switchtesting , (june13th2015:additional test by MRsDNF in post#44)

60 posts / 0 new
Last post

Pages

djozz
djozz's picture
Offline
Last seen: 7 hours 16 min ago
Joined: 09/07/2012 - 17:04
Posts: 17312
Location: Amsterdam
switchtesting , (june13th2015:additional test by MRsDNF in post#44)

 

I got these two 100W 0.5Ohm resistors in the mail yesterday, so I could make me a test-load to do current tests now, I wanted that for a while. I screwed them down on a MCU-heatsink with fan, and wired them parallel, now it is a maximum of 200W test-load, it should handle currents up to 28A.

This thread about a new 6A switch that I found on ebay made me want to test a few switches, I asked other BLF-members how a switch test should look like (the concensus was that long periods on high amps is what kills a switch, not the number of clicks), and tonight I started some testing on three switches that I had and will destroy for some greater good, I hope...  For completeness I would have liked to include the Tofty switch, but I don't have one, and Scaru did some pretty tough testing on that one already, actually more thorough than I am doing with these.

This is the test-rig (oven tray science!):

The circuit goes from powersupply to the test-load to the three switches in series back to the power supply, all thick copper wiring, almost everything soldered thoroughly together. The three switches are on a ceramic tile, not that I expect fire or anything serious when a switch blows, but it will run for many hours so I can not continuously guard the test.

The three switches are: a small Omten switch, a larger Omten switch, and the ebay-6A switch.

I started the test tonight at 3A (the power supply will give 3A regardless of any resistance changes, up to 16V), I think all three switches would survive lower currents than that. I have run the switches for four hours now and they all live. Tomorrow I will do four hours on 6A and four hours on 10A. Any switch surviving that will get 16A sometime this weekend.

Measuring the voltage over the switches at 3A already shows differences: the bigger the switch, the lower the voltage (and thus the resistance), I also measured the voltage over a piece of the copper wire:

     

Actually the voltage (resistance) varies after each switch-on, and after each switch-on it also changes a bit, usually it starts a bit higher, and it settles at a bit lower voltage after 10-20 seconds. Here's some voltages, at 3A I did 10 switch-on cycles on each switch and noted the voltage reading (you can calculate resistance and power generated in the switch yourself! Tongue Out)

*small Omten switch: 25 23 17 20.5 18.7 15.9 15.9 15.9 18.1 23 mV

*large Omten switch: 17 15 14.9 18 16 15.5 15.3 15.3 15.3 15.3 mV

*6A ebay switch: 5 5.5 6.4 5.4 7 7.5 6.2 5.2 5.8 6.3

 

After the 4 hours at 3A, I did another 10 clicks to see if the resistance had changed:

*small Omten switch after 4 hours at 3amps: 37.5 18 26 26 16 17.4 15.6 14.5 12 13.5 mV

*large Omten switch after 4 hours at 3amps: 13.2 13.5 12.3 18.6 15 14.4 13.5 11.5 13.2 12.9 mV

*6A ebay switch after 4 hours at 3amps: 6.2 5.3 5.2 5.6 5.1 6.1 4.9 6 5.2 5.5 mV

 

Nothing exiting yet, I thought that the small Omten had gotten a higher resistance, but after a few clicks it was back to what it was before the four-hour test.

6A test tomorrow! It was fun to build up the test-thing, but the testing itself turned out to be pretty boring to be honest Undecided.

 

Update: 6A results in post#14

Update: 10A results in post #17

Update: 16A results in post #25

Edited by: djozz on 06/13/2015 - 07:57
MRsDNF
MRsDNF's picture
Offline
Last seen: 11 hours 30 min ago
Joined: 12/22/2011 - 21:18
Posts: 13452
Location: A light beam away from the missus in the land of Aus.

What can I say but thanks. Looking forward to the results.

 

djozz quotes, "it came with chinese lettering that is chinese to me".

                      "My man mousehole needs one too"

old4570 said "I'm not an expert , so don't suffer from any such technical restrictions".

Old-Lumens. Highly admired and cherished member of Budget Light Forum. 11.5.2011 - 20.12.16. RIP.

