1. Home
  2. Electronics Design
  3. So, where's the catch?

So, where's the catch?

See the LTSpice circuit.

It started out with me asking myself "how can I make an integrator that works fairly well, and uses a honkin' big electrolytic capacitor?" And it ended up here.

R4/R3 must equal R1/R3 for proper integration action. Output comes off of the op-amp. The integrator gain is (1 + R2/R1)/(R4*C1). The way the circuit works is whacky, and the linear range of inputs is limited to 1/(1

+R2/R1), but aside from that I don't see any super big pitfalls.

Comments welcome.

Version 4 SHEET 1 880 680 WIRE 144 0 64 0 WIRE 352 0 224 0 WIRE 256 64 128 64 WIRE 352 64 352 0 WIRE 352 64 336 64 WIRE 224 160 224 128 WIRE 128 176 128 64 WIRE 192 176 128 176 WIRE 352 192 352 64 WIRE 352 192 256 192 WIRE -272 208 -352 208 WIRE -144 208 -192 208 WIRE -96 208 -144 208 WIRE 64 208 64 0 WIRE 64 208 -96 208 WIRE 192 208 64 208 WIRE 128 240 128 176 WIRE -96 256 -96 208 WIRE 224 256 224 224 WIRE -560 272 -560 224 WIRE -464 272 -464 224 WIRE -352 272 -352 208 WIRE 128 352 128 320 WIRE -96 368 -96 320 WIRE -560 400 -560 352 WIRE -464 400 -464 352 WIRE -352 400 -352 352 FLAG 224 256 0 FLAG 224 128 +3V3 FLAG 128 352 +1V75 FLAG -560 224 +3V3 FLAG -464 224 +1V75 FLAG -352 400 0 FLAG -464 400 0 FLAG -560 400 0 FLAG -96 368 0 FLAG -144 208 Vc SYMBOL Opamps\\UniversalOpamp2 224 192 R0 SYMATTR InstName U1 SYMBOL res 112 224 R0 SYMATTR InstName R1 SYMATTR Value 330k SYMBOL res 352 48 R90 WINDOW 0 0 56 VBottom 2 WINDOW 3 32 56 VTop 2 SYMATTR InstName R2 SYMATTR Value 330k SYMBOL res 240 -16 R90 WINDOW 0 0 56 VBottom 2 WINDOW 3 32 56 VTop 2 SYMATTR InstName R3 SYMATTR Value 330k SYMBOL voltage -560 256 R0 WINDOW 123 0 0 Left 2 WINDOW 39 0 0 Left 2 SYMATTR InstName V1 SYMATTR Value 3.3 SYMBOL voltage -464 256 R0 WINDOW 123 0 0 Left 2 WINDOW 39 0 0 Left 2 SYMATTR InstName V2 SYMATTR Value 1.75 SYMBOL voltage -352 256 R0 WINDOW 123 24 124 Left 2 WINDOW 39 0 0 Left 2 SYMATTR InstName V3 SYMATTR Value 1.75 SYMATTR Value2 AC 1 SYMBOL res -176 192 R90 WINDOW 0 0 56 VBottom 2 WINDOW 3 32 56 VTop 2 SYMATTR InstName R4 SYMATTR Value 330k SYMBOL polcap -112 256 R0 WINDOW 3 24 56 Left 2 SYMATTR InstName C1 SYMATTR Value 100µ SYMATTR Description Capacitor SYMATTR Type cap SYMATTR SpiceLine V=2.5 Irms=1.1 Rser=0.07 Lser=0 mfg="KEMET" pn="T520B107M2R5ASE070" type="Tantalum" TEXT -456 496 Left 2 !.ac dec 10 1m 1k TEXT -448 456 Left 2 !.ic V(vc)=1.75

--
My liberal friends think I'm a conservative kook.
My conservative friends think I'm a liberal kook.
Why am I not happy that they have found common ground?

