Strange Multi-Turn Potentiometer Problem

[Yellow Eyes]

We're trying to get some feedback on the various motions our robot's arm can make and two of them use the 100K 10 turn pots (Bourns 3590s). One of them is returning what seems to be fairly normal values to the RC (for rotating the arm), but the other (for extending the arm) just sits at max (255 with a bit shift, 1023 without) no matter where you turn it until about 1/8 to 1/4 of a turn from the end when it suddenly sinks all the way down. When tested on a multimeter though, the values are normal, 0 to about 100K ohms. The 2 pots are connected to rc_ana_in02 (extending) and rc_ana_in12 (rotating). If we swap them, rotating continues to work, and extending continues to just sit at max. Has anyone ever come across this before? Or does anyone know any potential fixes?
Thanks ahead.

[Jon236]

Originally Posted by Yellow Eyes

We're trying to get some feedback on the various motions our robot's arm can make and two of them use the 100K 10 turn pots (Bourns 3590s). One of them is returning what seems to be fairly normal values to the RC (for rotating the arm), but the other (for extending the arm) just sits at max (255 with a bit shift, 1023 without) no matter where you turn it until about 1/8 to 1/4 of a turn from the end when it suddenly sinks all the way down. When tested on a multimeter though, the values are normal, 0 to about 100K ohms. The 2 pots are connected to rc_ana_in02 (extending) and rc_ana_in12 (rotating). If we swap them, rotating continues to work, and extending continues to just sit at max. Has anyone ever come across this before? Or does anyone know any potential fixes?
Thanks ahead.

Sounds like a mounting/connection problem. We had that last year.

Jon
Team 236 Mentor

[Al Skierkiewicz]

Originally Posted by Yellow Eyes

We're trying to get some feedback on the various motions our robot's arm can make and two of them use the 100K 10 turn pots (Bourns 3590s). One of them is returning what seems to be fairly normal values to the RC (for rotating the arm), but the other (for extending the arm) just sits at max (255 with a bit shift, 1023 without) no matter where you turn it until about 1/8 to 1/4 of a turn from the end when it suddenly sinks all the way down. When tested on a multimeter though, the values are normal, 0 to about 100K ohms. The 2 pots are connected to rc_ana_in02 (extending) and rc_ana_in12 (rotating). If we swap them, rotating continues to work, and extending continues to just sit at max. Has anyone ever come across this before? Or does anyone know any potential fixes?
Thanks ahead.

It gives every indication of a bad connection to ground. The pot will always exhibit open circuit voltage i.e. 5 volts, until the lower side of the pot is attached to ground or the wiper is no longer contacting the element. The fact that numbers change near the end of the final turn may indicate that the wiper is no longer contacting the resistive element. Try running a test wire from the battery "-" terminal on the breaker panel to the lower pin on the pot and see what happens. (or measure the pot elements from the connector that plugs into the RC.) It should measure normal and when power is applied it should vary between +5 and 0 volts.

[coreyjon]

Thanks, it turns out that the pots where damaged by too much soldering, we had two, and both suffered from some type of failure (most likely too much heat). So after calling 20 places we finally tracked down a place not too far away with only 1 ten turn 100k ohm pot. We hopped in the Jake mobile, and in no time had the arm extend feedback working great. One question though, As I am mainly mechanical, and my quick dabble in electronics the other day was only due to the lack of more knowledgeable people, (I did take DE a few years ago) but how much heat can these pots handle, it seems like too much re soldering will cause the internals considerable damage. I'm sure This has happened before, I was just wondering if there is a rule of thumb..re-solder more than twice, expect a dead pot, or something along those lines.

[Joe Ross]

Originally Posted by coreyjon

One question though, As I am mainly mechanical, and my quick dabble in electronics the other day was only due to the lack of more knowledgeable people, (I did take DE a few years ago) but how much heat can these pots handle, it seems like too much re soldering will cause the internals considerable damage. I'm sure This has happened before, I was just wondering if there is a rule of thumb..re-solder more than twice, expect a dead pot, or something along those lines.

If you solder the pots correctly, there should be no heat related damage, no matter how many times you solder them. The problem is that most people who say they know how to solder really don't know the proper way, and even if they do haven't had enough practice.