 

FlashPilot
FlashPilot's picture
Online
Last seen: 2 min 52 sec ago
Joined: 05/10/2010 - 16:07
Posts: 4672
Location: USA

Thank you so much for doing this for us djozz! :bigsmile: This should put to rest many of my future switched build concerns.

RaceR86
RaceR86's picture
Offline
Last seen: 4 years 9 months ago
Joined: 02/22/2012 - 08:32
Posts: 3777
Location: Norway

This is great!

When I say the "typical large Omten swtich" i usually refer to this (PBS-101). PBS-101 is the one that is most common as far as I have seen. Not sure if its more durable than the one being tested (PBS-101C). Maybe we should start using proper names?

PBS-101

PBS-101C

Assuming those are the real names.

BLF LED database – collaboration spreadsheet and latest news about where to buy LEDs
http://budgetlightforum.com/node/19342

djozz
djozz's picture
Offline
Last seen: 7 hours 16 min ago
Joined: 09/07/2012 - 17:04
Posts: 17312
Location: Amsterdam

RaceR86 wrote:

This is great!

When I say the "typical large Omten swtich" i usually refer to this (PBS-101). PBS-101 is the one that is most common as far as I have seen. Not sure if its more durable than the one being tested (PBS-101C). Maybe we should start using proper names?

PBS-101

PBS-101C

Assuming those are the real names.

I know, I should have one of those somewhere but can not find it. If I manage to find it tomorrow it will join the party at 6A Smile

I did not know the Omten switches had type nr's, if I can also find a type nr with the small Omten I will update the OP with them. Now sleep...

wight
Offline
Last seen: 8 months 2 weeks ago
Joined: 11/27/2013 - 16:40
Posts: 4969
Location: Virginia, USA

Subscribed, thanks for the testing!

Still fine, still on a break. One day I’ll catch up with you folks! previous wight catchup Wink
list of my drivers & variants (A17DD, FET+1 stuff, WIP stuff, etc)

Ouchyfoot
Ouchyfoot's picture
Offline
Last seen: 8 months 2 weeks ago
Joined: 09/01/2012 - 06:15
Posts: 5089
Location: Canada

Watching and waiting. Thanks djozz.

likevvii
Offline
Last seen: 4 hours 23 min ago
Joined: 04/28/2014 - 10:12
Posts: 404
Location: USA, California

Thank you so much for doing this test! How come they rate these switches at 1.5A?

Hi-Beam
Hi-Beam's picture
Offline
Last seen: 5 years 1 month ago
Joined: 06/01/2014 - 22:37
Posts: 248
Location: TX

Nice, you are off to a great start here. Looking forward to the 6 amp and above test.

Cereal_killer
Cereal_killer's picture
Offline
Last seen: 11 months 7 hours ago
Joined: 07/22/2013 - 13:10
Posts: 4005
Location: Ohio
likevvii wrote:
Thank you so much for doing this test! How come they rate these switches at 1.5A?

Cause they rate them at 250vAC

 RIP  SPC Joey Riley, KIA 11/24/14. Now I am become death, the destroyer of worlds.

likevvii
Offline
Last seen: 4 hours 23 min ago
Joined: 04/28/2014 - 10:12
Posts: 404
Location: USA, California

Just to be sure that i am on the same page + other people who are still trying to learn electronics,
“*6A ebay switch after 4 hours at 3amps: 6.2 5.3 5.2 5.6 5.1 6.1 4.9 6 5.2 5.5 mV”

These numbers are the voltage drop?
To calculate heat dissipated from the switch, I multiply the voltage drop, 6.2mV X 3A = 0.0186W?

Really interesting. Do you think you can do a test of a mcclicky or some high end switch to get a better point of reference? or some examples of bad switches? 0.02W seems like nothing to me, but is it a problem?
As time passes, the voltage/resistance gets lower right? “(the concensus was that long periods on high amps is what kills a switch, not the number of clicks)” So at a magic number of amps, the switch would actually increase in voltage? How come the resistance drops, but when it hits a certain amp, it would rise? Molecules vibrate faster and faster because of more and more heat???? Assumptions on top of assumptions lol…
Can anyone please explain a little on how this works? thanks!

djozz
djozz's picture
Offline
Last seen: 7 hours 16 min ago
Joined: 09/07/2012 - 17:04
Posts: 17312
Location: Amsterdam

sorry RaceR, I did not find that square Omten switch, that one will have to wait. But the set-up is so simple, I can just do another session with a new string of switches.

