Yesterday I touched one of the heatsinks in my power supply and got a nice shock... it felt like AC since my arm was vibrating, but I'm not sure. Should I ground that heatsink, or would that short high voltage to GND and break something? I'll take a look later and see if any pins are touching it somewhere.
safe?
- Thread starter CTho9305
- Start date
no, static shocks you with a little lightning-bolt deal, and then its over, this continued the ~1/2 second until I realized and pulled my hand away.
<< no, static shocks you with a little lightning-bolt deal, and then its over, this continued the ~1/2 second until I realized and pulled my hand away. >>
Agreed, sounds a bit more serious than static. Definitely check for 'leaks'
Demon-Xanth
Lifer
- Feb 15, 2000
- 20,551
- 2
- 81
There's a reason why the PS is enclosed
The collectors and sources of transistors (where HSs are usually mounted to) are often NOT at ground and are actually rarely at ground. If you ground that heatsink the supply WILL NOT work and will most likely not work even if you remove the ground strap (ie: you fried something). This is not likely caused by a leak but is probably by design. Transistors with common collectors can easily handle high currents by using the heatsink itself as a conductor.
The collectors and sources of transistors (where HSs are usually mounted to) are often NOT at ground and are actually rarely at ground. If you ground that heatsink the supply WILL NOT work and will most likely not work even if you remove the ground strap (ie: you fried something). This is not likely caused by a leak but is probably by design. Transistors with common collectors can easily handle high currents by using the heatsink itself as a conductor.
Evadman
Administrator Emeritus<br>Elite Member
- Feb 18, 2001
- 30,990
- 5
- 81
I touched the top of one of those round things and got a shock
Yes, I actually did. I was so stupid. When I work on computers I leave them pluged in ( I gut the hot wire, but the neutral and ground are still conected, so I get a good ground ) but I forgot to use the correct cord. Owie did that hurt
I have a sparkle power supply that I have mounted in a CD-Rom drive, and the heatsink is soldered to the case. On that one anyway, the heatsinks are NOT carrying power.
Even if the heatsinks do have voltage, it should be reduced to somewhere around 20-24v as that is what the "popular" transisters take to get 12 or less volts out. ( some go higher, some lower. That is just the range on a few I pulled out of my parts drawer ) The transformers should be reducing the 120v into somewhere in the aformentioned range. And even if it did shock you, it would be with DC current, not AC.
Yes, I actually did. I was so stupid. When I work on computers I leave them pluged in ( I gut the hot wire, but the neutral and ground are still conected, so I get a good ground ) but I forgot to use the correct cord. Owie did that hurt
I have a sparkle power supply that I have mounted in a CD-Rom drive, and the heatsink is soldered to the case. On that one anyway, the heatsinks are NOT carrying power.
Even if the heatsinks do have voltage, it should be reduced to somewhere around 20-24v as that is what the "popular" transisters take to get 12 or less volts out. ( some go higher, some lower. That is just the range on a few I pulled out of my parts drawer ) The transformers should be reducing the 120v into somewhere in the aformentioned range. And even if it did shock you, it would be with DC current, not AC.
<< Yesterday I touched one of the heatsinks in my power supply and got a nice shock... it felt like AC since my arm was vibrating, but I'm not sure. Should I ground that heatsink, or would that short high voltage to GND and break something? I'll take a look later and see if any pins are touching it somewhere. >>
Come on CTho9305 - You should know better than opening up a power supply if you dont know enough about them... I dont bother replying to other clueless dimwits opening up power supplies anymore since every time I have attempted to suggest their life is worth more than a $25 power supply they have ignored it so I figure it might be better if they are cleansed out of the gene pool but since you have a good degree of intelligence and knowledge from what I've seen from your posts you really ought not to be opening it without a full understanding of SMPS systems and basic electronics....
Cheers and take care!!
