I suggest the first test would be to just find out if this is even a
problem. (Later on, the test could be changed to determine what is an
acceptable soldering dwell time for reliable operation.) For now, I would
just try abusing a switch with something like 5-8 seconds soldering time on
all four terminals. The switch can be inspected before and after soldering
to verify the contacts have been degraded. Then do your cycle testing to
see if the switch still works reliably. Then inspect again at the end of
the test.
One thing you might want to consider for your test is to measure the
resistance of the switch contacts with an ohm meter. This would be a more
sensitive test than a light bulb circuit. It is difficult to measure low
resistances accurately but if the contacts are degrading due to cycling or
soldering, then the resistance will start to climb. If it becomes
measurable then it is probably a problem.
Here is another thought. These switches have two small floating
spring-loaded mechanisms inside them to press the contacts together. If
one of those mechanisms gets bound-up or jammed then the switch will likely
fail to conduct. Therefore, it might be worth thinking about what would
cause that. In a rocketry environment, that might be fiberglass or carbon
fiber dust that finds its way into the switch. These switches are not
sealed. It might be interesting to dunk a switch into a bath of fiberglass
dust and then see how well it works.
It would be really valuable to autopsy an actual failure. Anyone have one?
Or is all the data we have so far just anecdotal hearsay?
Vern Knowles
From: R Dierking [mailto:applerocketry@xxxxxxxxxxx] ;
Sent: Thursday, January 19, 2017 11:29 AM
To: Vern Knowles; roc-chat@xxxxxxxxxxxxx
Subject: RE: [roc-chat] Re: Switch Warning - Schurter 110-220
Thank you Vern. I bought 3 switches and was wondering what I was going to
do with the 2nd switch. Switch #1 is at 150 cycles with no signs of trouble
btw. Im keeping Switch #3 as the control and will be disassembling that
one too. Thanks for the tip on disassembly!
I soldered the current test switch (#1) using my simple 25 watt soldering
iron, but I have a more sophisticated Weller soldering station with temp
control and readout. After I get to the end of checking switch #1, I can
start cycling a switch where the terminals have been subjected to
high/prolonged soldering temperature. What do you think this should be?
There are specifications for solderability and resistance to soldering heat
in the manufacturers documentation. However, the sheet is for the PCB
mounted switch. Not being a sparky guy I honestly didnt pay much
attention to this. Given in the spec sheet:
Resistance to soldering heat: 260 C/5 sec acc. to IEC 60068-2-20, Test Tb,
method 1A
Richard Dierking
From: Vern Knowles <mailto:knowles.vern@xxxxxxxxx>
Sent: Thursday, January 19, 2017 9:56 AM
To: roc-chat@xxxxxxxxxxxxx
Subject: [roc-chat] Re: Switch Warning - Schurter 110-220
I have used many Schurter switches over the years and not had any issues.
I am wondering if maybe the problems some people have had with them is
related to soldering. I just did a quick test and discovered that the
surface finish on the internal switch contacts does change visibly if a
terminal is heated to soldering temperature for more than a few seconds.
With the soldering iron I used, a couple seconds is okay but at about 4-5
seconds there was enough heat conducted into the switch to change the
contact surface finish. I have no proof the change in surface finish is
detrimental to the operation of the switch but frankly, under the
microscope, it does look somewhat degraded. The finish is not as nice
and smooth. It would not surprise me if the number of reliable switching
cycles is significantly reduced by this but I have not tested it. (By the
way, the switch is very easy to take apart for inspection. Just squeeze
together the two plastic posts sticking out the bottom. It is also very
easy to re-assemble. Just push it back together.)
It would be easy for someone to do some testing to see if this is a problem.
(hint, hint!).
Vern Knowles
www.multitronix.com
From: roc-chat-bounce@xxxxxxxxxxxxx [mailto:roc-chat-bounce@xxxxxxxxxxxxx]
On Behalf Of R Dierking
Sent: Wednesday, January 18, 2017 7:35 PM
To: kurtgug@xxxxxxxxx; roc-chat@xxxxxxxxxxxxx
Subject: [roc-chat] Re: Switch Warning - Schurter 110-220
I think the 3D printed switch arrangement that Mike brought to the meeting
is better. So are the screw-type Aerocon switches, and the one that holds
the PCB switch that someone posted. Im doing this mostly for people that
have the 110-220 switches already in their rockets. Also, of course, Im
looking for a reassurance for the switches Im using in my L3 project.
Ill definitely take the test switch apart after the testing concludes. Im
not sure what number of cycles to stop, but Im thinking about dropping the
test box on the garage floor every 100 cycles after 500. Really give the
switch a shock to see what happens.
Ill be taking one of the new switches apart too. I want to see if I can
pot one of the Aerocon PCB switches inside the 110-220 switch case. So,
retrofit with no re-wiring.
From: kurtgug@xxxxxxxxx
Sent: Wednesday, January 18, 2017 5:27 PM
To: roc-chat@xxxxxxxxxxxxx
Subject: [roc-chat] Re: Switch Warning - Schurter 110-220
Hey Richard. Great work you are doing there. You should also dissect the
switch at the end of the testing to see if you can see any wear and tear on
the housing or the contacts.
I'm also interested in making a switch like what Kramer came up with, a
micro switch and screw. That could also be a viable solution.
Kurt
Sent from my iPad
On Jan 18, 2017, at 2:37 PM, R Dierking <applerocketry@xxxxxxxxxxx> wrote:
After the group of posts on this switch and other types used to arm
electronics, I wanted to run a simple test. Maybe to provide some
reassurance to people that already have this switch installed for their
project. On the other hand, theres no way I can guarantee a favorable
result.
There were a couple of concerns voiced during the last thread for the
110-220 selector switch. One was the lifetime specified in the
manufacturers documentation: 300 operating cycles (without load). The
other was reports of failures including high resistance, unreliable
connection, and switches actually cracking during use.
I believe that at least some failures could be attributed to installing what
is intended to be a flat panel mounted switch on curved airframe tube. Im
thinking that some switches are being over-tightened creating stress on the
rotatory selector.
Please see the attached photos.
Installation:
The hole for these switches is not ½, but it is close. The manufacturers
dimensions for installation are metric. If you just drill a ½ hole, the
hole will probably be too large to use the small protrusion that should be
used to prevent the switch from rotating. Please see the arrow on the first
photo. The switch doesnt have to be tightened much at all on the mounting
panel. Unfortunately, Ive never seen any documentation on installation
when you buy them. Anyway, you should drill a smaller hole, and gradually
increase the size until the treads just clear. Also, the notch needs to be
filed for the small projection to prevent rotation.
Durability:
To test the lifetime, I installed a new switch that powers a small light
bulb (9 volt battery for power) and also has posts to check the voltage.
Im keeping track of how many times I cycle this switch on-off. Im
occasionally operating the switch 10 times when I go out into the garage. I
observe the light for intermittent connection, occasionally tap the test
box, and jog the selector. So far, Im at 80 cycles. Again, looking for
any problems. Im keeping the box on my clothes dryer because it vibrates
when operating, and even the temperature varies. This morning it was 45 F
in my garage, and after operating the dryer for a couple loads of laundry,
the test box was 80 F.
So, well see how many cycles I get from this switch. Please keep in mind
that this simple test is not designed to check mean time between failure
(MTBF) because those tests are a lot more involved.
Questions and comments are appreciated.
Richard Dierking
<IMG_0145.JPG>
<IMG_0146.JPG>
<IMG_0148.JPG>
