dell 270/280 polymer mods

[hardwareguy]

Re: dell 270/280 polymer mods

I should note that this board is the same or VERY similar to the board used in the Dimension 8300. Repair procedures are the same.

[JEWilson]

Re: dell 270/280 polymer mods

On the SX280 I have here, there are 5 of Rubycon MBZ 820uF 6.3V.
3 of them are bloated but test ok.
Are these on a 3.3V rail or used for bypass purposes?

Can I safely replace these with, e.g. polys or just lytics?

Thanks

[kc8adu]

Re: dell 270/280 polymer mods

On the SX280 I have here, there are 5 of Rubycon MBZ 820uF 6.3V.
3 of them are bloated but test ok.
Are these on a 3.3V rail or used for bypass purposes?

Can I safely replace these with, e.g. polys or just lytics?

Thanks

fire up the board and measure the voltage on them.
and mbz bloated?
PICS!
sure they are not mcz or hn/hm?
if they are mbz i would carefully check the psu and cooling.

[JEWilson]

Re: dell 270/280 polymer mods

Don't have a camera to hand
Y - bloated MBZ as well as one of three 16V 1000uF MCZ also

C382 16V 1500uF Rubycon MCZ
C176 6.3V 820uF Rubycon MBZ
C298 6.3V 820uF Rubycon MBZ

This board will POST ok.
The reported problem is the mobo locks up after being in use for a while.
Suspect, the obvious problem is heat. Caps evident enough of this.

Also, two of the 16V 100uF Nichicon ZL are reporting higher than average
ESR. As this mobo is getting a recap I want to eliminate my requirements in the first instance.

So.. I have the caps and the rest are on the way.

Another Q.
Will this mobo support 65nm Pentium 4s such as Intel 631, 641, 651 and 661?

Thanks

[stj]

Re: dell 270/280 polymer mods

how is that case cooled?

[JEWilson]

Re: dell 270/280 polymer mods

Two fans, push pull arrangement at the top of the case.
This is where the CPU VRM is located.

TC refers to all parts being south of these fans as fair game
for temp problems as this includes brands from Rubycon et al.

It is an ultra small form factor chassis

[PCBONEZ]

Re: dell 270/280 polymer mods

Heat can kill anything in those USFF cases.
Even the 220uF & 22uF go eventually.

[Topcat]

Re: dell 270/280 polymer mods

Heat can kill anything in those USFF cases.
Even the 220uF & 22uF go eventually.

I have never seen the 220uF caps fail in the GX620USFF and SX280, even the nichicon's. However, on the SX280, theres a 22uF 50v right next to the CMOS battery, I've seen a lot of those fail.

As I've said before, the caps in the 'north' side of teh case that are in the path of the fans also almost never fail (even nichicons), unless the fans have become obstructed with dust.

[JEWilson]

Re: dell 270/280 polymer mods

Y - in the SX280s I have all of the 16V 200uF caps are rated 85deg C
I consider this a problem waiting to happen, so replace them with
Pana FM in a slightly larger can.

What's the heads up on 65nm Pentium 4s such as Intel 631, 641, 651 and 661 working
with these boards. Have any member tried this?

[PCBONEZ]

Re: dell 270/280 polymer mods

I have never seen the 220uF caps fail in the GX620USFF and SX280, even the nichicon's.

Read up. Others have.

I've found bad & bloated 22uf, 100uF, & 220uF in a number of Intel built boards, even standard ATX retail boards.
Most of the time they are Nichicon VR or similar 85C caps and the board is 4+ years old.

You will see more of them in the USFF [anything actually] as time goes by because in hot boxes [or warm boxes with aging] the 85C just don't have the longevity of 105C caps.
.

[Topcat]

Re: dell 270/280 polymer mods

Read up. Others have.
.

I've seen bad 220uF as well, just never in an SX280 or GX620, which was the topic of this thread, and I see many dozens of these every week. Don't need to read up on what I've got plenty of hands-on with.

[PCBONEZ]

Re: dell 270/280 polymer mods

I guess I should have said read over.
I though we were in this thread.
https://www.badcaps.net/forum/showpost.php?p=83682

https://www.badcaps.net/forum/showpo...2&postcount=37
https://www.badcaps.net/forum/showpo...0&postcount=17
.

[Topcat]

Re: dell 270/280 polymer mods

I guess I should have said read over.
I though we were in this thread.
https://www.badcaps.net/forum/showpost.php?p=83682

https://www.badcaps.net/forum/showpo...2&postcount=37
https://www.badcaps.net/forum/showpo...0&postcount=17
.

Here's the correction I posted in that thread as well:
https://www.badcaps.net/forum/showpo...5&postcount=47

I'm not saying the 220uF caps don't fail in them, but of the many hundreds (maybe even thousands) of them I've seen/serviced, I've never seen one with a bad 220uF cap.

The board I recall seeing lots of bad 220uF, 330uF and 470uF caps in is an Epox P3 mATX that's used in some kind of video game.....they always come to me stale smelling and caked nicotine sludge, loaded with bad caps. i've also seen them on some s478 intel boards, but very seldom on those...

[PCBONEZ]

Re: dell 270/280 polymer mods

The little guys dry out and rarely bloat. Even when they do bloat it's often so slight you can barely see it.
- If you aren't pulling them to check them you won't know if they are bad or not.

I used to do what most other techs are apparently still doing.
[Small & not bloated, don't check.]
Then [maybe 1.5? years ago now] I found bloated VR on a uATX Skt 370 OEM & an ATX P4 retail board [both Intel boards] on the same day. A short while I got 7 more Intel P4 ATX retail with 4 & 5 mm bloaters from 22uF-220uF.
- That got me thinking abount the small caps again.
A couple weeks later I received a box of ~30 Dell SFF boards [mixed Skt 478 & 775] all with some 8mm & 10mm bloaters and most with 4 & 5 mm 85C caps. Out of curiosity I pulled all the 4 & 5 mm 85C caps. About 30-40% were dried out and measured bad [ESR or uF or both] on the meters.
-
Those can affect Vtt and other things to the Chipset, RAM, & CPU which can result in intermittent weird problems later on, so, since that box of Dells, if it's 85C I just replace them, no matter how small.

I don't consider in-circuit checks adequate for those because of possible parallel paths through IC chips [and I don't wanna trace the circuits through ICs], so pulling to check is required. With caps that cheap if you have to pull to check then you might as well just replace.
I'm not 'in a place' where an extra $2-$3 or the extra time on the work bench matters much, but the bother of chasing 'ghost' problems post work does, so I just get rid of the 85C caps.

The 105C go too, just not nearly as often. Interestingly the bad ones of those have been Nichicon or Rubycon but that's probably because that's just what happens to be there most of the time. I've had 4-5mm 105C Sanyo show up too sometimes but never found one of those bad [yet].
.

[kc8adu]

Re: dell 270/280 polymer mods

had 2 gx620's in a local titty bar where they were installed in a cabinet with little airflow.these were their digital dining terminals and they run 24/7.
all the sanyo 220@16 were high esr.this however is an extreme case.any system in this installation will die.
they are properly vented now.