<< Even if the heatsinks do have voltage, it should be reduced to somewhere around 20-24v as that is what the "popular" transisters take to get 12 or less volts out. ( some go higher, some lower. That is just the range on a few I pulled out of my parts drawer ) The transformers should be reducing the 120v into somewhere in the aformentioned range. And even if it did shock you, it would be with DC current, not AC. >>
That does not make sense mate
Most BJ transistors have the collector connected to the metal tab so if it does not use an isolated heatsink and you ground it then you are grounding the 160-odd volts rectified from the mains rectifier... FUN!!
Some SMPS circuits use the collector as the chopper output so it will have a pulsed voltage, ie AC. What you state for a transistor having 20-24 volts is most probably for a linear power supply using a transistor (or regulator) in power dissipation mode - the tab has the rectified secondary of the transformer, which in the case of what you may have measured is around 20-24 volts depending on load and it will regulate the output to a fixed voltage... this is VERY different to switch mode power supplies, low voltage linear supplies cannot kill you if you dont touch the primary of the power supply, in fact SMPS's cant do that either but linear supplies simply have a power cord and a transformer (with integral thermal fuse) and some have an external fuse as well for the primary circuit.
Cheers
Evadman
Administrator Emeritus<br>Elite Member
- Feb 18, 2001
- 30,990
- 5
- 81
Not the first time I have been wrong, and I am sure will not be the last
But serously, the heatsinks are attached to the case in the PWS that I hacked up. Is there a different way that the transisters could be packaged? It is a sparkle PWS if that means anything.
Hold on, let me get the Markings off the chips.
<edit>
Looking at the underside of the PCB, the heatinks are connected to the board with solder, and at least one side on both of the large heatsinks is soldered to the same section that all the black wires that go outside the case go to. ( one side of the heatsinks aer screwed on, the other is soldered. )
There are 2 other heatsinks that are very minimal in size, both of those I folded over so they would fit in the case, and did not attach either of them to that case. Both of those are also soldered to the PCB, and both have leads from caps going into the same trace ( both caps are orented so the marked side is the lead that goes into that paticular trace ) I assume those heatsinks are the "hot" ones you spoke of. Glad I didn't gound those, I never would have figured out why I was blowing fuses
The gounded heatsinks have a total of 5 transisters, 4 of which are the same size. Of all those 5 I can only read the information from one of them. ( it is one of the ones that are the same size as 3 others ) I know size has nothing to do with them being same or different, just wanted to give more info
The info on the one I can read, looks like this:
SRU 928 ( the "R" is backwards, and is part of the "U" I am assuming that is the company name/logo )
SF 10025 ( the 5 may be a 6, it is very hard to read )
~ + ~ ( those are right above the 3 leads. I assume is is a NPN, but am not sure 100% sure.
Do i just have a wierd setup on mine, and was lucky? I actually assumed that all heatsinks were gounded, or uncharged. I will not make that assumption again.
But serously, the heatsinks are attached to the case in the PWS that I hacked up. Is there a different way that the transisters could be packaged? It is a sparkle PWS if that means anything.
Hold on, let me get the Markings off the chips.
<edit>
Looking at the underside of the PCB, the heatinks are connected to the board with solder, and at least one side on both of the large heatsinks is soldered to the same section that all the black wires that go outside the case go to. ( one side of the heatsinks aer screwed on, the other is soldered. )
There are 2 other heatsinks that are very minimal in size, both of those I folded over so they would fit in the case, and did not attach either of them to that case. Both of those are also soldered to the PCB, and both have leads from caps going into the same trace ( both caps are orented so the marked side is the lead that goes into that paticular trace ) I assume those heatsinks are the "hot" ones you spoke of. Glad I didn't gound those, I never would have figured out why I was blowing fuses
The gounded heatsinks have a total of 5 transisters, 4 of which are the same size. Of all those 5 I can only read the information from one of them. ( it is one of the ones that are the same size as 3 others ) I know size has nothing to do with them being same or different, just wanted to give more info
The info on the one I can read, looks like this:
SRU 928 ( the "R" is backwards, and is part of the "U" I am assuming that is the company name/logo )
SF 10025 ( the 5 may be a 6, it is very hard to read )
~ + ~ ( those are right above the 3 leads. I assume is is a NPN, but am not sure 100% sure.
