IBM 15" LCD caps

[pflowers]

I found a thread on here with the same board for the monitor and I was wondering if I could get a suggestion for replacement caps for the suggested SAMXON used in that repair (Digi-Key doesn't carry them)

old cap Hermei 470uf 16v LT
SAMXON replacement 1000uf 16v GF

old cap Hermei 1000uf 16v LE
SAMXON replacement 1000uf 16v GF

so I need caps sold at Digi-Key for the 2 SAMXON caps

the board number is 715L1034-1

many thanks,
Paul

[sofTest]

Re: IBM 15" LCD caps

I wouldn't change the uF of the first cap, unless you are really sure of what you are doing, and you might also have trouble with the increased physical size. I would also normally used Panasonic FMs for both sizes, but since DigiKey is out of stock for the 16v/1000uF FM, the United Chemi-Con (UCC) KZE would be an alternative.

http://search.digikey.com/scripts/Dk...name=P12376-ND
http://search.digikey.com/scripts/Dk...me=565-1664-ND

[pflowers]

Re: IBM 15" LCD caps

The increased uf on the 470 was in a post I found on here.....I didn't understand that myself.....but from the looks of the thread there have been dozens of LCD repaired using those caps.......I would rather use a 470 myself (feel safer that way).....but I don't have an account with mouser so I need to stick to Digi-Key stock.

That UCC might be pushing the limits on length...it leaves me about 3 or 4 mm clearance....worth a shot I guess....the Hermei is a 10X15mm

Paul

[pflowers]

Re: IBM 15" LCD caps

heres a link to that old thread

https://www.badcaps.net/forum/showth...+16v+LT+series

Paul

[sofTest]

Re: IBM 15" LCD caps

It might not hurt (in this case) to increase from 470uF to 1000uF, but any benefits are doubtful. Also, as you are ordering from DigiKey, if increasing uF, you have to use the UCC capacitor I linked to, and with that also the increased size. In general, it will not hurt to increase the rated voltage of a capacitor, but the uF should be left as original.

As for the LE, the current datasheet specifies it as a 10x21 mm, but that might have changed, or they are made to order. Anyway, as long there is room, there should not be any problem with the increased hight. If DigiKey had Panasonic FM 16v/1000uF in stock, they would also be 10x20 mm. Also within a series, a bigger physical size lowers impedance and increases ripple, both a benefit.

Both the capacitors I linked to, have better specification than the originals, and are also of better quality.

[pflowers]

Re: IBM 15" LCD caps

Thanks Soft......BTW heres the board I am working on........it looks slightly different than the one mentiond in that other post



Paul

[PCBONEZ]

Re: IBM 15" LCD caps

I think he was trying to match Ripple/ESR out of stock on-hand at the time.
Those Hermei's have fairly decent ratings, especially the LT.
I haven't traced the circuit but in some apps ESR is much more important than uF.
Japlytic is very experienced. My guess is he got it right for what he was doing.

About your caps:

Hermei 470uf 16v LT -- 8mm,,, Ripple/ESR = 920/.050
Hermei 1000uf 16v LE -- 10mm,,, Ripple/ESR = 1520/.030

Of Digikey's stocked brand/series:

Hermei 470uf 16v LT -->
KZE 470 16v - 8mm,,, 995/.056 (Close but ESR is a little higher)
Best 8mm match without going under is an FM at 1240/.041.
http://search.digikey.com/scripts/Dk...name=P12376-ND

Hermei 1000uf 16v LE -->
Again KZE is the first/lowest grade even close at digikey but this time it's an upgrade at 1820/.023.
http://search.digikey.com/scripts/Dk...me=565-1664-ND

.

[pflowers]

Re: IBM 15" LCD caps

so the Pan FM would be an upgrade over that LT then right?

again the cost here is minute if compared to what we were being charged (150.00 per monitor) and I have a pile of these in the back room.........man I need a raise or something

Paul

[PCBONEZ]

Re: IBM 15" LCD caps

so the Pan FM would be an upgrade over that LT then right?

Yes ~~ But digikey is out of stock right now on that one.
http://search.digikey.com/scripts/Dk...name=P12366-ND


Higher number for Ripple is better.
Lower number for ESR is better.

Filter caps are used to pass noise/ripple voltages to ground.
- Like a drain, bleed-off, short ripple to ground.

Higher rating for ripple means it can pass more ripple without over heating.
- Similar to a max current rating.

[Resistance is not the correct term but gets the idea across.]
ESR is like the caps 'resistance' to ripple current.
Lower ESR means the cap is better at passing ripple to ground.

.

[pflowers]

Re: IBM 15" LCD caps

sorry PC I ment for the 470uf the Pan FM (P12376-ND) is slightly better on both ripple and ESR and I was planning on getting the UCC 1000uf KZE for the other.

Paul

[PCBONEZ]

Re: IBM 15" LCD caps

Sorry about what?
I'm just trying to 'learn ya' so you can figure this stuff out on your own.
[And so don't have to ask. And so you can help others.]

This is the part to burn into your brain.
~~~~~~~~~

Higher number for Ripple is better.
Lower number for ESR is better.

Filter caps are used to pass/short ripple voltages to ground.

Higher Ripple rating means it can pass more ripple without over heating.
- Similar to a max current or a watts rating.

Lower ESR means it's easier for ripple to go through the cap.
- Similar to a resistance value.

~~~~~~~~~


Once you got that down the only thing to know is how to find and read data sheets.
-
Finding can be impossible or crazy with some crap brand originals but most of the time it's at least do-able.
If not then go by what's typical in the application which can be asked or learned by experience.


One thing to watch in reading tables is that they use standard values.
-
Ripple should be: mA at 100kHz and at max rated temp. [Usually 105°C or 85°C]
- Some brands cheat and use 120kHz. [And some list both which is handy.]

ESR [a.k.a. Impedance in some tables] should be either
- mΩ at 100kHz and 20°C -or- Ω at 100kHz and 20°C
Some brands use a non-standard temp or frequency or both.

For instance:
Our buddy Fuhjyyu takes ESR at 25°C.
Elite takes Ripple at 120kHz.
Using those conditions inflates the numbers and makes the ratings 'look' better.
-
In a way that works out good because if you match their inflated bogus number in a chart for replacements that uses the correct standard conditions it's actually an upgrade.

I uploaded MBZ table to show how standard looks for comparison.