Mounting caps slightly off the board

[shovenose]

If you are installing a capacitor not completely flush to the board, for example if you have a taller cap and must installit sideways with leads bent at a right angle THAT IS OK!!! Certain members of this forum (im not going to say names) (including pcbonez) said that it raises the esr to lenghtn the leads. But it DOES NOT!!! Topcat was awesome enough to really test it out (i PMed him) and testing the esr right at the cap apparently registered no different on an esr meter than testing at the end of the leads.
Just be careful not to strain wherw the leads go into the capacitor becuase it may damage the seal. Also be aware that the leads are exposed thus more likely of them to be shorted by a foreign obkect or another component

[c_hegge]

Re: Mounting caps slightly off the board

I have done that in PSUs before when replacing 10mm Fuhjyyu caps with 12.5mm Panny FC. It works fine. To prevent short circuits, I usually use either heatshrink tubing or the plastic from a wire.

[Krankshaft]

Re: Mounting caps slightly off the board

Well lengthening any conductor will increase it's ESR to a degree.

The amount depends on wire gauge (thickness) and the resistance vs length of the metal or metal alloy of the conductor.

If the ESR meter can't detect it then it's too small to worry about.

I mounted some caps on 4 Antec power supplies around 1/2" off the board and they're still chugging. Mainly because REAL cap companies didn't manufacture caps in those tiny can diameters.

Just don't take it too far though and make the leads unnecessarily long. The goal is always to keep all component leads as short as possible.

Within reason though don't risk component damage by pressing it too close to the PCB to shorten the lead length for example.

[370forlife]

Re: Mounting caps slightly off the board

We have all these IBM designed MSI boards in our computers at school and they all use UCC and rubycon. Almost all the caps that are not in the VRM's are about 1-2mm off the board.

[Topcat]

Re: Mounting caps slightly off the board

We have all these IBM designed MSI boards in our computers at school and they all use UCC and rubycon. Almost all the caps that are not in the VRM's are about 1-2mm off the board.

That's just sloppy placement by the pick & place or the sweatshop child that's putting them in the board. On IBM boards, most all of them I've seen are like this too, especially on the 8mm caps with 3.5mm lead spacing, they've been fitted into through holes 5mm apart, and there's slack left in them so not to stress the leads and break the seal. That's the reason they do that, not some cobjob way of increasing ESR, which is false to begin with.

As for ESR increasing by leaving longer leads, its bogus. If there is any ESR increase, it wouldn't even register on my ESR meter when I measured 2mm from the bung and then again at the bottom ends of the leads. I tested this earlier today just to satisfy my own curiosity.

[Lagunapete]

Re: Mounting caps slightly off the board

Note that there is an ESR increase, absolutely. However, this amount would bear no difference to the performance of the referenced circuitry. In fact, the ESR differential between the first exit point of an average cap of this nature (electrolytic 105's typical) and the end of lead in probably not measurable at the voltages this equipment runs at nor would it be measured with typical equipment of the same nature. I'd say it would begin in the nano's somewhere above 20,000 which is way out of the league of this type of equipment since the coils are obviously on the end of the circuit.
Yes there is some difference. To the degree where you should shorten the leads? Nonsense.

[Professor Peach]

Re: Mounting caps slightly off the board

Slightly longer leads are handy to clip multimeter leads onto.

[PCBONEZ]

Re: Mounting caps slightly off the board

If you are installing a capacitor not completely flush to the board, for example if you have a taller cap and must installit sideways with leads bent at a right angle THAT IS OK!!! Certain members of this forum (im not going to say names) (including pcbonez) said that it raises the esr to lenghtn the leads. But it DOES NOT!!! Topcat was awesome enough to really test it out (i PMed him) and testing the esr right at the cap apparently registered no different on an esr meter than testing at the end of the leads.
Just be careful not to strain wherw the leads go into the capacitor becuase it may damage the seal. Also be aware that the leads are exposed thus more likely of them to be shorted by a foreign obkect or another component

- Sorry but it does raise the ESR.

Anyone with an ESR meter can test this by connecting the test leads right at the base of the cap [~1/8" cap lead where they are normally soldered] and then again with the test leads connected further [5/8"-3/4"] from the cap.
-
This is not new news and I didn't come with it.
I'm not the only one that has actually tested it and found it to be true.
-
On most caps getting readings .01 Ohm higher with the extra length as described above is consistent and repeatable.
.
Less added length will also raise the ESR but unless the meter has 3 digits to the right of the decimal it won't be enough for the meter to detect it.
.

