Power Supply POP! Why?

Status
Not open for further replies.
OK, so I finally found some time to do a destructive test...results, they DO NOT have short circuit protection. How difficult would it be to make my own short circuit protection circuit. In another thread (link below) some guys on here are already helping to come up with a way to make both supplies in series shut off if one goes out, I wonder if it could be incorporated into the same circuit design with short circuit protection and then my investment won't be lost. I simply cannot sell these, even used, to guys without having short circuit protection built in, it's just too risky that someone will touch the leads accidentally.

https://www.electro-tech-online.com/threads/series-power-supply-safety-circuit.140727/
 
Why do you say they don't have protection?
What was your test?
 
I touched the positive and negative leads together and saw fireworks. It actually still works this time because they only touched for a brief moment.
 
I touched the positive and negative leads together and saw fireworks.
Click to expand...
That may just be a cap discharging. It doesn't necessarily mean there's no protection. Have you tried cranking up a load on the supply and monitoring output voltage and current?
 
alec_t said:
That may just be a cap discharging. It doesn't necessarily mean there's no protection. Have you tried cranking up a load on the supply and monitoring output voltage and current?
Click to expand...
Haven't had time to try much of anything yet. So how would I monitor output voltage (with a voltmeter, scope?) and what would I be looking for, just voltage drop? Also, why? Sorry for all the questions, I guess you really need to spell it out for me...
 
Protection circuits I'm aware of would sense current and operate to reduce the voltage or disconnect the load if the current were excessive. An ammeter could monitor current. A voltmeter or scope could monitor voltage, which would show an abrupt drop for the excess current state. You need to know the maximum current you can safely draw from the supply before conducting this test.
 
I see. I think I need to finish my dummy load then I will have something with which to conduct these types of test.
 
jocanon said:
I touched the positive and negative leads together and saw fireworks. It actually still works this time because they only touched for a brief moment.
Click to expand...
This thread (**broken link removed**) has some very good info and advice (especially about floating the ground on one of the two PSs if you intend to cascade them for 24VDC) and also this about your overload concerns (failsafe):



Ignore the zener reference. Not a workable idea for voltage control in this case. These PSs have perfectly adequate (and somewhat adjustable) voltage control built into them.
Don't know at what current load the failsafe trips, but one or both will go, presumably prior to any damage to the PS, thus killing all power to your load.

<EDIT> Just found this on an HP site to corroborate the above "failsafe" comment:
**broken link removed**
While NOT your PS model, these are protection features I would expect to be present in all the HP network ("farm") related PSs of this sort.

Once you have your dummy load, your can test for these, if you so choose.
 
Last edited:
Ok, my bad...it does have short circuit protection. My other supplies spark when I touch the ends too. This time I connected up both pos and neg before power up, then when I powered up it just did not come on. Then I powered down, disconnected the short and it powered up fine with full rated voltage, no damage. Hopefully that first unit that popped was just a fluke. The only part inside that looked a little bit smoky was GSIB1580, the left most leg seemed like it was fried.
 
passed test...i don't, maybe the pop was just bad luck and it won't happen again, or at least not more than 1/100 X or so.
 
jocanon said:
passed test...i don't, maybe the pop was just bad luck and it won't happen again, or at least not more than 1/100 X or so.
Click to expand...
It mite have been one of the high power parts shorting to a heat sink, I have seen this happen before and I could not see the damage untill I removed to from the heat sink. Are any of the power transistors insulated from their heat sinks?
 
4pyros said:
... insulated from their heat sinks?
Click to expand...

I don't think I understand what you mean. If I interpret you literally, then I would have to say no, because that doesn't make sense...I mean if it was insulated from the heat sink wouldn't that mean that there was no connection between it and the heat sink and then how would it act as a heat sink if it had no connection? So I must be missing your meaning...just thinking out loud, are you saying connected to the heat sink by some paste or something that conducts heat but not electricity?
 
OK, random question, it is on the topic of the power supply, but a little off topic in the sense that it's not about the POP, but I don't feel like starting a whole new thread...

Anyway, so for connecting these in series, is it ok to have them hardwired connected (i.e. solder the series connection wire so it cannot be removed) and then for them to sometime be run as just the 12v supply on the lower supply. The upper would then be connect pos of lower to neg of upper, but pos of upper (24v) would not be connected to anything. I think this would be fine right, since there is no complete circuit on the upper 24v supply it is not like any stray electricity would travel through it, right? If this is a problem, then I will not hard wire the series connection, but instead make a jumper that people will have to connect themselves, with instructions on where to connect the jumper. Problem here is if they don't connect it right.
 
In your scenario above, would you still be accessing the 12V via the upper PSU positive terminal, or would you have the load connected directly across the lower PSU?

If your just connecting across the lower PSU and there is no current path for the upper PSU then this should be fine.
 

At this point it is an open book...meaning I could set it up however makes sense. I see what you are saying, so maybe the best answer is to hard wire the series connection and then only give the option to use the lower on 12v leaving the upper with no current path and then only use the upper when running 24v. However, I may still go back to just using a jumper for a couple reasons, the hard wired series connected wire will be a pain to deal with in shipping and if in the off chance that the lower unit fails on them but the upper unit is still working, if I use the jumper then they could still use the upper for 12v (does it matter that DC is isolated from ground on the upper - any problems using this as 12v if isolated?).
 
Have you considered/tested what happens when, with the two supplies stacked to give 24V and a heavy load connected, the overload shutdown operates on one of the supplies (it's unlikely both would shut down exactly simultaneously)? I suspect the sudden reduction to 12V will reduce the load current below the trip point, the supply that shut down will start up again, the trip will go again ....... etc, resulting in oscillation.
Would this be acceptable, or does this pose a risk of damage to the supplies or the load? .
 
Fortunately, once one shuts down it will not power back up again until the AC power to the PSU is cycled off and then on again.
 
Also, under 99.9% of instances, the power supply will be way more than is needed for the load so they will not be running up against the limit. It is possible if they hook up multiple loads, but then they would most likely need to be plugging each PSU into a different AC circuit or they would pop the breaker on their house. Point being, this would only come up in very rare instances anyway, but since you have to power cycle it to get the tripped PSU to come back on, I don't think there is a problem even in that rare extreme possible case, agreed?
 
Status
Not open for further replies.

Similar threads

G
  • Question
Why use this Capacitor
Replies
16
Views
947
aliarifat794
A
D
How fast should a power supply current limit cut in?
Replies
16
Views
1K
Nigel Goodwin
  • Question
Can the amperage of this power supply module be modified to increase?
Replies
11
Views
944
xmanhattan
X
  • Question
⚡️Why are pwm switching ICs called off-line power supplies?⁉️
Replies
10
Views
2K
Tony Stewart
F
ZK-4KX Buck-Boost Power Supply Module Documentation
Replies
4
Views
2K
For The Popcorn
F
Menu
Log in
Register
Cookies are required to use this site. You must accept them to continue using the site. Learn more…