Please stop using zero ohm resistors on Click boards

[MaGiK]

Hello there mE

First, I wish you guys a happy new year, and I hope that you guys enjoyed your off days, they are very well deserved

You've been improving a lot, and specifically in the Click boards department lately

I really enjoy using them, but not when they contain a zero ohm resistor that works as a switch.
It could possibly be used to determine which voltage level to be used (3.3 volts or 5 volts), or some other type of switching.

Kindly, could you replace those with male headers pointing upwards, and a female to female connector?
That would make life so much easier by making click boards work right out of the box, instead of having to perform a desolder and resolder action.

Best wishes for 2018. You guys are the best

[daflory]

I strongly agree with these suggestions. I regularly work with both 5V and 3.3V circuits and being able to select voltage with a jumper would be much more convenient.

[philippeb]

Totally agree.

[marko.stankovic]

Hi,

Thank you again for the suggestion,
I understand that it would be very useful.
I have forwarded to our developers to take in consideration.

Best Regards,

[oliverb]

FWIW if there's ever a "Click Mk.2" or an extended version it would be a good idea to have a CPU voltage pin, so boards equipped with level shifters could automatically shift to the right voltage without needing jumpering.

A lot of the time with I2C devices the issue could be eliminated simply by providing I2C pull-up resistors on the host board not the Click, then provided the Click's I2C port is 5v-tolerant no selection is required.

For example the MLX90640 is said to be compatible with 5V I2C even though it runs from a 3.3V supply.

[stefan.filipovic]

Hi Oliver,

Thank you for your feedback.

I will forward your suggestion to our developers.

Kind regards,

[oliverb]

Another suggestion would be for Click boards that do not need 5V at all to have that pin cut off or completely missing.

That way if a future generation of host boards are produced that are exclusively 3.3V and can't supply 5V at all then the 5V socket could be blanked off so incompatible click boards would not fit.

I realise that Mikroelektronika host boards probably can all supply 5v at present as even the 3.3V ones have a 5V rail but that might not always be the case.

[stefan.filipovic]

Hi,

Thank you for the suggestion.

I will forward it to our developers for consideration.

Kind regards,

[Electronza]

Happy New Year!

Ditching the 0-ohm SMD jumpers would be a significant improvement. Not only for 3.3V or 5V voltage selection but also for configuring other options, such as I2C address or switching from I2C to SPI.

I'm against 3.3V only click boards. Arduino Uno boards have 5V logic levels, and there are so many projects that can be done with Arduino.

What I would do is replace the 0-ohm SMD jumpers with solder pads. In the image below, you can see an excellent example of a 3.3V / 5V selector implemented with solder pads. Notice the PCB trace that connects the 3.3V power pad to the middle Vcc pad, effectively working as a factory setting. If one wishes to configure the board for 5V operation, he has to cut the PCB trace (a Xacto knife will do) and then place a solder bridge connecting the middle Vcc pad to the 5V pad.

This approach is better, as it doesn't require any external parts. It's just a minor PCB redesign. And it's easier to configure boards this way rather than moving SMD jumpers.

voltage-selector-default.jpg (1.35 MiB) Viewed 3950 times

[oliverb]

I'm still trying to think of a simple adaptation that would make it "plug and play". In the absence of a dedicated "Vl" pin could it detect a logic pin going above 3.3v and maybe use a CMOS switch to change from 3.3V to 5V in response?

Having said that I wonder if that's even needed. If inputs tolerate 5V but outputs output 3.3V that ought to be compatible with a lot of 5V devices anyway, though I'd have to check fine print of PIC specs.

For I2C addresses if there's ever a "Click 2.0" then it might be nice to have two address pins on the socket, each hardwired to the position, so the first would be 00, the second 01 etc. That way if two or more conflicting clicks are installed they'll be assigned nonconflicting addresses. I'd expect to still have a link or cuttable track for anyone wanting to overrule that.

FWIW will we be arguing about support for 1.8V-only devices in a few years time?