Started the 6A session almost an hour ago, no failures sofar. The small Omten measures 45mV at 6A, that is about 0.3W of heat it has to dissipate. We're getting somewhere Smile

At 6A, I just clamped the voltmeter over 1cm of the lead of the large Omten switch, and was a bit surprised to find 4.8mV, that stuff is not overly conductive!

djozz
djozz's picture
Offline
Last seen: 7 hours 16 min ago
Joined: 09/07/2012 - 17:04
Posts: 17312
Location: Amsterdam

likevvii wrote:
Just to be sure that i am on the same page + other people who are still trying to learn electronics, "*6A ebay switch after 4 hours at 3amps: 6.2 5.3 5.2 5.6 5.1 6.1 4.9 6 5.2 5.5 mV" These numbers are the voltage drop? To calculate heat dissipated from the switch, I multiply the voltage drop, 6.2mV X 3A = 0.0186W? Really interesting. Do you think you can do a test of a mcclicky or some high end switch to get a better point of reference? or some examples of bad switches? 0.02W seems like nothing to me, but is it a problem? As time passes, the voltage/resistance gets lower right? "(the concensus was that long periods on high amps is what kills a switch, not the number of clicks)" So at a magic number of amps, the switch would actually increase in voltage? How come the resistance drops, but when it hits a certain amp, it would rise? Molecules vibrate faster and faster because of more and more heat???? Assumptions on top of assumptions lol... Can anyone please explain a little on how this works? thanks!

You are correct, 0.0185W is nothing....yet. But the power increases with the current-square. I am now testing 6A, that generates 4 times the heat of 3A. The small switch does 0.3W already, and these switches are not designed for great heat dissipation.

At 3A, the resistance did not get lower over time, as was shown above , that was just about the first 20 seconds, and it could have something to do with the multimeter I am using (although both -very different- multimeters that I have show the same effect)

I am happy to torture any switch, but currently I don't have many different ones. If there's interest for it I can do a shopping session for a few more, but killing McClickies sounds a bit costly Undecided

likevvii
Offline
Last seen: 4 hours 23 min ago
Joined: 04/28/2014 - 10:12
Posts: 404
Location: USA, California

Thank you for confirming my guess. I dont want you to kill an expensive switch! XD I just want to see the switch run at around 5A-6A so we can decide if spending an extra 6$ for an expensive clicky is really worth the gain. With your results so far, I would definitely buy an expensive switch because .3W is horrendous!
Have you taken apart the omten switches before? You can replace the parts inside with a copper cut sheets. instead of the original ones.

I only made one switch but it is now in a light that is permanently sealed ):
I also stretched the spring, and sandpapered the contacts to hopefully get better results.
I have 2 otmen switches comming in from FT. I will definitely try to post what I have done. But I do not know if it is any good. ):

djozz
djozz's picture
Offline
Last seen: 7 hours 16 min ago
Joined: 09/07/2012 - 17:04
Posts: 17312
Location: Amsterdam

ok, the 6A results are in, we got merely survivors! Despite the 0.3W heat even the small Omten went strong. I measured voltages over the switches at the beginning 10 clicks, halfway 10 clicks and at the end 10 clicks. No increasing resistances here, the small Omten even did a bit better at the end.

Small Omten 6A begin: 38 41 49 45 49 41 41 52 45 40 mV

                   6A 2hrs  : 38 43 40 43 50 49 63 71 45 47 mV

                   6A 4hrs  : 29 42 33 31 32 33 26 38 32 30 mV

Large Omten 6A begin: 22 24 22 22 21 25 22 26 21 23 mV

                    6A 2hrs : 18 30 23 32 25 25 25 31 24 26 mV

                    6A 4hrs : 18 31 25 32 23 22 24 20 21 29 mV

6A ebay sw.  6A begin: 10 13 12 10 12 15 13 11 13 12 mV

                   6A 2hrs  : 9 13 13 12 12 11 14 12 11 12 mV

                   6A 4hrs  : 11 13 13 13 16 13 14 15 13 13 mV

 

Started the 10A:

djozz
djozz's picture
Offline
Last seen: 7 hours 16 min ago
Joined: 09/07/2012 - 17:04
Posts: 17312
Location: Amsterdam

1.5 hours underway at 10A, no failed switches as yet. Did 10 clicks on each switch again: everything works normal, no increasing resistances yet. I don't know what you guys do with your switches but it sure looks like it is not current that kills this type of switch (these are switches with metal plates that are pushed onto the leads by a spring: the current is not flowing through the spring, see likevvii's post #13).

