DW1000 ISR pin stays deasserted, no more interrupts

Hello Forum,

we have one problem that after some situation (which we have not figured out exactly)
the ISR pin does not activate anymore.

I will just kindly ask if someone observed similar behaviour (interrupt pin hangs in deasserted state) and what to do in that case.

We are using proprietary code, not using the decawave driver.
And the situation can not reproduced but happends spuriously.

Regards, Adib.

Hello Adib. I had a similar problem commented in thread https://www.decawave.com/decaforum/showthread.php?tid=311

Everything in software was OK (IRQs bit set), but the line remained low. Finally it was a hardware problem. The soldering was not correctly done and there was no continuity. If your problem occurs at random, there is a chance that a faulty soldering sometimes break the circuit, sometimes do not.

Hello Diego,

thanks for your input.
We investigated any further. And it turns out that we get strange SPI data back.
When reading the WHOAMI register we ecpect: 30h.01h.CAh.DEh
but we get this: 98h.00h.E5h.6Fh (and sometimes 60h.03.95h.BCh)
So the firmware driver does not initialize the dw1000. And hence no interrupts.

Basically the bitpattern is the same but latched at a bit too early or too late.
Clockfrequency is fine, 2MHz so below the magical 3MHz.
A poweron reset resolves the situation.

I could reproduce the more seldom case by playing with GPIO5 and GPIO6 on the EVK1000.
This will change the default SPI mode on the DW1000.
But I could never reproduce the first case.

We have GPIO5 and GPIO6 open and do not use those pins as outputs.
Our design is a “always on” wearable so there is no chance for power on reset.

Any ideas or hints?

Regards, Adib.