The end of the series capacitor saga.

If the capacitor values can be held to within 25% of each other, then the 
bleed resistors needed to set the capacitor voltages to 50% of VCC can be 
reduced to 1.6W each.

The rational is the typical leakage spec is a maximum value. For example,

I = 0.01 * C * V (in microamps)
  = 0.01 * 1000 * 450
  = 4.5mA

To allow worst-case conditions, the industry standard is to use 10 times 
this current for the bleed resistors. This results in high power dissipated 
in the bleed resistors.

However, manufacturers try to minimize the leakage current, and often 
produce capacitors with leakage currents one hundred times lower.

This means the bleed resistor values can be much higher and the power 
dissipated can be much lower.

This also allows a simple way to discharge the capacitor string within 2 
seconds after power off. Simply use a relay to switch a dump resistor 
across the supply voltage when power is turned off. When power is on, the 
dump resistor is switched off and does not load the supply. As soon as 
power is turned off, the relay turns off and the normally closed contacts 
switch the dump resistor across the supply.

Cheap dump diodes (1N4007) are still needed to prevent reverse voltage on 
the caps, which can damage them. 

Here is the ASC file that shows how this works.

Version 4
SHEET 1 1532 0
WIRE 752 -256 688 -256
WIRE 768 -256 752 -256
WIRE 864 -256 832 -256
WIRE 928 -256 864 -256
WIRE 1088 -256 928 -256
WIRE 1248 -256 1088 -256
WIRE 1472 -256 1248 -256
WIRE 688 -240 688 -256
WIRE 928 -240 928 -256
WIRE 1088 -240 1088 -256
WIRE 1248 -240 1248 -256
WIRE 1472 -176 1472 -256
WIRE 688 -144 688 -160
WIRE 928 -144 928 -176
WIRE 1008 -144 928 -144
WIRE 1088 -144 1088 -176
WIRE 1088 -144 1008 -144
WIRE 1248 -144 1248 -160
WIRE 1248 -144 1088 -144
WIRE 1248 -128 1248 -144
WIRE 928 -112 928 -144
WIRE 1088 -112 1088 -144
WIRE 1472 -48 1472 -96
WIRE 928 -32 928 -48
WIRE 1088 -32 1088 -48
WIRE 1248 -32 1248 -48
FLAG 688 -144 0
FLAG 928 -32 0
FLAG 1248 -32 0
FLAG 864 -256 D1C1
FLAG 1008 -144 C1C2
FLAG 752 -256 V1
FLAG 1088 -32 0
FLAG 1472 -48 0
SYMBOL voltage 688 -256 R0
WINDOW 123 0 0 Left 2
WINDOW 3 -61 161 Invisible 2
WINDOW 39 0 0 Left 2
SYMATTR Value PULSE(0 800 1 1 10m 50 100 1)
SYMATTR InstName V1
SYMBOL cap 912 -240 R0
SYMATTR InstName C1
SYMATTR Value 1250uf
SYMBOL cap 912 -112 R0
SYMATTR InstName C2
SYMATTR Value 1000uf
SYMBOL res 1232 -144 R0
SYMATTR InstName R2
SYMATTR Value 100k
SYMBOL diode 768 -240 R270
WINDOW 0 32 32 VTop 2
WINDOW 3 0 32 VBottom 2
SYMATTR InstName D1
SYMATTR Value 1N4007
SYMBOL diode 1072 -48 M180
WINDOW 0 24 64 Left 2
WINDOW 3 24 0 Left 2
SYMATTR InstName D3
SYMATTR Value 1N4007
SYMBOL diode 1072 -176 M180
WINDOW 0 24 64 Left 2
WINDOW 3 24 0 Left 2
SYMATTR InstName D2
SYMATTR Value 1N4007
SYMBOL res 1232 -256 R0
SYMATTR InstName R1
SYMATTR Value 100k
SYMBOL res 1456 -192 R0
SYMATTR InstName Rd
SYMATTR Value 1k
TEXT 720 -352 Left 2 !.tran 0 100 0 1m
TEXT 720 -384 Left 2 ;'Capacitor Discharge
TEXT 1016 -352 Left 2 !.ic V(c1c2) = 0

Steve Wilson <no@spam.com> wrote:

>>> Does ripple damage caps?

>> Of course it can if it too high, or causes reverse current through the
>> cap at or near zero volts. 

> A bit more explanation. When the cap voltage goes negative, the current 
> through the cap reverses. 

