Any kind of liquid resistors out there?

March 27, 2009 2:42 PM

What to coat connectors with to create a high resistance connection?

Hi! I am trying to observe what the effect is of a high resistance connection. I'm trying to coat the pins of a connector with something in order to achieve this. I'd like to get a couple ohms between the two connectors (like 3-10 ohms). I've played with super glue on the connector, and find that I get about 0.08 ohm (using 4-wire method of resistance measurement) or just an OL. The 0.08 heats up, but not as much as I calculate a 1 or more ohm connection would. I'm now playing with some old silver resin, but was wondering if the hive mind has any ideas. I've tried just making a loose connection and can't get it to be consistent enough. Ultimately, I believe that something I coat over the connection will produce the results I'm after. Some sort of liquid resistant material that will stick to metal and the more I put between the mating pieces of the connector the higher the resistance will be (but can only get a few mils worth of material).

Thanks!

posted by nickerbocker to sports, hobbies, & recreation (23 comments total)

Maybe I'm missing something here, but why don't you just calculate the resistance you desire and wire in a resistor? The cost should be negligible.
posted by torquemaniac at 2:46 PM on March 27

Because I want to see what happens within the connector due to a high-resistance connection. I want to get a fire.
posted by nickerbocker at 2:50 PM on March 27

Not for arson (lol). This is a laboratory experiment thing.
posted by nickerbocker at 2:50 PM on March 27

Getting a uniform thickness using a viscous liquid (i.e. superglue) will be nearly impossible with a simple dipping approach. Getting reliable resistive contacts between two metals is tough; perhaps you have a way of making a metal/semiconductor junction?
posted by JMOZ at 2:51 PM on March 27

Oh, and burning up a resistor is not hard at all.... Increasing the temperature is simply a matter of Joule heating (i.e. I^2*R or V^2/R), so increasing the current (or voltage) is sufficient to get a fire; simply exceed the power limit of your resistor. Many resistors are rated for as little as 1/4 Watt, so torquemaniac's idea is certainly workable.

For what it's worth, bad connections are a fact of life, and when I was in grad school, the copper connector on our chiller actually MELTED from a little series resistance. (Amazing was 208V and 30-50A of current can do with even an imperfect connection, eh?)
posted by JMOZ at 2:54 PM on March 27

I guess in the ballpark is good enough. Consistency, as in between 3-50ohms or something. The problem with Superglue is it drys as an insulator... i was relying on it limiting the metal-to-metal connection of the two pieces of my connector. Most of the time I would get a really low resistance...jiggle the connector...so high of a resistance no current would flow. The connection goes to a 3 ohm load with a 12VDC source....so a 3 ohm high resistance connection would provide max power at the connector and probably make my fire.
posted by nickerbocker at 2:56 PM on March 27

Oh... it is undesirable for me to blast the connection with current, btw. The most ideal thing is for me to have about a 3 ohm connection. Lol... how funny that when you want to make a high resistance connection it is so difficult, but when you don't want one you get it and it messes things up.
posted by nickerbocker at 2:59 PM on March 27

I suspect there's a ton of stuff you'll need to take into account if you're trying to simulate some kind of real-world equipment failure.
posted by le morte de bea arthur at 2:59 PM on March 27

Shhhhhh... le morte de bea arthur. You are going to get me in trouble.

Lets keep the answers to the point and not go much further than that. I've probably already said too much.
posted by nickerbocker at 3:01 PM on March 27

I think this is going to be very hard to do consistently, but I would suggest using some sort of acid to corrode the metal of the male connector.

The hard part here is how you avoid having the resistive layer (whatever it is) getting rubbed off when the male plugs into the female, and I don't see any way to avoid that happening some of the time. Connectors are made the way they are in order to do that deliberately.
posted by Chocolate Pickle at 3:03 PM on March 27

You might consider using the superglue, wiping it thin, and making pinholes in it. It might be possible to get to the desired resistance range in that way.

Alternatively, if possible, sand a pin until the fit is poor and measure resistance frequently.

Eventually, you'll get a bad connection and you can demonstrate that your competitor's power supply causes fires. (Or whatever you're trying to do.)
posted by JMOZ at 3:04 PM on March 27

Create an oxide layer on the terminals. This is what causes terminal ressistance in the real world.

Put them in ferric chloride or some strong acid to strip off the outer layer of metal, and with it any non-corrosive metal plating, exposing the cheaper underlying metal to the atmosphere. Either
- just leave them to corrode - it can happen surprisingly quickly - days instead of years (make the atmosphere oxygen richer if that's not too hard, to speed it along)
- for instant, controlled results, talk to someone in the chemistry about helping it along (such as the anodizing process used to colour Al)
- also for instant results, you could use terminals with a ferric base metal, and spend a few dollars on a cheap bottle of gun blueing liquid.

Alternatively, use the ferric chloride to strip off some metal simply to make the connectors looser, and then coat them in graphite lubricant ($3 at the locksmith stand of your local department store). I don't know how well that might work, but if you're going to coat the terminals in something, as you've suggested, you need to make the connection looser to accommodate the new material, and a resistive but conductive substance like graphite seems like a good thing to try. Perhaps mix it in with glue for a conductive but resistive glue.

