- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Greetings,
The EN6340QI and other buck regulators in the family exhibit strong EMI in the 180-220MHz (see attached) which violate RE102 requirements, this is also contrary to datasheet EMI performance chart which show low emission at that range of frequencies. Does anyone suffered similar experience?
The layout rules by the way were followed one for one.
Any insight will be appreciated!
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Hi Shalmon,
Thanks a lot for your feedback.
The datasheet measurement done under standard EMC test setup where inside the EMC lab, the test engineer shield all the used equipment during the measurement where most of the noise comes from that equipment.
Also they using a ferrite core on the input cables to reduce the common noise coming over the long wires between the input supply and the DUT.
So to make sure that the noise level that you see in your measurement comes only from the regulator, then you need do your normal setup, but make sure the regulator enable pin is low and then do your EMI measurement to check if there is any common noise in that setup that needs to takes in consideration.
If you can see high noise in your system even the regulator is off then:
- Try to use different input power supply that could be less noisy
- Use short wires
- Put a ferrite core over the input cables
Attached an example of the setup Enpirion uses in EMC lab.
Thanks,
Mostafa
Link Copied
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Hello Shalmon,
EN6340 has low EMI noise as shown in the datasheet page 10.
So probably the high noise could be related to the switching noise, but for the switching noise to be amplified on the PCB level, that's mean there is an issue with the layout and the decoupling caps.
i will send you personal email, so you can help to share EN6340 schematic and its layout for review.
Could you please help to share the output voltage ripple measurement with adjusting your scope bandwidth to 20MHz and 500MHz.
For the spectrum analysis; could you please share a wider range from 30MHz to 1G (log scale) for belter understanding if that noise related to the regulator switching noise.
Thanks,
Mostafa
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Hello Mostafa, thanks for replying.
I considered the possibility of inappropriate layout but it seems like the layout guy did a really good job following the datasheet layout instructions one for one. So we went through the hassle and purchased the EN6340 DC-DC regulator devkit, the EMI emission from that component is the same as on our design (see screenshots below).
Since the emission is so strong it is being coupled onto the main power line and conducted outside the box.
Any other insights?
Thanks,
Shalmon
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Hello Shalmon,
Thanks a lot for sharing the measurement using the Devkit.
Based on the measurement the fundamental frequency is 2MHz which is related to the switching noise of the regulator.
The basic capacitance input filter will not help alot to avoid that noise.
I would recommend to use pi filter and use the high switching frequency capacitor very near to the regulator input side.
But I don't know if your current schematic allows to use pi filter or not, so if you can't implement it with your current circuit, try to increase the input capacitance and also keep the high frequency caps near to the regulator side.
thanks,
Mostafa
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Thanks Mostafa.
Since the noise seems to be radiated rather than conducted the question would how would mitigate that?!
In any case, how come the datasheet portray's such good EMI performance where in reality it is not the case?
Best,
Shalmon
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Hi Shalmon,
Thanks a lot for your feedback.
The datasheet measurement done under standard EMC test setup where inside the EMC lab, the test engineer shield all the used equipment during the measurement where most of the noise comes from that equipment.
Also they using a ferrite core on the input cables to reduce the common noise coming over the long wires between the input supply and the DUT.
So to make sure that the noise level that you see in your measurement comes only from the regulator, then you need do your normal setup, but make sure the regulator enable pin is low and then do your EMI measurement to check if there is any common noise in that setup that needs to takes in consideration.
If you can see high noise in your system even the regulator is off then:
- Try to use different input power supply that could be less noisy
- Use short wires
- Put a ferrite core over the input cables
Attached an example of the setup Enpirion uses in EMC lab.
Thanks,
Mostafa
- Subscribe to RSS Feed
- Mark Topic as New
- Mark Topic as Read
- Float this Topic for Current User
- Bookmark
- Subscribe
- Printer Friendly Page