Swapping japanese/generic electrolythics with solid-states

[Snayperskaya]

Would it have any bad consequences, considering both to be of the same ESR rating, size and capacitance? On overall circuits, not only motherboards and VGAs.

[Per Hansson]

Re: Swapping japanese/generic electrolythics with solid-states

No, but it will never have the same ESR rating in equal sizes, the solid polymer caps always have much lower ESR.
And in some circuits this can cause problems.
In others it can be a benefit, depends on the circuit...

[divy]

Re: Swapping japanese/generic electrolythics with solid-states

Hi my TV power supply I have change capas 220uf 450vdc which capas to good to buy https://m.ebay.co.uk/itm/MAL21594722...QAAOSwEK9UCJ9R https://m.ebay.co.uk/itm/CAP-ALU-ELE...&ul_noapp=true

[Sparkey55]

Re: Swapping japanese/generic electrolythics with solid-states

Hi my TV power supply I have change capas 220uf 450vdc which capas to good to buy https://m.ebay.co.uk/itm/MAL21594722...QAAOSwEK9UCJ9R https://m.ebay.co.uk/itm/CAP-ALU-ELE...&ul_noapp=true

And you would actually buy from someone with 104 NEGATIVE and 126 NUETRAL FEEDBACK RATINGS???????????????????

[Agent24]

Re: Swapping japanese/generic electrolythics with solid-states

And you would actually buy from someone with 104 NEGATIVE and 126 NUETRAL FEEDBACK RATINGS???????????????????

Out of 400,000 feedbacks, that is not a lot, and over half of those are probably from idiot buyers who didn't read the description properly, or did something else stupid.

That said, I wouldn't go on eBay either, but that said, if all he wants are 450v mains filter capacitors and they're not in an Active PFC circuit, then even the cheap multicomp ones would probably be fine (I'd prefer the Vishay), IF they are not fakes or factory seconds etc - which on eBay you never know!

Much better to buy from a real electronics supplier, although some decent places (like Dipmicro) are on eBay and are fairly trustworthy.


Now, if divy had filled in his user-profile properly to include his actual country, we might be able to recommend a decent supplier near to where he lives...

[Sparkey55]

Re: Swapping japanese/generic electrolythics with solid-states

Out of 400,000 feedbacks, that is not a lot, and over half of those are probably from idiot buyers who didn't read the description properly, or did something else stupid.

That said, I wouldn't go on eBay either, but that said, if all he wants are 450v mains filter capacitors and they're not in an Active PFC circuit, then even the cheap multicomp ones would probably be fine (I'd prefer the Vishay), IF they are not fakes or factory seconds etc - which on eBay you never know!

Much better to buy from a real electronics supplier, although some decent places (like Dipmicro) are on eBay and are fairly trustworthy.


Now, if divy had filled in his user-profile properly to include his actual country, we might be able to recommend a decent supplier near to where he lives...

Sorry but I am not buying that excuse. Any seller on ebay that can consistantly have that much negative feedback is bad business practices and has not and will never learn from it. In other words they just dont give a damn about customers. It is people like you that give these type of scum sellers free range to do what ever they want. Ebay has turned into crap from it.

[Agent24]

Re: Swapping japanese/generic electrolythics with solid-states

Really? It's my fault that eBay is full of low-quality items and shifty sellers? I'm not buying THAT. Did you even read the feedbacks in question? Many of the negative ones were complaints about the parcel never arriving (Not seller's fault, parcels get lost or stolen, it happens), and a lot of the neutral ones were about the parcels being delivered slowly or the cost being too expensive. None of those are the seller's fault. Postal service mistakes, thieves, or people not reading the auction are to blame.

Think about that last one. A person orders an item, pays for it, knowing full well what the purchase price is, then leaves bad or neutral feedback complaining that the cost was too high - those are the kinds of people who make eBay stupid... If you can't read, don't bid!

And, be realistic. No shop will ever attain 100% positive feedback, because there will always be an idiot buyer somewhere who will complain no matter what. The postal service damaged their item, you give them a full refund, and they'll still complain because they were "inconvenienced".


Maybe I'm biased, because I've sold a lot of second-hand gear etc I didn't want\need online, but if you're complaining that eBay is full of crappy sellers (I'm not saying it's not - it is), then you've never tried selling anything, because as soon as you do, you'll find out just how many crappy buyers there are too.

I've had them all, the idiots who bid and never contact, the ones who never pay, the ones who change their mind 20 times and then decide they don't want the item, the ones that break it and try to blame you, the ones that say it was stolen when the tracking shows they signed for it, the ones that re-arrange 5 different pickup days over 3 weeks but never show up.... etc.


I'm not defending bad sellers, but there's always two sides, and the buyer is not always the innocent victim. When you have a high-volume seller attracting many buyers (and at the low price-point, there's going to be a higher percentage of cheapskates and idiots to start with), they are going to get a proportionally higher number of unfair bad feedbacks. They'll also get a higher amount of well-deserved bad ones too, because they are indeed selling cheap junk.

