KZG - KZJ - Same Electrolyte

[Wester547]

Re: KZG - KZJ - Same Electrolyte

I don't know if the origin (manufacturing plant) of the electrolyte and foil have a bearing on reliability, but it seems that some batches of KZGs don't fail as much as others.

I don't think it does in NCC's case, because according to NCC's website, they own 100% of most of their plants, except for Indonesia (who K-CC/Samyoung owns 10% of), Samyoung (whom they only own 33.4% of) and Marcon (whom they only own 20% of). I'm pretty sure NCC import most if not all their materials from Japan as well, and KDK Corp (Hitachi Electrolytic Foil Research Co. Ltd), their subsidiary which manufactures ultra-high purity anodic and cathodic aluminum foil (99.8-99.98%), is their source for all production of aluminum foil for capacitors. And according to UCC, they are completely autonomous regarding their materials (IE, all materials are produced in-house, essentially, except for the chemicals NCC use to create the electrolyte itself - those are sourced from Tomiyama Pure Chemicals).

As for why some may seem to fail less than others... I guess that depends upon several factors. One of which is the condition of the anodic dielectric. A faulty electrolytic composition cannot properly heal the dielectric (or provide enough oxygen to "wet" weak spots), so even if KZG and KZJ do measure good for ESR and capacitance, once a weak spot in the dielectric appears, it could cause a permanent short in any otherwise "good" capacitor. But if the dielectric's condition remains good, this may never happen. As the porous dielectric deteriorates in storage from the diffusion or penetration of ionic particles from the electrolyte, the issue is only exacerbated.

The other factor I can think of is the rate at which molecules of gaseous hydrogen escape by way of rubber seal diffusion. If it doesn't happen rapidly, there won't be enough vapor pressure to rupture the cans or cause a difference in appearance and they'll silently drift out-of-spec as the electrolyte vaporizes and the water electrolyzes. Of course heat will accelerate the diffusion rate or the rate of evaporation, and if the thermal gradient is high (the speed of the internal temperature rise, AKA high ripple current loads), that will significantly shorten the life of the cap.

KZE and KZN also appear not to fail at all.

Same goes for KY/KZM/KZH, at least in my experience. But KY, KZM, and KZN are rated to withstand up to 10,000 hours at 105ºC (with the maximum rated ripple current applied, for larger parts) - I would expect them to have greater chemical stability. KZE and KZH are rated up to 5,000 and 6,000 hours (under the same conditions) for larger parts, so I would expect them to do better.

The difference may be a crucial stabilizer/getter that was omitted in KZG/KZJ, or conversely, some trace contaminant in KZG/KZJ that is not present in KZE/KZN.

I think if the electrolyte was contaminated in any way or if some stabilizing neutralizer or scavenger was somehow absent from the electrolyte (that would cause hydrogen gas generation at the cathode), it should have been caught during factory testing (the electrolyte is synthesized in dry rooms where no human breath is permitted). Since KZG and KZJ degrade much faster (not just visibly but spec-wise) than all of NCC's other series, after the endurance test, IMO NCC should have seen that they were probably creeping towards the limits of their parameters (leakage current, dissipation factor, capacitance, ESR) and thereafter improvements could have been made upon the electrolytic composition, even if a small percentage of the caps are allowed to fail the endurance tests.

But I guess when you're producing thousand of series and hundreds of millions of capacitors per month (billions per year), only so much can be done. KZG/KZJ and all ultra-low-Z variants from NCC were discontinued a while back anyway, completely usurped by their NPCAP conductive functional polymer line.

[linuxguru]

Re: KZG - KZJ - Same Electrolyte

... What this probably means (in KZE's case), since KZE has a very good track record (unlike KZG and KZJ), is that the resemblance the electrolyte bears may be superficial.

I don't know if the origin (manufacturing plant) of the electrolyte and foil have a bearing on reliability, but it seems that some batches of KZGs don't fail as much as others. KZE and KZN also appear not to fail at all.

The difference may be a crucial stabilizer/getter that was omitted in KZG/KZJ, or conversely, some trace contaminant in KZG/KZJ that is not present in KZE/KZN.

[televizora]

Re: KZG - KZJ - Same Electrolyte

If you have a fully charged battery [as with the cap] and you apply AC to it's terminals, what happens?.

If you have a capacitor with AC applied on it's terminals...

No current thru the Capacitor if there is pure DC. The capacitor will charge initially and then nothing more will happen.
The reactance of the capacitor is infinity for a DC and decreases with the increase of the frequency.
The opposite is valid for the inductors. They are almost ideal conductor for the DC. They characterize only with their ohmic resistance.
And what is voltage? The voltage is the difference between the electrical potentials between two points. If one is taken as reference, then we measure what is the potential difference between the reference point and the other point. It can be positive or negative. The reference point is conditionally selected
Ripple is the part of the AC waveform that lasts after rectification by the diode bridge and smoothing of the capacitor... Usually the ripple is double the frequency of the input AC.

[Wester547]

Re: KZG - KZJ - Same Electrolyte

^ Yup, both KZG and KZJ employ an electrolytic formula that's so dreck it generates volatile amounts of gas even in NOS capacitors, a la GSC and Fuhjyyu.

And I think the OP missed something in the UCC KZG datasheet.

The KZG series is a new ultra low impedance series from United Chemi-Con. These capacitors are different from the standard low impedance capacitors, as they use a new low resistivity electrolyte. Compared to our KZE series that also uses this advanced electrolyte technology, the KZG series has lower ESR/impedance ratings, making them ideal for use in computer board circuits where very low impedance capacitors are required.

I know I've said this before. I just thought it should go in this thread. What this probably means (in KZE's case), since KZE has a very good track record (unlike KZG and KZJ), is that the resemblance the electrolyte bears may be superficial.