I did a very accurate temperature measurement (using my lips to sense the temperature Laughing), the housing of the two bigger switches are luke-warm, the small switch is a bit warmer, all the leads, including the copper wires are pretty hot, but not 60degC+ hot.

RaceR86
RaceR86's picture
Offline
Last seen: 4 years 9 months ago
Joined: 02/22/2012 - 08:32
Posts: 3777
Location: Norway

djozz wrote:

I don't know what you guys do with your switches but it sure looks like it is not current that kills this type of switch

 

The only switches I have killed have been some very junky stock switches that did not handle 10A.  Ive never had an Omten switch fail in the 10-12A range. I have not gone higher on those switches. But I believe others have run them at around 15A.

Im curious to see the difference between the two omten switches in this test. 

In my latest hot rod build, I have made my own "twisty switch". It should be good for lots and lots of amps, and then some..  I would not trust a typical Omten switch for what Im putting together..  J)

BLF LED database – collaboration spreadsheet and latest news about where to buy LEDs
http://budgetlightforum.com/node/19342

djozz
djozz's picture
Offline
Last seen: 7 hours 16 min ago
Joined: 09/07/2012 - 17:04
Posts: 17312
Location: Amsterdam

10A results. After four hours all three switches work fine. Over the small Omten there is a somewhat higher voltage variance in the end, but nothing I worry about. As said the small switch warms up more, but that does not seem to lead to failure yet. Of course for critical applications, like single li-ion cell lights it is nicer to have the larger switch because it causes a smaller voltage drop. In fact when using the small Omten at 10A the variance upon different clicks can be as high as 0.05V and that could be visible in the output of the light, or lead to a bit shorter period the driver stays in regulation.

What surpises me is that the voltage increase is not lineair with the current, it lags a bit behind, so the resistance decreases a bit when current increases (or because the temperature increases), which is good for the performance. Another thing that is good for the performance: as you can see with almost all 10 click voltage series, the first value is low compared to the average, it is the voltage reading before the first click, it means that if you leave the switch alone the resitance decreases somewhat over time.

Small Omten 10A begin : 54 56 70 59 57 62 60 55 59 55 mV

                   10A 1.5hrs: 39 62 54 64 61 68 51 56 60 63 mV

                   10A 4hrs   : 55 73 79 100 92 62 52 49 93 51 mV

Large Omten 10A begin  : 34 39 42 35 39 50 40 54 54 58 mV

                    10A 1.5hrs: 37 70 38 47 40 49 68 80 60 43 mV

                    10A 4hrs   : 34 46 51 53 81 88 59 50 54 52 mV

6A ebay sw.   10A begin : 19 18 18 18 20 22 20 22 20 18 mV

                    10A 1.5hrs: 15 16 16 19 17 18 18 24 20 27 mV

                    10A 4hrs   : 20 24 38 41 20 20 19 20 19 19 mV

 

All three switches make it to the 16A test! It has to wait a bit because my girldriend hates the sound of the fans, makes her think of bad airco days Undecided.

I can't help noticing that making a high amp switch is cheap and easy: a plastic housing with ballpoint click-mechanism, three pieces of metal, a spring. I can't comment much on the type of metal used, but these good performing Omten switches are not exactly using the metal with the best conductance.

MRsDNF
MRsDNF's picture
Offline
Last seen: 11 hours 30 min ago
Joined: 12/22/2011 - 21:18
Posts: 13452
Location: A light beam away from the missus in the land of Aus.

djozz wrote:
1.5 hours underway at 10A, I did a very accurate temperature measurement (using my lips to sense the temperature Laughing), the housing of the two bigger switches are luke-warm, the small switch is a bit warmer, all the leads, including the copper wires are pretty hot, but not 60degC+ hot.

djozz wrote:
All three switches make it to the 16A test! It has to wait a bit because my girldriend hates the sound of the fans, makes her think of bad airco days Undecided.

