#4 - Our customer is assuming the noise problem cannot be solved at the noise source. They do not have control of the motors and VFD motor drive that may be attached to their systems, so the variety of the noise sources may be infinite.
#3 - I do not assume the solution will be easy or even guaranteed, but the initial experiments in balancing the trade-offs shows promise. Implementation gets me into the realm of “the major cause of problems is . . . solutions.”
#2 – After I explain my job to somebody and how I enjoy my work week, I usually end with “. . . and they pay me for it!”
I’m glad it was a slow TechNet week, allowing me to receive a lot of attention and suggestions.
Thanks,
Carl
Carl B. Van Wormer, P.E., AE7GD
Senior Hardware Engineer
Cipher Engineering LLC
21195 NW Evergreen Pkwy Ste 209
Hillsboro, OR 97124-7167
503-617-7447x303
[log in to unmask]<mailto:[log in to unmask]> http://cipherengineering.com<http://cipherengineering.com/>
From: Inge Hernefjord [mailto:[log in to unmask]]
Sent: Friday, August 12, 2016 12:54 AM
To: TechNet E-Mail Forum <[log in to unmask]>; Carl Van Wormer <[log in to unmask]>
Subject: Re: [TN] looking for medium resistance material
Carl,
!. Admire you. What you do is real engineering.
2.You must have rare position and a generoujs boss.
3.I hope you are aware of how easily time and contract runs away. You may meet unexpected obstacles on the way to a tested an d approved component.
4. You did not react on my q wheater you could solve the problem by minimize the noise at the source.
5. You present , very openly, details that already may have been sent back to the Big Copiator. You may be on a hot spot, noise reduction is one of the most discussed topics in the SPACE industry, meterology, Medecine and other Hitech industries. Just a reminder....don't give 'them' to much goodies...
/ Inge
On 11 August 2016 at 18:06, Carl Van Wormer <[log in to unmask]<mailto:[log in to unmask]>> wrote:
Since it's a slow TechNet morning, I thought I'd give an update on my recent experiment: (feel free to delete if you have real work to do).
I obtained a sample of metalized Mylar that was hiding inside of a 3uF yellow capacitor. I prepared a test capacitor made from two 1" square pieces of Copper tape attached to two business cards (for insulation and alignment) and then ran some 30 gauge wires from opposite corners for measurement attachment leads. I named these two cards plate #1 and Plate #2. I prepared a third business card with 2" square of Copper tape (shield #1), a fourth card with a 2" circle of a shiny metal stick-on silver award star (shield #2), and a fifth with a 2" square of the metalized Mylar from the sacrificial capacitor (shield #3). I stacked up the test jig with the sandwich of plate #1, shield #1, and plate #2, and then measured about 10pF. When I connected the guard terminal of my ESI VideoBridge 2100 to shield #1, the capacitance reading dropped to about 0.2pF, showing that the shielding was effective. Repeating the measurements with the silver star and the capacitor film shields gave similar shielding effectiveness. It was difficult to make a reliable connection to the capacitor film, but when the contact was made, the shielding was effective.
Capacitive shielding effectiveness: Good!
To test the effect of the shield material in a magnetic field, I used the two ferrite cores and a 100-turn coil similar to the final product. The cup-cores are separated by about 1mm, and my business cards are about 1/3mm thick. Measuring the inductance of the coil in the cup-cores with a 3-business-card spacers in the gap, I got an inductance of about 780uH with Q=77. Inserting the Copper tape card (shield #1) in place of one of the spacers gave me an inductance of 255uH with Q=2.5 (the Copper is acting like a shorted turn). Similar testing of the silver star sticker (shield #2) gave an inductance of 630uH with Q=2.9 (I guess the metal on the sticker is thicker than I thought). When I inserted the card with the metalized Mylar (shield #3) into the gap-stack, the inductance measured 760uH with Q=68. This showed very little degradation to the magnetic system.
Magnetic transparency: Good!
Metalized Mylar shows great promise as a solution to my problem, probably because the thickness is quite thin compared to the calculated skin effect depth of about 0.08mm at 1MHz.