[Topcat]

Re: Mounting caps slightly off the board

^
My point exactly. My meter wouldn't even detect it if it did raise the ESR.......so anyone thinking that leaving extra leads on the cap will change circuit operation or raise the ESR enough to make up for using the incorrect cap type, it's not going to work.

[Per Hansson]

Re: Mounting caps slightly off the board

Tested a cap with my "Bop Parker" ESR meter, it has a red LED (so it's not the latest version)
It registered a capacitor as 0.03ESR at the bung, and at the end of the leads it was fluctuating between 0.03 and 0.04ESR
As PCBONEZ says a meter with a higher resolution is necessary to measure this
That said I've installed caps laying down with no problems, in those cases it was 12.5mm wide caps so wire length above the PCB was ca 7mm...

[MXM]

Re: Mounting caps slightly off the board

so a good argument could be that, it does raise the esr but it might not affect the component (or its operation)? then we'd have to consider where is that cap located at in the pcb and what is to do??

just wondering, nice thread imo.

peace.

[PCBONEZ]

Re: Mounting caps slightly off the board

If you are laying caps on their side you are typically adding 5/8" or more lead length [compared to standard mounting] making the length from the mounting point to the bung around 3/4".
At 3/4" an ESR increase of ~.01 ohms is detectable [and repeatable] even with a low res meter like the Bop Parker design units.
-
I've checked this on a number brands and sizes of cap.
It is consistent.

The Samxon RS I just bought from TC consistently cause the meter to bounce between .01/.02 at the bung and between .03/.04 near the end of the cap leads.
Saying a 0.01 Ohm increase for those would be conservative.

Picked on those because they happened to out for a PSU job.
.

[PCBONEZ]

Re: Mounting caps slightly off the board

so a good argument could be that, it does raise the esr but it might not affect the component (or its operation)? then we'd have to consider where is that cap located at in the pcb and what is to do??

just wondering, nice thread imo.

peace.

Actually it's a good thing to know because in a pinch you can de-rate an over-kill cap a little by mounting it high.
.
Not something you want to do normally. Just in a pinch where nothing else fits.
.

[Wizard]

Re: Mounting caps slightly off the board

What they have not mentioned is at load/current ripple the resistance goes up if the leads is long enough then capacitor is useless at smoothing the ripple and cause problems.

I tried that because all I had were 12.5mm caps pulled from mainboard (checked good) and used thick copper wires to extend the leads from caps. PC crashed intermittently. This was P4 2.8C.
Swapped out for 10mm capacitors, right on the PCB, PC now stable.

Cheers, Wizard

[Topcat]

Re: Mounting caps slightly off the board

I tried that because all I had were 12.5mm caps pulled from mainboard (checked good) and used thick copper wires to extend the leads from caps.

Now that could do it.

[Topcat]

Re: Mounting caps slightly off the board

The Samxon RS I just bought from TC consistently cause the meter to bounce between .01/.02 at the bung and between .03/.04 near the end of the cap leads.

Which RS's were they? the 3300uF 10/16v? I grabbed a couple rubycon's and tried this (tried it on a 3300uF 6.3v MBZ and a 2200uF 6.3v MCZ), and ESR increase over the span of the leads was not detectable by the meter. Looks like I need to check some of the others.

[PCBONEZ]

Re: Mounting caps slightly off the board

To Wizard.

I would think the solder joint between your copper lead and the stock cap lead would add more resistance [and possibly ESR] than just longer leads.

I considered doing something like that once using a solid poly when I had a nasty 'nothing fits' issue but I eventually found a good replacement and never did it.

All in all it's just better to hold out for an appropriate replacement.
BY FAR the best plan.
.

[PCBONEZ]

Re: Mounting caps slightly off the board

Which RS's were they? the 3300uF 10/16v? I grabbed a couple rubycon's and tried this (tried it on a 3300uF 6.3v MBZ and a 2200uF 6.3v MCZ), and ESR increase over the span of the leads was not detectable by the meter. Looks like I need to check some of the others.

