Low impedance measurement with solartron 1250/1260 and 1287

[Gavin Reade]

Dear All,

I am trying to measure impedance of low temperature SOFCs with the above kit.  Try as I might I cannot reduce high frequency inductance to low levels.  Measuring the impedance wit the leads shorted together results in an inductive component and a resistance which is approximately half the measured resistance of the SOFC (0.005 ohm leads shorted vs. 0.01 ohm SOFC).  A typical high frequency intercept value is between 0.8 and 1 kHz.  I believe I need to be able to get to at least 100 kHz!!!

I notice that when the system being measured has a much higher impedance (i.e. solartron  dummy cell) the impedance looks very good i.e. no inductive component.

Looking at the 1287 manual, it seems that at frequencies above 10 kHz and impedances in the range of a few mOhm, the machine is operating at the edge of it's specification.

I am wondering what the expert opinion is regarding where to go next.  For example would adding a known impedance in series with the SOFC help or is this the wrong way to go?

Thanks in advance,

 

Gavin

[Burak Ulgut]

Gavin, measuring low impedances at high frequencies is a challenge as you have noticed.

First and foremost, there is always a series inductance to your system due to shorted lead behavior of the measurement system. Best you can do is to take steps to minimize it. 

Gamry has a couple of nice application notes regarding this:

http://www.gamry.com/application-notes/low-impedance-limits-with-the-gamry-reference-30k-booster/

http://www.gamry.com/application-notes/testing-electrochemical-capacitors-part-3-electrochemical-impedance-spectroscopy/

http://www.gamry.com/application-notes/low-impedance-eis-using-a-1-mohm-resistor/

Adding a resistor sounds great at first, but you are still not getting rid of the inductance. Your measurement won't show the effect, but the subtracted spectrum will.

Good luck,

Burak

P.S For full disclosure, I used to work for Gamry.

--
You received this message because you are subscribed to the Google Groups "Impedance Spectroscopy" group.
To unsubscribe from this group and stop receiving emails from it, send an email to impedance-spectro...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Gavin Reade]

Thanks Burak,

I tried adding resistors but they either have no appreciable effect or they completely distort the spectrum.  I'll have a look at the stuff you sent me through.

cheers,

Gavin

[Benjamin Sanchez]

Hello Gavin,

For very low impedance measurement, the residual impedance of the shorting device should be less than 1/100 of the measured impedance of the DUT, otherwise it will directly affect the measurement results. Few hints. Use a test shorting bar with the same size and shape as the DUT (if possible) and in which the terminals can be connected directly. The shorter the cables used, the better. Also, consider using open/short/load compensation technique instead of open/short technique.

Best,

Benjamin

[Gavin Reade]

Hi Benjamin,

Can you explain what DUT is?

cheers,

Gavin

[Benjamin Sanchez]

Gavin

My bad, DUT means device under test

Regards

Benjamin

[Colin Gore]

One thing you can do is pay attention to how your leads are routed from the 1260 to the SOFC.  First off, ALWAYS use a 4-point connection to minimize the effects of the leads.  Secondly, ensure that the leads to cathode and anode are run parallel to each other, and from the same side into the testing chamber.  We're able to get acceptable resolution on cells with similar resistance to yours.  I assume you are testing cells with larger area, correct?  What temperature are you operating at?

A good resource is Barsoukov and J. Ross Macdonald's book "Impedance Spectroscopy: Theory, experiment and applications".  Fig 4.1.23 on p. 233 describes what I'm talking about, and the book holds a wealth of other information you will find valuable.



Regards,
Colin Gore
PhD Candidate, Laboratory Manager
University of Maryland Energy Research Center (UMERC)
Rm. 1207, Engineering Lab Building
College Park, Maryland 20742

[Gavin Reade]

Hi there,

I'm well acquainted with MacDonald et al.  I used to use a 1470 / 1455 combo for high temp SOFC (800 - 950C) and had minimal problems since the Rs value was of the order of 0.5 ohm.  A scan from 150 kHz was easy.  I even managed to get good data from within a pressurised system over 10m or so of wiring.  This was with cells of the order of 3 cm2.  The cells I am trying to measure are 10s of cm2 and yes I have tried routing the current wires parallel to each  other etc. etc. in fact using all the tricks I learnt with the high temp ones.  With the current (no pun intended :-) ), the impedance is of the order of 10 mOhm and I'm operating right on the limit of the 1260 / 1287 capability so it's somewhat outside my experience.

Anyway thanks to everyone for their inputs,

cheers,

greadey

--

[Colin Gore]

The only other relatively simple solution (without extra equipment) I can suggest is to reduce the sample area.  We've screenprinted the cathode on our cells in segments for some tests, with some smaller ~1 cm2 sample areas that allowed for increased resolution for EIS.  It also lets you see the performance near the fuel inlet vs near the fuel outlet, since you're obviously going to have different concentration effects at each location, though it can be difficult to infer the integrated cell performance from only a few spots.  Wiring up the rig was a bit tricky, but doable with proper isolation and cathode pattern.  Sounds like you're interested in specific high-freq info, so perhaps this still won't be any help.

On a side note, when asking my boss to buy a power booster I said "this unit has the potential to meet our current demands".  Only realized afterward that he must have thought I was trying to be a wise-guy.

Cheers,
Colin

To unsubscribe from this group and stop receiving emails from it, send an email to impedance-spectroscopy+unsub...@googlegroups.com.