supercap leakage

[John Larkin]

https://dl.dropboxusercontent.com/u/53724080/Parts/Caps/Supercap_Leakage.JPG

Looks like it would be fine to stack these in series without any
equalizing parts, like regular wet aluminum caps.

Measuring supercaps is an enormous pain; every step takes hours. The
equivalent dielectric absorption is huge.

Leakage tempco appears to be strongly negative, but that may be an
illusion.


--

John Larkin Highland Technology, Inc
picosecond timing precision measurement

jlarkin att highlandtechnology dott com
http://www.highlandtechnology.com

 

[Helmut Wabnig]

On Wed, 24 May 2017 14:34:23 -0700, John Larkin
<jjlarkin@highland_snip_technology.com> wrote:

https://dl.dropboxusercontent.com/u/53724080/Parts/Caps/Supercap_Leakage.JPG

Looks like it would be fine to stack these in series without any
equalizing parts, like regular wet aluminum caps.

Measuring supercaps is an enormous pain; every step takes hours. The
equivalent dielectric absorption is huge.

Leakage tempco appears to be strongly negative, but that may be an
illusion.

They self-discharge quickly, as shown in the diagram?

w.

 

[rickman]

Helmut Wabnig wrote on 5/25/2017 1:10 AM:

On Wed, 24 May 2017 14:34:23 -0700, John Larkin
jjlarkin@highland_snip_technology.com> wrote:


https://dl.dropboxusercontent.com/u/53724080/Parts/Caps/Supercap_Leakage.JPG

Looks like it would be fine to stack these in series without any
equalizing parts, like regular wet aluminum caps.

Measuring supercaps is an enormous pain; every step takes hours. The
equivalent dielectric absorption is huge.

Leakage tempco appears to be strongly negative, but that may be an
illusion.

They self-discharge quickly, as shown in the diagram?

Wow! That's enough leakage to make them useless for micropower energy
harvesting. Good to know.

--

Rick C

 

[John Larkin]

On Thu, 25 May 2017 07:10:30 +0200, Helmut Wabnig <hwabnig@.- ---
-.dotat> wrote:

On Wed, 24 May 2017 14:34:23 -0700, John Larkin
jjlarkin@highland_snip_technology.com> wrote:


https://dl.dropboxusercontent.com/u/53724080/Parts/Caps/Supercap_Leakage.JPG

Looks like it would be fine to stack these in series without any
equalizing parts, like regular wet aluminum caps.

Measuring supercaps is an enormous pain; every step takes hours. The
equivalent dielectric absorption is huge.

Leakage tempco appears to be strongly negative, but that may be an
illusion.

They self-discharge quickly, as shown in the diagram?

w.

Nothing happens quickly on a 10F capacitor.

At the rated 2.7 volts, leakage is spec'd at 30 uA max. That's a
discharge rate of 3 microvolts per second.

The leakage is specified after 72 hours at voltage. I didn't wait that
long.




--

John Larkin Highland Technology, Inc

lunatic fringe electronics

 

[o pere o]

On 25/05/17 07:53, John Larkin wrote:

On Thu, 25 May 2017 07:10:30 +0200, Helmut Wabnig <hwabnig@.- ---
-.dotat> wrote:

On Wed, 24 May 2017 14:34:23 -0700, John Larkin
jjlarkin@highland_snip_technology.com> wrote:


https://dl.dropboxusercontent.com/u/53724080/Parts/Caps/Supercap_Leakage.JPG

Looks like it would be fine to stack these in series without any
equalizing parts, like regular wet aluminum caps.

Measuring supercaps is an enormous pain; every step takes hours. The
equivalent dielectric absorption is huge.

Leakage tempco appears to be strongly negative, but that may be an
illusion.

They self-discharge quickly, as shown in the diagram?

w.

Nothing happens quickly on a 10F capacitor.

At the rated 2.7 volts, leakage is spec'd at 30 uA max. That's a
discharge rate of 3 microvolts per second.

The leakage is specified after 72 hours at voltage. I didn't wait that
long.

Thanks for sharing this!

