Original XBOX Cap ripple current range

[swofra]

Hey all,
Looking to recap a six or so XBoxs comprised of every major mobo variant 1.0-1.1, 1.2-1.4, & 1.6 and am trying to decide on which Caps to get.
A) For the Caps located by the CPU, I know they should be low ESR/impedance, but what is considered low, anything less than 100?

B) Additionally, what minimum low frequency and high frequency ripple should I be looking for? I know higher is better but with so many cap choices and price points out there, I'm just trying to narrow down my search window.

C) I've read some stuff on the various threads that 1.0-1.1 mobos are more dependent to low ESR but I've only seen that in one place so I'm not sure.

I'm planning on also limiting the search to Panasonic, Rubycon, NichiCon, ChemiCon, Werth, and Sanyo.

Thanks in advance for any insight you can provide!

[stj]

Re: Original XBOX Cap ripple current range

use polymer for the cpu caps

[swofra]

Re: Original XBOX Cap ripple current range

I know Polymer is preferred, but I'm trying to build up my knowledge regardless so thats why I'm looking to see what is considered low / not to exceed ESR and the ripple current low/high mins I should be considering.

Thanks!

[stj]

Re: Original XBOX Cap ripple current range

well the original box had 5x 1800uf in it, then they dropped to fitting 1500uf and later they had 3x 3300uf
so there is no exact original uf/esr value to go by.

more importantly, the original "low esr" caps used for cpu smoothing arent made anymore because of the availability of polymers.
you wont find nichicon HM or HN and you wont find rubycon MBZ or MCZ either.
if you do, they are decades old or fake

[momaka]

Re: Original XBOX Cap ripple current range

I've used Rubycon ZLH and a few other ZL-based series, along with Panasonic FR, FM, & FS, Nichicon HV and HW, and UCC KZH/KZM to replace UCC KZG and similar equivalents (i.e. Nichicon HM/HN and Rubycon MCZ/MFZ.) ZLH, FR, FM, FS, HV, and HW are not quite up on par with HM, HN, MCZ/MFZ, KZG, and KZJ... but they are close enough and should work fine.

So if looking for modern replacements that are still made, the above should work acceptably well. If the board you're working on has free/unpopulated cap spots, fill those too to lower the overall ESR. In any case, the hardware isn't that demanding... so as long as you have decent low-ESR caps in there, it will be OK.

[swofra]

Re: Original XBOX Cap ripple current range

Thanks for the guidance. I've looked and think I am going to go with the following, let me know if you see any issues with these:

1500: Rubycon 10ZLH1500MEFC10x16
1800: Rubycon 63ZLH1800MEFC10x16
2200: Nichi UHD0J222MPD
3300: Chemi EKZN6R3ELL332MJ25S

Another question - I know in a pinch I can use higher voltage capacitors without any negative effects, but, if I run out of caps at a certain capacitance can I sub a higher capacitance one if necessary (meaning can i put a couple of 2200's on in place of 1500s) without creating issues? And Could I, (again only if absolutely necessary such as I've run out of a certain capacitance) mix caps with different capacitance in the space near the CPU?

Thanks again for all the help!

[stj]

Re: Original XBOX Cap ripple current range

yes, but watch out for height clearance.
remember that dvd drive sits real low.
and try not to block access to the cpu heatsink locking lever.

[swofra]

Re: Original XBOX Cap ripple current range

Thanks, and just to be clear, did that yes answer apply to the fact those caps are ok AND both the follow on questions:

1) if I run out of caps at a certain capacitance can I sub a higher capacitance one if necessary (meaning can i put a couple of 2200's on in place of 1500s) without creating issues?

2) Could I, (again only if absolutely necessary such as I've run out of a certain capacitance) mix caps with different capacitance in the space near the CPU (as in 2 x 3300 and 1 x 2200 on a 1.6 board or 3 x 1500 and a 2 x 2200 on an earlier board, etc

Think the height should be good - I only plan to replace the CPU caps for now as long as that solves any issues that may be present and I've removed all the clock caps on the 1.0-1.4.

Thanks!

