Unconventional use of L9826 (erroneously before L9925)

[Fred4]

In my project I use a L9826, 6 Outputs are used conventionally 2 outputs are not used.

In the same project I need to have 2 digital inputs (SPI) and I would like to know if I could use the two unused outputs of the L9826 as HCMOS inputs via the Fault status register; would this type of unconventional use be possible? Thanks

[Peter BENSCH]

Welcome @Fred4, to the community!

You referred to L9825 in the title and L9826 in the text, so please correct one of them. Essentially, however, both devices have the same structure.

If you switch the unused outputs to non-active so that the driver MOSFET is disabled, you should indeed be able to use the relevant OUT as an input. Please remember to read the status register twice to be on the safe side, as the first value may still be incorrect. Please also remember that the threshold is 0.38*VDD, i.e. 1.9V for 5V, which differs from the usual value for CMOS inputs.

I'm curious to see what your test reveals.

Regards
/Peter

[Fred4]

Hi Peter,

thank you for your kindly reply.

Effectively I did a mistake while I'm using L9826.

I agree with your remark because I noticed in the datasheet that it is necessary to read the status register twice to have a reliable value, however the reading is circular and the values ​​are constantly updated. I will give feedback as soon as the tests are concluded.

Thanks, regards
Fred4

[Fred4]

Hi Peter, it seems that using an unused output as an input by reading the status twice is possible, it works as I would like.

I have another question but I don't know if I should open another post:

I have to interface the L9826 with a processor powered by 3.3V (logic levels 3.3V). In the data sheets of the L9826 it is not indicated in which VCC range the IC can work normally, if I power it at VCC=3.3V it does not seem to work, if I power it at VCC=5V the logic levels do not match because VINHI=0.7xVCC.

Is there a simple solution (apart from logic level translators) to solve the problem?

Thanks, regards,
Fred4

[Peter BENSCH]

Normally this would be a new thread, but we can do it here as well.

The L9826 has an undervoltage reset that becomes active at a VCC lower than 3...4V, which is why it can only work at voltages above 4V, if you take into account the tolerance of 10% of VCC, this is at least 4.45V.

Even at this voltage, you will have difficulty guaranteeing 70% VCC with CMOS outputs (3.115V at 4.45V), so you can either work with

• open-drain outputs that are slow due to the pull-ups (if your processor can do this),
• or you will indeed have to use level translators from 3.3V to 5V.

Regards
/Peter

[Fred4]

Hi Peter, Thank you very much, unfortunately it confirms what I already thought, I will see how to solve the problem.

Regards,
Fred4