Interesting: At 2.7V specified 30uA, measured 120uA. A x4 factor here...

Pere

 

[o pere o]

On 25/05/17 09:36, o pere o wrote:

On 25/05/17 07:53, John Larkin wrote:
On Thu, 25 May 2017 07:10:30 +0200, Helmut Wabnig <hwabnig@.- ---
-.dotat> wrote:

On Wed, 24 May 2017 14:34:23 -0700, John Larkin
jjlarkin@highland_snip_technology.com> wrote:


https://dl.dropboxusercontent.com/u/53724080/Parts/Caps/Supercap_Leakage.JPG


Looks like it would be fine to stack these in series without any
equalizing parts, like regular wet aluminum caps.

Measuring supercaps is an enormous pain; every step takes hours. The
equivalent dielectric absorption is huge.

Leakage tempco appears to be strongly negative, but that may be an
illusion.

They self-discharge quickly, as shown in the diagram?

w.

Nothing happens quickly on a 10F capacitor.

At the rated 2.7 volts, leakage is spec'd at 30 uA max. That's a
discharge rate of 3 microvolts per second.

The leakage is specified after 72 hours at voltage. I didn't wait that
long.

Thanks for sharing this!

Interesting: At 2.7V specified 30uA, measured 120uA. A x4 factor here...

Pere

Is this really going to be better after being 3 days there?

Pere

 

[Jan Panteltje]

On a sunny day (Thu, 25 May 2017 09:36:51 +0200) it happened o pere o
<me@somewhere.net> wrote in <og61ep$f9c$1@dont-email.me>:

On 25/05/17 07:53, John Larkin wrote:
On Thu, 25 May 2017 07:10:30 +0200, Helmut Wabnig <hwabnig@.- ---
-.dotat> wrote:

On Wed, 24 May 2017 14:34:23 -0700, John Larkin
jjlarkin@highland_snip_technology.com> wrote:


https://dl.dropboxusercontent.com/u/53724080/Parts/Caps/Supercap_Leakage.JPG

Looks like it would be fine to stack these in series without any
equalizing parts, like regular wet aluminum caps.

Measuring supercaps is an enormous pain; every step takes hours. The
equivalent dielectric absorption is huge.

Leakage tempco appears to be strongly negative, but that may be an
illusion.

They self-discharge quickly, as shown in the diagram?

w.

Nothing happens quickly on a 10F capacitor.

At the rated 2.7 volts, leakage is spec'd at 30 uA max. That's a
discharge rate of 3 microvolts per second.

The leakage is specified after 72 hours at voltage. I didn't wait that
long.

Thanks for sharing this!

Interesting: At 2.7V specified 30uA, measured 120uA. A x4 factor here...

Pere

If we need ANY regulations at all, then I think
it is about the highest time we impose regulations on China
to live up to their specs.
And force them to write decent English specs,
where they for example have at leat the slightest idea what they talk about.
It is in their own interests.
Else no choice other than to avoid that country's products,
no matter how much cheaper those are than say US or EU made stuff.

 

[default]

On Thu, 25 May 2017 01:30:22 -0400, rickman <gnuarm@gmail.com> wrote:

Helmut Wabnig wrote on 5/25/2017 1:10 AM:
On Wed, 24 May 2017 14:34:23 -0700, John Larkin
jjlarkin@highland_snip_technology.com> wrote:


https://dl.dropboxusercontent.com/u/53724080/Parts/Caps/Supercap_Leakage.JPG

Looks like it would be fine to stack these in series without any
equalizing parts, like regular wet aluminum caps.

Measuring supercaps is an enormous pain; every step takes hours. The
equivalent dielectric absorption is huge.

Leakage tempco appears to be strongly negative, but that may be an
illusion.

They self-discharge quickly, as shown in the diagram?

That isn't what the diagram shows.

Wow! That's enough leakage to make them useless for micropower energy
harvesting. Good to know.

They are excellent for micro power applications, they do not
self-discharge rapidly

Way more effective in some dawn to dusk solar powered lights where I
replaced the nicad or nmh batteries - especially when you take into
consideration that the batteries are often discharged well past their
safe levels in normal use.

