just use panasonic FR 3300uF 6.3v 10x25mm caps. the vrm design doesnt look like it uses a high switching frequency so esr is not that important. better to keep capacitance the same.
The inductor core looks like Micrometals Type 52. I'd guess the switch frequency is around 200KHz. Also, the original Xbox, IIRC, is a late 90s design, so ultra low impedance types probably would not have been used. Panasonic FR or FM look like good choices; from Nichicon, maybe HE or PA; from Chemicon, maybe KY or KZE; from Rubycon, maybe ZL.
PeteS in CA
Power Supplies should be boring: No loud noises, no bright flashes, and no bad smells.
****************************
To kill personal responsibility, initiative or success, punish it by taxing it. To encourage irresponsibility, improvidence, dependence and failure, reward it by subsidizing it.
****************************
if you search the forum i'v done a number of posts about this subject and how to upgrade to polymer caps
I saw your previous posts. Obviously polymers are really good but I feel modern day electrolytic's are more than capable of replacing capacitor-plague-era caps and lasting decades.
--------------------------------------------------------------
Ryzen 3600x
16GB Patriot 3600MHz
MSI B450 Gaming Plus
MSI Air Boost Vega 56
Acer 32" 1440P Freesync
Rosewill Capstone 750W
--------------------------------------------------------------
Hakko FX-888D Station
FX-8802 Iron
MG Chem .8mm 63/37 RA 2.2%
well,
you cant get ultra-low esr electrolytics anymore - because the cap makers realised that polymers are superior in most ways.
the original caps in that position are usually nichicon HN series btw.
and before they fitted 3x 3300uf ones, they fitted 5x 1800uf caps.
hence the 2 empty spaces
well,
you cant get ultra-low esr electrolytics anymore - because the cap makers realised that polymers are superior in most ways.
the original caps in that position are usually nichicon HN series btw.
and before they fitted 3x 3300uf ones, they fitted 5x 1800uf caps.
hence the 2 empty spaces
I actually found Nichicon polymers in 2200uf flavor. Would three of those suffice or would you still populate the extra 2 spots?
Needs to be 5mm lead spacing and no taller than 20mm, width can vary.
the first two u picked are too fat and may not fit as there are a couple of spots on the pcb where the caps are side by side and there may not be enough clearance to fit 12.5mm diameter caps in there.
the last one has too low ripple current handling and too high esr and is more meant for general purpose rather than low-esr applications.
so since it cant be taller than 20mm so it may interfere with the cpu heatsink clips, u can try panasonic fs 3300uF 6.3v 10x20mm caps instead.
Also, could I just load up on 1800uf and replace all the 3300's and 1500's with just those to keep things simple?
u must keep the total bulk capacitance as close to the original as possible. since the original has 3x3300uF it means the original total bulk capacitance is around 10,000uF or so. if u use 2200uF caps then use 5 caps to get as close to 10,000uF as possible.
I saw your previous posts. Obviously polymers are really good but I feel modern day electrolytic's are more than capable of replacing capacitor-plague-era caps and lasting decades.
it depends on how hot the hardware runs. polymers last better on hot running hardware but since this is a pentium 3 based system, it doesnt run hot at all unless u intend to put the console in a hot closed cabinet during summer. cost is also a factor as polymers are more expensive than electrolytics.
since this is a pentium 3 based system, it doesnt run hot at all unless u intend to put the console in a hot closed cabinet during summer. cost is also a factor as polymers are more expensive than electrolytics.
no, this is an xbox, the cpu and gpu are poorly cooled in a very confined space - it does get hot.
what you arent seeing is the dvd and harddrive trays about 5mm above the board!!!
and of coarse you have the switching psu on the right - also under the drive trays with no direct fan!
u must keep the total bulk capacitance as close to the original as possible. since the original has 3x3300uF it means the original total bulk capacitance is around 10,000uF or so. if u use 2200uF caps then use 5 caps to get as close to 10,000uF as possible.
I read STJ say somewhere that with polymers, you can cheat a good bit on capacitance since they are so robust.
