDO mounted to easy post-assembly mounting of the fiber module.

But the on-board 2.5V LDO was only designed in for the "option" that the fiber optical module is installed, there 2.5V output has no jumper or easy accessible-solderable test-point on the PCB so it is not as easy to "patch" the PCB to use the 2.5V

We will consider adding 0R jumper to power one or both banks from the 2.5V LDO in the next PCB revision.
But with existing PCB you need to supply 2.5V VCCIO either externally or then use thin wire-patch

[uncle_bob]

Hi

Ok, so the schematics are correct and the photo's are wrong. I guessed wrong

As a designer of this sort of thing as well as a user, I *do* understand that each added feature comes at a cost of board space and design risk. The TE0725 already has a bunch of features on it. I would certainly like the ability to use the 2.5 for LVDS inputs. I can't say if it is something other people would be interested in or not. I think I would take a look at the power dissipation of the regulator vs the bank i/o current before I went to far. Once the jumper is in there, some guy named Bob might put all the banks on 2.5V.

Now, if we just could do something about the size of the Atlantic ocean so shipping would not cost so much

Thanks!

Bob

[Antti Lukats]

if you get TE0725 you can use it instantly as 84 logic channel logic analyzer

https://wiki.trenz-electronic.de/display/PD/TE0725+Reference+Designs

well, on the fiber channel there is one channel 1GS/s capture unit

[uncle_bob]

Hi

That looks very nice. It also gives a base project to work from. Having a "working demo" that brings in all the pins usually lets me get a custom project up quicker.

Thanks for sharing !!

Bob

[dmitry_korablev]

Hello,

What is the maximum frequency can be transmitted via 2.54mm pitch headers?

[Antti Lukats]

it depends the standard and interface type and the other side that connects to the board.

one of the B2B connectors is routed differential 100ohm, if you have properly matched baseboard, I would say you can work up to 1GBit/s.

[uncle_bob]

Hi

If you are using a header soldered on both sides you will do better than using a header that plugs into a connector. A header in the middle of a 1 meter long trace pair will be more of a problem than a header in a 2 CM long trace pair. Grounding *does* matter even with differential signals, so it will eventually get into the act as well. Pile on top of that the fact that drivers and receivers are not ideal parts. You can combine driver X with receiver Y to get trouble at very low rates ...

So lots of variables, lots of corner cases where any one answer isn't going to be correct.  That said, with care and maybe a bit of tweaking, 1 Gb/s is not unreasonable. I seem to have a pile of 725's sitting here now, my application is down at 250 Mb/s. I do not expect any trouble at all there.

Bob

[Antti Lukats]

correct! running 250MBit over pin header is not a problem if all is done properly on both sides from the connector.

if you look the lab-X example, it spits out 125MHz into the POF and display the received data at 1GS/s, you can change the pin locations, and test your baseboard (to some extent) immediately.

[dmitry_korablev]

Simple question,
power supply (3.3V) is supplied via pins 5, 46 connectors J1, J2 ?

How to use the points PM1, PM2? Which of the appointment?

[Antti Lukats]

Simple question,
power supply (3.3V) is supplied via pins 5, 46 connectors J1, J2 ?

How to use the points PM1, PM2? Which of the appointment?

5,46 yes

PM1 and PM2 are designator for FIDUCIALS are used during manufacturing only

[uncle_bob]

Hi

Ok so here's a question:

Obviously the preferred approach is to put 3.3V on both pins. IF power was to be supplied to only one pin (the other pin floating): Is there a preferred pin ?

Thanks!

Bob

[Antti Lukats]

no all power pins are connected to same power plane, it is recommended to connect all power pins, if you choose 1, then any one is OK, with small preferences toward the pins that are closer to the power supply components on the board, you can identify the DCDC easily visually.