OK, we paved the faces of a machinist's vise with super-magnets.
We're getting about 10 mT, 100 gauss, in a reasonable volume, like this:
The little Fujitsu FTR-B3GA4.5Z relays, in the worst orientation, start to show effects at about 50 gauss and quit working entirely at about 100.
-- John Larkin Highland Technology, Inc jlarkin at highlandtechnology dot com http://www.highlandtechnology.com Precision electronic instrumentation Picosecond-resolution Digital Delay and Pulse generators Custom laser drivers and controllers Photonics and fiberoptic TTL data links VME thermocouple, LVDT, synchro acquisition and simulation
Got any mumetal foil (*)? Wrap the relay in that and see if it buys you any margin.
(*) Back in my OmniComp/GenRad days I had to cope with (nor influence) CRT's in close proximity to my electronics (portable testers). mumetal saved the day on many an occasion. ...Jim Thompson
--
| James E.Thompson | mens |
| Analog Innovations | et |
| Analog/Mixed-Signal ASIC's and Discrete Systems | manus |
| San Tan Valley, AZ 85142 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.
Cool! How big a field if you close it to, say, a 1.5 or 2 inch gap? I've got dreams of donut shaped magnet poles, and yokes that go around both sides.
100 gauss is a nice number. I did ~7 inch diameter helmholtz coils ~400 turns per side, ~ 10 ohms, that do 100 Gauss at 3 amps. I'd like to do ~2inch diameter at ~1 amp and 100 gauss. And then get to higher fields with permenant magnets.George H.
Roughly 30 mT (300 G) at 2" gap, 50 mT at 1". With the probe touching a magnet, we're seeing about 3 KGauss.
These are pretty dinky magnets, but about right for testing relays.
-- John Larkin Highland Technology, Inc jlarkin at highlandtechnology dot com http://www.highlandtechnology.com Precision electronic instrumentation Picosecond-resolution Digital Delay and Pulse generators Custom laser drivers and controllers Photonics and fiberoptic TTL data links VME thermocouple, LVDT, synchro acquisition and simulation
I once had problems with some relays that were near some electromagnet-actuated valves ... it'd make the contacts close sporadically. I just made a shield out of plain old galvanized sheet steel and that created enough margin to eliminate the problem. Mu-metal is a bit better (esp >60Hz) but costs a lot more than sheet 'tin'.
Changing the physical orientation of the relays might also serve. It just depends on how much extra "noise margin" is required. Finally, there's always solid-state "relays".
So, unshielded, you're about an order of magnitude short of the spec?
There's some old equations around for calculating shielding effectiveness (based on infinite hollow cylinders). I stuck them into a MATLAB script a few years ago- let me know if you need that or the paper they came from.
Best regards, Spehro Pefhany
-- "it's the network..." "The Journey is the reward" speff@interlog.com Info for manufacturers: http://www.trexon.com Embedded software/hardware/analog Info for designers: http://www.speff.com
OT: I once had a mechanical feedback problem.
A small relay controlled the coil of a very large contactor. The relays wer e mounted in the middle of the back cover of a rack cabinet. When the small relay energized, pulling in the armature of the large contactor in, it cre ated so much force that the back cover moved to disengage the small relay c ontacts and it would oscillate (made a H''' of a lot of acustic noise)
Cheers
Klaus
Looks like it.
I don't think it's feasible to shield 48 surface-mount relays on a VME board. What's weird is that the customer is buying four different boards from us for their system, and only one has this requirement. I think it's a legacy spec on one board, so we can probably talk them out of it. Political solutions beat hardware solutions every time.
-- John Larkin Highland Technology, Inc jlarkin at highlandtechnology dot com http://www.highlandtechnology.com Precision electronic instrumentation Picosecond-resolution Digital Delay and Pulse generators Custom laser drivers and controllers Photonics and fiberoptic TTL data links VME thermocouple, LVDT, synchro acquisition and simulation
Heh, heh ....... well ... people good at electrical engineering are not necessarily so sharp at mechanical engineering - and vice-versa. Feedback/oscillation can be a tricky foe, and only happen under very particular conditions - thus evading typical testing routines.
I once heard that the oxygen tank explosion that nailed Apollo 13 (an off-spec heater was inside) would only happen if the tank was x-percent full ... much more or less and it would have just run kinda hot, but OK. Of course the tests were done with the tank either full or nearly empty .... missing the critical interaction between x-mass of LOX and the heater coil.
I also once saw a large turbo pump that kept breaking. Odd thing was that it'd be repaired and then run just pefectly at every demand level. Trick turned out to be that at a *particular* RPM combined with a matching level of water from the river a whirlpool vortex would form around the intake of just the perfect size, shape and speed so that large gulps of air could be sucked in which caused serious uneven loading of the impeller turbine plus hammering from cavitation in the 250 horsepower pump.
