CkRtech wrote:

Of course from one line of caps to another, that term is rather relative...

Anything with an ESR (Impedance) less than 0.090 ohms @100kHz would be considered low ESR.
That is not an official standard but rather a rule of thumb used by old Electronics Techs like me.
====
Regarding Al-Lytic caps in the physical size range used for main power filtering on mobos and in PSUs.
.. meaning 6mm to 12.5mm diameter caps in typical lengths...
@100kHz ESR effectively = Impedance (It's within about 0.002 ohms @100kHz)
Exceptions are rare in that size range.
Outside that range it's still pretty close until you get far away in size.
Thus datasheets for mobo sizes of cap intended for 100kHz typically use the terms Impedance and ESR interchangeably.
===

You are speaking to the actual inventor of the technique used to compare ESR Grades.
I did it back during the Capacitor Plague while frequenting badcaps.net.
It's now used universally by techs in the caps repair industry but I never get credit for it because I did it openly on BCN instead of publishing it in a book or web page or something.
Most people that use it have no idea who created it.
Unfortunately I'm doing a file server upgrade and my comparison charts and tables have been offloaded to cold (shelved) hard drives. Can't readily access them. (And if I don't quit OCD'ing my favorite topics in forums it's going to be July before I finish the new file server.)
I will outline how it's done in another post. - Come to think I may have done that already on Vogons.
Wasn't recent. I don't remember for sure. Have to check. I might have posted some of my charts already.

CkRtech wrote:

The replacements I used were much higher quality and low impedance, but not from a series intended to be low ESR. (Panasonic KA and Nichicon UPW/UHE)

Oh boy did you screw up.
You used low ESR caps.
As described above Low Impedance effectively IS low ESR.

PW and HE are both low ESR grades.
PW are entry level but not the absolute bottom tier.
HE are few steps higher grade (lower ESR). Pretty close to the middle of the low ESR range.

PW is an exact grade match Panasonic FC. So much so that you can interchange the datasheets.
Intel used FC and PW in VRM on Intel SE440BX boards. I think that's what is on SE440BX-2 as well.
PW were common in final output filters in PSUs. I still see them occasionally in newer PSUs.

HE is an exact grade match Chemicon KY.
HE and KY saw use as VRM caps in socket 370 and as Field caps into socket 775/771.
KY were common in final output filters in PSUs. I still see them there in newer gear fairly often.

Both PW and KY are very common in final output of non-PC related PSUs and in better LCD screens.

KA were a bad choice because they are 85°C, not because their ESR is in the GP cap range.
The tiny caps on older boards do not need to be low ESR.
[IIRC] Those started becoming low ESR mid socket 478 era. Even then only sometimes,
Because the cost difference is nil it's common practice to use low ESR replacements if you have to replace those.
This also means you only need to stock one kind for that cap size/uF/volts, and its easier get the quantity discounts.

You might think the 85°C is not that important for the small caps but I've seen many a mobo that was completely dead totally due to bad 85°C 4 and 5 mm caps. Intel (brand) boards in sockets 370-423-478 had a lot of problems that way. (They were using quality Nichicon caps, but the VR series which is rated for 85°C.) Those are memorable to me because they had so freakin' many 4 & 5 mm caps needing replaced and they were a pita because they used extra tiny solder pads for those caps.
Anyway, 85°C caps + heat = dead mobo.
.

SpectriaForce wrote:

I have soldered on old tube radio's, tv's and really old home computers in the past. I have just enough experience and the right […]
Show full quote

PCBONEZ wrote:

@SpectriaForce Some people have the knack. Others don't. (That's not meant to be derogatory. There are things I suck at.) […]
Show full quote

@SpectriaForce
Some people have the knack. Others don't. (That's not meant to be derogatory. There are things I suck at.)

That's another reason not to buy the fancy equipment for the first project.
.

I have soldered on old tube radio's, tv's and really old home computers in the past. I have just enough experience and the right equipment to do that. I have some idea of what I do and own a pretty good Weller soldering station. That having said, these more recent motherboards were soldered in a factory with a computer and temperature controlled precision soldering machine using lead free solder (various methods exist). Maybe some of you can solder on a motherboard like this, but it probably takes years of experience and advanced tools. There's a good chance you get cold solder joints if you just try something.

