Jump to content

Replacing se/30 capacitors question


Recommended Posts

  • 68kMLA Supporter

Yes, under those capacitors are very fine circuit traces on the motherboard and if you overheat them or scrape them you'll cause a lot of damage. You can also lift the pads if you don't do it right. Also keep in mind if the electrolytic fluid has escaped from the existing capacitors, that stuff is like acid and eats away at those traces so just replacing the capacitors may not be enough.

 

Edited to add, the reason why tantalums are better is because they will not leak in the future so one fixed now will last forever. If you use the electrolytics, just remember to do it again in 10 or so years or when the fluid leaks out again it will dissolve the traces.

Edited by joethezombie
Link to post
Share on other sites
  • 68kMLA Supporter

Yeah, they should have! Funny thing is, some boards like the Quadra 700 is all tantalum, and the IIfx board has pads for tantalums, but then they used those crappy aluminum cans! Video cards aer even stranger, some of them have tantalums and some of them have aluminum cans. It seemed that Apple just used what they could get the cheapest at the time of production.

Link to post
Share on other sites

Yes, under those capacitors are very fine circuit traces on the motherboard and if you overheat them or scrape them you'll cause a lot of damage. You can also lift the pads if you don't do it right. Also keep in mind if the electrolytic fluid has escaped from the existing capacitors, that stuff is like acid and eats away at those traces so just replacing the capacitors may not be enough.

 

Edited to add, the reason why tantalums are better is because they will not leak in the future so one fixed now will last forever. If you use the electrolytics, just remember to do it again in 10 or so years or when the fluid leaks out again it will dissolve the traces.

Do you know about how much it will cost me to send my se/30 board to maccaps.com? I have 2 regular macintosh se computers that have never had to be recapped so im just wondering.

Link to post
Share on other sites
  • 68kMLA Supporter

No idea. I've never used his services as I do all my own work. I would guess the SE/30 is one of the more complicated boards with a high capacitor count, so probably at the higher end of his scale. Go to his website and email him directly. There's also a few here that are starting to offer this service, maybe at a cost savings.

Link to post
Share on other sites

No idea. I've never used his services as I do all my own work. I would guess the SE/30 is one of the more complicated boards with a high capacitor count, so probably at the higher end of his scale. Go to his website and email him directly. There's also a few here that are starting to offer this service, maybe at a cost savings.

Could you possibly refer me to a person on this forum who does se/30 recaps?

Link to post
Share on other sites

I tend to replace capacitors with the same type as used by the manufacturer if still available. Hobby machines don't get used the same as they were when new so even aluminum analytics should last a very long time.

 

Tantalum capacitors short closed (they explode), depending on the circuit that could be very bad. Also tants are not the best for audio circuits.

Link to post
Share on other sites
  • 68kMLA Supporter

Honestly, it's just not that hard. The biggest requirement is patience followed by care with a dose of gentleness. Get two 40ish watt pencils at Radio Shack (I used to use the grounded 15W pencils but they're too weak), let them heat properly, apply liquid flux to the old solder, and then with one pencil in each hand apply a pencil to each side of the cap. Don't drill or grind. Just wait. If the joint doesn't seem to be heating, retip your pencil with solder and reapply flux to the joint. Heat conduction is key. Ideally the cap will be loose is about 5 seconds but sometimes it takes longer. Then you just lift it with the pencils and wipe it off on a damp sponge.

 

All the warnings about lifted pads and traces underneath are why you don't drill or grind the pencils into the board. Just gently apply heat until the cap comes loose.

Link to post
Share on other sites
  • 68kMLA Supporter

I tend to replace capacitors with the same type as used by the manufacturer if still available. Hobby machines don't get used the same as they were when new so even aluminum analytics should last a very long time.

Eh, I'm not so sure if the usage characteristics of the machine has to do with leaky cans. Perhaps dried out or vented ones, but I would think the same chemical reaction that dissolves traces also dissolves the seals on the capacitor itself over time, regardless of use. Most of the machines we are repairing today haven't been in use for 20 years, yet still suffer. I'm not an expert in capacitor manufacturing, this is just personal observation.

 

Tantalum capacitors short closed (they explode), depending on the circuit that could be very bad. Also tants are not the best for audio circuits.

Apple used plenty of tants over the years. There's *far* more damage done by electrolytics than tants shorting. It'd be hard pressed to find a single instance of an exploded tant on a vintage mainboard, where conversely thousands of boards suffer trace damage from leaky cans. And I seriously doubt anyone, even the most seasoned audiophile, could detect any difference in a blind A B between a tantalum and electrolytic on a vintage mainboard.

 

But I guess if that seriously concerns you, there are solid polymer electrolytic caps available, at a premium, that resolve those "issues".

