Rubycon MCZ vs. SAMXON GC(M)?

[bw1]

I've got a board that has three of the same 1500uf/6.3v caps that are bad. It's an old board, barely worth fixing anyway, and all of the other caps (all different values/brands) look just fine while these three are leaking.

I know most of you would probably say to replace them all, but for this board (PC133, if that says anything) it's not worth it to me. It's for someone that doesn't use the computer much.

Anyway, my question is this: I've got one Rubycon MCZ cap that's the same size/shape as the original - it's my last one. I've got plenty of SAMXON caps that are the correct value, but instead of being the original short/fatter cap, these are taller and skinnier. They are the GC(M) series.

Should I use the Samxon's, or wait and order more Ruby's?

[pfrcom]

Re: Rubycon MCZ vs. SAMXON GC(M)?

If the board is labelled PC133, it may be a PC Chips board - I know they used to sell boards branded PC100

If it is PC Chips, any investment in it is throwing good money after bad - so use the Samxon if you already have them

[yyonline]

Re: Rubycon MCZ vs. SAMXON GC(M)?

The skinny GCs are *probably* okay to use. To know for sure, post the brand, series, and size of the capacitors you are trying to replace, and we can look up specs.

PC133 is a memory technology - old stuff. It doesn't suggest anything about who manufactured the board.

[bw1]

Re: Rubycon MCZ vs. SAMXON GC(M)?

It's an ECS motherboard -- yep, PC133/100 was referring to memory, indicating it was an old board.

1500uf/6.3v:
G-LUXON
LZ 105*
1127(M)

======

Also looks like I'll need to replace the 1000uf/10v's, currently the following:
G-LUXON
SM 105*
1128(M)

For these, the ones I've currently got that match the cap/volt value are Ruby MCZ's.


Starting to wonder if it's worth it. If I have to start ordering and waiting for caps, it's increasingly less worth it. Although I do need to order some anyway.

Thoughts?

[bw1]

Re: Rubycon MCZ vs. SAMXON GC(M)?

Also, would the Ruby ZLG series be a good replacement for a Su'scon SG 105* 0713(M) cap? (unrelated to above motherboard)

[PCBONEZ]

Re: Rubycon MCZ vs. SAMXON GC(M)?

It's a lot easier to help when you give the physical size of the caps..........

- ESR/Ripple

G-LUXON LZ 1500uf/6.3v
8x15mm - .085/980
8x20mm - .051/1070
10x16mm - .055/1070
10x20mm - .044/1250
HE or KY should be just enough depending on the can sizes available.
WX, KZE, or ZL should be slightly better.
Can't say if there are any actual matches because the can size affects ESR and Ripple ratings.
.

G-LUXON SM 1000/10v [not low ESR]
10x12.5mm - na/490
10x16mm - na/570

What size Su'scon SG?
At any rate SG are roughly comparable to WX, KZE, ZL
Can't say if there are any actual matches because the can size affects ESR and Ripple ratings.
.

[bw1]

Re: Rubycon MCZ vs. SAMXON GC(M)?

Thank you for the information, but it doesn't answer my question completely.

I've got some Samxon GC(M)'s on hand, and am wondering if they will work. I'm guessing not, since you didn't mention them. I've also got a Rubycon MCZ of the same uf/v values, but I'm not sure about the ESR/Ripple on either the Samxon or Ruby mentioned.

Would either of these work as a replacement? (again, I'm guessing not, since you didn't say so, but you also didn't say no)

Can size for the original looks to be 10x16mm for the 1500uf.
The 1000uf is 8x15mm (smaller than your guesses).

Thanks!

[PCBONEZ]

Re: Rubycon MCZ vs. SAMXON GC(M)?

If you are going to be replacing caps then you need to learn how to look up specs.
I gave you the hard ones to find.
Samxon and Rubycon specs are easy to find.

The simple fact that I told you the best spec on the old ones is comparable to ZL should have already told you that GC is overkill and MCZ is more overkill.
They'll -probably- work anyway but:
1: I personally wouldn't use them because -sometimes- overkill can lead to an unstable system and I don't like doing things twice.
2: You haven't told us what circuit they are going into on the board. "Some motherboard" isn't enough info to make a guess. There are many circuits on a mobo and they have different requirements. That's why photos are good.
.

[bw1]

Re: Rubycon MCZ vs. SAMXON GC(M)?

You make it sound as though I'm too lazy to look at the specs myself, but the truth is that I've looked up specs, but I don't know how lenient I can be, especially for the different parts/circuits I run into. Without asking those who know, there isn't a lot of option, unless you're going to order exact replacements for every cap you pull.