Are you not sure that it is not the first comment that your girlfriend wants the testing stopped though you did clarify by saying they weren't very hot.

Sometimes I wonder about members here. Orsm testing, still, by the way. Thanks again. Undecided

 

djozz quotes, "it came with chinese lettering that is chinese to me".

                      "My man mousehole needs one too"

old4570 said "I'm not an expert , so don't suffer from any such technical restrictions".

Old-Lumens. Highly admired and cherished member of Budget Light Forum. 11.5.2011 - 20.12.16. RIP.

 

djozz
djozz's picture
Offline
Last seen: 7 hours 16 min ago
Joined: 09/07/2012 - 17:04
Posts: 17312
Location: Amsterdam

MRsDNF wrote:

djozz wrote:
1.5 hours underway at 10A, I did a very accurate temperature measurement (using my lips to sense the temperature Laughing), the housing of the two bigger switches are luke-warm, the small switch is a bit warmer, all the leads, including the copper wires are pretty hot, but not 60degC+ hot.

djozz wrote:
All three switches make it to the 16A test! It has to wait a bit because my girldriend hates the sound of the fans, makes her think of bad airco days Undecided.

Are you not sure that it is not the first comment that your girlfriend wants the testing stopped though you did clarify by saying they weren't very hot.

Sometimes I wonder about members here. Orsm testing, still, by the way. Thanks again. Undecided

She says that she still loves me, I think mostly despite of the hobby even though I did give her a flashlight two years ago (and she refused any more flashlights ever since, glad it was the good old Fenix E01 and not some crappy Ultrafire Laughing)

djozz
djozz's picture
Offline
Last seen: 7 hours 16 min ago
Joined: 09/07/2012 - 17:04
Posts: 17312
Location: Amsterdam

16A started:

.....and after 10 minutes: 1 switch down. No, wrong, it was the larger Omten Undecided, electrical connection lost and it does not click anymore. oh, and something funny happened to the housing Laughing:

Two to go:

It was mounted in the middle, perhaps the two outer switches have some profit from heatsinking through the copper wires? Or the mounting of the small switch provides some extra heat path? Or just bad luck, the next switch would survive better.

djozz
djozz's picture
Offline
Last seen: 7 hours 16 min ago
Joined: 09/07/2012 - 17:04
Posts: 17312
Location: Amsterdam

There's not much wrong with the switch, just some plastic deformation caused it to fail.

MRsDNF
MRsDNF's picture
Offline
Last seen: 11 hours 30 min ago
Joined: 12/22/2011 - 21:18
Posts: 13452
Location: A light beam away from the missus in the land of Aus.

The heat would of been caused by a bad connection in the switch. is there any arcing on the contacts?

I would say your testing is working pretty good. Maybe testing on a batch of ten is next in order.

 

djozz quotes, "it came with chinese lettering that is chinese to me".

                      "My man mousehole needs one too"

old4570 said "I'm not an expert , so don't suffer from any such technical restrictions".

Old-Lumens. Highly admired and cherished member of Budget Light Forum. 11.5.2011 - 20.12.16. RIP.

 

djozz
djozz's picture
Offline
Last seen: 7 hours 16 min ago
Joined: 09/07/2012 - 17:04
Posts: 17312
Location: Amsterdam

MRsDNF wrote:
Are you not sure that it is not the first comment that your girlfriend wants the testing stopped though you did clarify by saying they weren't very hot.

Sometimes I wonder about members here. Orsm testing, still, by the way. Thanks again. Undecided

Ah, (language barrier), now I see your comment Laughing. ..Eehm, she wasn't there when I did that. (Shhhh, don't tell her, she still thinks that this hobby is just some innocent man thing)

djozz
djozz's picture
Offline
Last seen: 7 hours 16 min ago
Joined: 09/07/2012 - 17:04
Posts: 17312
Location: Amsterdam

MRsDNF wrote:

The heat would of been caused by a bad connection in the switch. is there any arcing on the contacts?

