2
\$\begingroup\$

I bought this power supply off Amazon, it is described as 5 Volts DC 60 ampere. I was able to adjust the output so it produces 5 volts at its terminals.

Here is a link to where I purchaed it from:

https://www.amazon.com/gp/product/B017YEOAPA/ref=oh_aui_detailpage_o03_s01?ie=UTF8&psc=1

Internally it consists of a power supply built around a TL494. The design seems typical, but I am unsure how it works. There appear to be two switching components labelled SPYM309. I can't figure out what these are. I expected them to be driven with some sort of signal 180 degrees out of phase with each other, but even using both probes I could not find such a thing.

enter image description here enter image description here enter image description here

The other unusual thing is the electrolytic capacitors on the output are rated for 25 volts. The larger components mounted to the case on the left hand side are just two pairs of rectifiers wired as a full bridge.

I know the TL494 internally is two error amplifiers and some other components. I probed around inside the power supply and eventually gave up on trying to figure out what was doing what. I could not even find the +5V from the internal reference regulator on the TL494. I added a single turn of 22 AWG magnet wire to the larger transformer and found this wild waveform. This is using a 2 ohm load. Note that I accidentally had the scope displaying 10x here, but the probe is actually set to 1x

enter image description here

The output waveform seems to just "wander" for about 65% of the time, but the other 50% is spent doing something square wave-ish.

The next test was the same thing, but with a 0.4 ohm load. The scope's setting was 1x, as well as the probes in this capture

enter image description here

The output waveform cleans up quite a bit with the larger load. Interestingly, it still spends about 1/2 of its time around 0 volts.

I have a couple questions about this power supply

  1. What does the smaller transformer do?
  2. What are the two switching components?
  3. Should I expect to find the two switching components in a push-pull configuration?

Any other suggestions on more interesting things to measure with my scope would be appreciated!

\$\endgroup\$
4
  • 1
    \$\begingroup\$ Did you check the TL494 datasheet - you didn't provide a link. My experience is that the circuit designers often copy directly from the datasheet rather than risk a new design. \$\endgroup\$
    – Transistor
    Commented Jun 19, 2016 at 22:39
  • \$\begingroup\$ We cannot help you without knowing the model number of the power supply. Some parts are obvious, others we would have to guess at. \$\endgroup\$
    – user105652
    Commented Jun 19, 2016 at 22:40
  • \$\begingroup\$ @Sparky256 I added a link to the seller's page. \$\endgroup\$
    – Eric Urban
    Commented Jun 19, 2016 at 22:43
  • \$\begingroup\$ Old thread... but I got one today, and wondered much the same thing.... google points to here when asked about the spym309.... it is a manufacturing mark, I think as the board is stamped spym400 as a part number..... 2sc2625 is the replacement for the main fets. It is a generic copy of the meanwell series. ( they use the 2sc2625 ) It is a half bridge, direct voltage feedback ( no opto)change R40 to suit) Uses assymnetric drive pulse on the drive transformer to get started ( no bootstrap) ( feedback loop through driver tranny gets wild pulses running until it gets to trigger the tl494 into life \$\endgroup\$
    – oztules
    Commented Mar 25, 2017 at 0:41

2 Answers 2

6
\$\begingroup\$

Looks like the manufacture is not giving up a schematic, so I will provide the best answer I can based on using the TL494 myself and working with many SMPS units.

1) The smaller transformer wrapped in yellow tape is a driver for the mosfets mounted on the chassis next to it. The TL494 needs a boost as it is not always the best choice to drive mosfets directly. The TL494 actually drives the 2 smaller transistors first, which boost the current to drive the smaller transformer in a push-pull mode.

2) The mosfets are your main switching components, driven in a push-pull mode (possibly). They drive the main transformer (wrapped in green tape) primary winding. Mounted to the chassis next to the green transformer are the 2 rectifier diodes that convert the secondary pulses back to a low DC voltage. The toroid and low voltage capacitors smooth out any ripple so the final 5 volt output is a clean DC current. I do not see it but somewhere on the board is a feedback path usually done with an opto-isolator that keeps the output voltage 'locked' at 5vdc.

3) The AC input fuse is to the right side, along with a inrush current limiter (black disc). These feed special filter capacitors designed for use on the AC line, then a common-mode filter transformer to prevent EMI noise from getting into the AC line, then into a bridge rectifier (black rectangle with one chiseled corner), which converts the AC input to a high DC voltage.

4) If the AC input voltage is 120 or 240, the DC voltage on the 2 large capacitors is about 300-340vdc, which in turn feeds current to the mosfets. Because of the matched parts connected to the mosfets, I do not think this is a H-bridge design, normally used for power supplies of 500 watts or more.

5) If it is a half H-bridge design, the 'appearance' would change very little. Only a schematic would prove which type it is. The TL494 is very flexible when extra parts are added, so it can work with buck / boost, flyback, push-pull or H-bridge designs.

6) This power supply comes with a 120/240vac switch. It does not affect the mosfet design directly but affects how the bridge rectifier and 2 large capacitors work. If this switch is closed then the neutral/L2 line is connected to the center of the 2 large capacitors, which are in series-not parallel. This and the bridge rectifier form a voltage doubler so you have 300 to 350vdc across the capacitors.

7) If you have a 208/240vac power source (neutral becomes L2), the switch is open and the bridge rectifier creates the 300 to 350vdc across the capacitors. The mosfets are being driven by 300 to 350vdc, regardless of the line source voltage. The product description (see link) warns the buyer to make sure this switch is set properly.

\$\endgroup\$
5
  • \$\begingroup\$ If those components are N-Channel power MOSFETs, I should be able to find a Gate drive signal with a peak of approximately 15 VDC above the low side of the voltage doubler, correct? \$\endgroup\$
    – Eric Urban
    Commented Jun 20, 2016 at 21:44
  • \$\begingroup\$ Yes, but be careful. The mosfet voltage is a floating supply. The mosfet source pins are connected to the low side of the 2 capacitors, which is -170 volts compared to earth ground. The high-side of the 2 capacitors is +170 volts compared to earth ground. One could say the center point of the 2 large capacitors is close to earth ground voltage-but not exactly. Across the 2 cap's you have 350vdc, very dangerous to touch, as it can burn your skin. Also it is not likely those are P-channel mosfets. Those are normally used as low voltage power switch's in cell-phones, etc. \$\endgroup\$
    – user105652
    Commented Jun 20, 2016 at 22:42
  • \$\begingroup\$ Do you have a spec sheet for those somewhere? \$\endgroup\$
    – Eric Urban
    Commented Jun 20, 2016 at 23:49
  • \$\begingroup\$ @EricUrban. Without a schematic from the manufacture the part numbers for the mosfets are unknown, as are many other parts. Any number of 700volt mosfets could be used, and I do not want to guess. Often the schematic is considered copyrighted material and the manufacture will not release it to the public. Get the parts numbers off of them if possible and then you can use Google-fu to get the spec sheet. \$\endgroup\$
    – user105652
    Commented Jun 21, 2016 at 2:10
  • \$\begingroup\$ How could the part number be unknown? I literally took the heatsink retainer off and photographed them. \$\endgroup\$
    – Eric Urban
    Commented Jun 21, 2016 at 2:14
2
\$\begingroup\$

Half bridge. Yellow transformer drives the two bipolars through C10. The toroid is the energy storage component since the green transformer is not gapped. As the load increases the pulse with increases, and vice versa.

\$\endgroup\$

Not the answer you're looking for? Browse other questions tagged or ask your own question.