 

[Tim Williams]

"John Larkin" <jjlarkin@highlandtechnology.com> wrote in message
news:80scic1nanh3ej836d41nl6ostnhogevde@4ax.com...

At the rated 2.7 volts, leakage is spec'd at 30 uA max. That's a
discharge rate of 3 microvolts per second.

The leakage is specified after 72 hours at voltage. I didn't wait that
long.

Well how long did you wait, then?

You could've saved us the time and deleted the picture, because it's
meaningless.

Tim

--
Seven Transistor Labs, LLC
Electrical Engineering Consultation and Contract Design
Website: http://seventransistorlabs.com

 

[Phil Hobbs]

On 05/25/2017 03:40 AM, o pere o wrote:

On 25/05/17 09:36, o pere o wrote:
On 25/05/17 07:53, John Larkin wrote:
On Thu, 25 May 2017 07:10:30 +0200, Helmut Wabnig <hwabnig@.- ---
-.dotat> wrote:

On Wed, 24 May 2017 14:34:23 -0700, John Larkin
jjlarkin@highland_snip_technology.com> wrote:


https://dl.dropboxusercontent.com/u/53724080/Parts/Caps/Supercap_Leakage.JPG


Looks like it would be fine to stack these in series without any
equalizing parts, like regular wet aluminum caps.

Measuring supercaps is an enormous pain; every step takes hours. The
equivalent dielectric absorption is huge.

Leakage tempco appears to be strongly negative, but that may be an
illusion.

They self-discharge quickly, as shown in the diagram?

w.

Nothing happens quickly on a 10F capacitor.

At the rated 2.7 volts, leakage is spec'd at 30 uA max. That's a
discharge rate of 3 microvolts per second.

The leakage is specified after 72 hours at voltage. I didn't wait that
long.

Thanks for sharing this!

Interesting: At 2.7V specified 30uA, measured 120uA. A x4 factor here...

Pere

Is this really going to be better after being 3 days there?

Pere

Could well be. AlPo caps behave like that.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs
Principal Consultant
ElectroOptical Innovations LLC
Optics, Electro-optics, Photonics, Analog Electronics

160 North State Road #203
Briarcliff Manor NY 10510

hobbs at electrooptical dot net
http://electrooptical.net

 

[John Larkin]

On Thu, 25 May 2017 09:40:09 +0200, o pere o <me@somewhere.net> wrote:

On 25/05/17 09:36, o pere o wrote:
On 25/05/17 07:53, John Larkin wrote:
On Thu, 25 May 2017 07:10:30 +0200, Helmut Wabnig <hwabnig@.- ---
-.dotat> wrote:

On Wed, 24 May 2017 14:34:23 -0700, John Larkin
jjlarkin@highland_snip_technology.com> wrote:


https://dl.dropboxusercontent.com/u/53724080/Parts/Caps/Supercap_Leakage.JPG


Looks like it would be fine to stack these in series without any
equalizing parts, like regular wet aluminum caps.

Measuring supercaps is an enormous pain; every step takes hours. The
equivalent dielectric absorption is huge.

Leakage tempco appears to be strongly negative, but that may be an
illusion.

They self-discharge quickly, as shown in the diagram?

w.

Nothing happens quickly on a 10F capacitor.

At the rated 2.7 volts, leakage is spec'd at 30 uA max. That's a
discharge rate of 3 microvolts per second.

The leakage is specified after 72 hours at voltage. I didn't wait that
long.

Thanks for sharing this!

Interesting: At 2.7V specified 30uA, measured 120uA. A x4 factor here...

Pere

Is this really going to be better after being 3 days there?

Pere

Could be. Things take a long time to settle, the equivalent of some
very slow dielectric absorption. If I step up the voltage, the power
supply current limits for a while, and when the voltage hits the
target, the current tapers off slowly, over hours and probably days.

These things take some getting used to. Temperature changes C, and
when you change a charged 10F cap connected to a power supply, things
happen. So measuring leakage is complex.




