EP2374141A1

EP2374141A1 - Method for reforming electrolytic capacitors

Info

Publication number: EP2374141A1
Application number: EP09765031A
Authority: EP
Inventors: Jörg Raimund HEMPEL
Original assignee: IMP GmbH
Current assignee: IMP GmbH
Priority date: 2008-12-04
Filing date: 2009-12-01
Publication date: 2011-10-12
Also published as: WO2010063453A1; DE102008060546B3

Abstract

The invention relates to a method for reforming electrolytic capacitors, comprising the following steps: a) if the electrolytic capacitor exhibits a load, slowly discharging via a suitably dimensioned load; b) charging of the electrolytic capacitor by applying a voltage increasing in steps, wherein each voltage increase step equals a fraction of the nominal voltage of the electrolytic capacitor, wherein the voltage is then increased by one step when the charge current at the current step has dropped substantially to zero, and wherein the stepped increase of charge voltage is continued until the charge current no longer drops to zero.

Description

Process for reforming electrolytic capacitors

The present invention relates to a method for reforming electrolytic capacitors.

Electrolytic capacitors are used when very large capacitance values are required. This is the case, for example, with power supplies, audio power amplifiers and the like. In automotive electronics applications include engine management systems for fuel injection, controls of radiator and wiper motors, electronic steering systems, airbags or multimedia devices, and the like.

In an electrolytic capacitor (also called Elko) is on the metal of the anode electrode by electrolysis (anodic oxidation, formation) generates a non-conductive insulating layer, which forms the dielectric of the capacitor. The electrolyte forms the cathode (counter electrode) of the electrolytic capacitor. It can consist of a liquid or pasty electrolyte (ionic conductor) or a solid electrolyte (electron conductor). The power supply to the electrolyte via films of the same metal as that of the anode or via a suitable contacting of the electrolyte.

Currently common types of electrolytic capacitors are: aluminum electrolytic capacitors with alumina as a dielectric, tantalum electrolytic capacitors with tantalum peroxide as a dielectric, niobium

669-68567PCT / CM Electrolytic capacitors with niobium pentoxide as a dielectric as well as so-called "gold caps" or "super caps".

The anode of the electrolytic capacitor is structured to increase the surface, in aluminum electrolytic capacitors it is a roughened anode foil, in tantalum electrolytic capacitors a sintered metal sponge. Due to the large surface area and the extremely thin dielectric, electrolytic capacitors with a small construction can achieve relatively high electrical capacitances of the order of 1 F.

For "Gold Caps" and "Super Caps" capacities reach the order of 2OF, but at relatively low nominal voltages of a few volts.

The electrolytic oxidation of, for example, a roughened aluminum foil as an anode material is called forming. This happens for the first time in the production of electrolytic capacitors at the end of the manufacturing process in the factory.

Electrolytic capacitors have the disadvantage of high leakage current losses due to the residual conductivity of the dielectric. If an electrolytic capacitor is not used for a long time, it discharges completely due to the leakage current. This condition is detrimental to the insulation resistance of the dielectric, which must be reformed or reformed to maintain or restore a high insulation resistance of the dielectric.

669-68567PCT / CM Previously known Formierungs- or Nachformierungsverfahren for electrolytic capacitors therefore always have the goal to regenerate the dielectric, so as to reflect the electrolytic capacitor its original qualities such as capacitance and breakdown voltage value.

From US 4,974, 118 A, a device is known, can be carried out with the Nachformierungsverfahren of the type in question here.

WO 2007/025057 A2 describes a Nachformierungsverfahren for built-in medical implants capacitors. Thereafter, empirically determined capacitor charge data of the capacitors used are used to determine the reformation time to be applied, in which the capacitors are charged to a voltage corresponding to 20% to 90% of the rated voltage.

DE 19 50 967 has a method for reforming of electrolytic capacitors to the object, in which during the Nachformierungsvorgangs this at least once, preferably several times by switching off a DC power source is interrupted and at the same time an electrical discharge resistor is connected in parallel to the capacitor over which the capacitor wenigsgtens partially discharges, so that a occurring in the previous charging of the capacitor polarization of the anode is sufficiently nullified by the mentioned discharge.