Also take a look over here: https://www.eevblog.com/forum/beginners/mothe … er-not-melting/

And here: https://www.google.nl/search?q=preheat+mother … 0&bih=882&dpr=1

I've been a DoD certified CET since 1981. (Retired from that.)
I have (literally) recapped 1000's of motherboards over the last 20 years.
I've also taught many people that have never soldered anything before how to do it.
Most of them do so well it becomes their hobby.
Most of them don't run out and buy fancy equipment.
- The RIGHT basic equipment works very well.
I would never recommend using hot air gear for radial caps. It's just the wrong tool for the job.
.
.
A have never liked Weller. Their irons always seem to run on the cold side.
Meaning less heat than should be there for the label plate watts.
Some people do just fine with them but I don't like them.
.

As long as you have a temp controlled solder station that is high enough wattage it is really easy to recap a motherboard like this.

I do them one-by-one.

Before I got my solder station, I used a crummy cheapo soldering iron. While it can be done, I do not recommend it at all. It is just way, way, way easier to use a good soldering station than it is to fight it with a cheapo.. which is what I used up until a few years ago.

The ones with digital displays and knobs to adjust the temperature are going to be much better, especially for a beginner as it will display the temp of the soldering iron and is super easy to adjust. I really dislike the idea of either just having a knob without a display or push button adjustments with a display.. oh yeah, I really want to push a button a bajillion times to get it to the temperature I want.

The first real soldering project I did was when I was about 12. My dad bought me a multimeter kit from a local electronics store and I built it all by myself. Pretty sure he still has it. I did do some soldering before that as well, but nothing as big as building something from scratch. The soldering iron that I built that multimeter with is still working over 25 years later.... but it too was a cheapo.

Definitely practice on junk beforehand. Remove every single last thing you can. By then you should have a pretty good feel for how to do stuff. If you are still not fully confident, find another junk item to practice on.

Make sure to get some solder-wick and some flux paste. The liquid stuff isn't as easy to use. Paste works so much better for regular soldering iron work.

A solder sucker is also a good idea for sucking the solder out of the holes.

Make sure to tin the tip of the soldering iron. Make sure to keep the tip clean or you will have a tough time. I wire brush and re-tin the tip very often when I am doing soldering work.
Don't just use the solder itself to tin the tip. Dip the solder in the flux paste before applying it to the tip and it will tin much better.

Also use flux paste on the joints your are soldering. You will get much cleaner/nicer looking joints and it will make the solder flow much easier and you won't end up with pokey bits from the solder not releasing from the soldering iron when you pull it away from the board.

Putting flux on the spots you are desoldering helps as well. You will get better heat transfer and it will tend to make you not near as likely to lift a pad or burn a trace.

The most important thing (to me) for a solder station (or even iron) is to get one with a ceramic heater element.
They have a shorter recovery time so there is less tip temp drop when you make contact with the joint.
They also last longer.
== That said, the older tech still works fine.

I would never fork over $$ for a station without first checking the availability and variety of replacement tips (and heaters).
.

cyclone3d wrote:

.. oh yeah, I really want to push a button a bajillion times to get it to the temperature I want.

HAHAHAHAHAHAHA
~ BEEN THERE! ~

PCBONEZ wrote:

Oh boy did you screw up. You used low ESR caps. As described above Low Impedance effectively IS low ESR. […]
Show full quote

CkRtech wrote:

The replacements I used were much higher quality and low impedance, but not from a series intended to be low ESR. (Panasonic KA and Nichicon UPW/UHE)

Oh boy did you screw up.
You used low ESR caps.
As described above Low Impedance effectively IS low ESR.

Was this a joke?

I found that old post.
Re: Recapping an abit bf6
That version of the chart was trimmed way down. A lot of series are missing.

CkRtech wrote:

PCBONEZ wrote:

Oh boy did you screw up. You used low ESR caps. As described above Low Impedance effectively IS low ESR. […]
Show full quote

CkRtech wrote:

The replacements I used were much higher quality and low impedance, but not from a series intended to be low ESR. (Panasonic KA and Nichicon UPW/UHE)

Oh boy did you screw up.
You used low ESR caps.
As described above Low Impedance effectively IS low ESR.

Was this a joke?

Since you had to ask... Yes, I was teasing. Being silly. Using sarcasm. However you want to put it.

PCBONEZ wrote:

Since you had to ask... Yes, I was teasing.

My bad. Long day. 😁

CkRtech wrote:

PCBONEZ wrote:

Since you had to ask... Yes, I was teasing.

