CMod A7 Datasheet Warning

[Eric Raymond]

Looking at the CMod A7 to design into a CCA of my simulator. there is a warning in the datasheet that worries me

"Warning: When a USB host is attached to the Micro-USB connector, the VU pin on the DIP connector (pin 24) is driven to the voltage being provided by the USB host (typically between 4.5V-5.5V). If you have a power source attached to the VU pin, you must disconnect it before attaching a USB host, or risk damaging it. This can be particularly dangerous if the power source is a battery"

Does that indicate that I shouldn't plug into a powered CMod, or that I shouldn't power a Plugged in CMod.

I would like to only power the A7 with my power source, can I simply remove D1 to do this?

Second I want to use the FTDI chip, but it is nowhere in the schematic???

Last (for this morning) I cannot find any pin dimensions to design a model to plug it onto my board? I

Any help is appreciated

Thanks

Eric

[JColvin]

Hi @Eric Raymond,

With regards to the first question, I would say it's best to go both ways (not plug into a powered Cmod and not power a plugged in Cmod), since you are relying on a single diode to prevent any back-powering problems that could arise (as opposed to a jumper setting to choose your power source).

You may be able to just remove the D1 and have it work out, though I don't know all of the ramifications with regards to powering the FTDI chip. You would also need to make sure any grounding between the input to the USB and the external power are connected to prevent voltage differences and issues there. However, if you are just wanting to power and have the Cmod A7 run a dedicated project, you can always load the flash on the Cmod A7 and every time the Cmod A7 is powered on it will automatically load the flash project. There are a couple of tutorials on this for the Cmod A7 Resource Center.

The schematic page for the FTDI chip (the "This page intentionally left blank" one) is proprietary. You can send an email to support dot digilent at ni dot com to request these details though.

We have a 3D model (.stp) for the Cmod A7 available on it's Resource Center on the right-hand side labeled as 3D STEP with a file name of cmod_a7.zip. There are some more dimensions about the Cmod A7 available on this thread as well.

Let me know if you have any questions.

Thanks,
JColvin

[YaUser]

Hi,

did anybody try removing D1 to check if it lets the user to use USB and external power simultaneously?

My use case requires the USB to be connected at all times for control of the board features, but I also need external power for +-12V rails (and lots of 5V and 3V3 rails). I could easily provide external power for the Cmod that would be cleaner, of the same ground (smaller groundloops), and of higher current capacity than any USB.

I need to know if removing D1 turns off FTDI chip, and since "this page is left blank" I cannot say on what supply rail the FTDI lives. Based on my experience with this chip family,  it still needs the 5V, and since the external power can be down do 3.5V for highest current drawn from Pmod, the FTDI is still on the USB power rail.

I respect the proprietary 'do not cross' line, and i dont need the full schematic of the FTDI page. Just need to know if it is possible to modify the Cmod7 for simultaneous power. I have tools and skill to do it (i guess ) as long as it doesn't require BGA rework. If possible I would only require a list of parts to be removed/changed, without disclosing the FTDI schematic page.

Can anybody help me?

Best regards,

YaUser

[PTSmith]

I'd like to know the answer to this question as well; does removing D1 work as expected?  Does the USB communication still work if D1 is removed?

How about a external diode in series with pin 24?  Shouldn't this protect an external power source?

[YaUser]

I haven't got the chance to do the modification yet, but my main concern is the module itself, as i can fix the external power source at almost no cost, and the modules are in very limited supply

[PTSmith]

I don't understand why it is so hard for Digilent to answer this question; really this information should be on the data sheet.

[JColvin]

Hi @PTSmith and @YaUser,

It obviously took way longer than necessary to get back to you, but shorting D1 will not work as that will invite potential backpowering into your PC's USB port. 

I asked one of our design engineers and they let me know you could use a schottky diode (we used a 20V one for D1) though in series with pin 24 (VU) to prevent backpowering to the external source. We haven't specifically tried it, but this should provide some current sharing when both sources are present and backup power otherwise.

You can remove the D1 (and leave that trace open and unconnected) so that the USB controller will be powered and VBUS won't get back powered. Naturally, the USB circuit will suspend when the usb cable isn't connected to the PC but if you remove D1 then the board will no longer be powered from USB and won't power on when VU isn't present.

Let me know if you have any questions.

Thanks,
JColvin

[PTSmith]

Yes, I think a schottky diode in series with pin 24 is the right way to go.

Shorting D1 vs Removing D1 are completely different. Shorting D1 will "back power" the USB port as you say. But if D1 is removed then no current can flow either from or to the USB connector. The question is whether the USB communication will still work.

Please ask your engineer whether removing D1 will cause the FTDI chip to lose power.

My guess is that the FTDI chip gets powered from the 3.3 volt regulator and the CMOD FTDI USB will still work from user-supplied VU if D1 is removed.

[JColvin]

Hi @PTSmith,

You are correct; apparently that bit didn't make it into my post. You can remove the D1 (and leave that trace open and unconnected) so that the USB controller will be powered and VBUS won't get back powered. Naturally, the USB circuit will suspend when the usb cable isn't connected to the PC but if you remove D1 then the board will no longer be powered from USB and won't power on when VU isn't present.

Let me know if you have any questions about this.

Thanks,
JColvin

[PTSmith]

Thank you, JColvin!  Might I suggest adding this very useful information to section 1.1 of the CMOD A7 data sheet.

Paul Smith

Indiana University Physics