669-68567PCT / CM From DE 23 60 688 A a method for reforming electrolytic capacitors is known, in which the voltage applied to the capacitor is increased in stages and a constant current per stage is selected so that the maximum voltage per stage is preferably less than 20 Seconds is reached.

The present invention has for its object to provide a method for reformulating electrolytic capacitors, with which the leakage current of the replenished electrolytic condensates can be reduced to a negligible degree and the dielectric strength and the nominal voltage can be increased, whereby higher amounts of charge can be stored in the electrolytic capacitor.

This object is achieved with the features of claim 1.

Features of preferred embodiments of the present invention are characterized in the subclaims.

With the method according to the invention nachformierte electrolytic capacitors showed in the test series even after several days no significant self-discharges and caused by the reforming breakdown voltage causes the rated voltage and thus the storable charge amount against non-reformed, otherwise identical electrolytic capacitors could be increased. In addition, the inventively nachformierten electrolytic capacitors could be charged in a remarkably shorter time compared to non-reformed, otherwise identical electrolytic capacitors to their (new) rated voltage.

669-68567PCT / CM In the following the invention is explained in more detail by means of the description of exemplary embodiments with reference to the drawing, wherein the single figure shows the circuit diagram of an apparatus for carrying out the method according to the invention for reforming of electrolytic capacitors.

In the following, an embodiment of a device for reforming electrolytic capacitors will be described first, followed by test runs on an aluminum electrolytic capacitor and a "gold cap" electrolytic capacitor, which were carried out with this device.

A laboratory power supply 1, in which the output voltage is adjustable separately from the output current and in which the output current can be limited, serves as a current source for controlled charging of the aluminum electrolytic capacitor 4 used here. For this purpose, the positive pole of the laboratory power supply 1 with the positive pole of the electrolytic capacitor 4 with interposition an ammeter 2 and a first and a second breaker contact 3, 10 connected. The negative terminal of the laboratory power supply 1 is connected to the negative pole of the electrolytic capacitor 4. Furthermore, the voltage across the electrolytic capacitor 4 is measured with a voltmeter 5. A parallel circuit of a second voltmeter 6, a third breaker contact 7 and a load resistor 8 is connected on the one hand to the negative terminal of the laboratory power supply 1 and on the other hand to the output of the ammeter 2. Immediately before

669-68567PCT / CM Load resistor 8, a fourth breaker contact 9 is arranged in series with this.

First embodiment of the method according to the invention

Four "gold cap" electrolytic capacitors are connected in series and introduced into the experimental setup shown in the figure instead of the electrolytic capacitor 4 shown there. The "Gold Cap" electrolytic capacitors used were capacitors with a nominal voltage of 2.2 V and a capacitance of 20 F ± 20%. In a first step, when the second breaker contact 10 is open, it is checked whether a (residual) charge is still present on the series connection of the capacitors and whether the voltmeter 5 accordingly indicates a voltage. If this is the case, the capacitors are slowly discharged by closing the breaker contacts 3 and 9 via the load 8. The load resistor 8 may have a resistance between 20 and 50 ohms. This unloading process can take a few minutes. If the voltmeter 5 shows no or only a low voltage in the millivolt range, to be sure by closing the second and the fourth breaker contact 3 and 7, the capacitor series circuit can be short-circuited.

Thereafter, the third and fourth breaker contacts 7 and 9 are opened and set a discharge voltage of 1 V on the laboratory power supply. Then, the second breaker contact 10 is closed, and the charging of the capacitor bank circuit starts. When closing the second breaker contact 10 of the charging current jumps to a

669-68567PCT / CM Value of 1.6 amps and then drops slowly until after about half a minute no charging current is displayed on the milliampere resolution ampermeter.

Then, the second breaker contact 10 is opened and the output voltage on the laboratory power supply 1 is increased by 1 V to 2 V. Then, the second breaker contact 10 is closed and the charging current behaves substantially as in the first charging stage.