Tim Wescott, Communications, Control, Circuits & Software
http://www.wescottdesign.com
Reply to
Tim Wescott
Loading thread data ...

It's a Howland current source (sort of), see...

formatting link

...Jim Thompson

--
| James E.Thompson, CTO                            |    mens     |
| Analog Innovations, Inc.                         |     et      |
| Analog/Mixed-Signal ASIC's and Discrete Systems  |    manus    |
| Phoenix, Arizona  85048    Skype: Contacts Only  |             |
| Voice:(480)460-2350  Fax: Available upon request |  Brass Rat  |
| E-mail Icon at http://www.analog-innovations.com |    1962     |
             
I love to cook with wine.     Sometimes I even put it in the food.
Reply to
Jim Thompson

I would not use an electrolyic there. 1.75/330k is 5.3uA, and the leakage current can easily be in the same range. this calls for something better than electrolytic if it's for measurement.

--
Regards, Joerg

http://www.analogconsultants.com/
Reply to
Joerg

Oops. I reinvented the wheel.

I can never remember the schematic for a Howland current source. Maybe now...

--
My liberal friends think I'm a conservative kook.
My conservative friends think I'm a liberal kook.
Why am I not happy that they have found common ground?

Tim Wescott, Communications, Control, Circuits & Software
http://www.wescottdesign.com
Reply to
Tim Wescott

In this case it's for some really slow DC finding, with a Tantalum in there.

Things are working barely well enough with some 4.7uF ceramics, but slower is better.

--
My liberal friends think I'm a conservative kook.
My conservative friends think I'm a liberal kook.
Why am I not happy that they have found common ground?

Tim Wescott, Communications, Control, Circuits & Software
http://www.wescottdesign.com
Reply to
Tim Wescott

These are listed at 4uA for leakage. Realistically you'll more likely see 1uA or so but at a max charge current of 5.3uA I don't think you'll be a happy camper with tantalums:

formatting link

You can get 47uF in X7R for well under a buck. Don't know the leakage data on those but I found it to be miniscule in my cases. Could go to church, come back, voltage still there. And when our pastor does the sermon that can be a while :-)

--
Regards, Joerg

http://www.analogconsultants.com/
Reply to
Joerg

You can't use a film cap, a big resistor, and a cmos opamp?

Somebody used to post here about the voltages he observed on some film caps over time, like months. The effective self-discharge time constants were, as I recall, many years.

--

John Larkin                  Highland Technology Inc
www.highlandtechnology.com   jlarkin at highlandtechnology dot com   

Precision electronic instrumentation
Picosecond-resolution Digital Delay and Pulse generators
Custom timing and laser controllers
Photonics and fiberoptic TTL data links
VME  analog, thermocouple, LVDT, synchro, tachometer
Multichannel arbitrary waveform generators
Reply to
John Larkin

Huh. So you can -- at least in the lower voltages.

I had it fixed in my head that it was a "no can do", because the last time I went searching for high-value SMD ceramics it was for 50V.

Gotta remember _all_ the parameters, Tim.

--
Tim Wescott
Control system and signal processing consulting
www.wescottdesign.com
Reply to
Tim Wescott

Do mind that they are microphonic, on the order of milivolts per fingertap on the board.

Tim

-- Deep Friar: a very philosophical monk. Website:

formatting link

Reply to
Tim Williams

Ceramic capacitors have come a long way since we were young. You can get multilayer ones with oodles of capacitance and remarkably small, but they are rated 4V or 6.3V.

--
Regards, Joerg

http://www.analogconsultants.com/
Reply to
Joerg

This is probably like electrolytic caps, where the size of the capacitor for any given product line is fairly proportional to the amount of energy the cap can store (i.e., proportional to the square of the rated voltage times capacitance). If I ever have the time, maybe I should do the math.

The dielectric layers must be exceedingly thin. I wonder how they lay them down? And do they make the caps in big sheets (or at least long ribbons) and then cut them up, or do they make them individually?

And if they make them in big sheets, how to they selectively cut back the conductors?