[coreyjon]

The students doing the work I beleive had little to no experiance soldering - which is how I beleive we ended up in the mad dash to find a new 10 turn 100k ohm pot, no one around here (including radio shack stocked them) had to go to EE Tailor, and they only had one. While I am mostly inept when it comes to sensitive electrics, I can hold my own in a pinch. Anywho - I think everyone involved learned that too much heat is a bad thing, I beleive they had the soldering iron on high, and held it there simply too long. You live, you learn. In the end everything worked out fine, so it wasnt that big of a deal.

[Al Skierkiewicz]

Originally Posted by coreyjon

One question though, As I am mainly mechanical, and my quick dabble in electronics the other day was only due to the lack of more knowledgeable people, (I did take DE a few years ago) but how much heat can these pots handle, it seems like too much re soldering will cause the internals considerable damage. I'm sure This has happened before, I was just wondering if there is a rule of thumb..re-solder more than twice, expect a dead pot, or something along those lines.

There is no rule of thumb on this. A careful job with the correct technique can rework parts many times without damage. Often it is a combination of heat and pressure that cause the components to be pushed out of alignment. Since the majority of pots you can buy are plastic body construction, it doesn't take much to melt the plastic and push a terminal a fraction of an inch. Without knowing the part in question, it is also possible you have a conductive plastic element. Heat is disastrous to these types of pots. You might consider some type of push on terminal in the future. That way no heat is applied to the pot and it is much easier to change if damaged.

[ConKbot of Doom]

Yes, soldering sensitive things can be challenging, but a little tip, most people say heat the object with the iron, and let the heat from the object melt the solder, that works for terminals and wire, and other larger things. But stuff that can be damaged by heat, that kills them most the time.
For those of you reading through this and wondering the correct way, what you do is put the iron on the joint, then add solder at the junction between the iron and the object. I know this sounds like it might produce a cold solder, but the flux lets the solder flow onto the work piece, and it transfers heat to it quicker than just holding the iron on would, letting the joint get done quicker. On ring board, I can get a pin on a DIP IC done in about 3 seconds. This is usually below the lead soldering time, so all turns out well.
If anyone else has a method that works, I'm not saying that is wrong, I'm just saying that this works for me.

[Dad1279]

You can also place a 'heat sink' in between the solder joint and the item you want to protect by soldering. I use needle nose pliers with a rubber band around the handle. The steel of the pliers will absorb the excess heat. They should also be using a soldering 'pencil' not a higher wattage gun.

[Al Skierkiewicz]

The most common mistake is to hold a soldering iron against a terminal when the tip is not transferring heat effectively. The correct procedure is to first tin the wire, then wrap the wire around the terminal on the pot. Clean the tip of the iron by wiping it on a wet sponge. Add a small amount of solder to the tip of the iron to "wet" it. Touch the iron tip to both the terminal and wire such that the molten solder on the tip of the iron contacts both. Add a little solder to the joint and remove both the iron and solder. This operation should take no more than a second or two. Without this method, the dry iron tip will transfer heat during the time it is in contact with the terminal, but it will not achieve solder melt point. (for 60/40, .031 solders this is about 450 degrees) By "wetting" the tip, you can get the terminal and wire up to the iron temperature (about 750 deg) before the heat can transfer down the terminal to the pot innards. Anytime you cannot get solder to flow within a few seconds remove the iron and let the terminals cool.

[coreyjon]

Theese have been some great tips. I did "tin" the wires before I connected them to the pot, I did not however wet the soldering iron, although it did have molten solder on it from a previous solder connection. we are using a pencil, not a gun. Perhaps either for competition, or at least for next year, I will bring in my soldering station I got off of ebay, I think it was used in some type of electronics production, because it is very hefty and well built and has a LED readout of the temperature, along with finite temp control - overall a very nice unit. Makes it nice to fix electronics down in my "lab" - where many evil creations are built or fixed. The problem was our main electrical guy, and thats the official title by the way, was not around, I think he left on a trip...so the programmers where left to do some of this stuff. Not a big deal, we got everything working before it got crated up, and it works good. See you guys at competition.