I've provided some pics below to help you get a better idea. Obviously, some specific help for this particular board would be appreciated, but I don't feel like I'm getting a very newcomer welcome here.

I'm not new to electronics at all, but I'm new to working directly with capacitors, and again, any helpful information is appreciated. Specifically, I'd like to know more about replacement compatibility for different circuits. A good example would be that you can use a higher voltage cap of the same capacitance, but not lower. Seems very common sense to me, but on the other hand I wouldn't have been surprised to learn that you *couldn't* do this.

====

The caps pictured are the 1500uf/6.3v and the 1000uf/10v, respectively.

Thank you.

[mockingbird]

Re: Rubycon MCZ vs. SAMXON GC(M)?

A good example would be that you can use a higher voltage cap of the same capacitance, but not lower. Seems very common sense to me, but on the other hand I wouldn't have been surprised to learn that you *couldn't* do this.

I've asked this before myself. It seems ostensible when you consider the extra longevity you'll get from the extra voltage rating which means the cap will withstand the extra heat for longer because it has to suppress less ripple. OTOH, to PCBONEZ' credit, he has described how capacitors work so one wonders if when the capacitor does not get enough stress placed on it, the 'self-healing' process doesn't work properly and the cap will fail prematurely, hypothetically.

I've not seen this personally. The screen I am using now was recapped with higher-rated Panasonic FCs and the screen is fantastic for a 2004 LCD that I have been using for the past half year 12 hours a day.

[yyonline]

Re: Rubycon MCZ vs. SAMXON GC(M)?

G-Luxon LZ series: 6.3V 1500uF 10x16 mm has the following specs:
Ripple: 1070 mA @ 100 @ 100 Khz
ESR: 0.055 ohms

For your replacements you have:

Rubycon MCZ 6.3V 1500uF 10x12.5 mm
Ripple: 1960mA @ 100Khz
ESR: 0.016 ohms

Samxon GC 6.3V 1500uF 8x20 mm
Ripple: 2220 mA @ 100Khz
ESR: 0.012 ohms


You want to have the same or more ripple.

You want to have the same or less ESR.

Of course, this is within reason. Drastically more ripple, or drastically less ESR may cause issues. It probably won't, but it has the potential to.

MCZ nearly doubles the ripple, and has 70% less ESR. This differs greatly from the specifications of the original. Will MCZ work? Probably. But the specs are very different, so there are no guarantees. If you don't want to spend the few dollars needed to get replacements closer to the originals - replace them and see what happens. Odds are it will work just fine - It looks the two LZ that are pictured are on the memory and chipset power circuits. Lowering ESR there shouldn't cause issues, I don't think. (Someone more knowlegeable feel free to correct me.) If it doesn't work after replacing them, get replacements that more closely match the original specs.

You mention there are 3 bloated 1500uF LZs, but I only see two in the photos...so I can only comment on the two I can see.

The GC are even further away form the original specs...use MCZ here if those are your only options.

SM are general purpose capacitors, they are not low ESR. When I run into GP caps on a motherboard, I will usually replace with an "entry level" low ESR cap like Panasonic FC. Will replacing with an ultra low ESR cap like MCZ or GC work? Again...probably, but no guarantees.

You're more than welcome on the boards. It's not a hostile reception at all, it's just that we just get the same questions on here a lot. It seems like the answers of some of our more knowledgeable members are on repeat answering the same questions again and again. You can learn the answers to a lot of your general questions by searching old posts. Sure, it takes awhile, but we all had to do that at first. Once you've done that, and have a basic understanding of the concepts, if you want information about your specific case, we need all the information we can get - that includes measuring the sizes, getting series info, taking detailed photos, etc. If we don't have enough info, we have to guess...give you some options...and you have to fill in the blanks.

[PCBONEZ]

Re: Rubycon MCZ vs. SAMXON GC(M)?

The self-healing process has to do with the voltage on the caps.

I was taught that the actual circuit voltage should be no less than 25% below the cap's rated voltage.
I have since learned that "that was in the 60's [and before]" and that caps are better now.
-
That said, the mechanism they were concerned about back then is still relevant to newer caps.
---
Ripple current strips the oxide layer off the plate.
DC voltage puts it back.
[[Much how electrolysis works to electroplate metals. ~ That's why they are called 'electrolytic' caps.]]
- Thereby the cap is continually 'healed'.
If you don't have enough voltage on the cap there may not be enough 'force' to move the oxide molecules
through the electrolyte back to where you want them faster than the ripple currents are stripping them off.
---
Even with modern caps I stick to the actual voltage being no less than 10% of the rated voltage,,,
,,, and that far only in a pinch..
That would be like a 50v cap on a 5v circuit.
I've never actually stretched it that far because caps with more realistic voltage ratings are easy to get,,
,, but that is where I set my limit.
.