My bad. Long day. 😁

No worries...

What for kind of solder, flux, solder tip and temperature setting do you guys use then? You make it look so easy but my experience is different.

Furthermore, does it matter mixing lead and leadfree solder? I’ve used my tips for both kinds..

By the way, the man that I’ve hired for my recent motherboard recap job used a device to preheat the motherboard on the specific place where the part had to be soldered.

Great info Bonez - you should defninitely write a book and pusblish it.

One more thing...
the dude i bought my last CRT from, said that if I put in a new cap which has not the exact same specs as the old one i should be looking at +20% capacity max and same or higher voltage.
Is this a good rule of thumb?

Just a quick reminder: higher frequency = more noticeable effects of esr(equivalent series resistance) and esl(equivalent series inductance).
Looking up the capacitor datasheet you can find out how it will perform on the specified task.
For a psu the switching speed will be about 40~400 kHz (more or less) so looking the datasheet will assure that the cap that you own will be suitable.

SpectriaForce wrote:

What for kind of solder, flux, solder tip and temperature setting do you guys use then? You make it look so easy but my experience is different.

"look so easy" = Are you watching me? = How you doing that? = Now I'm feeling paranoid! 🤣 🤣

My toolbox is a bit excessive for part-timers or a hobbyist and I know more than one winning combination.
It's late here and I don't want to go into all that now but I will tell you my go-to solder station for mobo work.
(So you can research it.)
I have several Hakko 926.
Bought 1 used just to try it. Loved it so much I bought a lot of 8 "parts" units. Got 6 working but a few are ugly as sin.
The Hakko 936 is identical except for the chassis and PCB shapes.
The iron/handle assemblies interchange between 926, 936, 937, 938. (Maybe more.)
Both (926/936) use marked temp dials and can be calibrated so it's accurate.
Neither has a digital display. Instead the LED flashes vs solid when it's heating vs up to temp. Works for me.
Wide variety of tips. Easy to find parts. Tips & parts are cheap. All the 900 series tips fit.
Replacement iron/cable assemblies are readily available and so cheap it's not worth buying new heaters.
There are actually 3 iron/handle options with the 907 (middle size) being best for mobo and general work.
I have the small (900) and large (908) handles too. Never use them.
There is also an SMD tweezers handle. I haven't tried that.
The next up digital display version (937) is a no go for me because you have to put in a digital security key to make it work.
The 938 is one chassis for two irons. (No key) I would rather have 1 for 1 in case one breaks.
I am aware of a reseller that has NIB 936's for $70 in both 220v and 110v versions. His tips are $1.95@.
His ads look scary like he's selling counterfeit. I've bought from him multiple times. It's all genuine parts so far.

SpectriaForce wrote:

Furthermore, does it matter mixing lead and leadfree solder? I’ve used my tips for both kinds..

Only matters to politicians, misguided environmentalists and some companies.
Mixing is fine and it makes removing lead-free easier.
I would never use lead-free unless required by a client.

SpectriaForce wrote:

By the way, the man that I’ve hired for my recent motherboard recap job used a device to preheat the motherboard on the specific place where the part had to be soldered.

If done properly it's fine. Done improperly SMD parts can slip off their pads.
I never felt a need for it and IMHO it would just makes the job take longer to finish.
.

amadeus777999 wrote:

Great info Bonez - you should defninitely write a book and pusblish it. […]
Show full quote

Great info Bonez - you should defninitely write a book and pusblish it.

One more thing...
the dude i bought my last CRT from, said that if I put in a new cap which has not the exact same specs as the old one i should be looking at +20% capacity max and same or higher voltage.
Is this a good rule of thumb?

Sorry. I missed this earlier.
He might not have explained it well but he knew what he was talking about.

===

The deal for uF is multi-part.
The steps (increments) of uF in standard cap sizes is intentionally close to 20%.
The most common uF tolerance is +/-20%. (Same as the increments.)
Circuits almost always have a uF tolerance wider than the caps. +/- 40% or 50% would not be unusual.
The uF value of caps always goes DOWN with age.

If the original is rated at 1000uF that means it's between 800uF and 1200uF.
The circuit's tolerances ~probably~ allow up to about 1500uF but you can't be 100% sure because they may have used a cap at the edge of the circuits range instead of in the middle.
If you replace that cap with a 1200uF the max it could be (+20%) is 1440uF.
- But they are very rarely that far off and even when they are off it's usually low, not high.
Even if it is that high the uF will go DOWN as it ages which just brings it closer to the original value.
[Going 20% lower is a bad idea though. That would drift further away from the original value.]

