HM or HN series?

The only advice i can give is to make sure you use a high power soldering station with a fat tip. These are 4-layer boards and any connection that's near a ground plane can be a pain in the ass to work with.

You will probably need to replace it with polymer capacitors.

These ultra low ESR aluminium electrolytic capacitors are no longer used or manufactured.

gdjacobs wrote:

HM or HN series?

What is this?

canthearu wrote:

You will probably need to replace it with polymer capacitors.

These ultra low ESR aluminium electrolytic capacitors are no longer used or manufactured.

Can you go into more depth about this? and give me an idea what I would need, the specifications I listed were the only things listed on the cap, I thought that was all that was needed.

Thanks for the information so far.

gryffinwings wrote:

gdjacobs wrote:

HM or HN series?

What is this?

I can actually see from the second photo that they're HM series caps. HM and HN are two series of Nichicon capacitors which are known to be problematic. There was a manufacturing problem for caps made between 2001 and 2004. Caps from late 2005 and newer should be fine.

Caps vary a little in how they're marked, but usually there will be a temperature spec, nominal capacitance, withstand voltage, series code, and date code. The series code is 2-3 characters (two letters for Nichicon) and the date code is a letter followed by the year (two digits) and the week (two digits) of manufacture.

You have several options for replacement with electrolytics:
https://www.badcaps.net/forum/showthread.php?p=82953

If you want exact specs for ripple and ESR, start with capacitance, voltage, casing diameter, and casing height and look up the numbers on the chart:
http://www.nichicon.co.jp/english/products/pdf/e-hm.pdf

If you want to replace the HM caps with polymers, as a rule of thumb, use polymer caps with approximately half the nominal capacitance.

gdjacobs wrote:

I can actually see from the second photo that they're HM series caps. HM and HN are two series of Nichicon capacitors which are […]
Show full quote

gryffinwings wrote:

gdjacobs wrote:

HM or HN series?

What is this?

I can actually see from the second photo that they're HM series caps. HM and HN are two series of Nichicon capacitors which are known to be problematic. There was a manufacturing problem for caps made between 2001 and 2004. Caps from late 2005 and newer should be fine.

Caps vary a little in how they're marked, but usually there will be a temperature spec, nominal capacitance, withstand voltage, series code, and date code. The series code is 2-3 characters (two letters for Nichicon) and the date code is a letter followed by the year (two digits) and the week (two digits) of manufacture.

You have several options for replacement with electrolytics:
https://www.badcaps.net/forum/showthread.php?p=82953

If you want exact specs for ripple and ESR, start with capacitance, voltage, casing diameter, and casing height and look up the numbers on the chart:
http://www.nichicon.co.jp/english/products/pdf/e-hm.pdf

If you want to replace the HM caps with polymers, as a rule of thumb, use polymer caps with approximately half the nominal capacitance.

Actually that post help me do some more identifying of the capacitors, I have found these capacitors marked H0139, so definitely really old and makes sense that these are not very good caps, also you are correct, these are labeled HM(M), not sure what that even means though. I think I would prefer to stick with electrolytic capacitors.

Does the physical height of the capacitor matter at all. The capacitors on my motherboard measure 10mm x 27mm.

I was thinking about these:

http://www.badcaps.net/store/product_info.php … &products_id=30

But found they are discontinued and these are recommended instead:
http://www.badcaps.net/store/product_info.php … &products_id=33

What should I do?

Rubycon LZH capacitors might work, but they are not motherboard capacitors or perform to the same degree as motherboard low esr capacitors like the HM series.

Virtually none of these motherboard grade ultra-low-esr capacitors exist on the open market, with any you see now being either extremely old stock or outright counterfeits.

Up to you, but I'd be using 1500uF polymer caps instead.

canthearu wrote:

Rubycon LZH capacitors might work, but they are not motherboard capacitors or perform to the same degree as motherboard low esr capacitors like the HM series.

Virtually none of these motherboard grade ultra-low-esr capacitors exist on the open market, with any you see now being either extremely old stock or outright counterfeits.

Up to you, but I'd be using 1500uF polymer caps instead.

Can you explain how a lower capacitance polymer cap would work better or actually work in the first place, I'm really interested in learning more about this stuff. How can you get away with more than half the lower capacitance? Would you be able to recommend me a substitute?