As an advantageous variant, the charging current can be limited to 100 or 200 mA on the laboratory power supply unit 1, if one wants to avoid the initial charging current peaks in the ampere range. This is advantageous in that it is important in the method according to the invention that the charging process proceeds relatively slowly in order to give the capacitors to be reformed sufficient time to form. It is also possible to reduce the initial high charging currents by making the voltage increasing steps lower, for example 0.5 V instead of 1 V, or even lower.

The gradual increase in voltage is continued until it is determined that the charging current no longer drops to 0, or a few milliamps, but a level greater than 0, in the measurements carried out here about 40 mA, einpegelt. At this stage, the electrolytic capacitor becomes a current consumer and reactions take place inside the electrolytic capacitor, which can destroy the capacitor or impair its capacity. When this stage is reached, the condenser chain is disconnected from the laboratory power supply 1 by opening the breaker contact 10

669-68567PCT / CM and slowly discharged via the load 8 by closing the fourth breaker contact 9.

If the capacitor bank circuit no longer has any charge, i. the voltmeter 5 indicates no voltage, the stepwise charging of the capacitor series circuit is repeated as described above. It is found that the time interval .DELTA.t.sub.i between the start of the charging process and the termination of the charging process at the voltage stage at which the charging current converges to a current value> 0 is already smaller than the time interval during the first charging passage.

At the same time, it is found that the voltage value at which a charging current not falling to 0 is detected is higher than in the previous charging pass.

In order to achieve an optimal result, the steps are: gradual charging of the capacitor until the charging current does not decrease to 0 and subsequent slow discharging over a suitable load repeated until the required time interval .DELTA.ti no longer decreases. Then the series connection of the electrolytic capacitors is completely reformed.

The so-transformed "Gold Cap" electrolytic capacitors have a new nominal voltage, which is about 50% higher than the nominal voltage specified by the manufacturer. It was also found that the post-formed electrolytic capacitors could charge much faster than non-reformed, otherwise identical

669-68567PCT / CM Electrolytic capacitors. It has also been found that after charging a reformed electrolytic capacitor to the new, ie higher rated voltage, an initial leakage current sets in, but at a voltage value which may be different depending on the capacitor, and about 10 to 20% below the new rated voltage, decreased to zero, ie no significant leakage current could be detected and the capacitor held the charge stored on it without significant charge losses over days.

The "gold cap" electrolytic capacitor treated with this embodiment of the inventive reforming method has a higher nominal voltage compared to commercially available, otherwise identical electrolytic capacitors, and thus a higher breakdown voltage value. Accordingly, more charge can be stored and the charging process takes place in a shorter time than this is not the case in accordance with the invention nachformierten capacitors.

Second embodiment of the method according to the invention

The electrolytic capacitor 4 to be treated is slowly discharged via the load resistor 8 in a first step. For this purpose, the laboratory power supply is disconnected by opening the second breaker contact 10 and the second and fourth breaker contacts 3 and 9 are closed. The electrolytic capacitor 4 used here is an aluminum electrolytic capacitor from Rescap (registered trademark) with the type number: 36DA105F016DJ2D with a nominal capacity of 1 F (± 20%) and a nominal voltage of 16 V.

669-68567PCT / CM The load resistor 8 may have a resistance value between 20 and 50 ohms.

After the voltage indicated on the voltmeter 5 or 6 has dropped to the millivolt range, the third breaker contact 7 is closed, whereby the electrolytic capacitor 4 is short-circuited. If the voltage measuring devices 5 and 6 no longer indicate voltage, the electrolytic capacitor 4 is deeply discharged.

In the next step, the laboratory power supply unit 1 is reconnected to the circuit and is first short-circuited by the closed third breaker contact 7. The fourth breaker contact 9 is open. Then the laboratory power supply is set to an output voltage of 10 V and the output current is limited to a value between 16 and 25 mA. Then, the third breaker switch 7 is opened and the electrolytic capacitor 4 is charged. It measures the time required for the capacitor to be charged to a voltage of 10V. The set voltage value of 10 V corresponds to approximately 2/3 of the rated voltage of the electrolytic capacitor 4. When the charging voltage of 10 V set on the laboratory power supply unit is reached, the charging current measured at the ammeter decreases to 0.