The secret was that the "perfect" vortex took a couple of minutes to form ... and the testers ran the unit through the RPM range faster than that.
The cure was a fairly small "fin" pounded in just beyond the intake, which blocked the vortex rotation and thus the issue. A $100 fix ... after several multi-$k repairs :-)
I also knew a guy who bought a used golf cart and "refurbished" it. However it had an odd habit of running out of control. This was an older design that used automotive-type starter relays, progressively engaged by the "gas" pedal. The current would pass through some copper-tubing "resistors" and then to the motor. Clearly an electrical issue ! Something's shorting !!!
Well, during the refurbishment, the guy mounted the relay bar in a "more accessible place" ... which happened to be to some handy screw holes on the rear axle. Every time he drove over rough ground the sharp vibrations would cause the relays to cut in and out sporadically. The factory mounting spot was on the body, protected from shaking by the suspension springs.
The moral of the stories ... nature ALWAYS has some sneaky trick it wants to play on short-sighted designers ... and self-reinforcing patterns, be they mechanical, electrical or electromechanical, are its favorite kind of trick.
What's the term "emergent behavior" ? I think that's what led to the great northeastern blackout a decade or so ago. One overload leading to an oscillating surge of overloads across the whole northeastern grid and POW !!! :-)
Mechanical things can have bizarre behavior just like electronics. Take a look at this, which happened to me...
...Jim Thompson
--
| James E.Thompson | mens |
| Analog Innovations | et |
| Analog/Mixed-Signal ASIC's and Discrete Systems | manus |
| San Tan Valley, AZ 85142 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.
Hmmm, seems like I've had this happen to my toilet float-valve system more than once ... usually manifests itself in the middle of the night :-)
Still thinking about a "better" system.
We just got an email from the customer. There was a typo in the spec; the limit is 60 uT, not 60 mT.
Oh well, I at least got a spiffy mag field meter and a lot of magnets.
-- 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
On a sunny day (Tue, 07 Jan 2014 10:19:03 -0800) it happened John Larkin wrote in
Yea, that is why US needs Russian taxies to the ISS.
On a sunny day (Tue, 07 Jan 2014 20:09:40 -0800) it happened John Larkin wrote in :
Well, ....
It SHOULD cost him, next somebody orders 10 MW something... that turns out to be 10mW. Seen that often, but this is really bad.
There ought to be a catchy name for compliance by negotiating the specs.. something along the lines of "percussive maintenance" or "rubber hose cryptanalysis".
Best regards, Spehro Pefhany
-- "it's the network..." "The Journey is the reward" speff@interlog.com Info for manufacturers: http://www.trexon.com Embedded software/hardware/analog Info for designers: http://www.speff.com
optimising the requirements?
-- For a good time: install ntp --- news://freenews.netfront.net/ - complaints: news@netfront.net ---
Great list! Thanks for sharing. Didn't know about that mission failure's EXACT cause.
Especially like the idea of pointing out the error of thinking in terms of extremes as the only source of damage. When switching power supplies first came out, the engineers only considered OFF and ON of the AC mains as two test conditions, not 'brownout', as in, somewhere in between OFF and ON. And, yes, those first ones used to burn up.
If you have enough of these excellent anecdotes, PUBLISH! [if not, maybe collect them and then publish] Causes one to think a bit.
For example, maybe the human body [also a system] gets discombobulated with a 'range' of perturbation, not extreme. And, your mentioning 'time' for bad results to occur would explain why clinicians have missed any links. Wow, food for thought when examining all those unexplained epidemiologic analyses.
I have a van that I use for business, I insulated it and built it out with plywood. I held the wood to the floor with drywall type screws. One of those screws was right over the connector that plugs into the fuel pump on the fuel tank. I didn't find this out for a few years. I loaded the van with 1500lbs of frozen shrimp, drove about 1/3 of a mile on a bumpy road and the van stalled. Had to transfer the shrimp to another truck until I found the problem and cut the screw shorter. Mikek
;)
Cheers
Phil Hobbs
-- Dr Philip C D Hobbs Principal Consultant ElectroOptical Innovations LLC Optics, Electro-optics, Photonics, Analog Electronics 160 North State Road #203 Briarcliff Manor NY 10510 hobbs at electrooptical dot net http://electrooptical.net
We're still bidding the project, and I had some fun and learned some stuff, so no harm done.
60 uT doesn't make sense either. That's 0.6 Gauss, about the nominal Earth's field. Why bother so specify that?-- 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
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.