| Electronics > Projects, Designs, and Technical Stuff |
| Choosing an inductor for a boost converter |
| (1/4) > >> |
| aiq25:
I’m designing a 5V/2.4A output boost converter, to work with a li-ion battery. In choosing an inductor, I can go with a approx 7x7 mm size inductor, with a DCR of 15-20 mOhm, versus 10x10 mm size inductor with a DCR of < 10 mOhm of DCR. I’m not sure what the trade off is in physical size of an inductor and lowering the DCR versus using a smaller inductor. The smaller inductor has more DCR but I’m worried a larger inductor might have more magnetic/AC losses. I would like to improve DCR but not if it means increasing AC losses. Can someone help explain if there is a con to using a larger size inductor in order to help lower DCR? |
| MagicSmoker:
Do both inductors have the same inductance and saturation current rating? If not (and likely they don't) then this isn't an apples-to-apples comparison. Note that off-the-shelf inductors need to be generously derated with respect to claimed saturation current if you don't want to accidentally blow up switches and scorch the board from overheating. I'm talking about maybe allowing a peak current of 5-6A in a part claiming a 10A saturation current. |
| Siwastaja:
No, physical size is not a good hint for AC losses. You need actual numbers for AC losses - some manufacturers give them, most won't. Otherwise, your only way is to test and measure. If the "typical applications" on the datasheet lists DC/DC conversion, and it's ferrite based, it's likely the AC losses are not catastrophically bad - but you still need to measure and verify. Say Inductor #1 may have 1W of DC loss and 0.5W of AC loss, while inductor #2 may have 0.5W of DC loss and 10W of AC loss, and inductor #3 may have 0.25W of DC loss and 0.25W of AC loss. A classical initial assumption that you double the total losses from the DC loss number (so that it ends up being 50%/50%) is something, but not sufficient. I almost exclusively use Wurth electronics inductors because they give real AC loss numbers through their web-based selector. Some others have similar tools as well. |
| splin:
Coilcraft have a really good tool that estimates the DC and AC coil and core losses: https://www.coilcraft.com/apps/power_tools/dc-dc Put your circuit parameters in and it will list suitable inductors which you order by price, losses, size etc. [Edit] Click on the 'Total losses' entry in the results list to see a breakdown of DC and AC losses. . |
| aiq25:
--- Quote from: MagicSmoker on September 15, 2019, 09:37:33 am ---Do both inductors have the same inductance and saturation current rating? If not (and likely they don't) then this isn't an apples-to-apples comparison. Note that off-the-shelf inductors need to be generously derated with respect to claimed saturation current if you don't want to accidentally blow up switches and scorch the board from overheating. I'm talking about maybe allowing a peak current of 5-6A in a part claiming a 10A saturation current. --- End quote --- No, one inductor rated current was ~6A and the other 10A. You are correct it's not a fair comparison but I was trying to get an idea. Thank you for pointing out the peak current handling capability, I didn't notice how much I would have to derate them by. In my design I noticed for 15% peak-to-peak current + 10% margin, I also added 30% margin to the inductance to account for DC current bias, frequency, initial tolerance, temperature, etc... Know I'm thinking I need to add maybe 40-50% margin after looking at some datasheets. Thanks for the suggestions Siwastaja and splin. I was considering Coilcraft and Wurth initially because I have used them for other projects and really like their products but for this project the assembly process is a big factor for me. Seems like a lot of overseas PCB assemblers charge a lot more for Coilcraft and Wurth versus TDK/Panasonic, etc... (this is with the PCB assembler buying all the components). But I'm looking for the best performance, so cost will have to take a back seat. |
| Navigation |
| Message Index |
| Next page |