[PCBONEZ]

Re: Rubycon MCZ vs. SAMXON GC(M)?

The whole circuit is not just a cap...
The ESR of the cap can affect other components just as the resistance of a resistor can.
In some cases the designer of a circuit may use the ESR of the cap in conjunction with other components to get something done.
[ESR is probably important in some oscillator/timing circuits for instance particularly if the clock is anywhere near 100kHz.]
If you change the ESR drastically without changing the values of the other components the result can be an unstable circuit.
-[ It would be rare for that situation to come up on a mobo,, but it -could-,, so it has to be said...]

Since I'm not willing to trace, draw-out, and do the math for every circuit that needs a new cap, I stick as close as possible to original specs.
-- And I'm not going to tell you it's safe to do something when I haven't done it myself and/or traced, drawn-out, and done the math for the circuit.

Me:
I would order something like WX, KZE, or ZL to replace those LZ.
I would order FC or LXZ [or similar] for the SMs.
- Because I don't like to do things twice....

Would the ones you have on-hand work.
Probably...
But -I- wouldn't use them.
.

[mockingbird]

Re: Rubycon MCZ vs. SAMXON GC(M)?

Even with modern caps I stick to the actual voltage being no less than 10% of the rated voltage

I thought so and this is what I practice as well. But 10 percent... Hmmm. 50V in place of 5V is a bit of a stretch, but what confuses me is that it is usually the microfarad rating which causes the increase in capacitor size, not the voltage.

[PCBONEZ]

Re: Rubycon MCZ vs. SAMXON GC(M)?

but what confuses me is that it is usually the microfarad rating which causes the increase in capacitor size, not the voltage.

True, but I'm not sure how that is relevant to what I was talking about.

[mockingbird]

Re: Rubycon MCZ vs. SAMXON GC(M)?

Probably something to do with my lack of understanding of the relativity of uF to voltage.

[bw1]

Re: Rubycon MCZ vs. SAMXON GC(M)?

I'm not able to find anything on the G-LUXON SM Series that matches the size I've got here -- only bigger. The one I'm looking for is 1000uf/10v SM 105*C 1128(M) roughly 8x16mm. The two SM's on the datasheet for 1000uf are bigger - 10 x 12.5/16.

Anyone have specs on the 8x16? (or 8x15... not exactly sure, but it's definitely 8)

[PCBONEZ]

Re: Rubycon MCZ vs. SAMXON GC(M)?

Not unusual to see sizes that aren't in the data sheet.
Happens with crap brands most often.

Here is how to cross reference...
Find an 8mm and a 10mm close to 1000uf/10v SM in the data sheet.
Take the ripple from those and find a good brand where the ripple matches for the same can size.
So:
SM - 1000uf/10v 10x16mm = 570 mA
SM - 470uf/10v 8x11mm = 305 mA

-
Keep in mind the uF and volts really doesn't matter here.
I'm just giving them so you know what cap I am talking about.
-

Nichicon VZ [closest to 1000uF/10v}
10x16mm[1000/16v] = 560 mA
8x11.5mm[1000/6.3v] = 390 mA
- - This is about as close to an exact match as you can get going from some junk brands to a good brand.

Personally I would use a slightly higher grade like LXY, LXZ, FC, PW or similar because that is what I keep on-hand...
... but the VZ should do fine.
.

[PCBONEZ]

Re: Rubycon MCZ vs. SAMXON GC(M)?

Probably something to do with my lack of understanding of the relativity of uF to voltage.

uF determines surface area. => Physical size.

The voltage rating has to do with how thick the oxide layer is across that surface.
Higher voltage rating -> thicker oxide layer.

The oxide layer is the actual dielectric. [Not the paper separator.]
If the oxide layer is too thick for the applied voltage it will act like an insulator and current [in this case ion current too] flow will stop.
.

[bw1]

Re: Rubycon MCZ vs. SAMXON GC(M)?

I've got an order started with Digikey, but I don't see any Nichicons in the 8x15-ish size available from them. Also, on the Nichicon SM datasheet, I don't see what the ESR rating is for the VZ you mentioned. (I'd like to keep to one order, from Digikey.)

Do you either have more info on the VZ's, or know of a compatible Panasonic or Rubycon?