fanpico showing massively wrong fan speeds

[gadg3ts]

Hi,
I have a fanpico 401D set with 1.7.1 firmware.
For the first few months everything was fine.
Now, the device is showing fan speeds of 117,000 or 235,000 instead of 1800.
If it makes any difference, it is normally connected to the PSU, but also a raspberry pi for console access.
What could I do to get the fan speeds displaying properly again, please?
In comparison, the temperatures are showing fine.

[tjko]

@gadg3ts, are you powering the board using the AUX connector or the MBFAN1 connector? Are all the power status LEDs lit (5V, 3.3V, AUX, D12)? Does resetting/rebooting the board make any difference? Does the RPM reading go to 0, if you disconnect the fan from fan output?

Are you able to send me the current configuration (you can use "CONF?" command to dump the configuration in JSON format)
Note, please remove potentially sensitive fields from the config (like wifi_ssid, wifi_passwd, mqtt_user, mqtt_pass, telnet_user, telnet_pwhash, ...) before sending it. (you can email the config to me, instead of posting here if you prefer...)

[gadg3ts]

I've tried reflowing all of the pins on the pico, along with the header sockets where that attaches on the main board. The RPM is still showing 235000.
There is a slight high-pitched whine though?
would pictures help?
I'm in the UK. I ordered the kit via ebay, then had it shipped via a mail-forwarding service from the US.
If we could somehow organise a replacement, that would be great.

[tjko]

Pictures might help. What kind of fan(s) you are using, have you tried with other fans and/or does the issue happen on all fan ports?
Whine sounds odd, as there is no inductors or multi-layer capacitors on the PCB, could the noise be coming from the Pico?
If you have an oscilloscope it would be helpful to see what the tachometer signal from fans look like at the Pico GPIO pins (GPIO18 is Fan1, GPIO21 is Fan4)

[tjko]

Tachometer signal should look something like this. This shows the signal on GPIO18, when there is Noctua fan running about 840rpm connected to Fan1 connector:

(signal should be between 0 and 3.3V at the GPIO pin)

While same signal measured at the fan connector can look bit different (if fan has internal pull-up resitor then it can show 12V pulses, while fans withouth one would show about 3V pulses):

[gadg3ts]

So...
it turns out it was indeed the fans.
Scythe SU1225FD12H-RP (3-wire)
I was using two of them, because they were the ones I had at the time with the highest airflow. They were fine for the first few months.
With trying a third one I'd not connected to the fanpico previously, now also giving the same readings. Which seemed a bit weird.

I've now rummaged in my box of spare fans, finding that all the working spare ones I've tried do in fact report the correct RPM.
The ones I'm settling on are a matching pair of Thermaltake TT-1225 as they run at 1800 (and were actually free with a computer case!) which are 4-wire fans.
Thank you all for the assistance and suggestions.

Incidentally, I've got a 3D printed case for the fanpico + screen that I designed. Is there somewhere I can submit that?

[tjko]

Surprising, I wasn't expecting fan to be the problem.

If you have created 3D printed case, you can files (or links to thingiverse, etc) here in the Discussion forum. I can then add it to the Wiki pages as well.

https://github.com/tjko/fanpico/wiki/FanPico-Enclosures-(3D%E2%80%90Printed)

[dmgeurts-mm]

I have some spare boards, I ordered a batch of five a few years ago. But they're not the latest version. I'm based in the UK so had my own boards made and some populated. Sorry to say I can't do it for free, but it'll probably be cheaper than international shipping.