The steps:

a) slow unloading via a suitably sized load followed by shorting; and

669-68567PCT / CM b) Charging by applying a voltage VLI, which corresponds to about 2/3 of the nominal voltage Vo, with a limited to 25 mA DC current Iu until the voltage across the capacitor VLI corresponds and the charging current has dropped to 0, while measuring the time interval .DELTA.tl between Applying the charging voltage and reaching the charging current ILI = 0

are repeated in this order as often (step c)) until no shortening of the time intervals .DELTA.t required for this is more noticeable.

In the employed measurement series, this condition was achieved in the case of several identical electrolytic capacitors in each case after four to five repetitions of method steps a) and b). The loading times in the first run of process steps a) and b) were on the order of about 20 minutes, in a second pass, this time interval decreased to 15 minutes to require in each case a same time interval of 5 minutes in a fourth and fifth run.

It has thus been found that the time required to charge the electrolytic capacitor to a voltage of 10 V has shortened substantially (to about ¹ A) compared with the first charging after several repetitions, with a constant one after a few repetitions for the required time interval Δt Value sets.

It has also been found that after several cycles of forming cycles after shorting the slowly discharged

669-68567PCT / CM Electrolytic capacitor builds up at a disconnected laboratory power supply to the electrolytic capacitor 4, a slowly rising voltage, which reached a value of 5 V after 3 to 4 hours. The slow build-up of a voltage on a previously rapidly discharged capacitor is explained by dielectric absorption. In this case, charges migrate in the charged state of the electrolytic capacitor partly into the dielectric and have with rapid discharge not the opportunity to leave this again and therefore only gradually migrate back to the electrodes. So it builds up a tension again. The effect is comparable to magnetic remanence and is sometimes called the remanence of capacitors. The magnitude of the absorption is given in relation to the originally applied voltage and depends on the dielectric used. For aluminum electrolytic capacitors, values for dielectric absorption of about 10 to 15% of the originally applied voltage have been measured so far.

The recovered voltage detected here in the series of experiments had reached about 50% of the originally applied voltage at 5 V. This can not be explained by the known dielectric absorption in electrolytic capacitors alone.

When an aluminum electrolytic capacitor post-formed in the manner described above is not more than 10 V, i. charged with the aluminum electrolytic capacitor used to about 2/3 of its rated voltage, it showed no significant self-discharge for several days already.

669-68567PCT / CM In the following, a further reforming of the aluminum electrolytic capacitor is described according to a preferred embodiment of the present invention, in which the aluminum electrolytic capacitor is charged to its nominal voltage.

Accordingly, after step c), i. the repetition of steps a) and b) until no shortening of the time interval .DELTA.t is more detectable, the electrolytic capacitor in turn

d) slowly discharged via a suitably dimensioned load and then short-circuited. It follows

e) charging by applying a voltage VL ₂ which is equal to the nominal voltage Vo, with a charging current IL ₂ limited to a maximum of 200 mA, until the voltage at the capacitor corresponds to the charging voltage VL2 = Vo and the charging current IL _{2 has} dropped to 0 mA, wherein the time interval .DELTA.t2 between application of the charging voltage VL2 and the achievement of the charging current IL2 = 0 mA is measured, and

f) the steps d) and e) are repeated until no shortening of the time interval .DELTA.t2 is more detectable.

In the test series carried out, a three-time repetition of steps d) and e) was generally sufficient until a shortening of the time interval Δt2 was no longer detectable.

669-68567PCT / CM The thus nachformierte aluminum electrolytic capacitor could be charged to its rated voltage and kept this for several days without significant losses.

