Protection of 1uF capacitor against short circuit

There are self-protecting SSRs, like the Ixys CPC1540. We sometimes use them always-on, as current limiters.

Teletypes are really blasts from the past. I used to hate them... slow, noisy, oily, always breaking. Then came dot-matrix printers, like Decwriters, slightly better.

--

John Larkin         Highland Technology, Inc 

lunatic fringe electronics

What's wrong with a resistor?

NT

NSC's LM317Hv rated to 60V, and better: TI's TL783C rated to 125V and 500mA. That should be fine.

Reminder, this info and more in Table 9.2, power-control Chapter 9. Also 108 pages free, sample AoE-III chapter.

--
 Thanks, 
    - Win

1uF regular aluminum electrolytic will have an ESR of a couple hundred ohms, so the current will be limited to < 1A. The optoisolator would also have an Idss which will limit the current. I don't see a spec for the maximum current.

You could add the classic two-BJT current limiting circuit, ensuring the SOA of the BJTs are not exceeded.

You lose a bit of voltage in the ccircuit but you have 120V to play with..

--sp

--
Best regards,  
Spehro Pefhany 
Amazon link for AoE 3rd Edition:            http://tinyurl.com/ntrpwu8

[snip]

What? A couple of hundred milliohms surely? The only time I bothered to measure one I got about 1 Ohm, and I think *that* was bad!

Jeroen Belleman

I'll throw my hat in with 10 ohms.

NT

Is this the same power supply?

If your 120VDC power supply is a 60 ma current source, and your selector solenoid only needs 3.3V to operate, then you have over 100 volts margin to play with. At 60 ma, that's: R = E/I = 100V / 0.060A = 1.7K ohms max. Unless your current loop has miles of wire in the loop, you can just add whatever resistance is needed to keep the inrush current down to reasonable levels when you short the capacitor.

Although I don't think it will be necessary, you might consider adding one or more PTC current limiting thermistors. The basic idea is the PTC thermistor is low resistance when cold. When shorted, the current goes over the normal 60 ma, the thermistor becomes warm and its resistance increases, thus limiting the peak current. Response time is slow, but the electrolytic doesn't need microsecond switching protection. A small series resistor will keep the peak current reasonable and protect the phone plug. The PTC thermistor will do the rest.

Last resort might be to install a fuse.

I don't have KiCAD installed. Could you post a JPG, PDF, or LTSpice schematic instead?

--
Jeff Liebermann     jeffl@cruzio.com 
150 Felker St #D    http://www.LearnByDestroying.com 
Santa Cruz CA 95060 http://802.11junk.com 
Skype: JeffLiebermann     AE6KS    831-336-2558

That might be a bit on the high side. It's not listed on most of the datasheets. I see 17 ohms for a 100V/1uF part from Nichicon that's claimed to be low impedance. I would expect a 250V part to be quite a bit higher, but it's not listed at all in the datasheet:

--sp

--
Best regards,  
Spehro Pefhany 
Amazon link for AoE 3rd Edition:            http://tinyurl.com/ntrpwu8

The small alum cap data sheets rarely mention ESR. I have found a weak correlation: for given C, higher voltage is lower ESR.

--

John Larkin         Highland Technology, Inc 

lunatic fringe electronics

Put a resistor in series, on the order of 55 ohm / 10 = 5.5 ohms. Say 4.7.

Or don't worry about it. 5mJ is a good ESD zap.

Seems like a pretty bad place to use an optoMOS part though, and kind of begs the question of why have an electrolytic there in the first place if it only makes things worse...

Tim

-- Seven Transistor Labs, LLC Electrical Engineering Consultation and Contract Design Website:

Two data points only at 1uF from the datasheet I linked.

1uF 5 x 11 50V 11 ohms 1uF 5 x 11 100V 17 ohms

Lots of points for bigger caps that support your weak correlation.

47uF 5 x 11 16V 0.8 ohms 47uF 6.3 x 11 25V 0.55 ohms 47uF 6.3 x 11 35V 0.34 ohms 47uF 6.3 x 15 50V 0.3 ohms 47uF 8 x 11.5 63V 0.25 ohms 47uF 8 x 15 80V 0.39 ohms 47uF 10 x 20 100V 0.28 ohms

But back to the oddball 1uF high voltage caps..

Then we have the Vishay MALREKA00PB110P00K which is 265 ohms ESR for

1uF/450V.

And the Illinois capacitor 105TTA450M which is a whopping 414 ohms for

1uF/450V.

And many other similar ones that fit my old experience with it being hundreds of ohms.

That's a pretty wide range!

--sp

--
Best regards,  
Spehro Pefhany 
Amazon link for AoE 3rd Edition:            http://tinyurl.com/ntrpwu8

I think you'll find better correlation to package volume.

--
 Thanks, 
    - Win

414 ohms gives it a 380 Hz internal time constant!

That's close to useless.

--

John Larkin         Highland Technology, Inc 

lunatic fringe electronics

For a constant body size and type, esr can be expected to rise with voltage rating for Al electrolytics.

Degrading or 'stabilizing' esr range with an external series R is a known method of getting predictable results from Al electrolytics, in small signal or pulsed applications.

A more common problem is obtaining a low enough worst case esr in an application where high esr doesn't work, usually in power conversion.

Getting anything below 10R in a

And for a given C, a higher voltage cap will be bigger.

But that wrecks the pulse current capability!

Polymer aluminum caps are great. Low and stable ESR... too low sometimes. They're available at over 100 volts now.

--

John Larkin         Highland Technology, Inc 

lunatic fringe electronics

Digital camera photoflash applications are exquisitely engineered.

You have a capacitor storing about 15 joules, enough to nuke any single semiconductor off the planet. You have a spark gap. You have an IGBT. You have a minimal handful of support components (ignition transformer and its switch, bypass and snubber caps and diodes, etc.).

And somehow you need to turn the thing on and off at will.

That they can pack 100A of reliable current commutation into a PDSSOP-8 style package is astounding!

Tim

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

...

Great idea! It turns out that availability of the CPC1540G (the DIP version) is poor, but the CPC1510G is readily available in DIP. The current limit is higher, but the goal here is for the device to protect itself. I'm also going to use a ceramic 1uF cap instead of the electrolytic. Those are less fragile. That should do it. I don't even have to change the PC board, which is currently on its way to me from the board house. (Seeed, in China.)

Someone wanted to see the circuit without having to install KiCAD. Here it is:

This is for antique Teletype hobbyists. Here's how we use it:

I restore those machines as a hobby. See "

". Right now, I have five of them.

John Nagle

I've been unable to view the OP's gerbers, for some reason, but the reasons people use electrolytics often include one size and cost. When troubleshooting performance in a fixed layout, one of the things that is seldom available is extra room. It is therefor more apt to consider radial electrolytic component solutions with the OD limitation in mind.

The OP is looking for 60mA pulse, with considerable headroom to provide it. His issue is predictable, non damaging, single fault behavior, in the event of load shorts.

A lower ESR will not solve the OP's problem.

RL

If everything goes right, nothing goes wrong. The OP's issue is survivability of selected components under single fault conditions.

RL