power supply repair

Status
Not open for further replies.

They are called 'RF burns' it's a well known occurance.
 
BTW, wasn't this topic was about repairing a PC power supply ?

PC power supplies are off-the-lime switching power supplies.

Extreme care should be taken when working in a switching power supply.

First, use an isolation transformer. If you don't, you won't be able to take any measurments with a scope. Some guys (i seen it !!!) had modified their scope to fit a live chassis by CUTTING THE SCOPE'S GROUND PRONG !!!! N-E-V-E-R do this unless you have a deep desire to have your wife inherit prematuerly from you ! If you plan to service switching power supplies, an isolation transformer IS A MUST !

Second, RF shocks will BURN you ! And having over 300 volts pk-pk 300 kHz to 1 MHz voltage on the switching transistor or MOSFET WILL DEEPLY BURN YOU !!! Even if you take care to isolate yourself electrically. This 300 volt RF voltage may even burn your oscilloscope !!! If you want to check the waveform with a scope, put your scope to high sensitivity (10mV/cm or so) and approach your probe tip to around 1/2 - 1 inch from the switching transistor WTHOUT TOUCHING IT ! The electric field generated by the switching action will induce a voltage in your probe tip and you'll be able to visualize the waveform.

To measure DC voltages on the primary, use an ordinary DMM with GOOD probes ( i mean well insulated and rated to 600 volts or more) and take great care keeping your hands away from the supply. AC or DC, 300-400 volts will ZAP you !
 
the safest method for using an oscope while testing an offline switching supply primary side is to use a 2-channel scope in differential mode (A+B, invert B, makes a differential "subtract mode"). then the scope probes are attached to the hot side "ground" and the scope gains adjusted for a straight line. once this is done, the A channel probe can be used normally with the B channel remaining on the hot "ground" the scope grounds are connected to the DUT (Device Under Test) chassis ground. what the scope is now measuring is the DIFFERENCE between A and B channels, and the hot "ground" is isolated from the scope ground.
 
Last edited:
Hi frequency RF tends to go around the outside of a conductor (human skin) depends on a whole bunch of factors, none of which are particularly easy to predict reliably all the time.
 
Hi frequency RF tends to go around the outside of a conductor (human skin)
That's not true with human tissue because it's less conductive than metals such as copper.

I'll post the link and quote again:

The Skin Effect Myth
 
I tend not to think of frequencies lower than 30mhz as 'real RF' I know kind of silly. I would like to point out that even that website Hero linked said at Tesla frequencies it still won't penetrate more than a few centimeters, mind you that is plenty far enough to do serious damage.
 
you might not tend to think that way, but generally anything above 5-10KHZ can be considered RF. anything that is radiated from an antenna is RF. actually if you wanted to do some digging, you'll find that 2khz is often used for a radio link in cave exploration, and the US Navy operates a radio transmitter at 75hz for communicating with submarines. the distinction between AF and RF is very "fuzzy", and always has been. the early wireless transmitters and receivers worked in the 5khz-300khz range, so at the time AF was (and still is) considered to be 20hz-20khz, and RF was 5khz and up. it really depends on the final destination of the signal. if it's intended to move air, it's AF. if it's intended to radiate electromagnetically, it's RF.
 
So frequencies bellow 20hz don't exist? =) J/k The main reason I don't considering anything bellow about 30mhz as RF is because the antennas are so messy for stuff that low.
 
messy antennas???? i've seen a lot of different types of antennas, but other than the occasional random longwire draped over tree branches and coiled up in spots on the ground, i don't think i've seen any that could be classified as "messy". i've even seen verticals with a loading coil tuned for 3.5Mhz (80 meter band) that other than the loading coil at the bottom could be mistaken for a flagpole.
 
Last edited:
So frequencies bellow 20hz don't exist? =) J/k The main reason I don't considering anything bellow about 30mhz as RF is because the antennas are so messy for stuff that low.

Large - Yes
Messy - No
Depends if you are receiving or transmitting.
Small loop for receiving - very unobtrusive.
Rhombic on 80 Mtrs - Need lots of acerage.
 
Status
Not open for further replies.
Cookies are required to use this site. You must accept them to continue using the site. Learn more…