> This causes the formation of a thin aluminum oxide layer on the cathode.

> This causes another capacitor to form on the cathode. This capacitor is
> in series with the capacitor on the anode, so the total capacitance is 
> reduced.

> So passing current through the cap when the voltage is near zero is a
> bad idea.

> The dump diode will prevent this from happening. It bypasses the current
> from going through the cap.

Best Supporting Reference:

https://www.mouser.com/pdfDocs/UCC_ElectrolyticCapacitorTechnicalNotes.pdf

Page 10

Applying a reverse voltage will cause chemical reactions (formation  of 
dielectric) to occur on the cathode foil, and, as is the case with  
overvoltage, the leakage current will rapidly increase with heat and  gases 
generating and thus the internal pressure increases. The reactions are 
accelerated by the voltage, current density  and ambient temperature. It 
may also accompany a reduction in  capacitance and an increase in tan h as 
well as an increase in  the leakage current. An example of capacitor 
reverse-voltage  characteristics is shown in Fig. 27. A reverse voltage of 
as small as 1V can cause the capacitance to  decrease. A reverse voltage of 
2 to 3V can shorten lifetime due to  a reduction in capacitance, increase 
in tan h, and/or increase in  leakage current. A reverse voltage of even 
higher value can open  the pressure relief vent or lead to destructive 
failures (Fig. 27).

NOTES:

There are many references on the web to reversed voltage on electrolytic 
capacitors. Many of them are just plain silly, such as the production of 
oxygen on the cathode. This is impossible due to the electrolysis equation:

2Al + 3H2O --> Al2O3 + 3H2 (Gas) + 3e

This shows that hydrogen is produced, not oxygen.

However, the web links almost universally indicate that reversing the 
polarity of an aluminum electrolytic cap will damage it.

Also, the original PDF would not allow copying the above paragraph. I 
simply passed it to "Remove restrictions in PDF files", at

https://online2pdf.com/remove-pdf-restrictions

Now I can copy information freely from the PDF.

Steve Wilson <no@spam.com> wrote:

> jlarkin@highlandsniptechnology.com wrote:

>> On Thu, 24 Oct 2019 16:06:54 -0000 (UTC), Steve Wilson <no@spam.com>
>> wrote: 
>>>> I'll put electrolytic caps in series and not worry.
 
>>>1. You have no discharge path when power is turned off. Plenty of loads
>>>do not have constant power drain. This can leave dangerous voltages on
>>>the caps. 
  
>> I have already noted the obvious fact that it's sometimes reasonable
>> to discharge the entire series string, as if it were a single cap. 
 
> Then you need a dump diode. 
 
>>>2. You can have reverse current on the lower cap when there is a
>>>constant load and power is turned off. Reverse current is bad for
>>>electrolytics. 
 
>> Reverse current? Caps always have "reverse current" during discharge.
>> The current is equal everywhere in a series circuit. All series caps
>> have the same current. 
 
> When you have mismatched caps with the higher value on top, the top cap 
> will try to push the voltage on the bottom cap negative. A dump diode
> will prevent the bottom cap from going negative.
 
> This doesn't help when the top cap has a lower value than the bottom
> one. 
 
> As I pointed out, the capacitor tolerance is -50% to +100%, so your
> chances of getting matched caps is very low.
  
>> Does ripple damage caps?
 
> Of course it can if it too high, or causes reverse current through the
> cap at or near zero volts.

A bit more explanation. When the cap voltage goes negative, the current 
through the cap reverses. 

This causes the formation of a thin aluminum oxide layer on the cathode.

This causes another capacitor to form on the cathode. This capacitor is in 
series with the capacitor on the anode, so the total capacitance is 
reduced.

So passing current through the cap when the voltage is near zero is a bad 
idea.

The dump diode will prevent this from happening. It bypasses the current 
from going through the cap.

jlarkin@highlandsniptechnology.com wrote:

> On Thu, 24 Oct 2019 16:06:54 -0000 (UTC), Steve Wilson <no@spam.com>
> wrote:
>>> I'll put electrolytic caps in series and not worry.

>>1. You have no discharge path when power is turned off. Plenty of loads
>>do not have constant power drain. This can leave dangerous voltages on
>>the caps. 

> I have already noted the obvious fact that it's sometimes reasonable
> to discharge the entire series string, as if it were a single cap.

Then you need a dump diode. 

