Shuttle SN85G4 Mobo FN85v1.2

[seanc]

By coincidence I have two of these systems - both with different issues.

Machine 1:
This one I'm most interested in and was full of bad caps. The PSU had every capacitor bulging (including mains cap) apart from two 470uf 16v, I believe this is what caused the caps on the mobo to bulge and cook. All caps on the motherboard seem to be 10mm or below.

I've replaced 4 bulging OST RLX 1500uf 16v 10mm with equivalent Panasonic FC but 12.5mm in diameter so they're sitting a bit crooked. I've also replaced 2 bulging OST RLX 1500uf 10v with equivalent Panasonic FM 10mm. - it still boots, so that's good.

I've got 6 visibly OK OST RLS 3300uf 10v 2150ma ripple I want to replace, but they have to be 10mm else they won't fit. I cannot for the life of me find anything that matches this. The closest I have come is Panasonic FK (the datasheet seems to suggest they're better than FC) 2700uf 10v 2140ma. I'm assuming this is the CPU VRM section, do you reckon I can get away with the lower capacitance?

Machine 2:
Everything capacitor wise in there looks good and it runs fine, when it's not overheating.
The CPU fan was connected one of the fan headers on the motherboard. The fan doesn't work on another PC, no fans work on any of the 3 fan headers on the motherboard.

Anyone have any ideas on where I would start looking to get the fan headers working? Any specific make of chip or is it likely to be a resistor?

[kaniki]

Re: Shuttle SN85G4 Mobo FN85v1.2

Machine 1:

I've got 6 visibly OK OST RLS 3300uf 10v 2150ma ripple I want to replace, but they have to be 10mm else they won't fit. I cannot for the life of me find anything that matches this.

Machine 2:
The CPU fan was connected one of the fan headers on the motherboard. The fan doesn't work on another PC, no fans work on any of the 3 fan headers on the motherboard.

Machine 1:
why not try the samxon brand.. they have the 3300 10v caps in 10 mm

https://www.badcaps.net/store/produc...products_id=57

and im sure you could get them from around where you are??

Machine 2:
Are you saying that the fan works on other computers, but not on any of the fan headers on the second computer? sorry but the way that you have that worded, it sounds like the fan does not work on any computer..

[stj]

Re: Shuttle SN85G4 Mobo FN85v1.2

no2.

meter the 12v on the psu connector - *not* the extra small connector, the big 20pin one.

if it's there then take the board out and look at that connector for damage & try to see how the power gets to the fans - it may go through a fuse or inductor or something.

[seanc]

Re: Shuttle SN85G4 Mobo FN85v1.2

Machine 1:

As far as I can see, no one stocks Samxon. Suppliers I've tried in the UK:
RS Components
Digikey
Farnell
Mouser

Any other suppliers who I can check for UK equivalents?

Machine 2:

The fan I removed, which was connected to fan header 2, does not work in another PC. None of the fan headers 1, 2 or 3 work with a standard 80mm fan.

It runs perfectly, except for the fan headers. It now has a brand new PSU in it. Machine 1 has the PSU from machine 2 and is working OK, apart from needing the mobo recap finishing.

I've replaced the fan with one from RS Components, 80x15mm with a max speed of 2440rpm, hooked up a 12v molex connector. As long as it's cool enough, I'll stick with this and ignore the fan headers.

[stj]

Re: Shuttle SN85G4 Mobo FN85v1.2

the missing 12v may effect some pci cards or the serial-ports,
it's worth finding the fault.

[seanc]

Re: Shuttle SN85G4 Mobo FN85v1.2

STJ, I now believe you have a very good point.

I booted the 2nd Shuttle and tried to browse the internet, extremely slow, same with file transfer. Tried to PXE boot, no go/very slow to get a lease and then fails to get the boot file. Definitely something wrong with networking, possibly related to the fan headers not working.

Testing the 1st Shuttle, it will get a DHCP lease and the boot file every time.

I'm thinking I'll do a full re-cap on both. With regards to the VRM caps, do you think I'll be safe dropping to Panasonic FK 2700uf from 3300uf? If I could get hold of the Samxon GD series I could go with the correct specifications, but no UK suppliers have Samxon. If not, I'll order this site, but shipping will take a while.

[seanc]

Re: Shuttle SN85G4 Mobo FN85v1.2

So, I've since re-capped both power supplies and they appear to be working fine, with the exception of a 11.37v (new one) vs 11.42v (older one) discrepancy on the 12v line, with no load.