Do i just have a wierd setup on mine, and was lucky? I actually assumed that all heatsinks were gounded, or uncharged. I will not make that assumption again.
<<
<< Yesterday I touched one of the heatsinks in my power supply and got a nice shock... it felt like AC since my arm was vibrating, but I'm not sure. Should I ground that heatsink, or would that short high voltage to GND and break something? I'll take a look later and see if any pins are touching it somewhere. >>
Come on CTho9305 - You should know better than opening up a power supply if you dont know enough about them... I dont bother replying to other clueless dimwits opening up power supplies anymore since every time I have attempted to suggest their life is worth more than a $25 power supply they have ignored it so I figure it might be better if they are cleansed out of the gene pool but since you have a good degree of intelligence and knowledge from what I've seen from your posts you really ought not to be opening it without a full understanding of SMPS systems and basic electronics....
Cheers and take care!! >>
jesus man that is one run on sentence!!!!!!
<< Yesterday I touched one of the heatsinks in my power supply and got a nice shock... it felt like AC since my arm was vibrating, but I'm not sure. Should I ground that heatsink, or would that short high voltage to GND and break something? I'll take a look later and see if any pins are touching it somewhere. >>
You will permanently damage the PSU by shorting out the heatsink to ground. With the current price of PSU's, I'm almost positive it will be a total loss.
Those heatsinks are often connected to hot(hot in electrical context means floating potential voltage from ground)collector and 160-250V DC. You'll blow out the fuse at minimum. Most likely taking alot of components with it.
Don't be tweaking with the inside of the PSU unless you know what the h@ll you're doing. You'll destroy the PSU if not yourself as well.
<< Those heatsinks are often connected to hot(hot in electrical context means floating potential voltage from ground)collector and 160-250V DC. You'll blow out the fuse at minimum. Most likely taking alot of components with it. >>
That strikes me as a stupid design. Any reason to do that?
For God Sakes...
Don't mess with power supplies if you don't know whats goin' on inside!
One of those Caps will send ya flyin' if you're not careful!!
Don't mess with power supplies if you don't know whats goin' on inside!
One of those Caps will send ya flyin' if you're not careful!!
<<
<< Those heatsinks are often connected to hot(hot in electrical context means floating potential voltage from ground)collector and 160-250V DC. You'll blow out the fuse at minimum. Most likely taking alot of components with it. >>
That strikes me as a stupid design. Any reason to do that? >>
Actually, most that I have taken apart do not do this - in fact I have never seen a live heatsink in a power supply. I have seen them in televisions, however. Most PC power supplies use isolated tab transistors and leave the heatsink floating, ie not grounded and not live. The bonus with a floating heatsink is if any ONE component somehows leans over and touches it then it will not destroy the power supply nor harm anything. The disadvantage with floating heatsinks is they produce a lot of electromagnetic interference (EMI) - This should only be a concern to the rest of the power supply, and if it has been engineered reasonably well it will not affect it. The EMI is not a problem outside the power supply since it's enclosed in a metal case and the openings are usually smaller than the wavelength of the radio waves trying to get out.
Cheers
Evadman: Is your power supply faulty?? If so take the fan out, replace the cover and screw it back on. Now throw it away. There are much safer devices to be learning electronics with than a computer power supply.
TRENDING THREADS
-
Discussion Zen 5 Speculation (EPYC Turin and Strix Point/Granite Ridge - Ryzen 9000)- Started by DisEnchantment
- Replies: 25K
-
Discussion Intel Meteor, Arrow, Lunar & Panther Lakes + WCL Discussion Threads
- Started by Tigerick
- Replies: 23K
-
Discussion Intel current and future Lakes & Rapids thread- Started by TheF34RChannel
- Replies: 23K
-
-
Facebook
Twitter