In a further preferred embodiment of the present invention, the electrolytic capacitor that has been reformed with the formation steps a) to f) has been further developed with the aim of increasing its maximum dielectric strength in order to increase its nominal voltage.

For this purpose, after completion of step f)

g) a charging voltage VL3 applied to the electrolytic capacitor, which is between 3 to 6% higher than the nominal voltage Vo, wherein the charging current IL3 is limited to a maximum of 170 mA until the measured at the ammeter charging current IL3 drops to 0 mA.

h) Subsequently, further charging steps are carried out analogously to step e), wherein in each case the charging voltage is increased by 3 to 6% of the rated voltage until the charging current IL3 no longer goes back to 0 mA.

The fact that the charge current IL3 no longer goes back to 0 is considered to indicate that an irreversible damage of the electrolytic capacitor takes place, in which the chemistry at the interface between the dielectric and the electrolyte is irreversibly changed. To avoid this or this harmful effect that the

669-68567PCT / CM Capacity of the electrolytic capacitor can be kept as small as possible, is in a final forming step

i) the charging current is switched off and the electrolytic capacitor is discharged to the voltage at which the charging current IL3 last decreased to 0 mA.

In the replicas thus made on aluminum electrolytic capacitors, the aluminum electrolytic capacitors could be charged to voltages of 20 to 22 V, so that the original rated voltage of 16 V had increased by 4 to 6 V, an increase of about 25 up to 30%.

By increasing the rated voltage, the electroformed capacitor thus reformulated can store 25 to 30% more charge.

Viewed the other way round, this means that you can dimension the electrolytic capacitor correspondingly smaller with a given capacity to be achieved.

In the first described embodiment of the inventive Nachformierungsverfahrens the charging voltage is increased gradually. It may be appropriate to limit the charging current at the same time. However, as explained above, this is not required under certain circumstances, such as very small voltage levels.

669-68567PCT / CM In the second Nachformierungsverfahren presented above, a high charging voltage is set at the power source, ie the laboratory power supply and limited from the outset, the charging current to a very low value (25 mA).

Both reforming methods have in common that it is important that the charging process is relatively slow. It can be assumed that this procedure favors a migration of ions or electrons into the dielectric and into the electrolyte and that the dielectric is positively influenced.

In the first presented Nachformierungsverfahren the voltage increase can also be continuous, wherein the criterion for the rate of voltage increase is the observed charging current. The voltage increase should be such that the charging current is very low, i. depending on the capacitor to be reformed on the order of a few tens of mA to a few hundred mA.

In addition, it should be noted that automated post-forming devices, which are integrated into the manufacturing process of the capacitors, can be designed to carry out the methods presented. The realization of such automated Nachformierungsvorrichtungen is a manageable task for the skilled person.

It should be noted that the process according to the invention for reforming electrolytic capacitors achieves that

669-68567PCT / CM the amount of time required to charge an electrolytic capacitor to a certain voltage is markedly shortened; lower charging currents are required to charge the electrolytic capacitor; the post-formed electrolytic capacitor has no appreciable leakage currents and dielectric losses, which are otherwise typical for conventional electrolytic capacitors, and the maximum withstand voltage of the nachformierten Elektrolytkondehsators can be increased, so that also the rated voltage is increased by 25 to 30%, whereby with the inventively nachformierten electrolytic capacitor 25th Up to 30% more charge can be stored, or for a given capacity of the electrolytic capacitor can be built smaller.

The above description of preferred embodiments of the present invention is provided for illustrative purposes only, and is not intended to be limiting. Various variations of individual features are possible within the scope of the appended claims. So it can be with the nachformierten

Electrolytic capacitors also tantalum electrolytic capacitors with tantalum pentoxide as a dielectric or niobium electrolytic capacitors with niobium pentoxide as a dielectric. Likewise, another suitable device for carrying out the method according to the invention for reforming electrolytic capacitors

669-68567PCT / CM be used as long as it allows the implementation of the method steps described above.