All of that said I would have to be in a real bind to actually do it.
= Because I don't know for certain if the original was at the edge or in the middle of the circuits acceptable range.
That and I haven't run into any 'unfindable' caps for long time.

====

As for the voltage rating. That is not a target, it's a limit.
A 10v cap is good for anything less than 10v.
A 16v cap is good for anything less than 16v.
.... and so on.
Going to a higher voltage limit is just fine so long as you don't get ridiculous.
When I went through ET school they taught us that the actual circuit voltage should be no less than 20% of the cap's rated voltage or the oxide layer wouldn't rebuild itself. [It may have been 30%. That was a long time ago.]
But they don't say that anymore anywhere.
I suspect the info was dated when I learned it and that it doesn't apply to modern cap technology.
.

.legaCy wrote:

Just a quick reminder: higher frequency = more noticeable effects of esr(equivalent series resistance) and esl(equivalent series inductance).
Looking up the capacitor datasheet you can find out how it will perform on the specified task.
For a psu the switching speed will be about 40~400 kHz (more or less) so looking the datasheet will assure that the cap that you own will be suitable.

You are not wrong.
That would really only matter to someone that is designing or modifying a PSU such that the switching frequency changed.
I never got into that personally but I know a few people that do that.
I just repair them as they are.
Using lower ESR output caps doesn't change the frequency of the switchers and the ESR is lower no matter what the frequency is.
.
Note to the peanut gallery.
The switching frequency is controlled by an IC chip that adjusts it based on the -DC- output voltage.
.

PCBONEZ wrote:

You are not wrong. That would really only matter to someone that is designing or modifying a PSU such that the switching frequen […]
Show full quote

You are not wrong.
That would really only matter to someone that is designing or modifying a PSU such that the switching frequency changed.
I never got into that personally but I know a few people that do that.
I just repair them as they are.
Using lower ESR output caps doesn't change the frequency of the switchers and the ESR is lower no matter what the frequency is.
.
Note to the peanut gallery.
The switching frequency is controlled by an IC chip that adjusts it based on the -DC- output voltage.
.

Within the context of a switch mode converter repair it's not a significant consideration, but it becomes exceedingly important for RF applications.

Not rebuilding radios here.
RF + Mobo = bad bad bad
It's EMI to a mobo.

Since we went there....
A good share of those tiny caps, especially those near ports that connect to external wires, are there to filter out RF EMI.
Some by the add-in slots are for the same in the event that the add-in card connects to external wires.
.

PCBONEZ wrote:

You are not wrong. That would really only matter to someone that is designing or modifying a PSU such that the switching frequen […]
Show full quote

.legaCy wrote:

Just a quick reminder: higher frequency = more noticeable effects of esr(equivalent series resistance) and esl(equivalent series inductance).
Looking up the capacitor datasheet you can find out how it will perform on the specified task.
For a psu the switching speed will be about 40~400 kHz (more or less) so looking the datasheet will assure that the cap that you own will be suitable.

You are not wrong.
That would really only matter to someone that is designing or modifying a PSU such that the switching frequency changed.
I never got into that personally but I know a few people that do that.
I just repair them as they are.
Using lower ESR output caps doesn't change the frequency of the switchers and the ESR is lower no matter what the frequency is.
.
Note to the peanut gallery.
The switching frequency is controlled by an IC chip that adjusts it based on the -DC- output voltage.
.

I used the smps as an example, on decoupling caps the lower esr caps is a *must have*
On one old keyboard (musical instrument) that i have the original power supply wnet missing and looking for a replacement i got a psu for guitar pedals(same voltage, same connector, same polarity) but only 300mA rated the keyboard kind of worked but it was unstable, it was a linear psu, the step down transformer seemed that could handle more than 300mA, the full bridge rectifier was rated for 2A, and the power regulator didnt had to drop a lot of power , the traces were thick enough so i figured out that it was probably ripple by the increased load, i added another 2200uf in parallel at the output of the regulator and guess what? Problem solved, now lo esr caps.
But on a motherboard recap, low esr for cpu vrm is important.

Should have done the Cap-Map.......
Identifiying burnt transistor on a Intel SE440BX-2 board
.