Given the pinout details for the UART, can I assume 3.3 V as voltage?

The pinout shows pins 4…1 for UART as: GND, UART0_TxD_D, UART0_RxD_D, +3V3

May I assume based on the +3V3 that it is appropriate to connect with that voltage level? I am asking, because this may be a convention, but I’m unaware of it. I only want to connect 4…2 anyway (i.e. “passive”).

Thanks!

The description is quite clear, I do not understand your doubts. Adapter on the UART side must be switched to 3.3V. On the other hand, there will probably be a USB with 5 V.

Example

2 Likes

“The description”? I don’t see any on the pinout. And my question was merely regarding the voltage level of UART.

So great, that it’s 3.3 V as I assumed. Means everything will work without level shifting.

I had no doubts, but I also didn’t want to fry the hardware I was about to connect.

omnia-pinout

1 Like

Alright. I’ve seen that. I’ve cited it.

What I was asking was whether I can assume based on that label +3V3 for pin 1 that I can connect to RxD (pin 2) and TxD (pin 3) with 3.3 V (or whether there is a convention that basically implies this). And I explicitly stated that I don’t want to connect to pin 1 at all (because that’s not needed for my use case). So what I was “doubting”, as you put it, was not whether the +3V3 label for pin 1 was correct (I can measure that), but whether that holds true for the other two pins as well. After all digital signals won’t show up as a discrete voltage level on a voltmeter and I am not equipped with an oscilloscope. So better safe than sorry …

This may be perfectly logical for you if you are aware of such a convention, to me it wasn’t and in fact I still don’t know if this is a convention (which I could apply in the future) or whether it was an assumption that you tried without frying your hardware, for example.

Thanks a bunch for the answer in any case :+1:

You could also look into a datasheet of the TTL-232R-RPI data cable
that is mentioned in the official documentation: https:// docs.turris.cz/hw/serial/

relevant part:

That should be enough to provide you with safe voltage values I think.

But if you want the maximum allowed voltages, your best bet is probably to find the datasheet for the relevant controller that is used in omnia, which the official omnia HW schematics should help you with

2 Likes

Thank you for the additional information.

The BTE13 converter mentioned here doesn’t use 3.3V TTL logic. It has a switch 5V/3.3V, but that only affects the VCC pin on the converter. TTL still uses 5V. Could somebody from the team make this clear? The official docs clearly say that MOX uses 1.8V, so why can’t we have a similar “sure” statement about Omnia? I know there’s one particular cable mentioned, but people might want to use their own solutions…

1 Like

omnia uses 3.3V TTL logic for UART

I double-checked this with the official HW schematics (schema for connector CN10) and also by measuring voltages* between GND and data pins (Tx & Rx).

Also, quick search for the Issue with BTE13-009 that you are mentioning suggests that this might not be that uncommon and might be something you should be aware of when buying similar cheap UART converters!

And since I can’t promise you that omnia will survive supplying 5V to the data lines, you should verify your adapter before use, possibly by the measurement of voltages on its data pins.


*This measurement is possible, because when fully booted, omnia doesn’t send any data over those lines until you press ENTER on the console and because for UART, data lines are kept “high” when connection is idle = max. voltage when no data is being transmitted.

1 Like

Thanks for verifying, I didn’t know UART keeps idle lines high, and I was wondering how to persuade it to output ones :wink:

Anyways, I remember I once attached the mentioned converter (switched to 3.3V, but as we know now, it was still using 5V logic) to my MOX, which works with 1.8V logic. It survived, and surprisingly, I was even able to receive data from it. I did not, however, send anything to it. But if the idle lines are kept high, it should probably fry it… Anyways, the MOX is alive and works well :slight_smile: But I wouldn’t suggest anyone trying this.

Knowing the problems with the UART converters, I now ordered the quality ones from FTDI. They cost over 20$ per cable, but it seems it’s worth it. And I’m just wondering why nobody is making universal cables with a switch for 1.8/3.3/5V logic… FTDI makes the universal cable TTL-232RG-VIP-WE which uses a logic level equal to a voltage you provide it on a pin. Maybe that one would be best as a universal cable, but having the power coming from the USB port is so convenient…

1 Like