BUT, bad sellers don't have free reign to do whatever they want - because it's YOUR decision to bid. They don't have a gun to your head, they're not forcing you to buy their trash.

And, nowhere in my previous post did I recommend that anyone should buy parts from such a seller!

[rugger]

Re: Swapping japanese/generic electrolythics with solid-states

You won't find a solid polymer capacitor at 450V.

http://br.rsdelivers.com/product/rub...CE%BCf/7672832

This capacitor would be a good match if it fits (might not, pretty tall package)

[Snayperskaya]

Re: Swapping japanese/generic electrolythics with solid-states

I've got another board with faulty NCC KZGs... What would be my best choice for replacing those?

[mockingbird]

Re: Swapping japanese/generic electrolythics with solid-states

I've got another board with faulty NCC KZGs... What would be my best choice for replacing those?

If you want exact equivalents, look into Suncon WG. It is still being produced and sold, AFAIK.

[Snayperskaya]

Re: Swapping japanese/generic electrolythics with solid-states

If you want exact equivalents, look into Suncon WG. It is still being produced and sold, AFAIK.

Thanks, but I'd like to have a superior model, actually. I'd rather pay a little more into quality/lifespan.

[mockingbird]

Re: Swapping japanese/generic electrolythics with solid-states

For the motherboard, you can safely use polymers to replace non-VRM caps. I recommend getting a bunch of 6.3V 820uF 8mm polymers for this purpose.

For the VRM, it depends. Which board do you have?

[Snayperskaya]

Re: Swapping japanese/generic electrolythics with solid-states

I've checked (eye only) two boards ATM:

ASUS A8V:

NCC KZG:
15x 820uF @ 6.3V

INTEL S5520HC:

NICHICON:
10x 470uf @ 10V

NCC KMG:
2X 470uF @ 16V

Apparently the failed/failing caps aren't responsible for the VRM. The Asus have some Panasonic and Rubycons there while the Intel have polymers.

The Asus got some bulged ones (so I think replacing all of them would be wiser) and the Intel have caps sizzling when operating at a high-power level (2x 130W CPUs).

[mariushm]

Re: Swapping japanese/generic electrolythics with solid-states

For both of these boards, they probably failed due to simply being a problem series (in the case of KZG) or maybe because the power supply had a bad 5v / 5v stand by circuit.

On the green board ... all but the one above the chipset and the one near the 24 pin atx connector are used as additional filtering for 5v in PCI slots and USB ports.

Not really worth spending extra money on polymer capacitors to replace those.. you can use any 820uF 6.3v .. 16v electrolytic capacitors with low esr ... can also be higher, 1000-1200uF would be safe to use. You can go up in voltage (if you have for example 820uF 16v capacitors left from another project for example you can use those), just make sure they're not too tall to block cards ... for example for the capacitors near the pci slots.

You can use any low ESR electrolytic capacitors .. panasonic fm, fr, nichicon hm, hn, nippon chemi con kze, ky , rubycon zlg , yxg, zlh, zlj

You can replace the 10v rated Nichicon capacitors with 6.3v rated capacitors, those are also used with 5v or less...

The second power supply also has some capacitors that only filter usb ports (they're near the usb connectors or headers) but that one has more of those near the cpu socket and memory sticks, probably that board uses a dc-dc converter that uses 5v to 1.8v or whatever the voltage used for memory is.
The first one uses 12v to produce the lower voltages for memory, because there's more sticks and makes more sense to use 12v for conversion.

[mockingbird]

Re: Swapping japanese/generic electrolythics with solid-states

I've checked (eye only) two boards ATM:

ASUS A8V:

NCC KZG:
15x 820uF @ 6.3V

INTEL S5520HC:

NICHICON:
10x 470uf @ 10V

NCC KMG:
2X 470uF @ 16V

Apparently the failed/failing caps aren't responsible for the VRM. The Asus have some Panasonic and Rubycons there while the Intel have polymers.

The Asus got some bulged ones (so I think replacing all of them would be wiser) and the Intel have caps sizzling when operating at a high-power level (2x 130W CPUs).

For the Asus board:

Re-cap the VRM. Those caps are approaching almost 15 years now. I have opened up 15 year old Panasonic FL/FJ. They do not look pretty inside (electrolyte concentrates on the bottom - for lack of a better way of putting it at the moment). The VRM high and low are 8mm. For the high use 16V 270uF-470uF polymers or better, and for the low use 2.5v-4v 680uF or better (You can get 2.5v 1500uF polymers in 8mm if you're prepared to pay).

For the Intel board:

Just replace the electrolytics with polymer equivalents (and do the same for the Asus - i.e. replace all the non-VRM caps).

If you want to save money, use something like Panasonic FR or equivalent for these non-VRM caps.

If you need help finding caps, let me know.

[Snayperskaya]

Re: Swapping japanese/generic electrolythics with solid-states

For both of these boards, they probably failed due to simply being a problem series (in the case of KZG) or maybe because the power supply had a bad 5v / 5v stand by circuit.