I would say your testing is working pretty good. Maybe testing on a batch of ten is next in order.

no arcing that I can see, it must have been pure the heat caused by the resistance over the whole metal part of the switch, i measured somewhere above that the metal used is not the best conductive kind (it looks like the disc has a nice silver coating though):

djozz
djozz's picture
Offline
Last seen: 7 hours 16 min ago
Joined: 09/07/2012 - 17:04
Posts: 17312
Location: Amsterdam

I stopped the 16A test after 2 hours because nothing much was happenening, I got bored, and I want to go to bed. The small Omten switch ran hot, but still works fine and so does the 6A ebay switch although the resistance seems to have gone up a little.

Here's the last voltage numbers of the two remaining switches:

Small Omten   16A begin: 73 101 125 92 95 91 145 106 93 100 mV

                    16A 1.5hrs: 70 113 110 80 132 81 115 77 92 104 mV

6A ebay sw.    16A begin: 26 29 37 27 29 29 33 37 31 40 mV

                    16A 1.5hrs: 34 51 54 47 45 37 40 42 35 38 mV

I did not expect the switches to work so well. Perhaps the Mcclicky and the Judco(Tofty) switch have a lower resistance (can someone measure that for me?, I'm curious about that), but it is not that these switches fail easily. My ideas on these switches: they eventually fail from heating up, and they get rid of the heat via the leads. If the heat can travel a short distance from the leads to a flashlight body (thick solder blobs all the way to where the leads go into the switch and the leads in direct contact with the aluminium, and such) the switch may well perform even better than in this test where the heat must directly dissipate into the air.

Hope you liked these tests, good night Smile

MRsDNF
MRsDNF's picture
Offline
Last seen: 11 hours 30 min ago
Joined: 12/22/2011 - 21:18
Posts: 13452
Location: A light beam away from the missus in the land of Aus.

There is but I could not measure the resistance in the Tofty switch with any of my DMM.

 

djozz quotes, "it came with chinese lettering that is chinese to me".

                      "My man mousehole needs one too"

old4570 said "I'm not an expert , so don't suffer from any such technical restrictions".

Old-Lumens. Highly admired and cherished member of Budget Light Forum. 11.5.2011 - 20.12.16. RIP.

 

djozz
djozz's picture
Offline
Last seen: 7 hours 16 min ago
Joined: 09/07/2012 - 17:04
Posts: 17312
Location: Amsterdam

MRsDNF wrote:

There is but I could not measure the resistance in the Tofty switch with any of my DMM.

A DMM, when measuring resistance, uses a small constant current and then measures the voltage (that is what I think it does). When the resistance is very low the voltage is below what the DMM can measure. The way to go with very low resistances is use a much larger current than your DMM does (like more than 1 amp, or 16 Evil ) and then the voltage is measurable, and the resistance can be calculated.

Hi-Beam
Hi-Beam's picture
Offline
Last seen: 5 years 1 month ago
Joined: 06/01/2014 - 22:37
Posts: 248
Location: TX

Thanks for doing these test.

I’ve had 3 switches fail recently. One at 3.04 amps, one a 4.8 amps and one at 5.1 amps. These switches were not Omten but generic switches in different lights. They all failed by dimming to barely visable light or going completely off.

It appears that the Omten are much better switches, I have 6 Omten 250v 1.5 amps switches on the way with full confidence in their ability thanks to your test.

Ledsmoke
Ledsmoke's picture
Offline
Last seen: 4 months 3 weeks ago
Joined: 08/08/2011 - 16:05
Posts: 1995
Location: Denmark

Thank you for soing these tests. Sure must have been boring though. So much more thank you Smile

~ Ledsmoke ~

Dutch humor:

[quote=djozz]

 I do not think that the BLF-community ben

led4power
led4power's picture
Offline
Last seen: 9 hours 55 min ago
Joined: 12/29/2012 - 09:48
Posts: 1284
Location: Croatia,EU

I'd like to see tests with the same currents but pwm-ed?Because constant current does nothing to switch,except generating heat,but maybe fast pulsating (for ex. 1khz pwm) current cause some additional "electrical wear" damage?I'm just guessing.

Pages