[zigi]

Re: KZG - KZJ - Same Electrolyte

KZJ are a really bad capacitors.
They blown up during storage in good conditions.
See photo.

[Pyr0Beast]

Re: KZG - KZJ - Same Electrolyte

Suit yourself.

[PCBONEZ]

Re: KZG - KZJ - Same Electrolyte

Sorry.
The ROCK training experiment is over for today.

You don't even understand how they work in a static condition.
I'll be damned if I'm going to get into dynamic conditions with you.
.

[Pyr0Beast]

Re: KZG - KZJ - Same Electrolyte

Invalid comment.

Valid comment.

There is no charging taking place in this discussion about a cap.

O really ?

Charging means changing of electrode material, same as with capacitor.

Just that capacitor is electrostatic device, way before electrochemical.

If you have a fully charged battery [as with the cap] and you apply AC to it's terminals, what happens?

It heats up due to its internal resistance and losses from constant charge and discharge.

But it won't charge further even one bit ! Sum of chemical reactions is 0.

Problem is you don't understand it [electricity & chemistry] well enough to realize that is what you are saying.

I do. You just don't let yourself to understand what I am saying.

[PCBONEZ]

Re: KZG - KZJ - Same Electrolyte

Heh. Can you charge a car battery with AC ?

Invalid comment.
There is no charging taking place in this discussion about a cap.

If you have a fully charged battery [as with the cap] and you apply AC to it's terminals, what happens?
- That discussion would valid.
- And its a very good analogy of how Ripple damages the oxide layer in caps with DC present.

Didn't say that.

Yes you did. - MANY times.

Problem is you don't understand it [electricity & chemistry] well enough to realize that is what you are saying.

.

[Pyr0Beast]

Re: KZG - KZJ - Same Electrolyte

No, it's not..

It's self evident that you don't understand chemistry and how how chemical reactions are affected by current [and vice versa].
.
If you did you would understand how caps [and batteries] work.

Heh. Can you charge a car battery with AC ?

If you 'float' that combined DSL + POTS line's signals on 200vdc , which you filter out with a cap before the device , the signal won't pass to the device.

Didn't say that.

Don't work like that,,, yet that's what you keep saying..

No, that is what YOU keep _reading_ and _thinking_.

[PCBONEZ]

Re: KZG - KZJ - Same Electrolyte

And its effect is limited by DC bias.

So, --> according to you...

If you 'float' that combined DSL + POTS line's signals on 200vdc , which you filter out with a cap before the device , the signal won't pass to the device.

Don't work like that,,, yet that's what you keep saying..
.

[PCBONEZ]

Re: KZG - KZJ - Same Electrolyte

And its effect is limited by DC bias.

No, it's not..

It's self evident that you don't understand chemistry and how how chemical reactions are affected by current [and vice versa].
.
If you did you would understand how caps [and batteries] work.

[Pyr0Beast]

Re: KZG - KZJ - Same Electrolyte

That means the only current passing -through- the cap that could possibly affect it's chemistry is the AC current, which is due to Ripple voltage.

And its effect is limited by DC bias.

[Pyr0Beast]

Re: KZG - KZJ - Same Electrolyte

DC current is not needed. DC potential is.

Same with lead acid batteries. They loose their potential - they degrade.

[PCBONEZ]

Re: KZG - KZJ - Same Electrolyte

You stick that scope on the DSL + POTS signal and it will display the SUM.
Neither the POTS nor the DSL would work on that signal.
They use filters in the down stream device so it only gets one signal.

You stick that scope on the DC + AC on a filter cap it will display the SUM.
But the cap -IS- a filter to DC. It blocks it. - There is no DC current.
That means the only current passing -through- the cap that could possibly affect it's chemistry is the AC current, which is due to Ripple voltage.
.

[PCBONEZ]

Re: KZG - KZJ - Same Electrolyte

Because that is only thing what matters at capacitor chemistry.

And the only way electricity affects chemistry is by CURRENT.

There is AC current. [And the DC voltage is does NOT affect it.]

With the exception of leakage there is no DC current at all.
- And in lytics the majority of leakage is due to the molecules that move in the healing the Oxide layer.. from the damage caused by the AC current.. .. the AC current which isn't affected by the DC voltage at all.

You are trying to add the AC and DC together as if they are one component and they are NOT the same component any more than the DSL and POTs signal are when they co-exist in one circuit.
.

[Pyr0Beast]

Re: KZG - KZJ - Same Electrolyte

YES, it does work.
- But if what you say about voltages on caps was true, it wouldn't.

- And THAT is what is ridiculous.

Too bad as it seems you don't get it. Eh. I have tried.

[PCBONEZ]

Re: KZG - KZJ - Same Electrolyte

I have VDSL wired that way with POT phone on the same line, WORKING !
32Mbit synchronisation !

YES, it does work.
- But if what you say about voltages on caps was true, it wouldn't.

- And THAT is what is ridiculous.

[Pyr0Beast]

Re: KZG - KZJ - Same Electrolyte

You are too lost to waste any more of my time with.

I would say quite the same for you.

You are trying to mix the AC and DC as if they are one component and that's not the proper way to do a circuit analysis.

Yes I am. Because that is only thing what matters at capacitor chemistry.

They way you think sending POT and DSL through the same wire would NEVER WORK because they would mix.
- That isn't how things work.

Now you are just ridiculous.

I have VDSL wired that way with POT phone on the same line, WORKING !
32Mbit synchronisation !

[PCBONEZ]

Re: KZG - KZJ - Same Electrolyte

~ jibberish ~

You are trying to mix the AC and DC as if they are one component and that's not the proper way to do a circuit analysis.

They way you think sending POT and DSL through the same wire would NEVER WORK because they would mix.
- That isn't how things work.

.