How best to connect small wires to a TO-92 package and pot it in a tube?
ElectricAye
Posts: 4,561
Hi all,
I've been trying to pot some digital thermometers (DS18B20 in a TO-92 package) inside small tubing so they can be immersed in salt water and washed in alcohol. But I'm sick of trying to solder wires to the tiny prongs and not short the connections. I've been using heat shrink tubing after soldering the wires but I keep thinking there must be a better way of doing this. I'm also not so sure about the best way to pot them after they've been soldered. I've been using off-the-shelf epoxy but I'm not so sure it likes to stick to the PVC-insulated wires. Any one have any good alternatives?
many thanks,
Mark
I've been trying to pot some digital thermometers (DS18B20 in a TO-92 package) inside small tubing so they can be immersed in salt water and washed in alcohol. But I'm sick of trying to solder wires to the tiny prongs and not short the connections. I've been using heat shrink tubing after soldering the wires but I keep thinking there must be a better way of doing this. I'm also not so sure about the best way to pot them after they've been soldered. I've been using off-the-shelf epoxy but I'm not so sure it likes to stick to the PVC-insulated wires. Any one have any good alternatives?
many thanks,
Mark
Comments
how hot are the water and alcohol ?
is this ocean salt water - as in >50 feet depth?
- H
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Using standard #30 Kynar wire-wrap wire and manual wirewrap tool ( Radio Shack), put a lead on each of the TO-92 terminals. Solder if you wish. Put short length of heat shrink on each lead to cover the exposed conductors.
Pot as desired. Epoxy, hot melt glue, silicone, hi-temp automotive RTV
· Also, industrial heat shrink has a thermosetting glue inside it. Put a piece over the assembly instead of a piece of tubing. Heat it up and crimp both the ends (gently) with pliers to seal.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Tom Sisk
http://www.siskconsult.com
Post Edited (stamptrol) : 9/17/2009 7:41:47 PM GMT
Perhaps using a simple K type thermocouple would be easier to deal with environmentally, but you'd have to design an analog front end for it.
edit -
I would lean towards the wire-wrap idea myself.· I have read that correctly applied wire wrapping is actually better than soldering.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Post Edited (Agent420) : 9/17/2009 7:43:37 PM GMT
Leon
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Amateur radio callsign: G1HSM
Suzuki SV1000S motorcycle
Understood.· My experience has been that many TO92 transistor type packages utilize squared leads, as do many led's.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
This is for a biological test done in a lab at atmospheric pressure. But the temperature ranges from 0 F (solid ice) to room-ish temperature (80F). The saltiness is a characteristic of the microbial growth media, which is very close to that of the ocean. Alcohol gets involved when we have to sterilize the devices... or when happy hour gets out of hand.
To Tom, Agent and Leon,
that's very interesting about wire-wrap. I never knew the pins had to be square.
...Mike Green,
a very interesting idea! I had never thought about using some kind of silicone to pot heat shrink tubing. That would definitely keep the size down. But I'm guessing the devil is in the details here. Would you mind providing some exact suggestions as to what heat shrink tubing to use, what size, and what sort of silicone to pot with? Heck, I could even use a suggestion for a good heat gun since I don't even own one of those, either. Many thanks.
thanks everyone,
today's depression has been lifted by your gracious suggestions,
Mark
Sometimes the sensor needs a couple of extra parts with it for ESD protection or power filtering, etc. In that case, a small narrow circuit board can hold the parts and arrangement can be made with holes or pads for easier cable connection at the other end.
The approach using #30 wire-wrap+solder+heat shrink is quick and easy for short runs where the solid wire is desirable.
Some types of dual wall heat shrink stick much better than others to PVC. I like Raychem type ES2000. It has ample sticky liner capable of filling and sealing voids and the critical sneak path along the cable. Clean the outside of the cable with an solvent to remove the oil that they use for extrusion. ES2000 is also quite rigid on the outside layer when cool, which is good to prevent damage to the sensor and connections. It doesn't hurt to coat the sensor with silicone resin or epoxy (cured) before the adding the outer shrink layer.
