Due to high demand, expect some shipping delays at this time - orders may not ship for up to 2-3 business days.
0

Switches of Great Resistance
Moderators: adafruit_support_bill, adafruit

Please be positive and constructive with your questions and comments.

Switches of Great Resistance

by USSJoin on Sat Jan 15, 2022 4:10 pm

I have an issue with the switches I ordered. (Order 2751522-6121516015)

The switches are https://www.adafruit.com/product/3307 and the other three colors. I got them to use as power switches for a ham radio portable box, so power efficiency is of some importance (not incredible importance--it's not like I can't tolerate four LEDs---but some).

However, after playing around with them a bit (they're very pretty! The switch is very tactile!), I happened to measure the resistance. I measured it with the same multimeter I've had and used for years, and which I fully trust. (It's not a $600 multimeter, but it's a decent one from Sparkfun and it hasn't steered me wrong.) The results, measuring across the two bottom pins (the ones the switch connects and disconnects), on four switches:

* 42 ohms
* 6 ohms
* 52 ohms
* 26 Mohms (yes, seriously)

Um... this seems odd, considering these are supposed to be switches. I'd expect them to be like 0.1ohm or so. At 42 ohms, four of them will take almost as much power as *the actual ham radio I'm using*---when it's transmitting! (It's a Lab599 TX-500, so it transmits at 2A-3A, and receives at 0.1A. By my calculations, a 42ohm switch at 14V (max
on the battery I'm using) will consume about 1/3 of an amp *per switch*.)

So: are these just badly broken? Am I completely misinformed about how switches ought to work? I've always loved Adafruit parts in the past, so I feel a bit like I must be wrong somehow, but... I think I'm not.

Advice would be greatly appreciated.

Thanks very much!
USSJoin
 
Posts: 10
Joined: Tue Dec 28, 2010 12:24 am

Re: Switches of Great Resistance

by danhalbert on Sat Jan 15, 2022 5:45 pm

I think the description on the product page may be wrong. Are you measuring resistance between the two bottom pins? I think you're measuring the LED resistor.

See the description in this guide, which says that the basic "switch part" is not the two bottom pins, but one side and one bottom pin:
https://learn.adafruit.com/mystery-box- ... -2194553-6

Let us know what you find out and we'll get the product description corrected.

danhalbert
 
Posts: 3158
Joined: Tue Aug 08, 2017 12:37 pm

Re: Switches of Great Resistance

by USSJoin on Sat Jan 15, 2022 10:35 pm

Unfortunately, I don't believe this is correct. I grabbed a green switch and measured resistance between all three pairs of contacts just now.

When the switch is in the "off" position: all three pairs (front to bottom-front, front to bottom-back, bottom-front to bottom-back): infinite resistance (no connection).

When the switch is in the "on" position: 7.8ohm between bottom-front and bottom-back. No connection between the other two pairs.

So the switch is definitely meant to switch the two bottom pins together, which is as the product description says.

Any chance you could pull a few of these out of your stock and check on your end? It seems unlikely I got eight bad switches (but not impossible!), but if yours are the same then there may be QC issues at your supplier? (Or I'm open to other suggestions as well.)

Thanks very much!
USSJoin
 
Posts: 10
Joined: Tue Dec 28, 2010 12:24 am

Re: Switches of Great Resistance

by danhalbert on Sat Jan 15, 2022 10:39 pm

Thanks for the comprehensive check. I don't have any of these, so I'll have others check what is going on.

danhalbert
 
Posts: 3158
Joined: Tue Aug 08, 2017 12:37 pm

Re: Switches of Great Resistance

by jps2000 on Sun Jan 16, 2022 3:31 am

I just can give a general comment:
The contact resistance of a switch/ relay at very low voltages / current is sometimes quite high due to an oxide layer on the contact surface.
This should be no problem once significant voltage or current is applied. These oxide layers then break through and resistance is low as expected.
So just switch a 12V bulb and measure the voltage across the contacts in on.
Also contact spring design plays a role. Looking a relay operation you see there is also a little horizontal movement of the contacts that rubs away oxide or dust.
If you need to switch very low currents / voltages then yo sould go for gold contacts or the like.
A reed contact is close to ideal as it works under protecting gas

jps2000
 
Posts: 728
Joined: Fri Jun 02, 2017 4:12 pm

Re: Switches of Great Resistance

by blakebr on Sun Jan 16, 2022 3:59 pm

jps2000 is 100% correct. Very low currents do not properly 'seal' the contacts of a switch. Try cycling them 30 or 40 times and retest.