I've half re-capped the board with the dead fan headers and tested with an Antec Earthwatts 380 - it's still running as it was, fan headers are still dead. Having done some more internet trawling, it seems a BIOS update may fix it

I'm still stuck with regards to replacing the VRM caps. I've tested them with a multimeter, they hit 3.3v on power on, then drop to 1.52v which matches the VCORE shown in the BIOS.
Is there any reason why I shouldn't replace them with 3300uf 6.3v (as opposed to the original 3300uf 10v), I can get 3300uf 6.3v 10mm Panasonic FK in that size?

I'm not going to overclock (they're replacing my parents' Dell GX270), so I can't see a reason for the VCORE to get any higher than the 3.3v at its initial power on.

Any help really is appreciated!

[stj]

Re: Shuttle SN85G4 Mobo FN85v1.2

a lot of people here talk of replacing vrm cap's with polymer solid caps.

[kc8adu]

Re: Shuttle SN85G4 Mobo FN85v1.2

I'm still stuck with regards to replacing the VRM caps. I've tested them with a multimeter, they hit 3.3v on power on, then drop to 1.52v which matches the VCORE shown in the BIOS.

if in fact vcore is overshooting that badly caps are the least of your problems.that will kill the cpu quickly!

[seanc]

Re: Shuttle SN85G4 Mobo FN85v1.2

if in fact vcore is overshooting that badly caps are the least of your problems.that will kill the cpu quickly!

Did some more poking around, cheap auto ranging meter is getting confused - it's not actually hitting anything above 1.5v at all.

Tested by setting meter to V only and tested with a second meter.

[seanc]

Re: Shuttle SN85G4 Mobo FN85v1.2

All caps replaced on both boards, except 3300uf 10v and 100uf 16v. Waiting on back order.

Broken network card on one motherboard turns on and off (much like it's restarting) when you try to boot from LAN. When you power it off via the power button, it still does the same thing even though nothing is is trying to use it. Definitely something wrong with it.
Tried re-flowing pins on board for the connector, and 2 x Attansic chips. I guess I should look for an nVidia chip since it claims to be nVidia onboard LAN.

I can't work out the fan connector issue. I tried replacing the small 68uf 16v caps next to each of them (3) with 47uf 35v capacitors. They still don't work. Voltages are either low or non-existent.

Just found out the "-" button on my soldering station is broken. I can only increase temperature!

[seanc]

Re: Shuttle SN85G4 Mobo FN85v1.2

Found a Realtek RTL8201BL chip. Will investigate poking this with a pencil/re-flowing (carefully).

I'm still skimming the datasheet, but I noticed this: "25MHz crystal/oscillator as clock source". I've spotted at least four crystals on the board. I wonder if one of these is faulty.

[swartz]

Re: Shuttle SN85G4 Mobo FN85v1.2

Yeah I'm replying to an old post, but whatever. Someone might find it useful.

Also recently acquired two of these systems. Exact same model and revision FN85 v1.2 on the mobo.

Before re-capping
-----------------
System #1:
Booted to BIOS, sometimes even booted to Ubuntu (via USB drive). But was not stable, random shut-offs. 3 visible bad caps (2 of OST RLX 10v 1500uF, 1 of can't remember now). Dead FAN1 connector, no voltage registering.

System #2:
No visibly bad caps. Sometimes turned on when power was pressed but never signal to monitor (neither with built-in nor AGP video). Would randomly shut off after a few moments. FAN1 and FAN2 header were not working, FAN3 did.

After recapping
------------------
System #1:
Just the 3 bad caps were replaced with panasonics. Now mobo is stable and passed memtest, and various stability tests (like pi, video encoding, etc) under Ubuntu. Everything works, PCI slot, AGP, onboard audio, video, and ethernet.
Fan#1 header still does not work. Fan2 and Fan3 headers work fine.

System #2:
After recapping the same three caps (just a shot), problems remained. After recapping all caps, still the same. Tried different and tested PSU, still nothing. Abandoned.

[Scenic]

Re: Shuttle SN85G4 Mobo FN85v1.2

check the tiny transistors around the fan headers.
they blow up really easy on those shuttle shitboxes if the fan draws too much current, resulting in no voltage at all on the fan header

[seanc]

Re: Shuttle SN85G4 Mobo FN85v1.2

Funnily enough I still have both of these boxes and was going to revive this thread...

Box 1 is running fine with Panasonic FK for the CPU and Panasonic FM everywhere else.
Replaced and re-capped the PSU since I screwed the first one.
Working fine apart from 2 of the fan headers.

Box 2 is running fine with the same cap combination and identical PSU recap as box 1, but none of the fan headers work and the onboard Ethernet STILL won't work.
I replaced the RTL8201BL chip and the Ethernet port & attached USB connectors.
Still just turns on when you plug a cable in, blinks the lights and reboots the network card. Weird.

Scenic - I will have a look at the tiny transistors - I can cross reference to the other box. Any thoughts about the network card?