Question about Low ESR capacitors

Keith Hilton · Post by **Keith Hilton** » 26 Apr 2000 9:12 pm

Is there anything in particular you don't like about a particular application, using Low ESR capacitors?

------------------

Mark Amundson · Post by **Mark Amundson** » 28 Apr 2000 4:51 am

Low Effective Series Resistance (ESR) and low ESL capacitors vary from general purpose capacitors in the amount of construction. For example, a general purpose electrolytic cap has only two foil-to-terminal tabs; one for the aluminum foil anode (+), the other for the aluminum foil cathode (-). Whereas Low ESR, ESL caps have multiple redundant tabs to lower connection resistance and drop the inductance of the rolled aluminum foils.

On the application side, the low ESR caps are acceptible if your circuit depends on less than ten ohm impedances at high frequencies (10kHz through 1MHz). For most audio decoupling uses, low ESR caps are overkill. But with switching power supplies, they are a necessity.

Also, do not ignore dry slug tantalum caps as they are inherently low ESR due to their construction. But they tend to be limited in value (<100uf) and not useful for audio decoupling due dielectric absorption/distortion. However, tantalums do not "dry out" like electrolytic caps. Conversely, electrolytics can sometimes heal if punched through by over voltage, whereas tantalums would short out.

Keith Hilton · Post by **Keith Hilton** » 28 Apr 2000 7:05 am

I wondered if you would reply Mark. I've been wondering what you were up to. I had built a form of switching power supply, and had been using regular electrolytic capacitors. I noticed in the literature the manufacturer of the DC/DC converter put out, it said to use Low ESR capacitors. The power supply didn't seem to work as good with the ESR capacitors as it did with regular capacitors. Maybe it was just my imaginationm ,but something was different. I can't pin down why it don't work as good. I know this sounds crazy, but I am a experimenter. Maybe it was something else I was doing. Actually, it seemed the ESR capacitors made everything work to good. This seemed to make my DC/DC converter have a difficult time getting charged up.

------------------

Mark Amundson · Post by **Mark Amundson** » 28 Apr 2000 9:23 am

In your business Keith, using low ESR electrolytic caps everywhere is just another statement toward superior quality. I believe performance over price is your current philosophy.

I like the Panasonic HF series of low ESR caps (dark blue covering) for my audio prototypes. The better nichicons are good too. I get the Panasonic caps from Digi-Key.

Rich Paton · Post by **Rich Paton** » 30 Apr 2000 1:15 pm

Keith...my guess would be that the low ESR caps' lower impedance caused a change in the frequency of the transformer primary signal, and nudged the circuit a bit out of its optimum operating parameters (?)

Keith Hilton · Post by **Keith Hilton** » 30 Apr 2000 8:43 pm

Two of the people I respect most on this FORUM are Mark and Rich. I wondered if you would reply to this topic Rich. The root of all this gets back to powering up a DC/DC converter, from a wall wart. All this is parts of a switching type power supply I am building. I am using a brick type DC/DC converter, and they have a charging period when first turned on. If you exceed the current, the DC/DC converter simply shuts down then tries to re-start. I have installed current limiting resistors, but it still seemed like the DC/DC converter was not charging correctly. Maybe the reason was not the low ESR capacitors, but the fact that I used a lot lower value of uF capacitors than I had been using.

------------------

Rich Paton · Post by **Rich Paton** » 1 May 2000 3:17 am

ABOVE: What NOT to waste your $30 Million Mad Money on! This is junk at its finest!

Thanks for the vote of confidence, Keith...
It's impossible to actually mentally troubleshoot such a problem based on written info. I'm not very familiar with "Brick" dc/dc units, must be a "black box" with I/O connections(?)
I have worked as a technician & as a greenhorn engineer around some both brilliant and grizzled engineers, and have "picked their brains" like a starving coyote.
Early on I learned that the power supply in a piece of equipment has to be 100% operational, no easily overlooked high frequency oscillations, "soft" load current paths, leakage paths, excess ripple, sag or cresting, thermal instability problems, etc.
I've ****ed off a few EE's who were right out of school trying to explain such mundane matters, but it's the wiser old dog who will get the bone.
I recently saw a problem in a dc regulator board used to power the heater element in a TWT thermionic valve, & saw the problem right off, but the systems engineer was sceptical. The 3 amp, TO-3, 3 terminal regulator could supply enough current to run the 1.9 amp heater in stride, but the original chip was a 5 amp unit. The tungsten/thorium filament's cold resistance "looked like" a short to the chip, forcing the less robust unit to shut down immediately.
BTW, do you need an R&D/production tech assistant...for a year or so? I seem to be getting the itch to travel and get off this fog coast for a spell!
Good luck on the project.<FONT SIZE=1 COLOR="#8e236b"><p align=CENTER>[This message was edited by Rich Paton on 01 May 2000 at 04:27 AM.]</p></FONT>