--------------------------------------------------------------
Ryzen 3600x
16GB Patriot 3600MHz
MSI B450 Gaming Plus
MSI Air Boost Vega 56
Acer 32" 1440P Freesync
Rosewill Capstone 750W
--------------------------------------------------------------
Hakko FX-888D Station
FX-8802 Iron
MG Chem .8mm 63/37 RA 2.2%
Nothing bulging on this V1.3 board and clock cap already removed. Would you proactively replace anything on this board?
Yes.
Replace all Nichicon HM and HN series caps with H01xx, H02xx, H03xx, and H04xx date codes. These WILL go bad, if not already (sometimes they don't bulge/leak, as I found out on an Intel 845 board that kept crashing with normal-looking HM caps, but they tested quite out-of-spec once removed.) No point to wait for them to fail. If you have the caps in stock, and since you already have the board out, just replace them.
^ That's completely normal... for BAD/FAILED/FAILING Nichicon HM and HN caps, anyways.
What you are seeing is high leakage current - a.k.a. the capacitor is loosing charge between [+] and [-] plates as if there is a small-value resistor placed in parallel with the cap. This leakage current is essentially fooling your meters into thinking the caps have larger capacity - which they don't!
This is why there is also a tolerance specified on the capacitance... and for electrolytics, this is typically 20%. So a 3300 uF cap can read as low as 2650 uF or as high as 3960 uF and still be considered in-spec.
My advice, however - at least for ultra-low grade electrolytic capacitors - is to be wary of them if they jump more than 15% high in capacitance. If they are at or close to 20% higher capacitance than printed on sleeve, they are probably EOL or nearly there and should be replaced.
This is why I suggested you change these caps, even if they didn't look bad. Nichicon HM and HN with H01xx, H02xx, H03xx, and H04xx date codes are known to go bad like this. In addition, some HM, HN, and HZ series with H05xx date codes (particularly 6.3V -rated caps) have had problems too. United Chemicon KZG and KZJ series, along with Sanyo WF usually fail in the same way. All of these are ultra-low ESR motherboard-grade capacitors.
It's not that all ultra-low ESR motherboard-grade capacitors have these problems, but the ones listed above often do. So if you see these anytime, always check them.
^ That's completely normal... for BAD/FAILED/FAILING Nichicon HM and HN caps, anyways.
What you are seeing is high leakage current - a.k.a. the capacitor is loosing charge between [+] and [-] plates as if there is a small-value resistor placed in parallel with the cap. This leakage current is essentially fooling your meters into thinking the caps have larger capacity - which they don't!
This is why there is also a tolerance specified on the capacitance... and for electrolytics, this is typically 20%. So a 3300 uF cap can read as low as 2650 uF or as high as 3960 uF and still be considered in-spec.
My advice, however - at least for ultra-low grade electrolytic capacitors - is to be wary of them if they jump more than 15% high in capacitance. If they are at or close to 20% higher capacitance than printed on sleeve, they are probably EOL or nearly there and should be replaced.
This is why I suggested you change these caps, even if they didn't look bad. Nichicon HM and HN with H01xx, H02xx, H03xx, and H04xx date codes are known to go bad like this. In addition, some HM, HN, and HZ series with H05xx date codes (particularly 6.3V -rated caps) have had problems too. United Chemicon KZG and KZJ series, along with Sanyo WF usually fail in the same way. All of these are ultra-low ESR motherboard-grade capacitors.
It's not that all ultra-low ESR motherboard-grade capacitors have these problems, but the ones listed above often do. So if you see these anytime, always check them.
Thanks for clearing that up. I was baffled. When I saw the low ESR rating, I thought I replaced them for nothing. I feel better now. I actually replaced 4 caps in total tonight. I plan on doing more or less all of the electrolytic's when I have more time. Here is what I replaced tonight...
The three 3300uf HM's got replaced with 3900uf Rubycon ZLQ's and the larger 3300uf 10v cap got replaced with two Panasonic 2200uf 10v FM or FC I forget.
Attached Files
Last edited by Per Hansson; 10-11-2020, 04:34 AM.
Reason: Offsite image uploaded
--------------------------------------------------------------
Ryzen 3600x
16GB Patriot 3600MHz
MSI B450 Gaming Plus
MSI Air Boost Vega 56
Acer 32" 1440P Freesync
Rosewill Capstone 750W
--------------------------------------------------------------
Hakko FX-888D Station
FX-8802 Iron
MG Chem .8mm 63/37 RA 2.2%
Comment