>>2. You can have reverse current on the lower cap when there is a
>>constant load and power is turned off. Reverse current is bad for
>>electrolytics. 

> Reverse current? Caps always have "reverse current" during discharge.
> The current is equal everywhere in a series circuit. All series caps
> have the same current.

When you have mismatched caps with the higher value on top, the top cap 
will try to push the voltage on the bottom cap negative. A dump diode will 
prevent the bottom cap from going negative.

This doesn't help when the top cap has a lower value than the bottom one.

As I pointed out, the capacitor tolerance is -50% to +100%, so your chances 
of getting matched caps is very low.

> Does ripple damage caps?

Of course it can if it too high, or causes reverse current through the cap 
at or near zero volts.

On Thu, 24 Oct 2019 16:06:54 -0000 (UTC), Steve Wilson <no@spam.com>
wrote:

>jlarkin@highlandsniptechnology.com wrote:
>
>> On Thu, 24 Oct 2019 06:01:55 -0000 (UTC), Steve Wilson <no@spam.com>
>> wrote:
>
>>>I noticed a problem with the sym. The line that reads
>
>>>TEXT 1016 -352 Left 2 !.ic V(c1c2) @0
>
>>>should read
>
>>>TEXT 1016 -352 Left 2 !.ic V(c1c2) =400
>
>>>If you try to run the sym, you will get an error:
>
>>>"Trouble parsing initial condition value for V(c1c2)"
>
>>>The fact you never reported an error says you never ran the sym.
> 
>> If you ran your own sim, why did you not fix the error?
>
>It didn't occur on my copy. Only on the posted version.
> 
>>>Hardly any point in discussing the results of the sym if you never run
>>>it. 
> 
>> I ran your first one and decided that it was silly. Which it certainly
>> was. -330 volts across an electrolytic cap!
>
>As I have already pointed out, the bottom cap had a typo. 
>
>Instead of 1,000uF, it was 100uF. Obviously something was wrong with the 
>sym. It took a bit of sleuthing to find the problem.
>
>You should have noticed this when you looked at it.
>
>> OK, you buy a bunch of giant wirewound resistors, mounting brackets
>> for same, diodes, fans, blast shields, warning signs, covers with
>> interlock switches, safety goggles, rubber glove dispensers, fire
>> extinguishers, and tranquilizers so you can sleep.
>
>I have since developed a method that requires no bleeder resistors, no 
>drain while power is applied, and will drain 800V off 2,000 uF of caps in 2 
>seconds when power is turned off.
> 
>> I'll put electrolytic caps in series and not worry.
>
>1. You have no discharge path when power is turned off. Plenty of loads do 
>not have constant power drain. This can leave dangerous voltages on the 
>caps.

I have already noted the obvious fact that it's sometimes reasonable
to discharge the entire series string, as if it were a single cap.

>
>2. You can have reverse current on the lower cap when there is a constant 
>load and power is turned off. Reverse current is bad for electrolytics.
>
>

Reverse current? Caps always have "reverse current" during discharge.
The current is equal everywhere in a series circuit. All series caps
have the same current.

Does ripple damage caps?



-- 

John Larkin         Highland Technology, Inc

lunatic fringe electronics

jlarkin@highlandsniptechnology.com wrote:

> On Thu, 24 Oct 2019 06:01:55 -0000 (UTC), Steve Wilson <no@spam.com>
> wrote:

>>I noticed a problem with the sym. The line that reads

>>TEXT 1016 -352 Left 2 !.ic V(c1c2) @0

>>should read

>>TEXT 1016 -352 Left 2 !.ic V(c1c2) =400

>>If you try to run the sym, you will get an error:

>>"Trouble parsing initial condition value for V(c1c2)"

>>The fact you never reported an error says you never ran the sym.

> If you ran your own sim, why did you not fix the error?

It didn't occur on my copy. Only on the posted version.

>>Hardly any point in discussing the results of the sym if you never run
>>it. 

> I ran your first one and decided that it was silly. Which it certainly
> was. -330 volts across an electrolytic cap!

As I have already pointed out, the bottom cap had a typo. 

Instead of 1,000uF, it was 100uF. Obviously something was wrong with the 
sym. It took a bit of sleuthing to find the problem.

You should have noticed this when you looked at it.

> OK, you buy a bunch of giant wirewound resistors, mounting brackets
> for same, diodes, fans, blast shields, warning signs, covers with
> interlock switches, safety goggles, rubber glove dispensers, fire
> extinguishers, and tranquilizers so you can sleep.