On the green board ... all but the one above the chipset and the one near the 24 pin atx connector are used as additional filtering for 5v in PCI slots and USB ports.

Not really worth spending extra money on polymer capacitors to replace those.. you can use any 820uF 6.3v .. 16v electrolytic capacitors with low esr ... can also be higher, 1000-1200uF would be safe to use. You can go up in voltage (if you have for example 820uF 16v capacitors left from another project for example you can use those), just make sure they're not too tall to block cards ... for example for the capacitors near the pci slots.

You can use any low ESR electrolytic capacitors .. panasonic fm, fr, nichicon hm, hn, nippon chemi con kze, ky , rubycon zlg , yxg, zlh, zlj

You can replace the 10v rated Nichicon capacitors with 6.3v rated capacitors, those are also used with 5v or less...

The second power supply also has some capacitors that only filter usb ports (they're near the usb connectors or headers) but that one has more of those near the cpu socket and memory sticks, probably that board uses a dc-dc converter that uses 5v to 1.8v or whatever the voltage used for memory is.
The first one uses 12v to produce the lower voltages for memory, because there's more sticks and makes more sense to use 12v for conversion.

Thanks for the tips. The green PCB is the Intel one, it's a server board that handles quite some load into it. I haven't seen any bulged (top) on it, just some sizzling. The board operates OK, I just want to recap BEFORE anything goes bad since I've heard the sizzling noise when using high power CPUs.

For the Asus board:

Re-cap the VRM. Those caps are approaching almost 15 years now. I have opened up 15 year old Panasonic FL/FJ. They do not look pretty inside (electrolyte concentrates on the bottom - for lack of a better way of putting it at the moment). The VRM high and low are 8mm. For the high use 16V 270uF-470uF polymers or better, and for the low use 2.5v-4v 680uF or better (You can get 2.5v 1500uF polymers in 8mm if you're prepared to pay).

For the Intel board:

Just replace the electrolytics with polymer equivalents (and do the same for the Asus - i.e. replace all the non-VRM caps).

If you want to save money, use something like Panasonic FR or equivalent for these non-VRM caps.

If you need help finding caps, let me know.

Thanks. So I should go for a full recap on the Asus. I'll check the specs on the VRM caps (Rubys next to the CPU side, Panas to the IO side) later. What bothered me is that the board hadn't a single cap bulging or leaking (from the top) last time I've played with it, a year and half ago. Looks like bad caps can go poof even when not used at all.

[momaka]

Re: Swapping japanese/generic electrolythics with solid-states

You can use any low ESR electrolytic capacitors .. panasonic fm, fr, nichicon hm, hn, nippon chemi con kze, ky , rubycon zlg , yxg, zlh, zlj

+1

For the non-VRM caps of the ASUS A8V board, none of the caps need to have very good ESR spec, despite the originals being KZG. This is because older ASUS boards from that vintage use linear regulators pretty much throughout the board. Thus ultra-low ESR caps are not a requirement at all.

probably that board uses a dc-dc converter that uses 5v to 1.8v or whatever the voltage used for memory is.

Actually the 3.3V rail. Many socket 775 and 939/AM2 boards use the 3.3V rail as a supply to the RAM VRM circuit. And with older ASUS and AsRock like this, typically the RAM is linearly-regulated... though that applies more towards board with DDR. For DDR2, they started using buck regulators for better efficiency (and they then even used part of the DDR2 voltage to generate the chipset Vcc).

The green PCB is the Intel one, it's a server board that handles quite some load into it. I haven't seen any bulged (top) on it, just some sizzling. The board operates OK, I just want to recap BEFORE anything goes bad since I've heard the sizzling noise when using high power CPUs.

That noise might be coil whine too.

What bothered me is that the board hadn't a single cap bulging or leaking (from the top) last time I've played with it, a year and half ago. Looks like bad caps can go poof even when not used at all.

That's Chemicon KZG for ya! (Same goes for KZJ too, tough it happens far less often.) That said, newer KZG caps (made past 2008/2009) haven't had too many problems, so looks like Chemicon might have actually fixed the issue.

[ChaosLegionnaire]

Re: Swapping japanese/generic electrolythics with solid-states

for the asus a8v, i think u missed a cap near the cmos battery right by the mosfet. i have this exact same board and the cap and mosfet there actually regulates the 5vsb supply for the board. i recommend replacing that as well so u have 16 kzg caps to replace not 15.

[Uranium-235]

Re: Swapping japanese/generic electrolythics with solid-states

I had an A8N, and I believe someone I knew had an A8V. Just remember one thing. These Asus boards have the capacitor half-moon mark on the POSITIVE side of the capacitor.

[mariushm]

Re: Swapping japanese/generic electrolythics with solid-states

Yeah, Asus and Asrock (for some time, I'm not sure they still do it) ... used the reverse notation for capacitors where the filled side of the circle represented the positive.

Good to keep in mind as HP if I remember correctly often uses OEM motherboards made by Asus