I am using as standard:
0603 - 4u7/16V X5R; 1u/25V X5R
0805 - 22u/6V3 X5R, 10u/16 X5R , 1u/50V X7R,
1206 - 10u/25V X5R, 1u/100V X7R
1 Like
I donât use tantalum. They can got in fire even with short overvoltage.
Most ceramic are tested with 2,5x overvoltage.
So the enginner rule is use tantal for 2x voltage you have and you can use ceramic for 1x voltage you have (I prefere more than 1x) So for VCC of 5V I would use 10V tantal or 6V3 ceramic.
2 Likes
This topic has been about selecting a suitable capacitor for the output of a low power LDO regulator. The requirements you specified were not even correct. Just about any X5R/X7R ceramic would do. All the information you needed to source the correct parts, just like anyone else would, is in the datasheet, not to mention the PDF that @maui linked above.
2 Likes
When I was a student at electronic Iâve designed and assembled an oscilloscope. It was not school project but my own - just because those times (1983) it was not possible for me to buy one and I wished to have it. I had no such source of info like internet. I suppose if sometime internet would be switched off many people would say we canât do enything.
The key parameters are value, dielectric type and voltage.
With ceramics and the right dielectric, the ESR is always low.
You have not said what your input and output voltages are.
There is no point using a 50V capacitor if the input is 9V and the output 5V
The datasheet implies that you might want to use a higher value on the output, 4.7uF at 10V would be a good choice
1 Like
So you claim, yet your behaviour says otherwise ?
Ok, so you need to actually read a data sheet. Disti sites do not have all the filters anyone wants.
Did you look at even one data sheet ?
General comments:
- 500mOhms is actually quite a high ESR. What frequency is that specâd at ?
- When searching, do not set exact values, use (say) 1uF~10uF and 50V~100V
- Dielectric matters, so avoid the ones that say -20%, +80%
- Once you have a Value.Voltage.Case.Dielectric choice, do not be afraid to look at the data sheet.
eg this data shows ESR, goes as low as 8mOhms at 1MHz
http://ds.yuden.co.jp/TYCOMPAS/ut/detail.do?productNo=UMK316B7225KL-T&fileName=UMK316B7225KL-T_SS&mode=specSheetDownload
Notice also C varies with temperature and voltage, so you need to consider that 1uF still exists at temperature and bias. What voltage does this operate at ?
1 Like
Not quite - with ceramic caps, you need to look at the C - vs - Voltage curves, as higher bias voltages do reduce the C.
That could mean using a seemingly high voltage, just to get the C value you need.
1 Like
The reduction in capacitance with voltage is why Analog specified to use the X7R or X5R dielectrics and noted not to use the HiK dielectrics like Y5V
If you look in RS most 50V, 1uF, 1206 are X7R, which is a stable and low ESR dielectric
X7R is somewhat temperature stable, but not very voltage stable according to the Taiyo Yuden data.
Google found this,
http://ds.murata.co.jp/simsurfing/mlcc.html?lcid=en-us
which is quite a nifty quasi-spice curve generator. Pick almost anything to graphâŚ
About 30% reduction for a Samsung X7R 50V part at 30V.
In 1206 size, there is plenty of choice at 4.7uF/50V
Ohms Law.
2 of whatever you can buy will halve the ESR.
Just to mention: both Digikey and Mouser have techs. If you call their ordering number, just ask the nice person you need to talk to a tech. They are very knowledgeable and can help you.
Pro tip: When you get a job as an EE (assuming you do), There are three sets of people you do not want to piss off: Techs, Admins, and the Mafia. They can make or break you.
Admins and Techs should be obvious. They are like NCOâs in the military. They keep the place running. They fix the mistakes. They know everybody. As an EE, you are dependent on the techs. As an employee, you are dependent on the Admins.
You are likely to be in a production environment at some point. The Mafia are usually a few little old ladies who have been there since God was in short pants. If you piss them off, you will never get any help to do anything. Be nice and polite to them. Just sit in the production area and see who goes to whom. Once you know who they are, approach politely and introduce yourself as the new guy. They will know that already. You might want to bring a nice ceramic bowl and a big bag of M&Mâs. Say you have this gift so that, in the future, if you get in a crunch (which you will), you just want them to remember you. They will know why.