I have since developed a method that requires no bleeder resistors, no 
drain while power is applied, and will drain 800V off 2,000 uF of caps in 2 
seconds when power is turned off.

> I'll put electrolytic caps in series and not worry.

1. You have no discharge path when power is turned off. Plenty of loads do 
not have constant power drain. This can leave dangerous voltages on the 
caps.

2. You can have reverse current on the lower cap when there is a constant 
load and power is turned off. Reverse current is bad for electrolytics.

On Thu, 24 Oct 2019 06:01:55 -0000 (UTC), Steve Wilson <no@spam.com>
wrote:

>John Larkin <jlarkin@highland_atwork_technology.com> wrote:
>
>> On Wed, 23 Oct 2019 20:27:32 -0000 (UTC), Steve Wilson <no@spam.com>
>> wrote:
>
>>>> In a series string of similar electrolytic caps, they automatically
>>>> voltage-equalize in both directions. 
>
>>>We have been through this. You previously said the capacitor leakage 
>>>current minimizes itself automatically after a few hours or days.
>
>>>How do they equalize voltages? You need bleed resistors to do that.
> 
>> Hopeless.
> 
>You stated:
>
>--------------------------------------------------------------------
>From: jlarkin@highlandsniptechnology.com
>Newsgroups: sci.electronics.design
>Subject: Re: Film capacitor as power-supply filter
>Date: Mon, 07 Oct 2019 19:45:57 -0700
>
>That is precisely the charm of wet aluminum caps. The series string
>optimizes itself for minimum possible leakage current.
>--------------------------------------------------------------------
>
>How does it equalize voltages?
> 
>>>> I did a leakage test on a real cap, beyond the positive rated voltage.
>
>>>Yes, we all remember your test. You also stated the reading was very
>>>noisy, which meant the measurement was bad. 
> 
>> My HP power supply and Fluke DVM are bad?
>
>You stated:
>
>--------------------------------------------------------------------
>From: John Larkin <jlarkin@highland_atwork_technology.com>
>Newsgroups: sci.electronics.design
>Subject: Re: Film capacitor as power-supply filter
>Date: Wed, 16 Oct 2019 16:48:02 -0700
>
>Here's possibly the only curve like this ever posted online:
>
>https://www.dropbox.com/s/i4wwttdgqycz9rv/Alum_Leakage_63u.JPG?raw=1
>
>The data is crude, because there is reforming and dielectric
>absorption going on, and I don't have a month to play with this. But
>the curve is clearly radically upward. Not a zener, more like an MOV.
>
>The current is very noisy above maybe 70 volts. I see what looks like
>brief high current spikes.
>--------------------------------------------------------------------
> 
>> No, the current leakage of a wet aluminum cap is noisy. There's
>> complex chemistry going on inside.
> 
>Not much current leakage below the rated voltage. The cap would filter it.
>
>>>> You might volunteer to do a similar leakage test, but for negative
>>>> voltage. I should have done that when I had it set up. 
>
>>>There is no point. You can damage electrolytics by applying reverse
>>>bias. That's what the dump diode is for.
> 
>> I didn't think you would actually measure anything.
>
>No point. you can damage the cap.
> 
>> Where will you get an ideal diode to prevent reverse bias on the cap?
>> At -0.6, it will explode and kill everyone in the building.
>
>Diode D2 is a 1N4007. It won't explode.
>
>I noticed a problem with the sym. The line that reads
>
>TEXT 1016 -352 Left 2 !.ic V(c1c2) @0
>
>should read
>
>TEXT 1016 -352 Left 2 !.ic V(c1c2) =400
>
>If you try to run the sym, you will get an error:
>
>"Trouble parsing initial condition value for V(c1c2)"
>
>The fact you never reported an error says you never ran the sym.

If you ran your own sim, why did you not fix the error?

>
>Hardly any point in discussing the results of the sym if you never run it.

I ran your first one and decided that it was silly. Which it certainly
was. -330 volts across an electrolytic cap!

OK, you buy a bunch of giant wirewound resistors, mounting brackets
for same, diodes, fans, blast shields, warning signs, covers with
interlock switches, safety goggles, rubber glove dispensers, fire
extinguishers, and tranquilizers so you can sleep.

I'll put electrolytic caps in series and not worry.