Now the problem comes down to making a reliable electrical connection to the metalized Mylar. I am open to any suggestions as to effective ways to physically connect the metalized Mylar to my system, for positional and electrical attachment of the shield. I'm hoping to hear of proven techniques (Conductive epoxy, ultrasonic welding, etc.) that have been successfully used for reliable electrical and mechanical mounting and attachments.
Thanks,
Carl
Carl B. Van Wormer, P.E., AE7GD
Senior Hardware Engineer
Cipher Engineering LLC
21195 NW Evergreen Pkwy Ste 209
Hillsboro, OR 97124-7167
503-617-7447x303<tel:503-617-7447x303>
[log in to unmask]<mailto:[log in to unmask]> http://cipherengineering.com
-----Original Message-----
From: Carl Van Wormer
Sent: Wednesday, August 10, 2016 7:42 AM
To: TechNet E-Mail Forum <[log in to unmask]<mailto:[log in to unmask]>>
Subject: RE: [TN] looking for medium resistance material
I found that a capacitive shield plate connected to ground through a 100 Ohm resistor would eliminate the VFD noise effects. Changing the resistor to 1000 Ohms started to show some interference. I'm hoping that a flimsy Mylar conductor (around 1 Ohm per square) would be a sufficient shield for the electrical field coupling (the gap between the 2 coils is a couple of mm).
As the old guy in the ESI repair department would say . . . "One experiment is worth a thousand expert opinions."
Thanks
Carl
Carl B. Van Wormer, P.E., AE7GD
Senior Hardware Engineer
Cipher Engineering LLC
21195 NW Evergreen Pkwy Ste 209
Hillsboro, OR 97124-7167
503-617-7447x303<tel:503-617-7447x303>
[log in to unmask]<mailto:[log in to unmask]> http://cipherengineering.com
-----Original Message-----
From: Louis Hart [mailto:[log in to unmask]<mailto:[log in to unmask]>]
Sent: Wednesday, August 10, 2016 7:28 AM
To: TechNet E-Mail Forum <[log in to unmask]<mailto:[log in to unmask]>>; Carl Van Wormer <[log in to unmask]<mailto:[log in to unmask]>>
Subject: RE: [TN] looking for medium resistance material
Carl, if the conductor thickness is less than the skin depth, would that mean the magnitude of the evanescent wave electric field would be great enough to allow propagation outside the shield? Sorry to say, my understanding of electromagnetic field theory is a lot less than what I think it should be, but I am curious.
Regarding aluminum and soldering, I knew a guy who had a roll of 'aluminum solder' on his desk although I never saw him use it. The wire was several millimeters in diameters, so may not have been for electronic purposes.
On the subject Drew brought up of mu-metal, I inherited a microwave spectrometer long ago which had a klystron housed in the stuff. The fellow who bequeathed it to me said the metal was great for shielding, but its desirable properties were degraded by mechanical forming.
Nice to see good activity on TechNet. (I may qualify as one of the old-timers, having joined when it first appeared back in 1994 or 1995.) Louis Hart
-----Original Message-----
From: TechNet [mailto:[log in to unmask]<mailto:[log in to unmask]>] On Behalf Of Carl Van Wormer
Sent: Wednesday, August 10, 2016 9:16 AM
To: [log in to unmask]<mailto:[log in to unmask]>
Subject: Re: [TN] looking for medium resistance material
My next attempt will be to try a metalized Mylar, hoping that the thin conductor (much thinner than the skin effect depth) will be good enough for the capacitive shielding, but too thin for significant eddy current flow from the magnetic field. If that work, I will be left with the "simple" matter of connecting to the shield. How do you solder a wire to a party balloon?
Thanks,
Carl
Carl B. Van Wormer, P.E., AE7GD
Senior Hardware Engineer
Cipher Engineering LLC
21195 NW Evergreen Pkwy Ste 209
Hillsboro, OR 97124-7167
503-617-7447x303<tel:503-617-7447x303>
[log in to unmask]<mailto:[log in to unmask]> http://cipherengineering.com
|