Similarly, off-the-shelf circuit-trace pens, and conductive glue, could be useful. These are a combination of fine metal dust with binders that set into a solid, and tend to have a bit of resistance to them.
posted by -harlequin- at 3:18 PM on March 27

Is ferric chloride used for etching circuit boards? I have ammonium persulfate... do you think that would work in its place. I wish I had a chemist handy lol.

Thanks for the suggestions harlequin, I'll give it a try.
posted by nickerbocker at 3:28 PM on March 27

Try dipping them in talcum powder-- the real thing, not the starch based stuff. Bike shops sell a completely unscented variety for prevent tubes from getting pinched in the tires.

If the resistance of that is too high, try mixing in some pencil graphite.
posted by jamjam at 3:39 PM on March 27

I don't think ferric chloride will work on iron. It reacts with copper by displacement (forming copper chloride and iron). Someone correct me if my chemistry is wrong.

I wouldn't risk ammonium persulfate either:

Incompatibilities:
Acids, alkalis, halides (fluorides, chlorides, bromides and iodides), combustible materials, most metals and heavy metals, oxidizable materials and other oxidizers, reducing agents, cleaners, and organic or carbon containing compounds. Contact with incompatible materials can result in a material decomposition or other uncontrolled reactions.
posted by le morte de bea arthur at 3:40 PM on March 27

The connectors are both made of brass coated with tin.
posted by nickerbocker at 3:49 PM on March 27

not sure what the resistance of it is, but it may be worth trying the rear window defrost repair stuff. it paints on over the old traces, and has enough resistance to heat the window.
posted by KenManiac at 4:06 PM on March 27

I've put tin-plated brass connectors in ferric chloride. It sort of works - it doesn't dissolve the tin so much as crack it and partly flake it off, at which point you'll want a wire brush (or better - a wire-brushhead and a dremel) to finish the job.
The brass underneath gets sanded to a shine by the wirebrush, but is unprotected and tarnishes quickly (days/weeks). I'm not sure how to accelerate the tarnishing of brass. However a lot of connectors that have a brass base metal can be purchased in other metals. But if they're a highly machined part like BNC plugs, it's going to be harder to find it in something other than brass.

Alternatively, perhaps there is a way to anodize the tin? The blueing liquid might be worth a shot because it doesn't cost much, and is semi-effective on a fair few alloys if soaked in it for a 15 minutes, instead of the instantaneous brush-on effect when used on steel or iron.
posted by -harlequin- at 4:21 PM on March 27

"I wouldn't risk ammonium persulfate either:
...Contact with incompatible materials can result in a material decomposition or other uncontrolled reactions."

A material decomposition is the desired outcome. :-)

(Also, regarding the ferric chloride sort-of works on brass+tin, I suspect the tin is bonded to the brass via a copper plating.)
posted by -harlequin- at 4:29 PM on March 27

J.B. Weld?
posted by torquemaniac at 4:38 PM on March 27

I was going to suggest anodizing also. As I understand it, the thickness of an anodized coating is controlled by the voltage you've applied in the anodizing bath— it grows until it's thick enough to resist further action. So it could be a good way to get a uniform, controllable thickness. But I have no clue what chemistry you'd need to anodize tin or brass.

Alternately, perhaps you could mix graphite and a very thin binder into a slurry (dilute glue? KenManiac's defroster-repair suggestion?), then dip the connector and let it dry, repeating as needed. If you can make the slurry consistent, each dip should add a consistent thickness of somewhat-conductive graphite+binder.

But I think Chocolate Pickle points out the most difficult aspect, which is that the connector is designed to scrape off any surface crud each time you connect/disconnect it.
posted by hattifattener at 5:20 PM on March 27

How about surface oxidation via flame (eg. match or lighter)? Should take just a few seconds to build a surface oxide.
posted by ZenMasterThis at 7:03 PM on March 27

When I was in graduate school, I had a full time job as a product manager for a company named CTS, Corp. (They use to be a big name in resistors.) My line was hearing aid volume controls.... largest .250 dia and smallest 0.100" diameter!

We made resistors from bakelite substrates and carbon bearing paint. The paint vehicle was some sort of normal paint VOC, and the carbon percentage yielded a specific Ohms-per-square value. Thickness impacted the values, but the paint carbon content dictated the Ohms-per-square.

It's an interesting concept... if you measure two opposite points on a square of material of any size, then look at an even larger square, you get the same reading. The additional series distance is offset by larger parallel paths and the resistance stays the same. Cool stuff.

Anyway... carbon paint is one option.

Another is cer-met. Probably not suitable for you because you need exotic materials for the metallic part and a high temperature substrate for the ceramic portion. We used glazed alumina substrates. Very stable thin films. Tricky, but stable.

I'd look into resistive paints. Diluting with normal solvents can adjust the Ohmic readings and repeated applications can reduce the values.

Have fun! Ain't science grand?
posted by FauxScot at 7:33 PM on March 27 [2 favorites]