How about any of these:

https://www.digikey.com/products/en/capacitor … e=1&pageSize=25

And thank you for the information so far, very helpful.

https://www.digikey.com/product-detail/en/nic … 4393-ND/2549489

These should do the job.

The reason why you don't need as much capacitance is that in these circuits, it the capacitors are not chosen for their capacity, more for their ESR and ripple current handling, Larger caps have lower ESR and higher ripple current handling, so you stick the largest you can when you are designing a high performance power circuit. Polymer capacitors have much superior ESR and ripple current to standard electrolytic, so you don't need nearly as large a capacitor to get the required values for the power circuit.

Edit: Change to PLG caps. Slightly higher ESR, a little bit of a closer match to the original HM caps

True. There is discussion about polymod rules here: https://badcaps.net/forum/showthread.php?t=12932

canthearu wrote:

https://www.digikey.com/product-detail/en/nic … 4393-ND/2549489 […]
Show full quote

https://www.digikey.com/product-detail/en/nic … 4393-ND/2549489

These should do the job.

The reason why you don't need as much capacitance is that in these circuits, it the capacitors are not chosen for their capacity, more for their ESR and ripple current handling, Larger caps have lower ESR and higher ripple current handling, so you stick the largest you can when you are designing a high performance power circuit. Polymer capacitors have much superior ESR and ripple current to standard electrolytic, so you don't need nearly as large a capacitor to get the required values for the power circuit.

Edit: Change to PLG caps. Slightly higher ESR, a little bit of a closer match to the original HM caps

So even if reduced by more than half of original capacitance, these should do the job even better than the original HM caps? Interesting, because the basic rule as I understand is to reduce by have which is 1650, which I am not sure would exist.

zyga64 wrote:

True. There is discussion about polymod rules here: https://badcaps.net/forum/showthread.php?t=12932

I've started reading this thread, very interesting read.

Yup. It's a rule of thumb, but you still want ~12 mOhms ESR and ~2.5A ripple (minimum) to replace those caps. So, make sure you check the charts when you buy the replacements.

gdjacobs wrote:

Yup. It's a rule of thumb, but you still want ~12 mOhms ESR and ~2.5A ripple (minimum) to replace those caps. So, make sure you check the charts when you buy the replacements.

So no more than 12 mOhms and no less than 2.5 amps ripple. Correct?

There might be some wiggle room, but those are your targets. Max ripple can be anything over the spec. ESR is circuit dependent but in VRM applications can usually be less than original due to simpler feedback.

gdjacobs wrote:

There might be some wiggle room, but those are your targets. Max ripple can be anything over the spec. ESR is circuit dependent but in VRM applications can usually be less than original due to simpler feedback.

Thanks a lot, I appreciate the information, I've learned quite a bit.

gryffinwings wrote:

gdjacobs wrote:

There might be some wiggle room, but those are your targets. Max ripple can be anything over the spec. ESR is circuit dependent but in VRM applications can usually be less than original due to simpler feedback.

Thanks a lot, I appreciate the information, I've learned quite a bit.

NP. Please, let us know how it goes.

gdjacobs wrote:

gryffinwings wrote:

gdjacobs wrote:

There might be some wiggle room, but those are your targets. Max ripple can be anything over the spec. ESR is circuit dependent but in VRM applications can usually be less than original due to simpler feedback.

Thanks a lot, I appreciate the information, I've learned quite a bit.

NP. Please, let us know how it goes.

Will do, but I'm just deciding on the caps, I'm not sure I'm comfortable with reducing the capacity by more than half of original capacitance, I think I will go with the high 1800uF instead, link below:

1800uF
6.3v
ESR - 8 mOhm
Ripple Current - 6.6A @ 100kHz

https://www.digikey.com/product-detail/en/nic … 3109-ND/1662495

You can go with these 16V rated UCC polymers.
https://www.digikey.com/product-detail/en/uni … 4315-ND/5824538

They have a little bit more capacitance and are cheaper than the Nichicons.

gdjacobs wrote:

You can go with these 16V rated UCC polymers.
https://www.digikey.com/product-detail/en/uni … 4315-ND/5824538

They have a little bit more capacitance and are cheaper than the Nichicons.

Also rated for 20000 hours instead of 2000 hours that the Nichicons are rated, 16v huh, I think I remember being ok to go higher voltage just not lower voltage. ESR is just as low and the ripple current capability is even higher. Sounds like it's good to go.

Have you used these before?