-- 

John Larkin         Highland Technology, Inc

lunatic fringe electronics

Winfield Hill <winfieldhill@yahoo.com> wrote:

> Steve Wilson wrote...
>>
>> A typical spec is 
>> I = 0.01CV or 3uA, whichever is greater
>>
>> Where
>> I : Max. leakage current (uA) at 20C after 2 minutes
>> C : Nominal capacitance (uF)
>> V : Rated voltage (V)
> 
>  That's a common, but horrible formula, especially if
>  the units aren't stated.  Should be farads and amps.
> 
>> Example for a 1000uf, 450V capacitor
>> I = 0.01 CV = 0.01 * 1000 * 450 = 4,500 uA
>> So the bleeder current should be 45mA, or
>> R = 450 / 45e-3 = 10,000 Ohms
>>
>> The bleeder dissipation is
>> P = 45e-3 * 450 = 20.25 Watts for each resistor.
> 
>  Wasting 40W in a pair of bleeder resistors is not an
>  attractive situation, leading one to change the "rule".
> 
>  I routinely make HV supplies, with big caps for high
>  pulse output currents, but with a maximum dissipation
>  of under 10W.  Wouldn't want to waste more than 4W.
> 
>> The bleeder time constant is
>> RC = 10,000 * 1000e-6 = 10 seconds.
>>
>> So the capacitor voltage should be reduced to a safe level one minute
>> after power off. 
> 
>  I designed a fairly simple HV discharge circuit, that
>  works automatically when the operating voltages are
>  removed, fully discharges the HV in under 5 seconds.

can you post the LTspice file?

John Larkin <jlarkin@highland_atwork_technology.com> wrote:

> On Wed, 23 Oct 2019 20:27:32 -0000 (UTC), Steve Wilson <no@spam.com>
> wrote:

>>> In a series string of similar electrolytic caps, they automatically
>>> voltage-equalize in both directions. 

>>We have been through this. You previously said the capacitor leakage 
>>current minimizes itself automatically after a few hours or days.

>>How do they equalize voltages? You need bleed resistors to do that.
 
> Hopeless.
 
You stated:

--------------------------------------------------------------------
From: jlarkin@highlandsniptechnology.com
Newsgroups: sci.electronics.design
Subject: Re: Film capacitor as power-supply filter
Date: Mon, 07 Oct 2019 19:45:57 -0700

That is precisely the charm of wet aluminum caps. The series string
optimizes itself for minimum possible leakage current.
--------------------------------------------------------------------

How does it equalize voltages?
 
>>> I did a leakage test on a real cap, beyond the positive rated voltage.

>>Yes, we all remember your test. You also stated the reading was very
>>noisy, which meant the measurement was bad. 
 
> My HP power supply and Fluke DVM are bad?

You stated:

--------------------------------------------------------------------
From: John Larkin <jlarkin@highland_atwork_technology.com>
Newsgroups: sci.electronics.design
Subject: Re: Film capacitor as power-supply filter
Date: Wed, 16 Oct 2019 16:48:02 -0700

Here's possibly the only curve like this ever posted online:

https://www.dropbox.com/s/i4wwttdgqycz9rv/Alum_Leakage_63u.JPG?raw=1

The data is crude, because there is reforming and dielectric
absorption going on, and I don't have a month to play with this. But
the curve is clearly radically upward. Not a zener, more like an MOV.

The current is very noisy above maybe 70 volts. I see what looks like
brief high current spikes.
--------------------------------------------------------------------
 
> No, the current leakage of a wet aluminum cap is noisy. There's
> complex chemistry going on inside.
 
Not much current leakage below the rated voltage. The cap would filter it.

>>> You might volunteer to do a similar leakage test, but for negative
>>> voltage. I should have done that when I had it set up. 

>>There is no point. You can damage electrolytics by applying reverse
>>bias. That's what the dump diode is for.
 
> I didn't think you would actually measure anything.

No point. you can damage the cap.
 
> Where will you get an ideal diode to prevent reverse bias on the cap?
> At -0.6, it will explode and kill everyone in the building.

Diode D2 is a 1N4007. It won't explode.

I noticed a problem with the sym. The line that reads

TEXT 1016 -352 Left 2 !.ic V(c1c2) @0

should read

TEXT 1016 -352 Left 2 !.ic V(c1c2) =400

If you try to run the sym, you will get an error:

"Trouble parsing initial condition value for V(c1c2)"

The fact you never reported an error says you never ran the sym.

Hardly any point in discussing the results of the sym if you never run it.