EMI Shielding Question and Sheet Resistance 1

[John Parnell]

First let me state that I am a mechanical engineer working to EMI shield an electrical instrument gauge. This gauge is much like the oil temperature gauge on a car. The shielding is in the form of a 2" diameter aluminum housing. The front of this housing is ITO coated glass for viewing the gauge, that is supposed to be well grounded to the aluminum housing with conductive epoxy. On the drawing we called out for the ITO to have a sheet resistance value of 20 Ohms/square or less. We then bought the housing assemblies from a vendor. The vendor sent us in a sample of housings that I had measured by an independent lab to have a sheet resistance of about 9 ohms/square. However, when I use a multimeter to measure the resistance from the center of the ITO coated glass to the housing I get values that range from 9 ohms to 150 ohms. We looked at the housings with the higher resistance values and noted that there seemed to be gaps in the conductive epoxy. This is not good for EMI shielding. During EMI qualification testing, we had used a case that had a measured value of about 10 ohms from the center of the 2" ITO coated glass to the case using a multimeter. That unit passed EMI testing and qualified us to begin production. Now I am wondering if I need to specify a resistance value of ITO coating back to the housing in order to properly qualify production units for use. If so, is this a proper method to specify that measuring using a multimeter between the case and the center of the ITO coating should be less than a certain value? What value should that be? What value should one expect, measuring resistance using a multimeter between the center of a 20 ohm/square ITO coating out to the 2" diameter of the aluminum housing if the housing is well grounded to the ITO coating? I cannot go back and retest cases for EMI, as these tests are hundreds of thousand of dollars. I also don't want to specify something that is impossible to deliver by using values that are too restrictive.

 

[VEBill]

"...much like the oil temperature gauge on a car..."

Of course I don't know the details of your application, but it seems odd to have such concerns about EMI shielding a gauge.

If the EMI Source is inside some large cabinet, then the wiring to the gauge could be shielded and filtered at the point of wiring penetration into the gauge housing. Then the gauge can be happy outside the overall shield. The signal arriving is clean. Done.

If it's the other way around, and you're trying to keep external noise from entering the cabinet, same approach works. Filter the wiring as it leaves the gauge housing to enter the larger system cabinet.

If the room environment is so EMI noisy that it's impacting the actual gauge itself, then what about the operator who is looking at the gauge? Are they exposed to such extreme EMI? A simple gauge itself shouldn't be that sensitive to (HF) EMI.

If it's a (DC) magnetic problem, then that's a whole different issue.

Apologies in advance for starting at a higher level, before getting to your more focused questions.

 

[IRstuff]

Is there electronics in the gauge, or is it purely an electromechanical gauge? IF the latter, then it is unlikely to be affected by EMI, as indicated VEBill. What's attached to the wires of the gauge; is it a low-impedance driver?

What was the EMI attenuation requirement and the specified EMI field? Did an actual EMI EE come up with the requirements, or is this just best practice, etc.?

From a practical perspective, you presumably have a conductivity requirement on the conductive epoxy, but might not have had a requirement for gaps and voids in the epoxy. Certainly, an acceptance test with a DMM measuring the resistance would seem to be at least a means of winnowing out the really bad assemblies, that could then be reworked.


TTFN (ta ta for now)
I can do absolutely anything. I'm an expert!

https://www.youtube.com/watch?v=BKorP55Aqvg

faq731-376 forum1529 Entire Forum list

http://www.eng-tips.com/forumlist.cfm

 

[John Parnell]

Sorry to be a little vague. It is for a military dashboard application and the EMI shielding is both for emitted and susceptibility issues. We have limits on both which are part of the design specification. We passed EMI testing with our design, but we did have one emission spike which was in a frequency about that of the AM radio band, so it was waived as insignificant. Of course all of the wiring is shielded, and there are EMI filters on the wiring into the housing. Thus I need to provide an EMI enclosure for the electronics inside, and part of that shielding is the ITO coating on the glass for the front.

 

[John Parnell]

There is quite a bit of electronics in the gauge, including a microcontroller and associated circuitry that is used to compensate for temperature variances. The specifications for EMI are covered by military specs on the vehicle.

 

[VEBill]

Good info.

"...gaps in the conductive epoxy..."

Mil projects don't usually accept poor QA.

It's often easier to set a tight limit than to spend months trying to determine the precise limit of acceptability. If 9 ohms is achievable, then the limit might be set to Max 10.0 ohms.

A simple screening test (single measurement with a DVOM) can be ideal for the Acceptance Test Procedure.

9 or 10 ohms is almost low enough to be cautious of measuring techniques. Probably don't need 4-wire, but do need caution to avoid being +/- 0.5.

 

[IRstuff]

The DMM contact resistance is often on the order of 0.5 ohms for the two probes.

Military EMI is typically 200 V/m, ignoring HEMP, etc.

TTFN (ta ta for now)
I can do absolutely anything. I'm an expert!

https://www.youtube.com/watch?v=BKorP55Aqvg

faq731-376 forum1529 Entire Forum list

http://www.eng-tips.com/forumlist.cfm

 

[John Parnell]

Thanks for the information. After doing some more testing, I am going to set my limits for ITO to housing based on a sampling of products that we received. It seems like the value across the ITO is very consistent, but the value of ITO to case varies widely. The higher ohm values of those indicate a poor assembly technique and breaks in the conduction path of the conductive adhesive.

 

[Compositepro]

Is it necessary to have any electrical contact with the ITO coating to provide EMI shielding? With a Faraday Cage it is only the the size of the gaps or openings that are important.

 

[IRstuff]

If the ITO is floating, then it acts like an antenna.

TTFN (ta ta for now)
I can do absolutely anything. I'm an expert!

https://www.youtube.com/watch?v=BKorP55Aqvg

faq731-376 forum1529 Entire Forum list

http://www.eng-tips.com/forumlist.cfm

 

[Compositepro]

Metals reflect RF because they act as an antenna, where the surface currents counteract the electric field. I cannot see what difference electrical bonding makes. Now, there are electrostatic benefits to bonding and grounding, but that generally requires only slight conductivity, unless you are talking about lighting protection.

 

[waross]

I can see the indicated resistance being influenced by the contact between the probe and the ITO surface.
That is, a probe rounded on the end may give a different indication than a pointed probe.
Do you specify the procedure for making the multi-meter measurements?

As an example:
There are strict specifications for the conductivity of the floor in a hospital operating theater.
The resistance must be in a range that is high enough so that there is no shock hazard in the event of some type of equipment failure and yet low enough to discharge any static charges.
I don't remember the exact numbers but the electrode area and contact pressure are both specified for acceptance testing.
Something like a 1 ft[sup]2[/sup] copper plate weighted by 50 lbs, evenly spaced on the plate electrode.

Alternately you may consider a four wire measurement scheme.
eg: two electrodes pass a small current through the film and a second pair of electrodes measure the voltage drop caused by the current.
Another possible issue is a junction voltage being generated that throws off your resistance measurement.
Try reversing the multi meter leads and see if the reading changes.

Given that ITO films are used for heating elements to defrost mirrors and windshields, I am sure that a suitable uniformity of film thickness is achievable.
Have a talk with an engineer who works applying ITO films and find out what is achievable and at what cost.

Bill
--------------------
"Why not the best?"
Jimmy Carter