669-68567PCT / CM

Claims

claims

1. Method for reformulating electrolytic capacitors with the following steps:

a) if the electrolytic capacitor has a charge, slow discharge through a suitably sized load;

b) Charging the electrolytic capacitor by applying a stepwise increased voltage, wherein the voltage increase per stage is in each case a fraction of the nominal voltage of the electrolytic capacitor, wherein the voltage is increased in each case by one stage, when the charging current at the current stage has dropped substantially to zero and wherein the stepwise increase of the charging voltage is continued until the charging current does not decrease to zero.

2. The method according to claim 1, characterized by the following further steps:

c) measuring a time interval Δti between the beginning of the charging process and the time when the charging current no longer drops substantially to zero;

d) repeating steps a) and b) until the duration of the time interval Δti no longer decreases.

669-68567PCT / CM

3. The method according to any one of claims 1 or 2, characterized in that the stages of the voltage increase are not greater than 1/5 of the nominal voltage of the electrolytic capacitor.

4. The method according to any one of claims 1 or 2, characterized in that the stages of the voltage increase are not greater than 1/10 of the nominal voltage of the electrolytic capacitor.

5. The method according to any one of the preceding claims, characterized in that the charging current is limited by a current limiter.

6. The method according to claim 5, characterized in that the charging current is limited to a value between a few tens of mA to a few hundred mA.

7. The method according to any one of the preceding claims, characterized in that the charging of the electrolytic capacitor takes place at a voltage level until the charging current has dropped to less than a few milliamperes and then the charging voltage is increased by one stage.

8. The method according to any one of the preceding claims, characterized in that the voltage levels are all the same size.

669-68567PCT / CM

9. The method according to any one of claims 1 to 7, characterized in that the voltage levels are greater from stage to stage.

10. The method according to any one of the preceding claims, characterized in that the electrolytic capacitor is short-circuited after step a).

11. A method of reforming electrolytic capacitors, comprising the steps of:

a) slow unloading via a suitably sized load;

b) charging by applying a voltage VLI which corresponds to about 2/3 of the nominal voltage Vo with a limited to some 10 mA to a few hundred mA limited direct current Li until the voltage across the capacitor VLI and the charging current has dropped to 0, the time interval Δtl is measured between application of the charging voltage and reaching the charging current 0 mA; and

c) repeating steps a) and b) until no shortening of the time interval .DELTA.tl is more detectable.

12. A method for reformulating electrolytic capacitors according to claim 1 1, characterized in that

669-68567PCT / CM After completion of step c), the following further steps take place:

d) slow unloading via a suitably sized load;

e) charging by applying a voltage VL2 which is equal to the nominal voltage Vo, with a charging current IL2 limited to a maximum of 200 mA, until the voltage at the capacitor corresponds to the charging voltage VL2 = Vo and the charging current IL2 has dropped to 0 mA, the time interval Δt2 is measured between application of the charging voltage VL2 and reaching the charging current IL2 = 0 mA;

f) repeating steps d) and e) until no shortening of the time interval .DELTA.t2 is more detectable.

13. The method according to claim 12, characterized in that

g) after completion of step f), a charging voltage VL3 is applied which is between 3 to 6% higher than the rated voltage Vo, the charging current IL3 being limited to a maximum of 170 mA until the charging current IL3 drops to 0 mA,

h) subsequently, further charging steps take place analogously to step e); wherein each of the charging voltage is increased by 3 to 6% of the rated voltage until the charging current IL3 no longer goes back to 0 mA;

669-68567PCT / CM i) Switch off the charging current and discharge the electrolytic capacitor to the voltage at which the charging current IL3 last decreased to 0 mA.

14. The method according to claim 4, characterized in that the electrolytic capacitor is charged to a new nominal voltage of 20 to 22 volts.

15. The method according to any one of the preceding claims, characterized in that at the end of the slow discharge of the electrolytic capacitor, this is short-circuited.

16. The method according to any one of the preceding claims, characterized in that the charging current is limited to a maximum of 25 mA.

17. The method according to any one of claims 1 to 15, characterized in that the charging current is limited to a maximum of 200 mA.

69-68567PCT / CM