SPI Chip Select

Hi,

Unfortunately, I don't see any workaround.

Regards,

Hyun

Hi Hyun,

I am sure that there is a work around, because I bought the ECG board for the TMS320C5515EVM. There is an application note which describes using the ECG board with the TMS320C5515EVM, but there is no details.

Can you please let me know how I can send an official request to TI?

Regards,

Simon

Connecting multiple devices to a single SPI port is a very common situation. Here's how it might be done in a "from scratch" design: Assuming device 1 and device 2 each have active-low chip selects, each can be driven by the logical OR of the CS0 and a general purpose i/o line. In this way, the i/o lines do the device selection, and CS0 provides the correct timing.

In order to implement this technique on an existing evaluation board, you'll need to cut a few traces.

Hi David,

Thanks for your reply. I understand what you said, but I need to use the SPI on an existing evaluation board.

What I cannot understand is, how TI markets the TMS320C5515EVM and the ADS1298ECG Front End as a medical development kit, but then, they do not seem to be compatible.

I am almost sure that there is a clean way of working around this problem. In this application note, TI describes the general setup required to do what I am trying to do. However, there are no details about the SPI communication.

http://www.ti.com/lit/an/sprab36b/sprab36b.pdf

I wish someone from TI could clarify this issue.

Regards,

Simon

Simon,

The documentation for the Front End board references a 5505 MDK rather than the 5515 EVM. It wouldn't be surprising if these two boards had the same expansion headers, but somewhat different features. But I gather you want to use the 5515EVM and the Front End board, so I looked at the published schematics. On the 5515EVM, J13 is a 22-pin header, while on the Front End board,  J11 has 20 positions. Assuming that you can get these to physically mate, it looks like the SPI lines are in compatible places. To resolve the CS conflict, you might consider simply desoldering the SPI EEPROM from the 5515 EVM. (It would still have a I2C EEPROM.)

David

David,

Thanks for your help and effort! I really appreciate it.

It is true that the documentation mentions a 5505MDK, however it is clear in the title (and also other parts of the document) that they used the TMS320C5515EVM.

For this reason, I would love to have a response from a TI personnel, to explain to me how they did interface the ECG FE to the EVM board.

Having said that, if I do not find any solution to this problem, I would have to find a crude way how to go about it, such as desoldering the SPI EEPROM (although I would hate to do that to a $400+ device!).

Thanks again for your help David.

Simon

Hi all,

I have an idea how I can go around this problem. The ECG FE board has the possibility to connect the SPI chip select to the DSP's GPIO pin 4. (This is done by using jumper 25 on the FE).

I was thinking if it is possible to use the GPIO port instead of the normal SPI chip select pin to select the ECG FE for SPI communication.

Do you think this is possible?

Simon

Simon Attard said:

Hi all,

I have an idea how I can go around this problem. The ECG FE board has the possibility to connect the SPI chip select to the DSP's GPIO pin 4. (This is done by using jumper 25 on the FE).

I was thinking if it is possible to use the GPIO port instead of the normal SPI chip select pin to select the ECG FE for SPI communication.

Do you think this is possible?

Simon

Hi Simon,

Did you manage to work with this GPIO configuration, I am also stuck with this EVM5515 + ADS1298 FE issue. I have tried everything to get some meaningful data. I can take data from two leads only, below are my readings. Please let me know if you get any progress.