Each time a switch closes it must dig through any protective oxide or debris that may have accrued on the contacts and create a micro-weld that actually does the conducting. Those micro-welds build up debris after a while and must be pushed out of the way be subsequent connections. That is why in switches that are designed for low current applications have a wiping or scrubbing action when they close. Along with the micro-weld debris on the surface areas of the contacts (except gold) can and do develop an oxide layer. Most oxides are insulators and must be penetrated for the micro-weld to take place. The wiping action will generally penetrate this layer. In some cases the voltage is high enough to break through the oxide layer. If you can touch it, it ain't high enough to break through.

If you want to see the oxide layer disintegration in process, put some light oil on aluminum and rub your finger back and forth on it. The black residue you start to see is the aluminum oxide that is being dissolved by the oil. As soon as it is dissolved it reforms from fresh oxygen and the freshly exposed aluminum.

Bruce

blakebr
 
Posts: 269
Joined: Tue Apr 17, 2012 6:23 pm

Re: Switches of Great Resistance

by USSJoin on Tue Jan 18, 2022 1:56 pm

These switches are designed for low current, according to the documentation. I'd appreciate a followup answer from an Adafruit employee.
USSJoin
 
Posts: 10
Joined: Tue Dec 28, 2010 12:24 am

Re: Switches of Great Resistance

by danhalbert on Tue Jan 18, 2022 2:50 pm

I am getting some to test. In the meantime, could you try switching with some significant current (no more than rated) to see if there is a difference in contact resistance afterwards?

danhalbert
 
Posts: 3158
Joined: Tue Aug 08, 2017 12:37 pm

Re: Switches of Great Resistance

by danhalbert on Sun Jan 23, 2022 9:42 pm

I got five switches yesterday and tested them all. Several showed initial higher resistance, but then I put them in series with 100 ohm resistor, and fed the series arrangement with 5V (so ~50mA current). Then I measured the voltage drop across the switch. Afterwards I removed the power and measured the resistance again. I was able to use a multimeter that could zero out the resistance of the leads.

I also measured a bunch of other switches I have, some rated for 12V (NOS from Radio Shack and similar), and some ones that are probably 50 years old.

For four of the five Adafruit switches, after switching the 5V a few times, their resistance fell to a few tenths of an ohm, and there was no significant voltage drop across the switch. For one, I saw about 0.2V drop across the switch, and it still measured a few ohms afterwards.

I had a similar experience with my older switches, which showed noticeable resistances until actually put to use, and then they went to similar tenths of an ohm.

So based on the comments of blakebr and jps2000, I would say except for the one flaky one, the switches were acting as expected. If you try something like this and still see significant resistances, we can of course do a refund or replacement.

danhalbert
 
Posts: 3158
Joined: Tue Aug 08, 2017 12:37 pm

Re: Switches of Great Resistance

by USSJoin on Thu Feb 03, 2022 6:13 pm

@danhalbert

Thanks for this, and apologies for my delay. Tried a similar experiment, switching 0.8A@12V with a regulated bench power supply, and indeed, resistance dropped from (e.g.) 66 Ohm to a few tenths of an ohm. So the switches work as designed; thank you!

I might suggest that to help people like me who didn't know this about switches you update the product page with how to do this? (Maybe I'm the only person who didn't know, however, in which case it's fine as-is. :-) )

Thanks again!
USSJoin
 
Posts: 10
Joined: Tue Dec 28, 2010 12:24 am

Please be positive and constructive with your questions and comments.