.
Hmm - MBZ certainly do it.
Not even hard to reproduce on MBZ.

6.3 MBZ in 1000/1500/2200 were the first batch of caps I ever tested the phenomena on.
- That would be like a couple years ago now.
I checked based on what someone else at BCN said they did and found and I wanted to see it for myself.
Someone else was right.........
Like I said, this is not new news.

I often do that check on one or two caps during receipt inspection out of general curiosity and it's consistent no matter what the caps are.
[Except, when the ESR is too low in the first place to bump it over .02 with the added ESR. Most polys won't show it.]

Per Hansson just repeated it.

Why is the guru of craps having problems?
- Perhaps you have sheet rock dust in you meter? - HAHAHAHA

~~

RS @ 3300uF 16v.

The RS seem to have smaller than standard diameter leads.
[That's by eye-ball. I didn't break out a mic or anything.]
That or the metal alloy could explain why the condition is slightly more pronounced on Samxon.

~~

I pulled one each MBZ, ZL, FM and one of you Samxons out to do a photo expose'.
It's was easy enough to detect the condition and it's consistent across all of them, but alas the camera's battery died before I finished.

Since I have to wait.
Anyone know how to patch short MOV files into one long video for YouTube?


With my current stock [if I wasn't so lazy about using cameras] I could easily show the phenomena on something over 100 different size/shape/brand caps.

I'm thinking I'll eventually suck it up and actually do a video of maybe 10 examples so people can see the truth for themselves [and/or learn how to check.].
Might be a while. - *I* need to learn how to do a video first... LOL

This one is OLD people!
Already been answered more than once.
- - There IS a search here,, you know...
Even though over the last few years several different and knowledgeable BCN members have posted they have verified it occurs, the argument that it doesn't happen cycles through BCN once or twice a year.
It's getting like attempting to convince ALL the cavemen that the earth isn't flat - one caveman at a time.
- Google is your friend too...
.

[Topcat]

Re: Mounting caps slightly off the board

^
If they do, I was unable to detect it on the two I tried. For the sake of argument, why would someone even contemplate using this tactic anyway? If the trivial amount of ESR that would increase with the longer leads wouldn't make or break a PSU or mainboard, so leaving additional lead on the cap in order to use the wrong type in an application is just a silly thing to do....instead of just getting something as close to the original as possible to begin with. ESR tolerances on pretty much any circuit we'll encounter will be within specs of the cap that's intended to go there, regardless of the lead length. I exempt Wizard's mod from this, as he extended the lead, which creates an additional joint and a whole other can of worms.

FWIW, I'd be ashamed of myself if I returned a customer's board to them with extra lead and the cap sticking way off the PCB. That's just shoddy looking.

[PCBONEZ]

Re: Mounting caps slightly off the board

I agree.
I've said may times in many posts [including this one] that it's better to wait for the right caps.
-
But in some parts of the world any kind of low ESR cap is hard to come by.
.
And you know as well as I do that some people are just too impatient.
- - Here's a fine example. . [And he's a Mod...]
https://www.badcaps.net/forum/showpo...3&postcount=10

Someone that is going to try that little number should AT LEAST be aware of the effects on ESR so they can counter by using better than original spec caps.

......

When one can't find the right caps,, there CAN BE such a thing as too little ESR.
-
The following blurb is right out of a post you 'stickied' in 2006.
~~~~~~~~
Being a resistance it acts as a means of removing the ripple energy in terms of heat. If the ESR was reduced to zero there would be no way of removing the energy, if this were possible a near lossless output filter would have what is termed a high Q factor and damped oscillation is likely to occur (the ripple energy needs to be absorbed), this would be at a frequency outside of the regulator control. It is therefore possible that if ESR is reduced to a very low value, voltage spikes will develop on the regulator output. This would be critical in a VRM. Output capacitance total ESR should remain at an effective ratio to the load resistance.
~~~~~~~~~

That doesn't even take into account the coordinating the values of a cap's [or caps] ESR with the Inductor[s] in a filter.
- The famous jonnyguru PSU recap that went bad is a fine example of 'de-tuning' the output filters using caps with too low of ESR.
----

So someone in Timbuktu [or where ever] is 'stuck' using an MBZ/MCZ in place of a Fuhjyyu w/ESR @0.045 would probably be making a good move by mounting the cap high.
.