I want to see a movie on the process...

--
Tim Wescott
Control system and signal processing consulting
www.wescottdesign.com
Reply to
Tim Wescott

Voila, but only a brief graphics flow chart:

formatting link 

--
Regards, Joerg

http://www.analogconsultants.com/
Reply to
Joerg

If you decide to go that route, check the cap stability vs voltage. With those increasingly thin sheets, non linearity goes to hell. Yep, even for X7R, to the point that I once thought an X7R cap was Y5V. Checking the order, bobbin label and finally measuring tempco convinced me that the caps were X7R despite what the C-V curve indicated... Also you have some hysteresis and relaxation and the C-V curve isn't closed and don't retrace itself... Weird.

If you want SMD, I'd choose some PPS cap and a small low bias CMOS opamp which should do what you need.

--
Thanks,
Fred.
Reply to
Fred Bartoli

Agree, if you need it to be very precise use a film cap. But 47uF or

100uF only come in fairly large through-hole packages. 
--
Regards, Joerg

http://www.analogconsultants.com/
Reply to
Joerg

We don't know the dv/dt Tim needs, so it's hard to say, but my guess is that a 1uF may be enough... I've done really good loooong hold time S/H with 100n PP caps and good layout.

WIMA has SMD PPS caps up to 6.8u in 63V flavor in 6054 size: almost

15x15. mm, that is :-)

And no, I won't tell Tim to parallel 10 of those :-)

--
Thanks,
Fred.
Reply to
Fred Bartoli

If you have zero pollution in the air, no cosmic radiation, nobody within a 50ft radius lights up a Gauloises, and so on :-)

You can get one of these, not much bigger but much higher capacitance:

formatting link

Unfortunately they'll also set you back more than $5 in the BOM budget, in quantities.

Here you get at least 15uF, for around $3:

formatting link 

--
Regards, Joerg

http://www.analogconsultants.com/
Reply to
Joerg

Precision I do not need -- just long, slow, DC-finding.

--
My liberal friends think I'm a conservative kook.
My conservative friends think I'm a liberal kook.
Why am I not happy that they have found common ground?

Tim Wescott, Communications, Control, Circuits & Software
http://www.wescottdesign.com
Reply to
Tim Wescott

Then a cheapo ceramic material is probably ok. 100uF in really small for around 30 cents:

formatting link

Since it's rated 6.3V the capacitance loss at 1.75V won't be so dramatic. You can also get 10V or higher if your BOM can stomach over

50c for this cap. 
--
Regards, Joerg

http://www.analogconsultants.com/
Reply to
Joerg

integrator

comes

inputs

in

likely

leakage

to

get

=20

energy=20

=20

math.

the=20

20+ years ago they made them in sheets (I saw it). They may print them semi individually now. Thick film and [silk] screen printing have come a long way since then. Hell, they may even be "ink" jet printed now.

?-)

Reply to
josephkk

Using Kemet's own subcircuit model for the capacitor, things don't quite look so rosy:

.SUBCKT T520B107M2R5ATE070 1 8

  • Capacitance of 100. µF, Rated @ 2.5 Vdc, SMD with Cond-Polymer Cathode
  • Temp@ 25°C, Bias@ 0Vdc , Center Freq@ 300.000kHz
  • KEMET Model RLC Tant5RC.JPG / Spice Version 3.8.0 L1 1 2 1.80E-09 R6 2 8 100.00E+03 R1 2 3 49.27E-03 C1 3 8 3.23E-06 R2 3 4 4.24E-03 C2 4 8 6.45E-06 R3 4 5 4.24E-03 C3 5 8 12.90E-06 R4 5 6 4.24E-03 C4 6 8 25.81E-06 R5 6 7 4.24E-03 C5 7 8 51.61E-06 .ENDS
--
"For a successful technology, reality must take precedence 
over public relations, for nature cannot be fooled."
                                       (Richard Feynman)
Reply to
Fred Abse

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.