[stj]

Re: Original XBOX Cap ripple current range

yes you can go up in uf and/or voltage - just remember the cap will be bigger

no, dont mix caps round the cpu.
i think they are all in parallel on the xbox, but if you mix they will have different current/esr and the discharge curve will be all screwed up
on newer pc designs they are actually in 3 to 6 groups that must match

[momaka]

Re: Original XBOX Cap ripple current range

Thanks for the guidance. I've looked and think I am going to go with the following, let me know if you see any issues with these:

1500: Rubycon 10ZLH1500MEFC10x16
1800: Rubycon 63ZLH1800MEFC10x16
2200: Nichi UHD0J222MPD
3300: Chemi EKZN6R3ELL332MJ25S

No issues, looks good.

Though if there is enough height above the 1500 uF and 1800 uF caps, just go with Rubycon ZLH 6.3V 2200 uF for all of these to save with bulk-buying in 10's. Which leads me to your next question:

Another question - I know in a pinch I can use higher voltage capacitors without any negative effects, but, if I run out of caps at a certain capacitance can I sub a higher capacitance one if necessary (meaning can i put a couple of 2200's on in place of 1500s) without creating issues?

Absolutely!
The capacitance rating on most caps is +/-20%. So most circuits are able to tolerate AT LEAST that much change in capacitance... and in practice, even more (even 2x or 1/2 x capacitance is not overly excessive.) I rarely even bother anymore to try to match values exactly on a VRM circuit. So long as the overall ESR matches and the capacitance is "reasonably" close (I'd say don't go more than +/-250-300%), it will work. At least I never had a case where it didn't.

And Could I, (again only if absolutely necessary such as I've run out of a certain capacitance) mix caps with different capacitance in the space near the CPU?

If the capacitors are connected in parallel, yes.
Again, it's the overall capacitance and ESR of the circuit that matters.
In fact, I do this very often on older motherboards that might actually be slightly sensitive to a decrease in capacitance (in terms of OC-ing.) I usually do this when partially poly-modding the CPU VRM. I'll put polymers with smaller capacity (because that's what I usually have on hand) and then mix in electrolytic caps with high capacitance to balance out the total capacity. Here's one example:
https://www.badcaps.net/forum/showthread.php?t=63041
Basically for that one ^, I subbed 2x 820 uF polymers and 3x 2200 uF caps in place of the original 5x 1500 uF caps.

yes you can go up in uf and/or voltage - just remember the cap will be bigger

And to add to that, you CAN also go down in the voltage SOMETIMES.
Example: in ATX PSUs and standard ATX/mATX/ITX PC motherboards, the only main rails that come out of a power supply are 3.3V, 5V, and 12V. Both 6.3V and 10V caps can be used on the 3.3V and 5V rails, but NOT on the 12V rail. Thus, if you see a 10V cap anywhere, it can often be replaced with a 6.3V cap and vice-versa.
Sometimes, a manufacturer will use a higher voltage -rated cap simply because the higher voltage rating will increase the capacitor's size, and that will also make it have lower ESR (compared to a capacitor from the same series of the same capacity, but lower voltage.)

no, dont mix caps round the cpu.
i think they are all in parallel on the xbox, but if you mix they will have different current/esr and the discharge curve will be all screwed up
on newer pc designs they are actually in 3 to 6 groups that must match

Not sure where you've read this, but it's NOT true at all.
Buck-regulated circuits (CPU VRMs, to name the most obvious one) only care that a certain capacitance is present on the output of their inductor(s) and that the circuit also has a certain overall low-enough ESR. So whether you achieve this with 90x small 100 uF caps (to get 9000 uF total capacitance) or 4x 2200 uF caps or 3x 3300 uF caps or 2x 4700 uF caps, it's all the same to the buck circuit. Again, so long as the total ESR of these caps in parallel meets the design spec of the circuit.

[swofra]

Re: Original XBOX Cap ripple current range

Thank you SO much for this information, its greatly appreciated and great knowledge to have. Think I'll just 2200's to keep it simple. Will use:
3 x 2200 on a 1.0 board in place of the 5 x 1.5's and
4 x 2200 on the other boards in place of the 3 x 3300s (if I run out of 3300s).

Thanks again!