--

John Larkin Highland Technology, Inc

lunatic fringe electronics

 

[John Larkin]

On Thu, 25 May 2017 08:16:40 -0500, "Tim Williams"
<tiwill@seventransistorlabs.com> wrote:

"John Larkin" <jjlarkin@highlandtechnology.com> wrote in message
news:80scic1nanh3ej836d41nl6ostnhogevde@4ax.com...
At the rated 2.7 volts, leakage is spec'd at 30 uA max. That's a
discharge rate of 3 microvolts per second.

The leakage is specified after 72 hours at voltage. I didn't wait that
long.

Well how long did you wait, then?

You could've saved us the time and deleted the picture, because it's
meaningless.

Tim

My apologies for wasting your valuable time. Send me an invoice.


--

John Larkin Highland Technology, Inc

lunatic fringe electronics

 

[Chris Jones]

On 25/05/2017 07:34, John Larkin wrote:

https://dl.dropboxusercontent.com/u/53724080/Parts/Caps/Supercap_Leakage.JPG

Looks like it would be fine to stack these in series without any
equalizing parts, like regular wet aluminum caps.

How do you conclude that? Wouldn't it depend on
* The variation in leakage current between units, at the working voltage
* How much leakage charge inflicted at above rated voltage is required
to do damage to the least leaky cap in the string
* What voltage you are using the string at, compared to the sum of the
individual voltage ratings
* the capacitance matching (assuming you are going to charge and
discharge them rather than just letting them sit there)

Put another way, could you not draw the same conclusion if you measured
the self-discharge of a single lithium battery cell vs. voltage? Yet
people keep putting balancing circuits on them. I'm sure there is a reason.

 

[rickman]

default wrote on 5/25/2017 7:01 AM:

On Thu, 25 May 2017 01:30:22 -0400, rickman <gnuarm@gmail.com> wrote:

Helmut Wabnig wrote on 5/25/2017 1:10 AM:
On Wed, 24 May 2017 14:34:23 -0700, John Larkin
jjlarkin@highland_snip_technology.com> wrote:


https://dl.dropboxusercontent.com/u/53724080/Parts/Caps/Supercap_Leakage.JPG

Looks like it would be fine to stack these in series without any
equalizing parts, like regular wet aluminum caps.

Measuring supercaps is an enormous pain; every step takes hours. The
equivalent dielectric absorption is huge.

Leakage tempco appears to be strongly negative, but that may be an
illusion.

They self-discharge quickly, as shown in the diagram?

That isn't what the diagram shows.

Wow! That's enough leakage to make them useless for micropower energy
harvesting. Good to know.

They are excellent for micro power applications, they do not
self-discharge rapidly

Way more effective in some dawn to dusk solar powered lights where I
replaced the nicad or nmh batteries - especially when you take into
consideration that the batteries are often discharged well past their
safe levels in normal use.

That's not micropower. The issue is not if they self-discharge rapidly.
The issue is if they can be charged up at all given the low power
source they are being charged from. JL's diagram shows higher leakage
current than the application I would be using these in.

--

Rick C

 

[John Larkin]

On Fri, 26 May 2017 00:30:00 +1000, Chris Jones
<lugnut808@spam.yahoo.com> wrote:

On 25/05/2017 07:34, John Larkin wrote:

https://dl.dropboxusercontent.com/u/53724080/Parts/Caps/Supercap_Leakage.JPG

Looks like it would be fine to stack these in series without any
equalizing parts, like regular wet aluminum caps.

How do you conclude that? Wouldn't it depend on
* The variation in leakage current between units, at the working voltage
* How much leakage charge inflicted at above rated voltage is required
to do damage to the least leaky cap in the string
* What voltage you are using the string at, compared to the sum of the
individual voltage ratings
* the capacitance matching (assuming you are going to charge and
discharge them rather than just letting them sit there)

Put another way, could you not draw the same conclusion if you measured
the self-discharge of a single lithium battery cell vs. voltage? Yet
people keep putting balancing circuits on them. I'm sure there is a reason.

