USB 2.0  4-PORT HUB -  FE 1.1 S  IC

I bought this 4 port hub to expand the USB ports of my Raspberry Pi model B. When I tried to connect two flash drives at the same time it stopped working. It is a cheap one so I was not expecting great performance from it but I hoped that it would work. All things considered it was worthwile to try a couple of things to make it work as it should be. I looked for the datasheet here. 

I compared the info on the datsheet and how it had been implemented in the circuit that I had in front of me. I found a discrepancy/difference on a capacitor (pin20)  with a much lower value, I replaced the 10uF16V  (very cheap one) with a 100uF 10V low ESR electrolytic. While looking for the datasheet, I found a recommendations page which suggests a ceramic type. (hard to find)

I tried now and works better but fails when I connect the fourth device. This seemed to me that there was a power problem. So I tried a 47uF/100uF 16V electrolytic cap as close to each Vdd upstream connection next to each connector, with a ceramic multilayer 0,1 uF 50V NPO cap for decoupling and noise filter. Tried again and now it became far more stable. Still, every now and then and in particular when the fourth device is connected, it resets itself but it is a huge improvement compaired to how it worked at the beggining. Cheap stuff, hard to make it work, poor results.

Electrolytics (I used what I had) + ceramics....

News !! I kept working in improving the stability and eventually I find out that cutting the length of the cable to about 15cm, made the hub STABLE and works correctly without re-setting the ports !!!

Job done !!

ATX Power Supply - LPG9-25 Generic Labeled 450W (Power consumption - delivers 250w?) - Repair

WARNING: BE AWARE AND WARNED THAT THIS PS WORKS WITH LETHAL VOLTAGES !!!! IF YOU DO NOT KNOW HOW TO HANDLE VOLTAGES IN EXCESS OF 300V DC/220V AC DO NOT ATTEMPT TO EVEN OPEN THE PS BOX !!! IF YOU DO KNOW, BE VERY CAREFUL AND TAKE ALL THE PRECAUTIONARY MEASSURES. THE BOARD IS CONNECTED TO THE MAINS POWER GRID (HOT SIDE)

This ATX power supply was installed in a "white box" PC. It is labeled LPG9-25. As the main element to characterize this power supplies is its PWM IC i have to say that is an AT2005, which it turns out to be actually a WT7520. Click for datasheet. I highly recommend this page with a worth of infromation, ATX PS schematics in particular. I started looking to this PS after the computer was not booting at all. Doing a preliminary visual inspection, I found a blown capacitor in a video card. I remove the video card but the Pc still didn't want to boot. I focused my attention to the PS. I open it and immediately noted that the fuse was blown which made evident based on my experience that a massive short circuit had ocurred further inside the PS.

This is the video card with the blown cap. I replace it with a 1000uF 16V HT Rubycon. After that it was up and running.

This is the PS after doing a bit of inspection and components removal. Yeah... I forgot to take a picture in the original destroyed condition. I supposed that you have a similar electronics knowledge than me or better. I consider myself a hobbyist. Don't look for deep electronic physics knowledge. I am practical and straight to the point. So if you want to follow a litlle bit more in deep what I did go to the website mentioned above and get the schematic otherwise just have a look to how and why I did it and hope this information would help you. Back to business. Fuse blown. One 2,2 ohm resistor that goes to the base of one of the power tr toasted meaning high current ... almost certainly the power tr's (2SC2334) were blown. This was confirmed with a 000 reading of the B-E junction with the multimeter. All the cables maze that was inside the PS was stiff showing signs of high temperature. I assumed thereafter that all the components had experienced a high temperature environment. Next step was to check the fan who was also difficult to make it turn which should be easily turned with a light tip of a finger "nudge". Going back to the high current issue, I also checked the bridge rectifier diodes (RL205). Two of them were open. I made the decision to replace all the components with ones that could withstand higher currents. At this point I replaced: Fuse, ALL the rectifier diodes with a RS605 6A bridge rectifier (the PCB had empty holes to fit it), the power tr's (2sc2334) with E13009. The two base 2,2 ohm resistors, the two 1uF 50V cap's in the power tr's section for two 1uF 50V HT. I tried to start the PS with no load, just in case, but nothing happened. Next step was to look for the 5V STBY which are required to start the PS. The 5V STBY where there. You can check this measuring in the ATX main power connector. Then having a look at the oscilator elements, the WT7520 and the two 2SC945 I found that the voltages on the bases of the 945's where way too low 0,55V to make them work. I removed this tr's from the board and one of them had the B-E junction fusioned (Diode test reading 000). I replaced the two. I tried to start up the PS. Now the FAN which I had replaced with one I found in my "old parts" box slightly moved. GOOD SIGNAL! But that was it. Next step the Schottky rectifiers. One of them, (1640PT) the one who rectifies the 5V line was short. I replaced all of them with similar ones. I couldn't find higher current ones but in my opinion (expected MAX output currents vs rectifiers datasheets) this were the right ones anyway. I also replaced ALL the filter capacitors with similar ones but of good/higher quality. (Rubycon HT) Now I tried again. Voila! the fan started to spin. voltages checked, were OK. I loaded the PS with 2 ohms 50W resistors the 5 and 12 V outputs. The PS started with no problems. Voltages checked again and were OK. Next step, to try to boot a spare motherboard that I had laying around, just in case. ALWAYS before trying a repaired PS on a good motherboard make all the possible tests to make sure that is working fine and that the outputs have the right voltages. It booted up with no problem.