Simple Power Supply Fault

Thread Starter

aalauzier

Joined Aug 19, 2021
2
Hello all, I'm assuming this will be a very simple answer, however since I don't have a strong background in electrical theory I am fairly baffled. I take care of the majority of repairs at a small live event company, and recently we had a medium sized 10,000 lumen projector die. After digging in to check fuses I saw that the 110V-230V selector switch was cooked after someone repeatedly toggled it while the unit was on. This switch is housed in a module along with the IEC Power Input and on off rocker switch. Apologies for not having a schematic, but there was surprisingly little going on. Both the hot and neutral of the IEC passed through the rocker switch where the neutral then passes through a ferrite core and out to a binding post. The hot leg then leads to 2 fuses, one rated 20A and the other 15A. These two fuses feed the two 'outputs' of the voltage selector switch, while the common then runs through the other side of the ferrite core and out to its own binding post. Once the voltage selector switch cooked, all three legs contacts were shorted no matter its position. I made my way through the circuit with a meter and had 120V until I hit the binding post, at which point it dropped down to an unstable 8V-9V. In order to get the unit operational again until replacement parts can come in, I replaced the voltage selector switch with a simple jumper on the 20A fused side, and we're back to 120V on the binding post. This brought up a few questions for me, the first being, why did a parallel path of 120V drop down to 8V, or was I just getting a noisy reading from my meter? I also thought that it was curious that the voltage selection switch was only used to select a more appropriately sized fuse, and not changing the path through the windings of a transformer. Is this a fairly common design choice in PSU's? Thanks for any insight, I am always trying to learn more about the physics behind power.
 

LowQCab

Joined Nov 6, 2012
1,596
Many Switching-Power-Supplies can readily accept wide Input-Voltage-Ranges,
but the Input-Current will also vary quite a bit,
half the Voltage will require twice the Current, and vice-versa,
so changing the Fuse size is completely appropriate.
If this is a fixed installation, the switch is completely unnecessary,
just install an appropriately sized Fuse, .... done.

The problem that You really have is the Person who is F%&king around with
expensive equipment that they know nothing about.

There are some people who can F-up a wet-dream, and do so routinely,
but mysteriously, it seems that no one ever sees who was responsible.

This won't be the last time that this sort of "accident" "just-happens" when
that sort of Person is "on-the-loose", and not being continuously watched.
They have too much time on their hands, and nobody watching them.
I don't care how much You may like them, they are a liability,
and may even get someone seriously injured or killed.

I've seen this sort of thing happen far too many times when
I was not in a position of choosing the involved Personnel.
.
.
.
 

Thread Starter

aalauzier

Joined Aug 19, 2021
2
Luckily people with that ability to break things are what give me job security right now, along with a reason to drink. We use 3 phase power daily so I'm familiar with the efficiency advantages, as well as the potential problems that can occur when some devices are not a fan of anything over 120V. Luckily for this device I assumed the actual power supply section would be fine after this particular fault since that switch was only selecting a different fuse and not re-tapping a transformer. The main part I am trying to understand is how opening a parallel path through both fuses and out to the output terminal caused such a major drop in voltage. Since there are no diodes present my only assumption was the angry pixies were more than happy to stick to a small loop instead of pushing through to the actual power supply. (Apologies for being so technical)

I don't mean to sound ignorant, I am just trying to learn as much as I can to both help diagnose problems like this more easily in the future, as well as teach others. As you mentioned, you have to keep a close eye on some people, so I am trying to take every opportunity to learn so that I have a structured argument when shooting down people's tomfoolery. For example we recently had a new power service installed in part of our building, originating at a 480V panel, running through a step down transformer, and down to a 100A panel. After it was installed I verified the voltages and resistance between ground and neutral, saw that the two were bonded, and was happy. A few days later I serviced one of our mobile power distro's and took it over to this panel to test it after having it buttoned up, at which point I discovered that someone had gone and broken the bond between ground and neutral. This lead to a multi day debate where I adamantly stated that the bond should be re-made. It's reasons like this why I am trying to learn as much as I can so I can try and pass that knowledge on in the never ending fight against misinformation, as well as the safety of people who are interacting with this equipment.
 

LowQCab

Joined Nov 6, 2012
1,596
I don't know the exact situation with the Switch in question, but ......
If I were to speculate, .......
When a Switching-Power-Supply is under Load,
and it's Input is abruptly switched on and off,
a huge "Inrush-Current" can be caused,
this is caused by the charging and discharging of the large Bulk-Input-Capacitors.
This can create a huge amount of Arcing across the Switch Contacts,
which may weld themselves together,
or create a very high-Resistance connection, or zero connection.
It's a crap-shoot as to exactly how the Switch may fail.

The Switch may have been barely adequate to start with.
.
.
.
 
Top