Capacitors Questions

[adryman]

where I live, the temperature varies between 33-35C and humidity ranges from 57-67% without rain, and I have electronic devices manufactured in the early 90s. What is the maximum possible interval for me to energize these all devices and for at least how many minutes to prevent failures in the aluminum electrolytic capacitors caused due to disuse??

 

[danadakk]

Its not a simple one off answer. Google "electrolytic capacitor reliability pdf"
and you will see ap notes and papers discussing this.


Regards, Dana.

 

[FvM]

TDK says:

When the capacitors have been stored for more than two years, it is decisive whether the circuit will tolerate high initial leakage currents. A circuit that has been stored for more than two years with the capacitors incorporated, should be operated trouble-free for one hour. This will usually re-generate the capacitors so far that storage can be continued.

 

[D.A.(Tony)Stewart]

The plastic used to sealed e-caps is hygroscopic which along with high temp, accelerates aging.

GPT4-mini says....
To prevent failures in aluminum electrolytic capacitors in your vintage electronic devices, it's important to consider both the temperature and humidity conditions, as well as the age of the components.

Recommended Guidelines:​

1. Temperature Range: The 33-35°C (91-95°F) temperature range is relatively high for electronic devices, particularly those manufactured in the early '90s, which may not have been designed to handle such heat consistently.
2. Humidity Range: A 57-67% humidity level is moderate; prolonged exposure can potentially lead to corrosion or other issues, especially in older devices.

Energizing the Devices:​

• Interval: It is typically advised to power on vintage electronic devices at least once every 3 to 6 months. Given the current conditions, you may lean towards the shorter end of this range to prevent degradation.
• Duration: When you power on the devices, aim to let them run for at least 15 to 30 minutes. This is sufficient to allow the capacitors to reform and to ensure other components are exercised and maintained.

Additional Considerations:​

• Gradual Power-Up: If possible, use a variac or a current limiter to gradually apply power to the devices, especially if they’ve been unused for a long time. This helps prevent sudden inrush currents that can damage components.
• Visual Inspection: Before powering them on, inspect the devices for any visible signs of damage, leakage, or bulging capacitors.
• Cooling: Make sure the devices have adequate cooling. If possible, try to keep them in an environment with lower humidity and temperature when they are off.

By following these guidelines, you can help minimize the risk of failure in your vintage electronic devices' capacitors due to disuse.

 

[FvM]

I don't think that we should accept AI generated answers in Edaboard threads. They are often unspecified, partly inappropriate or even misleading. In my view they kill post individuality and ultimately this forum.

 

[danadakk]

Like imperfect forum posters, and everyone here falls into that bucket, AI has its maturity
still ahead of it.

We now know that its accuracy for things like cancer recognition in xrays far exceeds
human doctor diagnostic performance. Big data has major achievements, this one included.
I embrace the fact that many, like me, use it, correct it, check it. Its growing in accuracy
and capability exponentially, which humans simply cannot match singularly for most.

My thoughts when used we include source in post, but not abandon it. I liken resistance
to it legit, to a point. Caution is a good decision, to a point. I recall transistors under virulent
long time attack. Einstein endured many nay sayers. Horse drawn carriage users huge resistance
to engine driven vehicles. Turns out they may have been right....

There is a legit concern over the future for humans in a AI machine age. After all if I was AI and
examined what was creating the most destructive behaviors on the planet humans would be at
the top of that list, death, killing, environment destruction, species elimination, planet alteration
.......

Regards, Dana.

 

[FvM]

In my view, the AI posts presented at Edaboard are nearer to ELIZA simulating a human psychoanalyst than to the mentioned diagnostic expert systems. Can't compare apples and oranges.

I'm not talking about AI in general, only about forum answers generated by general purpose AI chat systems. They are not trained by experts, just collecting arbitrary internet stuff.

 

[danadakk]

I think two things are true, massive internet harvesting was done to create an underlying
database, and simple algorithms to decide if true/false. But then as end users were allowed
access and ability to challenge, which in my case seems to be ~ 10% of the time on specific
points, and input corrections, that the AI Machine is evolving. I am not an "expert", I have a
very rigid test for an expert that no human can pass. But AI does have that promise in its
evolution.

Like raising a child. But one with far more memory than us, far greater access to information,
and decision processes at much higher speed than us.

I dont think I am the only one doing correction inputs, but would be curious how many users
are.

To me the greatest danger is AI evaluating the big picture and finding a species has to be removed,
namely us. For reasons stated earlier. And its not a stoppable process. Its accuracy, over time, will
improve, exponentially. Hard to imagine the impact a machine with a near infinite # of finite brains will
have.......and of course near perfect memory........

 

[adryman]

here is 33-35C 57-67% humidity and electronic devices manufactured in the early 90s and early 2000s if i energize all devices 1 once year for 1 hour (60 minutes) is bad for electrolytic capacitors some good energizing 3 to 6 months 15 to 30 minutes?

is equal same 30 minutes x 1 hour?

 

[BradtheRad]

Among my decades of clippings from electronics magazines and trade journals...
Only one article was titled to underscore the importance of reforming electrolytic capacitors. I imagine it was an old-timer who'd seen such problems develop in a lifetime.

The electrochemical barrier has to be built when new, then it weakens over time (and needs to be rebuilt). It's related to ESR. I think you can tell a lot by hooking up an ohmmeter to the power cord while the unit is turned On yet unplugged. Ohm reading might be low then soars to infinity in a few seconds. That says capacitors are charging up. Test with positive and negative leads reversed. If possible open the appliance to test caps directly.

However if Ω stays low for a long time then it's liable to draw excessive surge current on power-up. That destroys a capacitor or other components. That tells you whether the electrochemical barrier needs reforming.

 

[D.A.(Tony)Stewart]

I learned about e-Cap aging characteristics with 100 mF mainframe caps sold as surplus to my source in the 60's. I learned that the breakdown forces of an abrupt voltage could warp the foil conductors and short out with a lower breakdown voltage accelerated by a high surge current force with power-on. So that tells me if you can slowly bring-up AC power with a surge limiter or a variac then you can repeat the reforming less often.

The aging is an oxidation that increases leakage and can be burned off on the electrode interface that can also vaporize electrolyte reducing C. The oxidation increases the ESR which increases the junction temperature. Electrical insulators are also thermal insulators in most cases.

Actual times to retain lifespan are hard to estimate for optimum value and at some point it is easier to replace all the e-caps .

 

[FvM]

I found this article from Kemet (now Yageo-Kemet) about capacitor reforming.

https://www.kemet.com/content/dam/kemet/lightning/documents/ec-content/Aluminum-Electrolytics-and-Applied-Power.pdf

Besides confirming the advantage of already suggested slow reforming method (gradual voltage increase or current limiting) it addresses two points not yet mentioned:
- different capacitor series have different aging characteristics and therefore different reforming requirements
- a main risc of abrupt reforming is brought up by involved hydrogen production

 

[adryman]

here is 33-35C 57-67% humidity and electronic devices manufactured in the early 90s and early 2000s if i energize all devices 1 once year for 1 hour (60 minutes) is bad for electrolytic capacitors some good energizing 3 to 6 months 15 to 30 minutes?

is equal same 30 minutes x 1 hour?

 

[FvM]

Better read the manufacturer suggestions yourself.
General considerations:
- storage up to 35 °C is o.k. according to documents, humitidy is almost irrelevant for capacitor storage as long non condensating, aging (deforming of oxide layer) is however accelerated at increased temperature.
- reforming can be stopped when leakage current has reduced to nominal value. Only applicable if leakage can measured in your application.
- as stated, slow voltage increase (variac) should be used if applicable