Also remember they will be watching you. If you are an ass to others, they will remember. If you are a hard worker, professional, and polite, they will know. They are the center of an efficient rumor and information network. And, if you complain to their boss, you should start looking for a new job.
5 Likes
Interesting statement and all too true. However in my experience there are mostly two types of people, Those who intuitively know what you are saying and those who will never will.
I first read this thread after post #7 and quickly fell into the old lady group you mentioned.
1 Like
And sometimes youâre better off not knowing what they did to keep things running. (Especially the techs.)
In which branch did you serve? Comparisons like this are unlikely to be accurately understood or appreciated, since few engineers or managers ever wore a uniform.
Dale
1 Like
Never did - 4F - I have one hand, thus the âone-armed banditâ. But - my folks were Navy (met in the Navy, mom was honorably discharged, because that is what happened in those days.) Also - I married a Marine. (Her âOver your dead bodyâ is not a figure of speech :^)
Plus, I have lived in Albuquerque since 2003, and worked for several govt contractors. Spent a bit of time on a couple of bases plus been on a couple of weapons research facilities (WSMR, Yuma - in June - joy). If you are a nerd in Albuquerque, it is basically impossible to not work with the military.
I have seen NCOs pull otherâs chestnuts out of the fire. True story. Was on a SBIR gig to gather data to create a drone refueling system. Flight tests at Niagara AFB. Lear jet pretending to be an F-16. Tests were the Lear pretending to be refueled by a KC-135. They brought the boom within a couple of feet of the Lear. This was some serious piloting. Air Force test pilots - some of the best you will ever find. The boom hitting the Lear would have been a Bad Thing. I had a very fancy video recorder. Camera was on the Learâs âdashâ. I was on two flights. Never had two hours pass so fast.
Another team had a GPS system - but forgot to be able to set it up for military precision. Panic ensued. A Sgt mentioned his brother, another Sgt, had the clearance and the majic incantations. Saved the GPS teamâs ass. We are talking a $1Mill project just for the data.
Whenever I am on a base, everybody is âSirâ or âMaâamâ. NCOs let it slide because I am a civilian. Also, âI know I am not supposed to call you Sir, Sirâ works. Donât want to call a Senior Master Sgt just âSgtâ.
I was pretty aggressive in High School - football, wrestling, etc. I probably would have gone Marines or Army. As it was, nobody mentioned the intelligence branches - a logical place - they care about brains.
+dchisholm
Which branch were you?
U.S. Air Force communications officer, 1975 - 1981.
In Sept 1969 I started college and set aside a 2S student deferment to pursue a commission through the ROTC program. I expected that would include a Southeast Asia tour at some point. By the time I stood to receive the Presidentâs commission in June 1973, the U.S. was beating feet out of SE Asia and drastically reducing military size. The Air Force didnât have manning slots for all the people enrolled in commissioning programs. I took 18 months to make my first attempt at grad school, and eventually spent 6 years behind desks doing various headquarters-support jobs.
Dale
1 Like
I donât doubt that!
One of my most prominent memories from years in the Air Force was one ride as a passenger on a KC-135. I also gained an appreciation for some of that âserious pilotingâ when the boom operator let me watch over his shoulder while refueling a B-52. Later, several of us crammed into the boomerâs pod to take in the scenery when the flight path followed the Grand Canyon.
Like you, most of my interesting âmilitary experiencesâ came after I left service and was working for defense contractors. Itâs especially sobering to realize that some GI may be putting his life may on the line, trusting in some design decision you made. Occasionally you really DO remember back to some difficult design problem such as selecting an output capacitor, and hoping that what went into the final version is really up to the performance specs. (Well, at least I had those thoughts a few times.)
Dale
3 Likes
What you donât know is that they may work perfectly fine when mostly new.
20 years later, they catch on fire when power is applied; and they make a horrid stench when in the box sent back for re-work.
1 Like
I can recomand you this nice web tool from Kemet :
http://ksim.kemet.com/
It allows to compute a lot of parameters with curves depending of frequency, voltage, âŚ
3 Likes