Edited by joethezombie
Link to post
Share on other sites

Eh, I'm not so sure if the usage characteristics of the machine has to do with leaky cans. Perhaps dried out or vented ones, but I would think the same chemical reaction that dissolves traces also dissolves the seals on the capacitor itself over time, regardless of use. Most of the machines we are repairing today haven't been in use for 20 years, yet still suffer. I'm not an expert in capacitor manufacturing, this is just personal observation.

 

Apple used plenty of tants over the years. There's *far* more damage done by electrolytics than tants shorting. It'd be hard pressed to find a single instance of an exploded tant on a vintage mainboard, where conversely thousands of boards suffer trace damage from leaky cans. And I seriously doubt anyone, even the most seasoned audiophile, could detect any difference in a blind A B between a tantalum and electrolytic on a vintage mainboard.

 

But I guess if that seriously concerns you, there are solid polymer electrolytic caps available, at a premium, that resolve those "issues".

Heat has quite a bit to do with the lifespan of the capacitors in question.

 

https://en.wikipedia.org/wiki/Aluminum_electrolytic_capacitor

 

The testing time and temperature depend on the tested series. That is the reason for the many different lifetime specifications in the data sheets of manufacturers, which are given in the form of a time/temperature indication, for example: 2000 h/85 °C, 2000 h/105 °C, 5000 h/105 °C, 2000 h/125 °C. This figures specifies the minimum lifetime of the capacitors of a series, when exposed at the maximum temperature with applied rated voltage.

 

Referring to the endurance test, this specification does not include the capacitors' being loaded with the rated ripple current value. But the additional internal heat of 3 to 10 K, depending on the series, which is generated by the ripple current is usually taken into account by the manufacturer due to safety margins when interpreting the results of its endurance tests. A test with an actual applied ripple current is affordable for any manufacturer.

 

A capacitor's lifetime for different operational conditions can be estimated using special formulas or graphs specified in the data sheets of serious manufacturers. They use different ways achieve the specification; some provide special formulas,[53][54] others specify their capacitor lifetime calculation with graphs that take into account the influence of applied voltage.[39][55][56] The basic principle for calculating the time under operational conditions is the so-called “10-degree-rule”.[57][58][59]

 

This rule is also well known as the Arrhenius rule. It characterizes the change of thermic reaction speed. For every 10 °C lower temperature, evaporation halves. That means for every 10 °C lower temperature the lifetime of capacitors doubles.

L x = L Spec ⋅ 2 T 0 − T A 10 {\displaystyle L_{x}=L_{\text{Spec}}\cdot 2^{\frac {T_{0}-T_{A}}{10}}} L_{x}=L_{{\text{Spec}}}\cdot 2^{{\frac {T_{0}-T_{A}}{10}}}Lx = life time to be estimated

LSpec = specified life time (useful life, load life, service life)

T0 = upper category temperature (°C)

TA = temperature (°C) of the case or ambient temperature near the capacitor

 

If a lifetime specification of an electrolytic capacitor is, for example, 2000 h/105 °C, the capacitor's lifetime at 45 °C can be "calculated" as 128,000 hours—roughly 15 years—by using the 10-degree-rule. Although the result of the longer lifetime at lower temperatures comes from a mathematical calculation, the result is always an estimation of the expected behavior of a group of similar components.

 

Capacitor seals are chemically resistant to the liquids used (circuit boards are not). I have seen blown tants, last one being on a Daystar 040 PDS board.

Link to post
Share on other sites
  • 68kMLA Supporter

I understand capacitor lifetime ratings. What I am getting at is there is a huge difference between baking at 105 for 5000 hours (the rating, which can and is tested), and sitting idle for 20 or 30 years, the latter of which has no design or real test data. Just because a formula extends life out to 40 years with minimal use does not mean the components of the capacitor won't start to break down earlier. I do know that the Nichicon description documents for electrolytic cans specify a gradual degradation after only a 2 year shelf.

 

But really, its opinion either way. No one has a room of 30 year old capacitors that they are actively testing, so in the end, do what you want. It probably doesn't matter.

Link to post
Share on other sites

Being idle can probably be as problematic as being over used (over heated). I try to fire up each machine at least a few times a year. You have to admit not all of the millions of machines that were made in the 80's and 90's are dead, so there is real life data on aging capacitors. Another interesting thing about those capacitors is every other manufacture used them, and outside of some issues with IBM PS/2 floppy drives and hard drives from PS/2 I never had a reason to replace them.

Link to post
Share on other sites
  • 68kMLA Supporter

Agreed. It does seem the bad leakers are always the 47uF 16v smd alu cans. One can only hope that new (better?) manufacturing techniques have increased the reliability of those guys. That's additional ignorance I'm introducing, because I'm comparing manufacturing processes from 30 years ago to something that you would buy today. And I haven't even discussed the capacitor plague or counterfits yet!

Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

Loading...
×
×
  • Create New...