Low voltage conducted emissions setup(10-40V)

[eriksen]

Hi,

We have an automotive product using a switching power supply(10V-32V) as the power source. We passed the EMC conducted emissions tests fine.
However we had to add some capacitance to the input to avoid a scenario where the PS would burn if it got a voltage spike 40V+, and we are unsure how this might affect our EMC performance.

We would rather not spend lots of money on new tests from a test house and want to buy a DSA815-TG with EMI addon so we can do our own tests.
The question here is; do we need a LISN as well? Or can we measure passively on the power lines directly with the analyzer?

Any suggestions for either a cheap LISN or a good-enough diy way of measuring if it's not possible to measure directly?

We have also gotten some indications that we have conducted noise in the 300MHz range, DAB+ interference which was solved with moving the antenna away.
So we want to measure from KHz range to at least 300Mhz.

Any input here would be greatly appreciated.

Thanks.

EDIT: Wrong forum category, don't know how I managed to do it... Can I change this myself?

 

[Warpspeed]

Its a long time since I did any of this, but I cannot see how adding shunt capacitance could degrade conducted EMI, unless other major changes to physical layout are also involved.

You will need a LISN through which to power the device, and make meaningful conducted noise measurements. Its not a terribly difficult thing to make yourself, and the standard defines exactly what is required.

 

[D.A.(Tony)Stewart]

I would think load dump suppression devices should be satisfied with a bipolar TVS and conducted noise improved with low ESR C on input or series choke.

A LISN (50Ω/5µH) is used for AC power line conducted emissions measuring voltage for AC line external connections.

Battery conducted emissions , can be done with a current probe or 50mV current shunt.

 

[eriksen]

Its a long time since I did any of this, but I cannot see how adding shunt capacitance could degrade conducted EMI, unless other major changes to physical layout are also involved.

You will need a LISN through which to power the device, and make meaningful conducted noise measurements. Its not a terribly difficult thing to make yourself, and the standard defines exactly what is required.

Thanks. I agree that a bit of capacitance shouldn't matter much. There are some small changes to some traces as well though. So we want to be sure before moving forward.

I would think load dump suppression devices should be satisfied with a bipolar TVS and conducted noise improved with low ESR C on input or series choke.

A LISN (50Ω/5µH) is used for AC power line conducted emissions measuring voltage for AC line external connections.

Battery conducted emissions , can be done with a current probe or 50mV current shunt.

Thank you for the suggestions.
The tests are run from either a 12V car battery og two 12V car batteries in series.
Would the measurements be accurate enough on both low and high frequencies with the same current probe?
After what I've been reading it seems a current probe would need very much calibration as I step up in frequency.
I guess I could use the tracking generator on the scope to calibrate each range.. Any thoughts?

Also the test house we tested at initially used a modified LISN for the battery measurements, it was accurate for a very large frequency range.

Thanks.