In these caps, current increases nonlinearly, radically upward, with
voltage, so a series pair will tend to equalize. In fact, a series
pair that I tried did equalize nicely. And overvoltage makes them leak
a lot but doesn't appear to damage them.

It's dangerous to put film caps in series to increase voltage rating.
Ditto some polymer aluminums and ceramic caps. Wet aluminums, and this
supercap, look to be self-equalizing.

Batteries are different, because the first one that runs out of
coulombs can be reverse-biased by the others, and damaged. These
supercaps are really unpolarized, so won't be damaged by reverse bias.
And they are real capacitors, so have a linear discharge curve, not
flat-topped like batteries; that makes the caps in the stack tend to
have the same voltage across them.




--

John Larkin Highland Technology, Inc

lunatic fringe electronics

 

[Jan Panteltje]

On a sunny day (Thu, 25 May 2017 08:23:31 -0700) it happened John Larkin
<jjlarkin@highlandtechnology.com> wrote in
<d5tdic1fvqinvqvcjupvk6ggrc801uom6p@4ax.com>:

On Fri, 26 May 2017 00:30:00 +1000, Chris Jones
lugnut808@spam.yahoo.com> wrote:

On 25/05/2017 07:34, John Larkin wrote:

https://dl.dropboxusercontent.com/u/53724080/Parts/Caps/Supercap_Leakage.JPG

Looks like it would be fine to stack these in series without any
equalizing parts, like regular wet aluminum caps.

How do you conclude that? Wouldn't it depend on
* The variation in leakage current between units, at the working voltage
* How much leakage charge inflicted at above rated voltage is required
to do damage to the least leaky cap in the string
* What voltage you are using the string at, compared to the sum of the
individual voltage ratings
* the capacitance matching (assuming you are going to charge and
discharge them rather than just letting them sit there)

Put another way, could you not draw the same conclusion if you measured
the self-discharge of a single lithium battery cell vs. voltage? Yet
people keep putting balancing circuits on them. I'm sure there is a reason.

In these caps, current increases nonlinearly, radically upward, with
voltage, so a series pair will tend to equalize. In fact, a series
pair that I tried did equalize nicely. And overvoltage makes them leak
a lot but doesn't appear to damage them.

It's dangerous to put film caps in series to increase voltage rating.
Ditto some polymer aluminums and ceramic caps. Wet aluminums, and this
supercap, look to be self-equalizing.

Batteries are different, because the first one that runs out of
coulombs can be reverse-biased by the others, and damaged. These
supercaps are really unpolarized, so won't be damaged by reverse bias.
And they are real capacitors, so have a linear discharge curve, not
flat-topped like batteries; that makes the caps in the stack tend to
have the same voltage across them.

Indeed, lipos will just go up in voltage if over charged.
The 'equalizers' are sometimes just a parellel load system,
like a zener, per cell, preventng the cell from going over 4.2 V.
That way everything gets charged.

 

[rickman]

Tim Williams wrote on 5/25/2017 9:16 AM:

"John Larkin" <jjlarkin@highlandtechnology.com> wrote in message
news:80scic1nanh3ej836d41nl6ostnhogevde@4ax.com...
At the rated 2.7 volts, leakage is spec'd at 30 uA max. That's a
discharge rate of 3 microvolts per second.

The leakage is specified after 72 hours at voltage. I didn't wait that
long.

Well how long did you wait, then?

You could've saved us the time and deleted the picture, because it's
meaningless.

It's far from meaningless. It is just a measurement under different
conditions. If your usage won't be for storing charge for over 72 hours
this is a measurement that is more to your conditions.

--

Rick C

 

[John Larkin]

On Thu, 25 May 2017 15:27:49 GMT, Jan Panteltje
<pNa0nStpealmtje@yahoo.com> wrote:

On a sunny day (Thu, 25 May 2017 08:23:31 -0700) it happened John Larkin
jjlarkin@highlandtechnology.com> wrote in
d5tdic1fvqinvqvcjupvk6ggrc801uom6p@4ax.com>:

On Fri, 26 May 2017 00:30:00 +1000, Chris Jones
lugnut808@spam.yahoo.com> wrote:

On 25/05/2017 07:34, John Larkin wrote:

https://dl.dropboxusercontent.com/u/53724080/Parts/Caps/Supercap_Leakage.JPG

Looks like it would be fine to stack these in series without any
equalizing parts, like regular wet aluminum caps.