If the sensor is going to be inside a tube, epoxy is okay, but it does not stick well to PVC, due to the plasticizers that make the cable flexible. I think there are primers available that can help greatly with the bonding but those AFAIK are not readily available. If you can use a gland nut on the cable and seal that to the tube with the epoxy, that could do the trick.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Tracy Allen
www.emesystems.com
Right - it also doesn't stick to PVC itself very well as PVC is a low energy surface - the pipe cleaners and primers help, but the MEK based ones can chemically interact with the epoxy.
> I think there are primers available that can help greatly with the bonding but those AFAIK are not readily available.
They don't work very well - and to get them, you usually have to buy a large quantity. Sanding the cable would probably work almost as good (which isn't very good).
> If you can use a gland nut on the cable and seal that to the tube with the epoxy, that could do the trick.
Yes - that's why I asked about the depth... (which probably doesn't matter, as I'm guessing E.A. is using this in the lab, not in the ocean [noparse]:)[/noparse]
- H
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
That's totally cool about the preform tubing. Also, any ideas on where to get those "small narrow circuit boards"? Is such a thing commercially available?
thanks,
Mark
Small "flea clips" (tiny alligator clips) on each leg of the device is a must and baffling the hot air from hitting the device package would be a good idea.
And to be on the safe side, check the action of the sensor before potting it.
As for potting material - it has to be non-conducting so anything with a metal composition wouldn't be a good idea. Someone mentioned silicon rubber (RTV?). Good idea, but I'd recommend the type used for aquariums as it's water proof.
Later,
DJ
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Instead of:
"Those who can, do.· Those who can't, teach." (Shaw)
I prefer:
"Those who know, do.· Those who understand, teach." (Aristotle)
·
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Post Edited (Agent420) : 9/18/2009 4:03:53 PM GMT
Dave, that stuff is OK but can have some problems [noparse][[/noparse]EDIT: including the anti-microbials!]
Silicone shrinks a little over time, can become brittle, and might be chemically affected by the alcohol. Also it works better at higher temps than lower. To get a consistent seal, you have to goop it on heavy - with a caveat. If there's a crack or openning created at initial install along a mating surface, you can have the stuff inches thick, and the liquid will wick in along the crack. I learned all that the hard way: yes, the radiosonde is still down some where about 80 feet below, "sleeping with the fishes."·:-|
cheers
- Howard
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Post Edited (CounterRotatingProps) : 9/18/2009 4:17:42 PM GMT
RE clear silicone in tube vs cauling-gun tube -
@Mike, you have to look at the ASTM numbers in addition to the intended application. Several of Dow's clear adhesives have the microbials Agent420 mentions because it's intended for sinks and bathtubs, etc. You have to get "100% pure".
I've used the little tube stuff for 200+ deg. C. applications, but it failed in a -20 to 10 deg. C. app. (Failure took ~4 months, however.)
- H
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
your comments about sealants, etc. being anti-microbial are definitely an eye-opener! Now I'm finding possible such problems even with some of the epoxies we were planning on using for other equipment - that's what I get for trying to take the cheap route, I suppose. But nothing would have been as costly as going down this road only to find our microbes are pushing up daisies because of the darn glue!
Once again, y'all saved my rear!
You asked about the small narrow circuit boards. I make my own. I attached a couple of pictures. One has a slit cut down the center so that when assembled it makes a very rigid "X" form with 4 V-grooves. The wires of the cable and the TO46 leads lie in grooves, as do components in between. The second type is a 2-sided flat board for a TO92 three lead package, with pads on top and bottom for the components and the wire connections. Send me a PM if you'd like to try a few of those.
Here is a construction article that uses the solder preforms as in the Parallax AD592 for Applied Sensors. Another construction article for the LM34 is here.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Tracy Allen
www.emesystems.com
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
MOORE'S LAW: The capabilities of electronics shall double every 18 months.
cloyd's corollary: Hardware is easy, software is hard.
Hi Tracy,
a thermowell sounds like a smart idea and is probably easy to do with my application. As for narrow circuit boards, I don't need any right now but I'll certainly keep this bookmarked for future reference. Sounds like something sensor people would need all the time - maybe you can supply Digikey, etc???. As for those solder preforms, I think I saw those at my local Home Depot. I didn't even know what they were so I'm glad you pointed out how nifty they are.
thanks!
Mark