Hey Baby, Bigger is Better...(?)

[Elfen]

With help with from Uniserver, I decided to recap my SE/30 PSU as he determined and evidence shows that is where the Humongus Fish Smell is coming from. And on my thread, he posted up a pic of the caps that need replacement in the SE/30 PSU. So I started to look them up and ran into an interesting situation: Caps for the same rating by different and even the same manufacturer come in different sizes. Thus it comes to a situation - is a physically bigger cap better than a smaller one with the same rating? Is a thin tall cap better or worse than a short fat one with the same rating? Note - there are a few that are the same size, so is this one a "standard size" for such a cap?

Here's a partial listing from Mouser.com which I was looking at and found this:

(Note: I am not affiliated with Mouser in any way, I was using them as a resource to look up the caps in question. In this list, I removed other details such as price per unit as we are discussing cap sizes for the same ratings and not its pricing. Sometimes buying cheap is a bad thing. So here it goes...)

Capacitor Type:

Aluminum Electrolytic Capacitors, Radial - Can Type : Leaded, 4,700uF, 16volts
 
Manufacturer: 
Mouser Part # : Manufacturer Part #
Description: (Size in mm)
 
Nichicon
647-UVY1C472MHD : UVY1C472MHD
Size: 16x25
 
Nichicon
647-UVR1C472MHD : UVR1C472MHD
Size: 16x25
 
Nichicon
647-UPW1C472MHD6 : UPW1C472MHD6
Size: 18x25
 
Panasonic
667-EEU-FR1C472L : EEU-FR1C472L
Size: 12.5x35
 
Panasonic
667-EEU-FR1C472 : EEU-FR1C472
Size: 16x25
 
Nichicon
647-UHE1C472MHD : UHE1C472MHD
Size: 18x25
 
Panasonic
667-EEU-FR1C472B : EEU-FR1C472B
Size: 16x25
 
Nichicon
647-UPS1C472MHD : UPS1C472MHD
Size: 16x31.5
 
Panasonic
667-EEU-HD1C472 : EEU-HD1C472
Size: 12.5x25

 

[techknight]

You have to look at the ratings. They are all going to have different ESR, ripple current, temperature, and life-hours ratings.

 

[Elfen]

Looking at the Alum. Electro. Caps again, not all that information is posted unless I read each cap's data sheet.

Where it is posted, in just a few of them, perhaps a third of the list, are all more or less the same:

Temp: -5C to 105C

Tolerance: 20%

In reading a few data sheets, ESR is not mentioned, though there is a lot of technical information where the ESR could be deduced. But the Average Joe is not going to do this and needs something more simplier to grasp on to. And they all state 1000 hours as a testing benchmark and not as an actual bench mark of the product's life ("After 1000 hours of operation, the capacitor is still within its 20% tolerance blab blab blab..."), there is no mention of actual or predicted shelf or operational life span for the device. This makes the information statement is pretty much useless. 1000 hours is about a month and a half of continuous use or a couple of years of on/off usage.

I know my electronics but am sometimes confused by situations like this one. Things are not always what they seem to be. I understand what you are saying TechKnight, but not all that information is posted, not even on their data sheets.

 

[techknight]

I know digikey lists alot of this information when looking up capacitors. 

I know not all datasheets list this information, but when using capacitors in switching power supplies, this information is critical! Coming from an electronics engineering background, I tend to know things like this. In case your wondering, thats what I went to school for. Electronics engineering. Granted its only a 4 year degree, so I know theres a ton more stuff I can learn, But I do know alot.

Crazy thing is, I graduated before the whole AVR/PIC Arduino craze, and those other robotics stuff thats out now. Only thing I learned in school was on 68HC11 and Z80 trainers. What a mess...

I learned if the capacitor doesnt provide the information you need, Then dont buy it. You need a low-ESR high temperature 105C cap for SMPS units. If you want it to last a long time.  The higher the ESR, the hotter the capacitor will run, along with a higher ripple current will be generated within the capacitor leading to more heating.. See where this is going? Now granted this is an EXTREME case scenario and may not always apply. But it is there, and you should be aware of it. 

Ever noticed on cheap chinese power supplies, the popped caps always scorched the PCB? Well thats because the ESR of the capacitor was too high causing a thermal runaway condition which ultimately blew the capacitor. That, or the ripple current coming through was too high for the capacitors ESR rating, and ripple rating. 

All the caps I look at on digikey, show me hours at rated temperature of life, ESR, ripple current, etc. 

 

[Elfen]

Thanks for that information. I'll take it to heart.

I am not as gifted with the wealth of knowledge you have, TechKnight, but I did wet through years of experience with electronics as far back as 1969, pulling tubes out of dead TVs and radios to fix others. Things were simple then, esp. for a 7 year old who like looking at a glowing get of tubes instead of the cartoon at the crt's face. From there to CB and Ham radio, and then computers with the Commodore PET. Electronics was a bit more than a hobby, as was computers. I became one of First Star's #1 programmers, earning a couple of million on the video games I did then, the money would be gone before 1999. Thing is, the Computer industry in those days were raw. No one knew how to fix them, and those that did were self taught. That is what I did - tech myself how to fix dead or broken machines with a multimeter, a logic probe and a soldering iron. I became good at it. Two 74LS125 on an Atari 800 and XL Series blowing out would kill a keyboard. Thing is, why are they blowing out at the same time together? And the Apple II/II+ used a separate microprocessor for its keyboard decoding, but that is not part of the IIe/IIc system. 

I became apple certified and fixed the classic macs in those days and then taught students who to do it. I would be hired to be a comp tech and instructor for the Bored of Ed and stayed with them for almost 30 years. That is where I learned all my Mac stuff as it evolved. Joined the NYC Mac User Group and earned a rep as one of the best, though I did not consider myself as such.

20 years later here I am- BOO!

A lot of the things changed. One did not expect for caps to fail in those days. Now its part of the scenery. There is a lot for me to learn. SMD/SMT I got a general grasp of. But electronics today, everything is modulized, if a module fails, throw out the whole thing and buy a new one.

So thanks for the information.

 

[techknight]

Yea its all modulized these days. Luckily in most cases I can fix them component level. Not easy though.

You were around wayy before me. I didnt come into this world until 85. Didnt know or use computers until 95 or so.

 

[Gorgonops]

 And the Apple II/II+ used a separate microprocessor for its keyboard decoding, but that is not part of the IIe/IIc system.

Actually the IIe has a dedicated keyboard controller as well, it just happens to be on the main board instead of stuck to the back of the keyboard itself.

(Same part is used in the IIc and the Apple III. The IIgs uses a custom 6502-based MPU to both emulate this part and to drive other aspects of the ADB bus. One of my two IIgs has a bad one of those unfortunately; it's pretty much unobtainium short of desoldering one from a IIgs that's broken in some other way.)