How do you conclude that? Wouldn't it depend on
* The variation in leakage current between units, at the working voltage
* How much leakage charge inflicted at above rated voltage is required
to do damage to the least leaky cap in the string
* What voltage you are using the string at, compared to the sum of the
individual voltage ratings
* the capacitance matching (assuming you are going to charge and
discharge them rather than just letting them sit there)

Put another way, could you not draw the same conclusion if you measured
the self-discharge of a single lithium battery cell vs. voltage? Yet
people keep putting balancing circuits on them. I'm sure there is a reason.

In these caps, current increases nonlinearly, radically upward, with
voltage, so a series pair will tend to equalize. In fact, a series
pair that I tried did equalize nicely. And overvoltage makes them leak
a lot but doesn't appear to damage them.

It's dangerous to put film caps in series to increase voltage rating.
Ditto some polymer aluminums and ceramic caps. Wet aluminums, and this
supercap, look to be self-equalizing.

Batteries are different, because the first one that runs out of
coulombs can be reverse-biased by the others, and damaged. These
supercaps are really unpolarized, so won't be damaged by reverse bias.
And they are real capacitors, so have a linear discharge curve, not
flat-topped like batteries; that makes the caps in the stack tend to
have the same voltage across them.

Indeed, lipos will just go up in voltage if over charged.
The 'equalizers' are sometimes just a parellel load system,
like a zener, per cell, preventng the cell from going over 4.2 V.
That way everything gets charged.

Zeners protect them in the reverse direction, too.


--

John Larkin Highland Technology, Inc

lunatic fringe electronics

 

[Jan Panteltje]

On a sunny day (Thu, 25 May 2017 08:32:12 -0700) it happened John Larkin
<jjlarkin@highlandtechnology.com> wrote in
<96udicd921blfhqj744ho2aoguc9hvrg5n@4ax.com>:

On Thu, 25 May 2017 15:27:49 GMT, Jan Panteltje
pNa0nStpealmtje@yahoo.com> wrote:

Indeed, lipos will just go up in voltage if over charged.
The 'equalizers' are sometimes just a parellel load system,
like a zener, per cell, preventng the cell from going over 4.2 V.
That way everything gets charged.

Zeners protect them in the reverse direction, too.

Real zeners would, although with -.7 or so volts I would not want to use those cells again...
But usually you are talking about several amps charge current,
and like Tim posted here some year ago,
a transistor with a voltage sensor shorting out / limiting the voltage can be used.
If it was a MOSFET with a reverse body diode, yes.
Would have to be a logic level type that works far below 4.2 V Vgs on.
Bipolar is nice here.
No GHz

 

[Jan Panteltje]

On a sunny day (Thu, 25 May 2017 11:29:53 -0400) it happened rickman
<gnuarm@gmail.com> wrote in <og6t35$cjp$1@dont-email.me>:

Tim Williams wrote on 5/25/2017 9:16 AM:
"John Larkin" <jjlarkin@highlandtechnology.com> wrote in message
news:80scic1nanh3ej836d41nl6ostnhogevde@4ax.com...
At the rated 2.7 volts, leakage is spec'd at 30 uA max. That's a
discharge rate of 3 microvolts per second.

The leakage is specified after 72 hours at voltage. I didn't wait that
long.

Well how long did you wait, then?

You could've saved us the time and deleted the picture, because it's
meaningless.

It's far from meaningless. It is just a measurement under different
conditions. If your usage won't be for storing charge for over 72 hours
this is a measurement that is more to your conditions.

You may like this, guy leaves supercaps in over the weekend,
car starts with it:
https://www.youtube.com/watch?v=z3x_kYq3mHM

No battery..